JPH0937247A - Image coder and image coding method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the quantity of data obtained by coding. SOLUTION: An expansion circuit 112 generates an expanded binary image resulting from a binary image obtained by converting points around a characteristic point into the characteristic point among binary images consisting of the characteristic point of the image and other points than the characteristic point, and a change point detection circuit 114 detects change points of the expanded binary image. Furthermore, a proximity degree calculation circuit 115 calculates proximity degree denoting a degree of proximity of the points forming the image to change points, and a selector 116 selects only the image characteristic point whose proximity degree is a prescribed threshold level or over. Then only information relating to the characteristic point of the selected image is subjected to chain coding.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image coding device and an image coding method. In particular, the present invention relates to an image coding apparatus and an image coding method capable of reducing the data amount of coded data obtained by coding information about feature points of an image.

[0002]

2. Description of the Related Art Conventionally, in an apparatus for transmitting an image under a limited transmission band or an apparatus for recording on a recording medium having a limited storage capacity, the amount of image data is reduced. A high-efficiency coding method for efficiently compressing with code words is adopted. As a high-efficiency image coding method, for example, an input image is orthogonally transformed by DCT (discrete cosine transform) and adaptive quantization is performed for each frequency band according to human visual characteristics, and a wavelet basis ( A method is known in which an image is divided into subbands by wavelet transformation, and each band is weighted and encoded. According to these encoding methods, distortion is visually inconspicuous, and
A high compression rate can be realized.

However, in these encoding systems, if the compression ratio is further improved, the visually unfavorable effects such as so-called block distortion become remarkable.

Therefore, as an encoding method which does not cause visually harmful distortion even at a high compression rate, for example, characteristic points (characteristic points) of the structure of an image (for example, points (pixels) forming edges) are used. 2. Description of the Related Art There is known a structure extraction coding method by extracting a characteristic point of an image and efficiently encoding image data at the characteristic point.

FIG. 28 shows the configuration of an example of a conventional image coding apparatus for compressing and coding an image by the structure extraction coding method. The digitized image data is input to the quantizer 11 and the two-dimensional change point detection circuit 10. In the quantizer 11, the image data is quantized, and the quantized coefficient is obtained. This quantized coefficient is supplied to a chain (chain) encoding circuit (Chain Coder) 12. Further, the two-dimensional change point detection circuit 10 detects whether or not the input image data is a feature point, and when the input image data is a feature point, for example, the value is 1
When the feature point data of 1 is not the feature point, the feature point data having a value of 0 is output to the chain encoding circuit 12, respectively.

When the quantization process in the quantizer 11 and the feature point detection process in the two-dimensional change point detection circuit 10 are completed for one frame of image data,
In the chain encoding circuit 12, the position information (position (coordinates) of the feature point) about the point (pixel) having the feature point data of 1 is chain-encoded, and further the quantized coefficient of the feature point (detected as the feature point It is multiplexed with the quantized coefficient of the image data in the pixel) to form chain encoded data. The chain-encoded data is supplied to the buffer (buffer memory) 14, temporarily stored therein, and then recorded in a recording medium (not shown) or output to the transmission path. Note that the chain encoded data is temporarily stored in the buffer 14 to smooth the data amount,
As a result, the chain encoded data is output to the recording medium or the transmission path at a substantially constant data rate.

[0007]

The detection of feature points in the two-dimensional change point detection circuit 10 is basically performed by using an operator (differential operator) that performs differential processing on image data. According to the differential operator, even a slight undulation in the image is detected as a feature point.
Therefore, for example, as shown in FIG. 29A, an area of a texture in which slight undulations are continuous (such an area is hereinafter referred to as a texture area as appropriate) (FIG. 29).
In (A) (the same applies to FIG. 24A described later),
For example, in a dark or light portion represents high or low brightness, respectively, a large number of points (pixels) are marked with diagonal lines in the figure as shown in FIG. There is a problem that is detected as the part).

That is, although the points (pixels) forming the inside of the texture area are not so important visually, the information about such points is included in the chain encoded data. There was a problem that the amount of data increased.

The present invention has been made in view of such a situation, and makes it possible to reduce the data amount of data generated by compression-encoding an image by the structure extraction encoding method. Is.

[0010]

According to a first aspect of the present invention, there is provided an image coding apparatus, wherein a point around a feature point is a feature point in a binary image including feature points of the image and points other than the feature point. Generating means for generating an expanded binary image which is a binary image converted into
A change point detection unit that detects a change point of the dilated binary image, a calculation unit that calculates a degree of proximity that represents the degree to which the points that make up the image are close to the change point, and a feature point of the image have a predetermined degree of proximity. It is characterized by comprising selection means for selecting a value greater than or equal to the threshold value of, and encoding means for encoding information regarding the feature points selected by the selection means.

In this image coding apparatus, the calculating means can calculate the proximity of the points forming the image based on the number of change points existing around the points.

According to a third aspect of the present invention, there is provided an image coding method, wherein a binary image obtained by converting points around a feature point into feature points in a binary image composed of feature points of the image and points other than the feature points. Is generated, the change point of the expanded binary image is detected, the degree of proximity of the points forming the image to the change point is calculated, and the proximity degree is calculated from the feature points of the image. Is selected to be greater than or equal to a predetermined threshold value, and information regarding the feature points of the selected image is encoded.

An image coding apparatus according to a fourth aspect of the present invention comprises coding means for coding information relating to feature points of an image and outputting coded data, feature points of the image, and points other than the feature points. Of the binary image, generating means for generating an expanded binary image which is a binary image obtained by converting points around the characteristic point into characteristic points, and change point detecting means for detecting a change point of the expanded binary image, A chain detecting means for detecting a chain connecting the feature points of the image based on the encoded data, a calculating means for calculating a degree of proximity indicating the degree of proximity of the chain to the change point, and encoding the encoded data from the encoded data. And a selecting means for selecting and outputting a chain whose degree of proximity corresponding to the data is a predetermined threshold value or more.

In this image coding apparatus, the calculating means can calculate the proximity of the chain based on the number of change points existing around the chain. In addition, the calculating unit can calculate the proximity of the chain based on the number of feature points that form the chain and have a change point around the feature points.

An image coding method according to a seventh aspect of the present invention is a binary image obtained by converting points around a feature point into feature points in a binary image consisting of feature points of an image and points other than the feature points. Is generated, the change point of the expanded binary image is detected, the information about the feature point of the image is encoded, the encoded data is output, and the feature of the image is calculated based on the encoded data. A chain that connects points is detected, and a proximity that indicates the degree of proximity of the chain to the change point is calculated.After that, from the encoded data, the proximity of the chain corresponding to the encoded data is greater than or equal to a predetermined threshold. Is selected and output.

In the image coding apparatus according to the first aspect, the generating means sets a point around the feature point as a feature point in the binary image including the feature points of the image and points other than the feature points. An expanded binary image that is a converted binary image is generated, and the change point detection means detects the change point of the expanded binary image. The calculating means calculates the degree of proximity, which indicates the degree to which the points forming the image are close to the changing point, and the selecting means is configured to select, from the characteristic points of the image, those whose degree of proximity is equal to or greater than a predetermined threshold value. ing. The encoding means is adapted to encode the information regarding the feature point selected by the selecting means.

In the image coding method according to the third aspect, a binary image obtained by converting points around the feature point into a feature point in a binary image including feature points of the image and points other than the feature points. An expansion binary image, which is an image, is generated, a change point of the expansion binary image is detected, a point that constitutes the image is calculated as a proximity indicating the degree of proximity, and the proximity point is calculated from the feature points of the image. It is configured such that a degree whose degree is equal to or higher than a predetermined threshold value is selected, and information regarding the feature points of the selected image is encoded.

In the image coding apparatus according to the fourth aspect, the coding means codes the information relating to the characteristic points of the image and outputs the coded data. The generating means is composed of feature points of the image and points other than the feature points.
In the value image, the points around the feature points are converted into feature points. 2
A dilation binary image, which is a value image, is generated, and the change point detection unit is configured to detect a change point of the dilation binary image. The chain detecting means detects the chain connecting the characteristic points of the image based on the encoded data, and the calculating means calculates the proximity degree indicating the degree of the chain approaching the change point. The selection means is configured to select and output, from the encoded data, a chain whose degree of proximity corresponding to the encoded data is a predetermined threshold value or more.

