JPH10285601A - Coder, its method and storage medium storing the method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the coding efficiency by inverting interpretation of a positive or negative sign of a prediction error in the prediction coding mode to reduce the code amount after the coding. SOLUTION: A prediction conversion circuit 105 predicts a coding object pixel value from surrounding pixels a, b to generate a prediction error Diff, and provides outputs of the prediction error Diff, a parameter (k), a state number S and two parameters Bα, N in a state of the state number S according to processing by a k-parameter generating circuit or the like. An inversion discrimination circuit of a prediction code conversion circuit 106 provides an output of a signal '1' that instructs sign inversion in the case of k=0 and Bα>N/2 and an output of a signal '0' that instructs no inversion in other cases. A sign inversion circuit based on the signal '1' inverts the sign of the Diff corresponding to the coded object pixel to provide an output of a -Diff to a signal line. A selector selects and outputs the Diff received from the signal line in the case that the output of the inversion discrimination circuit is '0' and selects and outputs the - Diff being an output of the sign inversion circuit in the case that the output of the inversion discrimination circuit is '1'.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an encoding apparatus and method for encoding image data and a storage medium storing the method.

[0002]

2. Description of the Related Art Conventionally, there is a predictive coding method as one of image coding methods used in an image coding apparatus.

The predictive coding method predicts the pixel value of a pixel to be coded from surrounding pixels and entropy-codes a prediction error. This prepares several prediction methods,
Various improved systems based on the prediction coding system have been proposed, such as adaptively selecting a prediction system according to surrounding pixels and performing entropy coding.

[0004]

However, in the case where the prediction error between the predicted value and the pixel value to be coded is entropy coded, the code length may be different even though the prediction error Diff has the same absolute value. There is still room for improvement in entropy coding to improve coding efficiency.

The present invention has been made in view of the above conventional example, and is intended to further improve the coding efficiency when entropy coding is performed on a prediction error between a predicted value and a pixel value to be coded. It is intended to provide a configuration.

More specifically, in the case where a plurality of encoding modes including a prediction encoding mode for entropy encoding a prediction error between a predicted value and a pixel value to be encoded are switched and used, efficient encoding is performed. The purpose is to do.

[0007]

According to the encoding apparatus of the present invention, a prediction value is generated based on peripheral pixels of an encoding target pixel, and the prediction value and the encoding target pixel are generated. A coding apparatus having a prediction coding mode for entropy coding a prediction error of a pixel value, and a run-length coding mode for run-length coding a pixel value without obtaining a prediction error, wherein the prediction coding mode Is characterized in that it has a sign inversion processing means for inverting the interpretation of the sign of the prediction error in.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION

(First Embodiment) Next, a first embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1 shows a configuration diagram of a first embodiment according to the present invention. In FIG. 1, reference numeral 100 denotes an input unit for inputting a signal, 101 denotes a signal line, 102 denotes a predictive conversion circuit,
04 is a Golomb-Rice encoding circuit, 105 is a prediction conversion circuit, 106 is a code prediction conversion circuit, 107 is a Golomb-Rice
Encoding circuit, 108 is a run-length counter, 109 is
A Melcode encoding circuit, 110 is a state determination circuit, 111 is a mode selector, 112 is a switch, 113 is a switch, 114 is a signal line, and 115 is a buffer for storing image data of two lines.

In this embodiment, the encoding process is performed by switching the three modes. More specifically, the first mode is a normal mode in which the predictive conversion circuit 102 and the Golomb-Rice (Golomb-Rice) coding circuit 104 perform an encoding process, and the second mode is a normal mode. Run length counter 1
08 and a run-length mode for encoding processing by a Melcode (Melcode) encoding circuit 109. The third mode is a prediction conversion circuit 105 and a code prediction conversion circuit 106.
And a run end mode in which the Golomb-Rice encoding circuit 107 performs an encoding process.

Next, the operation of each section in this embodiment will be described in order, taking as an example the case of encoding a monochrome image signal of 8 bits per pixel (value from 0 to 255). However, the present invention is not limited to this.
Alternatively, the present invention can be applied to a case where a multi-valued color image composed of eight-bit luminance and chromaticity components of Lab is encoded, and each component may be encoded in the same manner as the monochrome image signal.

[0012] Further, the coding switching unit of each component can be performed for each screen, whereby it becomes possible to check the state of the entire image on the decoding side earlier. Also, the coding switching unit of each component can be performed for each pixel, for each line, or for each band composed of a plurality of lines, whereby a partial but complete color image can be viewed earlier.

The initial values of the internal values of the buffer 115 are all set to 0.

First, a pixel to be coded (pixel to be coded) is input from the input unit 100 in raster-scan order, and the switch 112 and the buffer 1 are input through the signal line 101.
15 is input.

The buffer 115 stores two lines of image data sequentially input from the signal line 101. These two lines are the line of the encoding target pixel and the line immediately before the line. The state determination circuit 110 generates a state number S indicating the state of the peripheral pixel of the encoding target pixel and a phase flag R indicating the phase thereof, and performs the following processing. First, pixel data corresponding to the peripheral pixels a, b, c, and d of the encoding target pixel are read from the buffer 115.

FIG. 2 shows peripheral pixels for the pixel x to be encoded.
The positional relationship between a, b, c, and d is shown. In FIG. 2, peripheral pixels
a, b, c, and d are pixels that have already been encoded for the encoding target pixel x. Next, the read image data a, b, c, d
Are used to determine db, bc, and ca, and according to the correspondence shown in FIG. 3, -4, -3, -2, -1, 0, 1, 2, 3, 4
, And quantized values q (d−b), q (b
−c) and q (c−a).