In the image coding method according to the seventh aspect, in a binary image composed of feature points of the image and points other than the feature points, binary values obtained by converting points around the feature points into feature points. An expanded binary image that is an image is generated, and a change point of the expanded binary image is detected, while information about a feature point of the image is encoded, encoded data is output, and based on the encoded data,
The chain connecting the feature points of the image is detected, the chain calculates the degree of proximity representing the degree of proximity to the change point, and then
From the encoded data, a chain whose degree of proximity corresponding to the encoded data is a predetermined threshold value or more is selected and output.

[0020]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below, but before that, in order to clarify the correspondence between each means of the invention described in the claims and the following embodiments. The features of the present invention are described as follows by adding a corresponding embodiment (however, an example) in parentheses after each means.

That is, the image coding apparatus according to the first aspect is a feature point detecting means (for example, a feature point detecting means for detecting a feature point of an image.
Of the two-dimensional change point detection circuit 10 shown in FIG. 1), the feature points of the image, and the points other than the feature points,
A generation unit (for example, the expansion circuit 112 shown in FIG. 23) that generates an expanded binary image that is a binary image obtained by converting points around the characteristic point into the characteristic points, and a change point of the expanded binary image is detected. The change point detection means (for example, the change point detection circuit 114 shown in FIG. 23) and the calculation means for calculating the degree of proximity of the points forming the image to the change point (for example, the proximity shown in FIG. 23). Degree calculation circuit 115), selection means for selecting a feature point of the image having a degree of proximity equal to or higher than a predetermined threshold value (for example, selector 116 shown in FIG. 23), and the feature point selected by the selection means. A coding means for coding information (for example, the chain coding circuit 12 shown in FIG. 1) is provided.

An image coding apparatus according to a fourth aspect of the present invention is a feature point detecting means (for example, a feature point detecting means for detecting a feature point of an image.
A two-dimensional change point detection circuit 10 shown in FIG. 26), and coding means for coding information about image feature points and outputting coded data (for example, a chain coding circuit 12 shown in FIG. 26), Generation means for generating an expanded binary image which is a binary image obtained by converting points around the feature point into feature points, out of the binary image consisting of the feature points of the image and points other than the feature points (for example, FIG. The expansion circuit 122 shown in FIG. 27), the change point detection means for detecting the change point of the expanded binary image (for example, the change point detection circuit 124 shown in FIG. 27), and the characteristics of the image based on the encoded data. A chain detection unit that detects a chain connecting points (for example, the chain detection circuit 126 shown in FIG. 27) and a calculation unit that calculates the degree of proximity of the chain to the change point (for example,
27 and the like) and a selection means (for example, in FIG. 26) that selects and outputs, from the encoded data, a chain whose degree of proximity corresponding to the encoded data is a predetermined threshold value or more. The chain select circuit 16 and the like shown) are provided.

Of course, this description does not mean that each means is limited to the above.

FIG. 1 shows the configuration of a first embodiment of an image coding apparatus to which the present invention is applied. In addition, in FIG.
The same reference numerals are given to the portions corresponding to the case in.

Therefore, in this image coding apparatus,
The two-dimensional change point detection circuit 10 outputs feature point data corresponding to whether the input image data (pixels having the image data) is a feature point, while the quantizer 11 outputs an image. The quantized coefficient obtained by quantizing the data is output.

The feature point data is supplied to the chain coding circuit 12 via the change point threshold circuit 13, and the quantized coefficient is supplied to the chain coding circuit 12 as it is. Here, in the change point threshold circuit 13, the value of the feature point data for the feature point of low visual importance in the feature point data is changed from 1 to 0, whereby the feature point data having the value of 1 is changed. The number of is reduced.

In the chain encoding circuit 12, as described above, the position information about the point (pixel) whose feature point data is 1 is chain encoded, and further multiplexed with the quantized coefficient of the feature point to obtain the chain code. Data.

Therefore, in the change point threshold circuit 13, the number of feature point data having a value of 1 is reduced, so that the amount of chain encoded data can be reduced.

FIG. 2 shows the two-dimensional change point detection circuit 10 of FIG.
The example of composition of is shown. The image data is supplied to the frame buffer 41 and temporarily stored. When one frame of image data is stored in the frame buffer 41, the image data is read from the frame buffer 41 and supplied to the filter (low-pass filter) 42.
There, it is filtered (smoothed). After that, the image data is, for example, in the so-called line scan order, the multiplier 43, the one-sample delay circuit (D) 44, the multiplier 53,
And a 1-line delay circuit (LD) 54.

The multiplier 43 multiplies the image data by -1, outputs the result to the adder 45, and outputs the 1-sample delay circuit 4
In 4, the image data is delayed by one sample and output to the adder 45. In the adder 45, the output of the multiplier 43 and the output of the 1-sample delay circuit 44 are added. Therefore, assuming that the image data of the nth sample is the current image data and is represented by x (n), the multiplier 43, the 1-sample delay circuit 44, and the adder 45 can obtain the image data x (n- 1) (In this embodiment, the image data is filtered by the filter 4 in the line scan order as described above.
Since the data is output from 2, the image data x of one sample before x
(N-1) means the image data of the pixel adjacent to the left of the pixel of the current image data x (n)) and the difference x (n-1) -x between the current image data x (n). (N) is calculated.

Hereinafter, the output of the adder 45 will be referred to as a lateral amplitude gradient as appropriate.

The lateral amplitude gradient output from the adder 45 is divided into two and both are input to the multiplier 46. In the multiplier 46, the two lateral amplitude gradients are multiplied, that is, the lateral amplitude gradient is squared (hereinafter, the squared value of the lateral amplitude gradient is appropriately referred to as lateral amplitude gradient power), the adder 47 and It is supplied to the 1-sample delay circuit 48.

In addition to the lateral amplitude gradient power, a predetermined threshold value -T1 is supplied to the adder 47, where the threshold value -T1 is added to the lateral amplitude gradient power. That is, the adder 47 takes the difference between the lateral amplitude gradient power and the threshold value T1. This difference value is calculated by the multiplier 5
0 is supplied.

In the one-sample delay circuit 48, the lateral amplitude gradient power is delayed by one sample, and the adder 4
9 (hereinafter, appropriately, the lateral amplitude gradient power delayed by one sample is referred to as delayed lateral amplitude gradient power). The adder 49 is supplied with a predetermined threshold value -T1 in addition to the lateral amplitude gradient power, and the threshold value -T1 is added to the delayed lateral amplitude gradient power. That is, in the adder 49, as in the case of the adder 47, the difference between the delayed lateral amplitude gradient power and the threshold T1 is obtained. This difference value is supplied to the multiplier 50.

The multiplier 50 multiplies the output of the adder 47 and the output of the adder 49 and outputs the result to the comparator 51. In the comparator 51, the output of the multiplier 50 is compared with 0, and when the output is 0 or more or less than 0, 0 or 1 is output to the OR gate 52, respectively.

Here, the difference value between the lateral amplitude gradient power output from the adder 47 or 49 and the threshold value T1 or the difference value between the delayed lateral amplitude gradient power and the threshold value T1 is (x (n- 1) -x (n)) ^2- T1 or (x (n-2) -x (n-1)) ^2- T1. Therefore, the lateral amplitude gradient power (x (n-1) -x (n)) ² and the delayed lateral amplitude gradient power (x (n-2) -x (n-1)) ² are output from the comparator 51. ,
0 if both are greater than the threshold T1, less than the threshold T1, or if one of them is 0
Is output and one is smaller than the threshold T1 and the other is the threshold T
If it is larger than 1, 1 will be output.

That is, when the pixel is viewed in the horizontal direction (horizontal direction) (x-axis direction) from the comparator 51, when the rate of change of the pixel value (spatial differential value of the pixel value) greatly changes. 1 is output.

On the other hand, in the multiplier 53, the image data
1 is multiplied and output to the adder 55, and the 1-line delay circuit 54 delays the image data by 1 line,
It is output to the adder 55. In the adder 55, the multiplier 53
And the output of the 1-line delay circuit 54 are added. Therefore, assuming that the number of pixels on one line is LH, the multiplier 53, the one-line delay circuit 54, and the adder 55 display the image data x (n-LH) of one line before (in this embodiment, the image data is As described above, the image data x (n-LH) of one line before is output from the filter 42 in the line scan order as the image data of the pixel adjacent to the pixel of the current image data x (n). Will be)
And the difference between the current image data x (n) and x (n-LH) -x
(N) is calculated.

Here, hereinafter, the output of the adder 55 will be referred to as the vertical amplitude gradient as appropriate.

The vertical amplitude gradient output from the adder 55 is divided into two and both are input to the multiplier 56. In the multiplier 56, the two vertical amplitude gradients are multiplied, that is, the vertical amplitude gradient is squared (hereinafter, the square value of the vertical amplitude gradient is appropriately referred to as vertical amplitude gradient power), and the adder 57 and It is supplied to the 1-line delay circuit 58.

In addition to the vertical amplitude gradient power, a predetermined threshold value -T1 is supplied to the adder 57, where the threshold value -T1 is added to the vertical amplitude gradient power and the multiplier is added. 60. Further, in the 1-line delay circuit 58, the vertical amplitude gradient power is delayed by one line and supplied to the adder 59 (hereinafter, the vertical amplitude gradient power delayed by one line is appropriately delayed. Direction amplitude gradient power). A predetermined threshold value -T1 is supplied to the adder 59 in addition to the vertical amplitude gradient power, and the threshold value -T is set to the longitudinal horizontal amplitude gradient power.
1 is added and supplied to the multiplier 60.

The multiplier 60 multiplies the output of the adder 57 and the output of the adder 59 and outputs the result to the comparator 61. In the comparator 61, the output of the multiplier 60 is compared with 0, and when the output is 0 or more or less than 0, 0 or 1 is output to the OR gate 52, respectively.

Here, the difference value between the vertical amplitude gradient power output from the adder 57 or 59 and the threshold value T1 (addition value of the threshold value −T1) or the delayed vertical amplitude gradient power value and the threshold value T1. Difference value (threshold value-added value with T1)
Is (x (n-LH) -x (n)) ^2- T1 or (x
It can be expressed as (n-2LH) -x (n-LH)) ² -T1. Therefore, from the comparator 61, the vertical amplitude gradient power (x (n-LH) -x (n)) ² and the delayed vertical amplitude gradient power (x (n-2LH) -x (n-LH)) ²
Are both larger than the threshold T1, smaller than the threshold T1, or one of them is 0, 0 is output, and if one is smaller than the threshold T1 and the other is larger than the threshold T1, 1 is output. Will be output.

That is, 1 is output from the comparator 61 when the rate of change of the pixel value greatly changes when the pixel is viewed in the vertical direction (vertical direction) (y-axis direction).

The OR gate 52 calculates the OR (logical sum) of the output of the comparator 51 and the output of the comparator 61, and outputs this as feature point data. Therefore, feature point data having a value of 1 is output for a pixel whose rate of change in pixel value in the horizontal direction or the vertical direction changes significantly, and 0 for a pixel whose rate of change in pixel value does not change significantly. The feature point data will be output.

Next, FIG. 3 shows another configuration example of the two-dimensional change point detection circuit 10 of FIG. The two-dimensional change point detection circuit 10 includes a two-dimensional change point strength / maximum change direction calculation circuit 71 and a change point detection circuit 72. Two
In the dimension change point strength / maximum change direction calculation circuit 71, for each pixel forming the image, the direction in which the pixel value changes most from the image data (maximum change direction) and the change in the pixel value in the maximum change direction. Value corresponding to the amount (change strength)
And are calculated and output to the change point detection circuit 72. The change point detection circuit 72 determines whether or not each pixel forming the image is a feature point based on the maximum change direction and the change intensity, and 1 is determined for the pixel determined to be the feature point, For pixels determined not to be feature points, 0 is output as feature point data.

FIG. 4 shows a configuration example of the two-dimensional change point strength / maximum change direction calculation circuit 71 of FIG. The image data is supplied to the frame buffer 81 and temporarily stored. When one frame of image data is stored in the frame buffer 81, the image data is read out from the frame buffer 81 and is filtered (low-pass filter).
It is fed to 82, where it is filtered. After that, the image data is supplied to the multiplier 83, the one-sample delay circuit (D) 84, the multiplier 93, and the one-line delay circuit (LD) 94, for example, in the so-called line scan order.

Here, the block consisting of the multiplier 83, the one-sample delay circuit 84, and the adder 85 of FIG.
It has the same configuration as the block including the multiplier 43, the one-sample delay circuit 44, and the adder 45 shown in FIG. In addition, the block including the multiplier 93, the 1-line delay circuit 94, and the adder 95 in FIG. 4 corresponds to the multiplier 5 shown in FIG.
The block has the same configuration as the block including the 3 and 1 line delay circuits 54 and the adder 55.

Therefore, as in the case of the adder 45 or 55 of FIG. 2, the adder 85 or 95 outputs the horizontal amplitude gradient or the vertical amplitude gradient, respectively.

The horizontal amplitude gradient and the vertical amplitude gradient are supplied to the maximum change direction calculation circuit 87. In the maximum change direction calculation circuit 87, if the horizontal amplitude gradient or the vertical amplitude gradient is expressed as A or B, respectively, C = a
rctan (B / A) (tan ⁻¹ (B / A)) is calculated, and according to the calculation result (angle) (direction) C, FIG.
A value as shown in (A) is output as the maximum change direction.

Here, from FIG. 5 (A), the maximum change direction represents the direction as shown in FIG. 5 (B). That is, the value is 0
Represents the left and right directions of the pixel currently being processed (the pixel to be processed) (the hatched portion in FIG. 5B), and the maximum changing direction with a value of 1 is , Represents the lower left and upper right directions of the pixel to be processed. Furthermore, the maximum change direction with a value of 2 represents the upper and lower directions of the pixel to be processed, and the maximum change direction with a value of 3 represents the lower right and upper left directions of the pixel to be processed. That is, the maximum change direction of the value 0 represents the direction of the pixel P1 adjacent to the left and the pixel P6 adjacent to the right, and the maximum change direction of the value 1 is the pixel P2 adjacent to the lower left and the pixel adjacent to the upper right. It represents the direction of P5. Further, the maximum change direction of the value 2 represents the direction of the pixel P3 adjacent to the upper side and the pixel P4 adjacent to the lower side, and the maximum change direction of the value 3 is the pixel P7 adjacent to the lower right side.
And the direction of the pixel P0 adjacent to the upper left corner.

Next, in FIG. 4, the block consisting of the multiplier 83, the one-sample delay circuit 84, the adder 85, and the multiplier 86 is the multiplier 43, the one-sample delay circuit 44, and the adder shown in FIG. The block is composed of 45 and a multiplier 46. In addition, the multiplier 9 of FIG.
3, the block including the 1-line delay circuit 94, the adder 95, and the multiplier 96 is similar to the block including the multiplier 53, the 1-line delay circuit 54, the adder 55, and the multiplier 56 shown in FIG. It is configured.

Therefore, as in the case of the multiplier 46 or 56 of FIG. 2, the multiplier 86 or 96 outputs the horizontal amplitude gradient power or the vertical amplitude gradient power, respectively.

The horizontal amplitude gradient power and the vertical amplitude gradient power are supplied to the adder 97. In the adder 97, the horizontal amplitude gradient power and the vertical amplitude gradient power are added, and the addition result is output as the change intensity.

The maximum change direction and the change intensity obtained as described above are determined by the change point detection circuit 7 as described above.
2 (FIG. 3).

FIG. 6 shows a configuration example of the change point detection circuit 72 of FIG. The change intensity or the maximum change direction is supplied to and stored in the frame buffer 101 or 102, respectively. In the frame buffer 101 or 102,
When the change intensity or the maximum change direction for one frame is stored, the selector 103 sequentially reads the maximum change direction from the frame buffer 102, for example, in the line scan order, and performs the following processing for each maximum change direction. .

That is, when the selector 103 reads the maximum change direction of the pixel to be processed from the frame buffer 102, the selector 103 selects the pixels shown in FIG. 7 based on the maximum change direction as the first and second maximum changes. Change direction candidate points. That is, when the maximum change direction is 0, which indicates the left and right directions of the pixel to be processed, the pixel P1 adjacent to the left or the pixel P6 adjacent to the right is, respectively, as shown in FIG. 5B. The candidate point is the first or second maximum change direction candidate point. When the maximum change direction is 1 indicating the lower left and upper right directions of the pixel to be processed, the pixel P2 adjacent to the lower left or the pixel P5 adjacent to the upper right is the first or second maximum change direction candidate point, respectively. To be done. Furthermore, the maximum change direction represents the upward and downward directions of the pixel to be processed.
In the case of, the pixel P3 adjacent to the pixel P3 above or the pixel P4 adjacent to the pixel below is set as the first or second maximum change direction candidate point, respectively. Similarly, when the maximum change direction is 3, which represents the upper left and lower right directions of the pixel to be processed, the pixel P0 adjacent to the upper left or the pixel P7 adjacent to the lower right is
Each of the candidate points is the first or second maximum change direction candidate point.

Further, the selector 103 outputs to the frame buffer 101 the addresses at which the change intensities of the pixels that are the first and second maximum change direction candidate points are stored, so that the change intensities of those pixels are calculated. , Read from the frame buffer 101. The change intensity of the first or second maximum change direction candidate point is supplied to the adder 104 or 105, respectively.

The adder 104 is supplied with not only the change strength of the first maximum change direction candidate point but also a predetermined threshold value -T2, in which the change strength of the first maximum change direction candidate point is supplied. Is added with the threshold value -T2. That is, the adder 10
4, the change strength of the first maximum change direction candidate point and the threshold value T2
And the difference is taken. This difference value is supplied to the multiplier 106.

On the other hand, in addition to the change intensity of the second maximum change direction candidate point, a predetermined threshold value -T2 is also supplied to the adder 105, in which the second maximum change direction candidate point is selected. The threshold value -T2 is added to the change intensity. That is, also in the adder 105, as in the case of the adder 104, the difference between the change intensity of the second maximum change direction candidate point and the threshold value T2 is obtained. Then, this difference value is also calculated by the multiplier 106.
Is supplied to.

In the multiplier 106, the output of the adder 104 and the output of the adder 105 are multiplied and the comparator 107
Is output to In the comparator 107, the multiplier 106
Is compared with 0, and when the output is 0 or more or less than 0, 0 or 1 is output as feature point data, respectively.

Therefore, in this case, from the comparator 107, when the change intensities of the first and second maximum change direction candidate points are both larger than the threshold value T2, smaller than the threshold value T2, or one of them is 0. In this case, 0 is output, and if one is smaller than the threshold T2 and the other is larger than the threshold T2, 1 is output. That is, feature point data with a value of 1 is output for a pixel whose rate of change of the pixel value in the maximum change direction changes significantly, and the value of a pixel whose rate of change of the pixel value in the maximum change direction does not change significantly The feature point data of 0 will be output.

Next, FIG. 8 shows a configuration example of the chain encoding circuit 12 of FIG. The quantized coefficient from the quantizer 11 is supplied to and stored in the coefficient frame buffer 21. In the coefficient frame buffer 21, the quantization coefficient Qij of the image data at the point (i, j) on the image is
For example, as shown in FIG. 9, an upper address or a lower address is stored at an address represented by i or j, respectively (however, each i or j is, for example, a pixel (point ) In the horizontal direction (x-axis direction) or the horizontal direction (y-axis direction)).

The feature point data from the change point threshold circuit 13 is supplied to the mask frame buffer 22 and stored therein. The mask frame buffer 22
Also in the above, the feature point data is stored in the same manner as in the case of the coefficient frame buffer 21 described above.

When one frame of quantized coefficients or feature point data is stored in the coefficient frame buffer 21 or the mask frame buffer 22, respectively, the direction searcher 23 performs the processing according to the flowchart shown in FIG. Be seen.

That is, first, in step S1, the status register 24, the X coordinate register 25, and the Y register.
The coordinate register 26 is initialized to 0, and the process proceeds to step S2 to determine whether the stored value of the state register 24 is 0 or not. When it is determined in step S2 that the value stored in the state register 24 is 0, the process proceeds to step S3, and it is determined whether there are unprocessed coordinates. That is,
The direction searcher 23 determines X in step S4 described later.
The coordinate register 25 or the Y coordinate register 26 is configured to set x or y coordinates of pixels forming one frame image in, for example, a line scan order (from left to right and from top to bottom), Step S
3, the X coordinate register 25 or the Y coordinate register 26
When the images are viewed in line scan order, it is determined whether or not the x or y coordinate of the pixel located at the end is set (for example, if one frame image is 6
In the case of 40 × 400 pixels, 640 or 40 is set in the X coordinate register 25 or the Y coordinate register 26, respectively.
It is determined whether 0 is set).

When it is determined in step S3 that there is an unprocessed coordinate, the process proceeds to step S4, and the unprocessed coordinate (x, y) that appears first in the line scan order of the image appears.
The respective coordinates are set in the X coordinate register 25 and the Y coordinate register 26. Therefore, immediately after the quantized coefficient for one frame or the feature point data is stored in the coefficient frame buffer 21 or the mask frame buffer 22, respectively, immediately after the X coordinate register 25 and the Y coordinate register 2 are stored.
In 6 is set 0 in all cases.

Here, in the chain encoding circuit 12, X
The pixel of the coordinates corresponding to the stored values of the coordinate register 25 and the Y coordinate register 26 is the target of the process first, but the pixel which is (is) the target of the process is hereinafter referred to as the pixel of interest as appropriate. Say.

Thereafter, the process proceeds to step S5, where the feature point data corresponding to the pixel of interest is mask frame buffer 22.
Is read out (the feature point data is read out from the address having the x-coordinate or the y-coordinate of the pixel of interest as the upper address or the lower address, respectively), and it is determined whether or not it is 1. If it is determined in step S5 that the feature point data is not 1 (if it is determined to be 0), that is, if the pixel of interest is not a feature point, the process returns to step S2. If it is determined in step S5 that the feature point data is 1, that is, if the pixel of interest is a feature point, the process proceeds to step S6, and the search X coordinate register 27 or the search Y coordinate register 28 stores the X coordinate. The value stored in the register 25 or the Y coordinate register 26 is set. Further, in step S6, the 3-bit counter 29 is initialized to 0. The 3-bit counter 29 can count a number that can be represented by 3 bits, that is, 0 to 7 (2 ³ −1), for example.

Then, in the direction searching device 23, step S
At 7, the valid data selection signal whose value is 1 (H level) is output. The valid data selection signal is usually
It is set to 0 (L level). Further, the valid data selection signal is supplied to the latch circuit 34, the selector 36, and the multiplexer 37. further,
The valid data selection signal is set to 0 again when the time for one clock elapses after it is changed from 0 to 1.

Further, in step S7, the coordinates stored in the search X coordinate register 27 and the search Y coordinate register 28 are read out, and the selector 3 is set as the start point coordinate.
6 is output. In step S7, 1 is set in the status register 24, and the feature point data having a value of 1 and corresponding to the pixel of interest stored in the mask frame buffer 22 is rewritten to 0.
Return to 2.

Here, the selector 36 has a direction searching device 2
3, the valid data selection signal and the start point coordinates are supplied, the stored value of the state register 24, the ROM (Direction
The output of the ROM 33 and the output of the direction change signal generator 35 are supplied. When the selector 36 receives the valid data selection signal whose value is 1, the selector 36 refers to the value stored in the state register 24, and when it is 0, the start point coordinate from the direction searcher 23 and the output of the ROM 33. , Or the start point coordinate from the direction searching device 23 among the outputs of the direction change signal generator 35, and outputs it to the multiplexer 37.

When the multiplexer 37 receives the valid data selection signal having a value of 1, it reads the quantized coefficient of the pixel of interest from the coefficient frame buffer 21, multiplexes it with the output of the selector 36, and outputs it. There is.

Therefore, in step S7, when the value stored in the status register 24 is 0, the valid data selection signal becomes 1 (when the pixel of interest is a feature point).
Includes the chain encoding circuit 12 (multiplexer 37)
Outputs chain coded data in which the coordinates (start point coordinates) of the pixel of interest as the feature point and the quantized coefficient at the feature point are multiplexed.

When the value stored in the state register 24 is 0 (hereinafter, appropriately referred to as the state 0), the fact that the valid data selection signal becomes 1 is a characteristic feature of the chain. Then, when the starting point was found. Then, in this case, as described above, the stored value of the state register 24 is set to 1 in step S7.

On the other hand, if it is determined in step S2 that the storage value of the state register 24 is not 0, the process proceeds to step S8, and it is determined whether or not the storage value is 1. When it is determined in step S8 that the value stored in the status register 24 is 1, step S11 in FIG.
Then, the characteristic point data stored in the address generated by the address generator 32 is read from the mask frame buffer 22.

Here, the address generator 32 is the adder 3
An address having a value output from 8 or 39 as an upper address or a lower address is output. The adder 38 has a search X coordinate register 2
X coordinate stored in 7 and ROM (Search X-ROM) 3
An output value of 0 is supplied, and both are added and output there. Further, the Y coordinate stored in the search Y coordinate register 28 and the output value of the ROM (Search Y-ROM) 31 are supplied to the adder 39. As in the case, both are also added and output.

The ROMs 30 and 31 have a 3-bit address space in which the values shown in FIG. 12 are stored. The ROMs 30 and 31 use the output value of the 3-bit counter 29 as an address and output the stored value stored at that address.

Therefore, the outputs of the adders 38 and 39 are
(Output of adder 38, output of adder 39), and the value stored in search X coordinate register 27 or search Y coordinate register 28 is expressed as x or y, respectively, a 3-bit counter. When the output value of 29 is 0 to 7, the outputs of the adders 38 and 39 are
(X-1, y-1), (x-1, y), (x-, respectively)
1, y + 1), (x, y-1), (x, y + 1), (x
+1, y-1), (x + 1, y), (x + 1, y + 1)
Becomes

Therefore, as the output value of the 3-bit counter 29 changes from 0 to 7, the address generator 32 uses the stored values of the search X coordinate register 27 and the search Y coordinate register 28 as shown in FIG. Eight pixels P0 adjacent to the pixel of the coordinates shown (the hatched portion in the figure)
The address corresponding to the coordinates of P7 to P7 is output.

From the above, in step S11, the pixel adjacent to the pixel of the coordinates represented by the stored values of the search X coordinate register 27 and the search Y coordinate register 28 (pixel corresponding to the count value output by the 3-bit counter 29) The feature point data in any of P0 to P7 will be read from the mask frame buffer 22. here,
The process described below is performed by the search X coordinate register 27 and the search Y.
Based on the pixel of the coordinate represented by the value stored in the coordinate register 28, the pixel adjacent to the pixel is searched (sequentially as a target pixel), so the search X coordinate register 27
The pixel at the coordinate represented by the stored value of the search Y coordinate register 28 will be hereinafter referred to as a search center pixel as appropriate.

When the characteristic point data is read, the process proceeds to step S12, and it is determined whether or not the value is 1.
If it is determined in step S12 that the feature point data is not 1, that is, if the pixel of interest is not a feature point, the process proceeds to step S13, and it is determined whether the count value (output value) of the 3-bit counter 29 is 7. To be done. If it is determined in step S13 that the count value of the 3-bit counter 29 is not 7, that is, if processing has not been performed for all pixels adjacent to the search center pixel, the process proceeds to step S14. Is incremented by 1, and step S2 in FIG.
Return to In this case, since the value stored in the status register 24 remains 1, the steps S2 and S8 are performed.
The processing from step S11 onward is performed again via.

If it is determined in step S13 that the count value of the 3-bit counter 29 is 7, that is, all the pixels adjacent to the search center pixel are processed, and as a result, the search center pixel is adjacent to the search center pixel. When it is determined that the pixel has no feature point, the process proceeds to step S15, 0 is set in the state register 24, and step S2 is performed.
Return to

On the other hand, if it is determined in step S12 that the feature point data is 1, that is, if the target pixel is a feature point, the process proceeds to step S16, and the target pixel is newly set as the search center pixel. It That is, the stored value of the search X coordinate register 27 or the search Y coordinate register 28 is
Each is updated to the x or y coordinate of the pixel of interest. Then, in step S17, the valid data selection signal is set to 1 and the value stored in the status register 24 is set to 2. Further, in step S17, the 3-bit counter 29 is initialized to 0, and the feature point data having a value of 1 and stored in the mask frame buffer 22 corresponding to the pixel of interest is rewritten to 0. Return to S2.

Here, when the selector 36 receives the valid data selection signal having a value of 1, it refers to the value stored in the state register 24, and if it is 1, the start point coordinate from the direction searcher 23. , The output of the ROM 33 from the outputs of the ROM 33 or the output of the direction change signal generator 35, and outputs the selected output to the multiplexer 37.

The ROM 33 has a 3-bit address space, like the ROMs 30 and 31 described above, in which values as shown in FIG. 14 are stored. And
Similarly to the ROMs 30 and 31, the ROM 33 uses the output value of the 3-bit counter 29 as an address, and outputs the stored value stored at that address (hereinafter, this stored value will be referred to as direction data as appropriate). Has been done.

Therefore, when the output value of the 3-bit counter 29 is 0 to 7, the direction data C is read from the ROM 33.
0 to C7 are output respectively. This direction data C0
C7 to C7 represent eight directions in which pixels adjacent to the search center pixel are located, that is, upper left, left, lower left, upper, as shown in FIG. 15, for example.
The directions are downward, upper right, right, or lower right. Therefore, the direction data C0 to C7 respectively indicate the direction in which the pixels P0 to P7 adjacent to the search center pixel are located.

Therefore, the ROM 33 outputs direction data indicating the direction of the pixel of interest with respect to the search center pixel. In addition to the selector 36, the direction data is
The latch circuit 34 and the direction change signal generator 35 are also supplied.

From the above, in step S17, when the value stored in the state register 24 is 1, and the valid data selection signal becomes 1 (when the pixel of interest is a feature point), the chain encoding circuit 12 (Multiplexer 3
From 7), direction data indicating the direction in which the pixel of interest adjacent to the search center pixel (pixel located at the coordinates of the start point) and serving as the feature point is located, and the quantization coefficient of the pixel of interest. Is output as chain encoded data.

When the value stored in the state register 24 is 1 (hereinafter, appropriately referred to as the state 1), the fact that the valid data selection signal becomes 1 is a characteristic feature of the chain. Then, the one adjacent to the starting point is found. In this case, as described above, the value stored in the status register 24 is set to 2 in step S17.

Returning to FIG. 10, when it is determined in step S8 that the value stored in the status register 24 is not 1,
It progresses to step S21 of FIG. Here, the status register 24 is configured to store an integer of 0 to 2. Therefore, it is determined that the stored value of the state register 24 is not 0 in step S2 of FIG. 10 and is also not 1 in step S8 of FIG. 10, which means that the value is 2. become.

In step S21 of FIG. 16, as in the case of step S11 of FIG. 11, feature point data of any pixel adjacent to the search center pixel (pixel corresponding to the count value output by the 3-bit counter 29) Is read from the mask frame buffer 22, and the process proceeds to step S22, and it is determined whether the value is 1. If it is determined in step S22 that the feature point data is not 1, that is, if the pixel of interest is not the feature point, the process proceeds to step S23, and it is determined whether the count value of the 3-bit counter 29 is 7. Step S
23, the count value of the 3-bit counter 29 is 7
If not, that is, if all pixels adjacent to the search center pixel have not been processed, the process proceeds to step S24, the count value of the 3-bit counter 29 is incremented by 1, and the process returns to step S2 of FIG. . In this case, the value stored in the status register 24 remains 2, so that the processing from step S21 onward is performed again through steps S2 and S8.

If it is determined in step S23 that the count value of the 3-bit counter 29 is 7, that is, all the pixels adjacent to the search center pixel are processed, and as a result, the search center pixel is adjacent to the search center pixel. When it is determined that the pixel has no feature point, the process proceeds to step S25, 0 is set in the state register 24, and step S2
Return to

On the other hand, if it is determined in step S22 that the feature point data is 1, that is, if the pixel of interest is the feature point, the process proceeds to step S26 and the search X coordinate register 27 or the search Y coordinate register 28 is searched. The stored value is
The target pixel is newly set as the search center pixel by updating to the x or y coordinate of the target pixel. Then, the process proceeds to step S27, and the valid data selection signal is set to 1. Further, in step S27, the 3-bit counter 29 is initialized to 0, and the feature point data, which has a value of 1 and is stored in the mask frame buffer 22, is rewritten to 0. Return to S2. In this case, the status register 24
Since the stored value of 2 is still 2, the processing from step S21 onward is performed again through steps S2 and S8. However, in this case, step S26
At, the pixel of interest that was the target of the previous processing is
Since the pixel is newly set as the search center pixel, the pixel adjacent to the new search center pixel is processed as the pixel of interest.

Here, when the selector 36 receives the valid data selection signal having a value of 1, it refers to the value stored in the state register 24, and when it is 2, the start point coordinate from the direction searcher 23. , The output of the direction change signal generator 35 from the output of the ROM 33 or the output of the direction change signal generator 35, and outputs the selected output to the multiplexer 37.

The direction change signal generator 35 includes a ROM 33.
And the output of the latch circuit 34 are supplied there, and based on these inputs, direction change data, which will be described later, is generated.

That is, as shown in FIG. 17A, when the valid data selection signal becomes 1, the latch circuit 34 causes the ROM
The direction data (FIG. 17B) output by 33 is latched, delayed by one clock, and then the direction data is output until the valid data selection signal becomes 1. To continue to output (FIG. 17 (C)). That is,
When the feature point is found (when the effective data selection signal becomes 1), the latch circuit 34 continues to output direction data indicating the direction from the search center pixel to the feature point until the next feature point is found. . Here, the direction data output from the latch circuit 34 will be referred to as forward direction data (FIG. 17C) as appropriate.

When the valid data selection signal becomes 1, that is, when the characteristic point is found, the direction change signal generator 35 outputs the direction indicated by the forward direction data output from the latch circuit 34 and R.
The direction indicated by the direction data output from the OM 33 is compared, and in accordance with the comparison result, the direction change data indicating the change in the direction indicated by the direction data with respect to the direction indicated by the previous direction data is output. There is.

That is, the direction change signal generator 35 outputs D0 as the direction change data when the direction indicated by the forward direction data and the direction indicated by the direction data are the same as shown in FIG. 18A, for example. To do. In addition, the direction change signal generator 35 is, for example, as shown in FIGS.
When the direction indicated by the direction data differs from the direction indicated by the forward direction data by 45 degrees, 90 degrees, or 135 degrees counterclockwise, D1 to D3 are respectively output as the direction change data. Further, the direction change signal generator 35 has the same configuration as that shown in FIG.
As shown in (E) to (G), the direction indicated by the direction data is 90 degrees clockwise from the direction indicated by the forward direction data by 90 degrees.
In the case of a difference of 135 degrees or 135 degrees, D4 to D6 are output as the direction change data.

As for the relationship between the direction indicated by the direction data and the direction indicated by the forward direction data, in addition to the case described above, FIG.
As shown in FIG. 8 (H), the two directions may be different by 180 degrees, but in the chain encoding circuit 12 of FIG. 8, once the feature point becomes the processing target, as described above, the corresponding feature Since the point data is rewritten to 0, the direction indicated by the direction data and the direction indicated by the forward direction data are 180
There will be no occasions where they differ. Therefore, as shown in FIG. 18H, when the direction indicated by the direction data and the direction indicated by the forward direction data are different by 180 degrees, no code is given as the direction change data (it is necessary to give the code). Absent).

Actually, the direction change signal generator 35 stores a correspondence table of the direction change data and the forward direction data and direction data as shown in FIGS. 19 and 20, and the latch circuit 34 stores the correspondence table. The forward direction data being output,
A line matching the combination with the direction data output from the ROM 33 is searched from the correspondence table, and the direction change data described in the right column of the line is output. The code words as shown in FIG. 21, for example, are assigned to the direction change data D0 to D6, and the code words are actually output.

From the above, in step S27, when the value stored in the state register 24 is 2 and the valid data selection signal becomes 1 (when the pixel of interest is a feature point), the chain encoding circuit 12 (Multiplexer 3
From 7), the direction about the feature point found last time (the direction from the feature point found two times before to the feature point found last time) and the direction about the feature point found this time (from the feature point found last time, found this time) The direction change data indicating the difference between the direction change data) and the quantized coefficient at the feature point (pixel of interest) found this time is multiplexed, and chain encoded data is output.

When the value stored in the status register 24 is 2 (hereinafter, appropriately referred to as the status 2), the valid data selection signal becomes 1 when the number of consecutive characteristic points is 3 or more. This is the case when a feature point that constitutes a chain that is other than the start point and ones adjacent thereto (third feature point and subsequent ones) is found. In this case, the value stored in the status register 24 remains 2.

Returning to FIG. 10, if it is determined in step S3 that there are no unprocessed coordinates, the process ends.
Then, after waiting for the quantized coefficient or feature point data for the next frame to be stored in the coefficient frame buffer 21 or the mask frame buffer 22, the processing from step S1 is performed again.

The output of the selector 36 will be further described with reference to FIG. Now, for example, as shown in FIG. 22, when there is a chain consisting of five consecutive feature points (hatched portions in the figure), the coordinates (i, The feature point located at j) appears first. Since the state when the feature point is first found is state 0, the selector 36 outputs the coordinate (i, j) of the feature point output from the direction searcher 23.
Is output as the start point coordinates. After that, the state is state 1
Then, the feature point (i + 1, j) adjacent to the feature point located at the coordinate (i, j) (hereinafter, appropriately described as feature point (i, j)) is detected. This feature point (i + 1,
j) is located to the right of the previously found feature point (i, j), and the state is state 1. Therefore, the selector 36 outputs the direction data C6 (FIG. 15) is output. Then, in this case, the state is set to state 2.

Next, the feature point (i + 2, j + 1) adjacent to the feature point (i + 1, j) is detected. In this case, since the state is the state 2, the direction change data indicating the difference between the direction of the previously found feature point and the direction of the currently found feature point is output from the selector 36.
That is, the direction of the feature point (i + 1, j) found last time is the right direction represented by the direction data C6 as described above, and the feature point (i + 2, j +) found this time.
The direction of 1) is the characteristic point (i + 2, j + 1)
Is located in the lower right direction of the feature point (i + 1, j),
It is represented by the direction data C7. Therefore, the forward direction data or the direction data in the correspondence table shown in FIG. 20 is C6, respectively.
Alternatively, the direction change data D4 described in the right column of the row of C7 is output from the selector 36.

Further, the feature point (i + 2, j + 1) is
Since the feature points (i + 3, j + 2) are adjacent to each other, this feature point (i + 3, j + 2) is detected. The direction in which the feature point (i + 3, j + 2) is located is the direction data C for the previously found feature point (i + 2, j + 1).
It is the same lower right direction as the direction indicated by 7, and therefore the direction for the feature point (i + 3, j + 2) found this time is
It is represented by the direction data C7. In addition, since the state remains the state 2, in this case, from the selector 36, the forward direction data and the direction data in the correspondence table shown in FIG. Direction change data D0 being output.

The feature point (i + 2, j + 3) adjacent to the feature point (i + 3, j + 2) is also associated with the feature point (i
+3, j + 2), and as a result, the selector 36 outputs direction change data D6.

Therefore, when there is a chain on the image, the chain encoding circuit 12 first multiplexes the coordinates of the feature point which is the starting point of the chain and the quantized coefficient at the feature point. Stuff is output. Then, the direction data of the feature point adjacent to the start point and the quantized coefficient at the feature point are multiplexed and output. If there is a feature point (excluding the start point) adjacent to the feature point, that is, if there is a third feature point or later, the direction change data about the feature point and the feature point And the quantized coefficient in 1 is multiplexed and sequentially output.

Next, FIG. 23 shows a configuration example of the change point threshold circuit 13 of FIG. The feature point data from the two-dimensional change point detection circuit 10 (FIG. 1) is supplied to and stored in the frame buffer 111. In the frame buffer 111, the mask frame buffer 22 shown in FIG.
As in the case of (coefficient frame buffer 21),
The feature point data for each pixel is stored at an address corresponding to the coordinates of the pixel.

When one frame of feature point data is stored in the frame buffer 111, the expansion circuit 112 refers to the frame buffer 111 and outputs a binary image including feature points of the image and points other than the feature points. Of these, an expanded binary image that is a binary image in which points around the characteristic point are converted into characteristic points is generated. That is, the expansion circuit 112 uses the frame buffer 1
The feature point data stored in 11 is regarded as a binary image having binary values of 1 and 0, and among the pixels forming the binary image, a pixel having a value of 1 (feature point) is set as the center. By converting the pixel value of the pixel in the range of L1 × L2 pixels to 1, the pixel having the value of 1 is expanded, so to speak, to generate an expanded binary image.

Specifically, the expansion circuit 112 reads out the characteristic point data from the frame buffer 111 in, for example, the line scan order, supplies the characteristic point data to the frame buffer 113, and, as in the case of the frame buffer 111, the characteristic point data. It is stored in the address corresponding to the coordinates of the pixel corresponding to. At this time, when the read characteristic point data is 1, the expansion circuit 112 sets all addresses corresponding to the coordinates of pixels in the range of L1 × L2 pixels centered on the pixel corresponding to the characteristic point data. Stores 1,
When the characteristic point data is 0, the characteristic point data is transferred as it is to the address of the frame buffer 113 corresponding to the coordinates of the pixel corresponding to the characteristic point data (however, in this case,
If 1 is already stored at that address, then 1
Will be remembered).

As a result, image data including a texture area, for example, as shown in FIG. 24A is input, and the two-dimensional change point detection circuit 10 receives the image data, for example, as shown in FIG. As shown in B), when feature points (hatched portions in the drawing) are detected not only from the outer peripheral portion of the texture region but also from the inside thereof, each feature point is centered. A range of L1 × L2 pixels (range surrounded by a dotted line in the figure) is also set as a feature point, and as a result, the texture region is expanded and the inside is filled with the feature point as shown in FIG. 24 (C). Dilated binary image as shown (in the figure, the shaded area is 1
In (characteristic points), the non-hatched portions correspond to 0, respectively). In addition, in FIG.
L1 = L2 = 3.

When the expanded binary image as described above is stored in the frame buffer 113, the changing point detection circuit 114 detects the changing point of the expanded binary image. That is, for example, the change point detection circuit 114 is
3, when the pixel value of each pixel forming the dilated binary image is different from the pixel adjacent to the right of the pixel or the pixel adjacent to the pixel below, the pixel is detected as a change point, and the value is 1 The change point data of is output to the proximity calculation circuit 115. Further, the change point detection circuit 114 has a value of 0 when the pixel values of each pixel forming the dilated binary image and the pixel adjacent to the right of the pixel or the pixel adjacent to the pixel below are not different from each other. The change point data is output to the proximity calculation circuit 115.

Therefore, as shown in FIG.
When an expanded binary image similar to that in (C) is obtained, the binary image composed of the change point data surrounds the outer periphery of the expanded binary image as shown in FIG. 25 (B). (Fig. 25
In (B), the shaded portion corresponds to the change point data having a value of 1, and the non-hatched portion corresponds to the change point data having a value of 0).

In the proximity calculation circuit 115, for each pixel forming an image, that pixel is changed point detection circuit 114.
The degree of proximity that indicates the degree of proximity to the change point detected in
It is calculated as representing the visual importance of the pixel. That is, the proximity calculation circuit 115 determines, for each pixel forming the image, the number of change points existing within the range of L1 × L2 pixels centered on the pixel as the change point detection circuit 114.
Counting is performed based on the change point data from, and the count value is set as the proximity of the pixel.

Then, the proximity calculation circuit 115 compares the proximity with a predetermined threshold T, and when the proximity is equal to or higher than the threshold T, the select data having a value of 1, for example, and the proximity having a threshold T When it is less than T, select data having a value of 0 is output to the selector 116, respectively.

The selector 116 has a proximity calculation circuit 115.
When the select data corresponding to a predetermined pixel is received, the feature point data corresponding to the pixel is read from the frame buffer 111. And the select data is 1
If it is, the feature point data read from the frame buffer 111 is selected and output as it is. If the select data is 0, select data, that is, 0 is selected,
This is output as feature point data.

Therefore, even if the feature point data has a value of 1, the value is set to 0 when the proximity (visual importance) of the pixel is less than the threshold value T.

As described above, since the number of feature point data having a value of 1 supplied to the chain encoding circuit 12 is reduced, the data amount of the chain encoded data can be reduced. Further, in this case, the pixel corresponding to the feature point data whose value is changed from 1 to 0 is not of high visual importance, so that the image quality of the decoded image is hardly deteriorated.

That is, for example, a feature point existing in the texture area as shown in FIG. 24B and having a low visual importance (such a feature point is considered to correspond to a noise-like edge). ) Is not a feature point (thresholding), and chain encoding is performed, so that the data amount of chain encoded data can be reduced with almost no deterioration in image quality.

Next, FIG. 26 shows the configuration of the second embodiment of the image coding apparatus to which the present invention is applied. In the figure, the same reference numerals are given to the portions corresponding to the case of the image encoding device of FIG. That is, this image encoding device is the same as the image encoding device of FIG. 1 except that the change point threshold circuit 13 is deleted and the texture chain removing circuit 15 and the chain select circuit 16 are newly provided. It is configured.

Therefore, in this image coding apparatus,
The feature point data output from the two-dimensional change point detection circuit 10 is directly supplied to the chain coding circuit 12, where all the feature point data detected by the two-dimensional change point detection circuit 10 are detected as in the conventional case. Chain encoding is performed on the feature points. Then, the chain encoded data obtained as a result is output to the chain select circuit 16.

The chain select circuit 16 causes the texture chain removing circuit 15 to calculate the visual importance of each chain corresponding to the chain encoded data output from the chain encoding circuit 12. The texture chain removing circuit 15 is configured to calculate the visual importance of the chain designated by the chain select circuit 16, and to determine whether to output the chain according to the calculation result. Has been done. Then, the chain select circuit 16
Selects, from the chain-encoded data output from the chain-encoding circuit 12, one having a high visual importance according to the determination result of the texture chain removing circuit 15 and outputs the selected one.

Therefore, in this case, the chain select circuit 16 removes the chain-encoded data of the chain which is not so important in visual sense, so to speak, so that the data of the chain-encoded data output from the image encoding apparatus is removed. The amount can be reduced.

Next, FIG. 27 shows a configuration example of the texture chain removing circuit 15. The feature point data output from the two-dimensional change point detection circuit 10 is supplied to the texture chain removal circuit 15, and the feature point data is input to the frame buffer 121. Frame buffer 121, expansion circuit 122, frame buffer 12
3 or the change point detection circuit 124 is configured similarly to the frame buffer 111, the expansion circuit 112, the frame buffer 113, or the change point detection circuit 114 shown in FIG. 23. Therefore, the feature point data is Through the frame buffer 121, the expansion circuit 122, the frame buffer 123, and the change point detection circuit 124, the change point data is obtained and supplied to the approximation degree calculation circuit 125.

On the other hand, the chain detection circuit 126 is supplied with the chain encoded data from the chain select circuit 16 (FIG. 26). When the chain detection circuit 126 receives the chain encoded data from the chain select circuit 16, it is based on the chain encoded data, that is, the start point coordinates, the direction data, and the direction change data included therein (FIG. 22). The chain connecting the feature points is detected by referring to. The detected chain is supplied to the approximation calculation circuit 125.

In the proximity calculation circuit 125, the degree of proximity of the chain supplied from the chain detection circuit 126 to the change point detected by the change point detection circuit 124 is the visual importance of the chain. Is calculated. That is, the proximity calculation circuit 125 determines the number of change points existing within the range of L1 × L2 pixels centering on each pixel (feature point) forming the chain from the chain detection circuit 126, as the change point detection circuit 124. The average value (the total number of change points existing in the range of L1 × L2 pixels centering on each pixel forming the chain as the number of pixels forming the chain) Divide) and use this as the proximity of the chain.

The proximity calculated as described above is L1 × L2 centered on each pixel forming the chain.
It can be said that the density of change points existing in the range of pixels, but as the degree of proximity, an amount other than the density of such change points can be adopted.

That is, for example, among the pixels forming the chain, the number of pixels having a change point in the range of L1 × L2 pixels around the pixel is divided by the number of pixels forming the chain, and the result is obtained. The value can be the proximity of the chain.

After calculating the degree of proximity, the degree-of-proximity calculation circuit 125 compares the degree of proximity with a predetermined threshold value T4, and when the degree of proximity is equal to or higher than the threshold value T4, the select data having a value of 1, for example, , If the proximity is less than the threshold value T4, the select data having a value of, for example, 0 is output to the chain select circuit 16 (FIG. 26).

The chain select circuit 16 outputs the encoded data to the texture chain removing circuit 15 and then receives the select data from the texture chain removing circuit 15 (proximity calculation circuit 125). In some cases, the chain encoded data output to the texture chain removing circuit 15 is selected and output as it is. When the select data is 0, the chain encoded data is deleted. Therefore, the chain select circuit 16 selects only the coded data from the chain coding circuit 12 for which the select data corresponding to the chain is 1, that is, the chain having a high visual importance. Is output.

As described above, since the chain encoded data output from the chain encoding circuit 12 is reduced, the data amount of the chain encoded data output from the image encoding device can be reduced. Further, in this case, since the chain having select data of 0 is not of high visual importance, the image quality of the decoded image is hardly deteriorated.

The image coding apparatus to which the present invention is applied has been described above, but such an image coding apparatus can be incorporated in, for example, a video tape recorder or a video conference system.

In this embodiment, the image data is chain-encoded, but the present invention is not limited to this.
It is applicable to the case where the image data is subjected to structure extraction coding other than chain coding.

The present invention can be applied to both moving images and still images.

Further, in the present specification, "when the degree of proximity is equal to or larger than a predetermined threshold value" means that the degree of proximity is high when the value of the degree of proximity is large. . Therefore, in the case where the proximity is defined such that the smaller the value is, the higher the proximity is, "when the proximity is equal to or larger than a predetermined threshold value" in the present specification.
That is, "when the degree of proximity is equal to or less than (smaller than) a predetermined threshold".

[0138]

According to the image coding apparatus of the first aspect and the image coding method of the third aspect, among the binary images composed of the feature points of the image and the points other than the feature points, Expansion 2 which is a binary image in which points around the feature point are converted into feature points
A value image is generated, and a change point of the expanded binary image is detected. Further, the degree of proximity, which indicates the degree of proximity of the points forming the image to the change point, is calculated, and those having a degree of proximity equal to or higher than a predetermined threshold value are selected from the characteristic points of the image. Then, information about the feature points of the selected image is encoded. Therefore, the amount of data generated as a result of encoding can be reduced.

According to the image coding apparatus of the fourth aspect and the image coding method of the seventh aspect, among the binary images composed of the feature points of the image and the points other than the feature points, the feature points An expanded binary image, which is a binary image obtained by converting the points around it to a feature point, is generated, and a change point of the expanded binary image is detected. On the other hand, the information about the feature points of the image is encoded, and the chain connecting the feature points of the image is detected based on the encoded data obtained as a result. Furthermore, the proximity, which represents the degree of proximity of the chain to the change point, is calculated,
After that, from the encoded data, those whose chain proximity corresponding to the encoded data is greater than or equal to a predetermined threshold value are selected and output. Therefore, the data amount of encoded data can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment of an image encoding device to which the present invention has been applied.

FIG. 2 is a diagram showing a configuration example of a two-dimensional change point detection circuit 10 in FIG.

3 is a block diagram showing another configuration example of the two-dimensional change point detection circuit 10 of FIG.

4 is a diagram showing a configuration example of a two-dimensional change point strength / maximum change direction calculation circuit 71 in FIG.

5 is a diagram for explaining the operation of the maximum change direction calculation circuit 87 of FIG.

6 is a diagram showing a configuration example of a change point detection circuit 72 in FIG.

FIG. 7 is a diagram for explaining the operation of the selector 103 in FIG.

8 is a block diagram showing a configuration example of a chain encoding circuit 12 in FIG.

9 is a diagram for explaining the storage contents of the coefficient frame buffer 21 of FIG.

10 is a flowchart illustrating an operation of the direction search device 23 of FIG.

11 is a flowchart following the flowchart of FIG.

12 is a diagram showing stored contents of ROMs 30 and 31 of FIG. 8. FIG.

FIG. 13 is a diagram for explaining an address output by the address generator 32 of FIG.

14 is a diagram showing stored contents of a ROM 33 of FIG.

15 is a diagram showing a direction represented by direction data output from the ROM 33 of FIG.

16 is a flowchart following the flowchart of FIG.

17 is a timing chart showing the output timings of the valid data selection signal from the direction searcher 23, the direction data from the ROM 33, and the forward direction data from the latch circuit 34 in FIG.

18 is a diagram for explaining direction change data output from the direction change signal generator 35 of FIG.

19 is a diagram showing a correspondence table of forward direction data and direction data stored in the direction change signal generator 35 of FIG. 8 and direction change data.

20 is a diagram showing a correspondence table of direction change data and forward direction data and direction data stored in the direction change signal generator 35 of FIG.

FIG. 21 is a diagram showing code words assigned to the direction change data output by the direction change signal of FIG. 8.

22 is a diagram for explaining the data output from the selector 36 of FIG. 8. FIG.

FIG. 23 is a block diagram showing a configuration example of a change point threshold circuit 13 of FIG. 1.

FIG. 24 is a diagram for explaining the operation of the expansion circuit 112 in FIG. 23.

FIG. 25 is a diagram for explaining the operation of the change point detection circuit 114 of FIG. 23.

FIG. 26 is a block diagram showing a configuration of a second embodiment of an image encoding device to which the present invention has been applied.

27 is a block diagram showing a configuration example of the texture chain removing circuit 15 in FIG. 26. FIG.

FIG. 28 is a block diagram showing a configuration of an example of a conventional image encoding device.

FIG. 29 is a diagram for explaining feature point detection in the two-dimensional change point detection circuit 10 in FIG. 28.

[Explanation of symbols]

 10 Two-dimensional change point detection circuit 11 Quantizer 12 Chain coding circuit 13 Change point threshold circuit 14 Buffer 15 Texture chain removal circuit 16 Chain select circuit 111 Frame buffer 112 Expansion circuit 113 Frame buffer 114 Change point detection circuit 115 Proximity calculation Circuit 116 Selector 121 Frame Buffer 122 Expansion Circuit 123 Frame Buffer 124 Change Point Detection Circuit 125 Proximity Calculation Circuit 126 Chain Detection Circuit

Claims (7)

[Claims] 1. A feature point detecting means for detecting a feature point of an image; a feature point of the image; and a point other than the feature point.
Generation means for generating an expanded binary image that is a binary image obtained by converting points around the characteristic point into the characteristic points in the value image, and change point detection means for detecting a change point of the expanded binary image. And a calculating unit that calculates a degree of proximity that indicates the degree to which the points forming the image are close to the change point, and a selecting unit that selects, from the characteristic points of the image, those whose degree of proximity is a predetermined threshold value or more. And an encoding unit that encodes information about the feature point selected by the selection unit. 2. The calculation means calculates the proximity of points forming the image based on the number of change points existing around the points. Image coding device. 3. A feature point of an image is detected, and the feature point of the image and a point other than the feature point are detected.
Of the value images, points around the feature points are converted into the feature points to generate an expanded binary image, and a change point of the expanded binary image is detected to configure the image. Calculates a degree of proximity representing the degree of proximity to the change point, selects from the feature points of the image that the degree of proximity is equal to or more than a predetermined threshold value, and encodes information about the selected feature point of the image. An image encoding method characterized by encoding. 4. A feature point detecting means for detecting a feature point of an image, an encoding means for encoding information about the feature point of the image and outputting encoded data, a feature point of the image, and the feature point. 2 with other points
Generation means for generating an expanded binary image that is a binary image obtained by converting points around the characteristic point into the characteristic points in the value image, and change point detection means for detecting a change point of the expanded binary image. A chain detecting means for detecting a chain connecting the feature points of the image based on the encoded data; a calculating means for calculating a proximity degree indicating a degree of the chain approaching the change point; An image coding apparatus comprising: a selected unit that selects and outputs, from the encoded data, the degree of proximity of the chain corresponding to the encoded data is a predetermined threshold value or more. 5. The image coding apparatus according to claim 4, wherein the calculating unit calculates the proximity of the chain based on the number of the change points existing around the chain. 6. The calculating means calculates the degree of proximity of the chain based on the number of characteristic points forming the chain where the change point exists around the characteristic points. The image encoding device according to claim 4. 7. A feature point of an image is detected, and is composed of a feature point of the image and a point other than the feature point.
Among the value images, points around the feature points are converted into the feature points to generate an expanded binary image, and a change point of the expanded binary image is detected, while a characteristic point of the image is detected. Information on the encoded data is output, encoded data is output, a chain connecting the feature points of the image is detected based on the encoded data, and the chain indicates the degree of proximity to the change point. An image coding method, wherein the image data is calculated and then selected from the coded data and having a proximity of the chain corresponding to the coded data of a predetermined threshold value or more and outputting the selected data.