Next, q (d−b) × 81 + q (b−c) × 9 +
The state number S representing the state of the peripheral pixel is generated by the calculation formula of q (c−a). Next, the sign of the state number S is checked. If the sign is positive, the phase flag is set to 0. If the sign is negative, the phase flag is set to 1 and the sign of the state number S is inverted (positive or negative). Value). Through the above processing, the state number S having a value from 0 to 365 and the phase flag R having a value of 0 or 1 are generated and output.

The mode selector 111 is a state determination circuit 11
Generated state number S of 0 and run length counter 10
8 based on the run length RL and the coding control signal Q,
An encoding mode is selected, and switching of the switches 112 and 113 is controlled. Run length counter 108
The initial value of the run length RL from is set to 0.

FIG. 8 shows the overall flow of the encoding mode control by the mode selector 111 for one pixel to be encoded. The detailed operation of each of the following steps will be described later.

First, in step 801, the mode selector 1
In step 11, a state number S and a run length RL corresponding to a pixel to be encoded are first input, and if it is determined that S ≠ 0 and RL = 0, the process proceeds to STEP 802; otherwise, the process proceeds to STEP 803. .

In STEP 802, switch 112 is connected to terminal C
The switch 113 is connected to the terminal C '
Encoding of the code. On the other hand, in STEP803, switch 1
12 is connected to the terminal A, and the switch 113 is connected to the terminal A 'to perform run-length mode encoding, that is, encoding of run length.
Although the detailed encoding process of the run-length mode will be described later, the code is actually output in the run-length mode only when the run length is determined, not for each pixel, that is, when the continuation of the same pixel value is discontinued. It is. in this case,
The value of the coding control signal c changes from '0' to '1'.

After step 803, the value of the encoding control signal c is checked in step 804, and if c = 1, the process proceeds to step 805. In step 805, the switch 112 is connected to the terminal B, and the switch 114 is connected to the terminal B '.
The encoding process is performed in the run end mode. The above control is repeated for each pixel, and three encoding modes are selected for encoding.

The reason for switching the above three encoding modes, that is, the role of the three encoding modes will be briefly described.

The normal mode is applied to encoding of a pixel which is estimated to have a Laplace distribution when a prediction error is caused by predictive conversion. However, when the same luminance level is expected to be continuous (when the peripheral pixels have the same luminance level), the coding efficiency is further increased by switching to the run-length mode.

The run length mode is a mode used in a case where the same luminance level is expected to be continuous.
Instead of encoding the prediction error, the run length of the pixel value is encoded (run-length encoding). As a result, in the normal mode (when Huffman coding of prediction error, Golomb-Rice coding or the like is used), coding cannot be performed with 1 bit or less per sample, but in the run length mode, 1 bit cannot be coded.
Encoding with less than 1 bit per sample is also possible,
Encoding efficiency is improved.

The run termination mode encodes a pixel to be encoded when a run of the same luminance level ends. Also, since the value of this pixel is the end of the run length in run-length encoding, it is known that it is not the immediately preceding pixel value. Therefore, when the immediately preceding pixel value is used as the predicted value as in the normal mode, a special state is set in which no prediction error 0 occurs and the coding efficiency is not improved.
In consideration of such a state, the pixel for which the run-length mode has ended, that is, the encoding target pixel whose pixel value has changed from the immediately preceding pixel, performs encoding in the run end mode instead of the normal mode. In addition, when the encoding target pixel in the above state is predictively encoded, the distribution of the prediction error is often discrete, and it is estimated that the distribution is far from the Laplace distribution. Under such conditions, the later-described correction of the predicted value used in the normal mode is not very effective, and therefore the encoding is performed without using this. Even if the immediately preceding pixel value is encoded as the predicted value from the next encoding target pixel, the encoding efficiency is sufficiently improved.

Hereinafter, each of the above three encoding modes will be individually described.

First, the prediction conversion circuit 102 and the Golomb-Rice
The normal mode encoding process in which the encoding circuit 104 performs the encoding process will be described.

The prediction conversion circuit 102 shown in FIG. 1 predicts the pixel value of the pixel to be encoded from the surrounding pixels a, b, and c of the pixel to be encoded, and calculates the predicted value generated by this prediction and the actual value of the pixel to be encoded. A difference (prediction error) Diff from the pixel value of (i) and a parameter k described later are generated. FIG. 4 shows a prediction conversion circuit 102.
FIG.

In FIG. 4, 401 is a predictor, 402 is an error feedback circuit, 403 is a prediction error generation circuit,
404 is a memory, 405 is a parameter update circuit, 406
Is a k-parameter generation circuit. Although not shown in FIG. 1, the prediction conversion circuit 102 includes the buffer 1 shown in FIG.
15, the pixel value data of the peripheral pixels a, b, and c are input, and the state S and the phase flag R are input from the state determination circuit 110. The pixel value x of the pixel to be encoded is input from the terminal C of the switch 112.

The memory 404 stores four parameters N, A, Bγ and Cα for each state number S. These values are N = 1, A = 4, Bγ = Cα = 0 at the start of encoding.
Is initialized to N is the number of occurrences of state S, and A
Is the total of the absolute values of the prediction errors in the state S, Bγ is the total of the prediction errors in the state S, and Cα is a correction value for correcting the prediction value.

The operation of the predictive conversion circuit 102 will be described below with reference to FIG. First, the estimator 401
, The pixel values of the peripheral pixels a, b, and c (see FIG. 2 for the position) of the pixel to be encoded are input, and a predicted value p is generated based on the pixel values a, b, and c. The predicted value p is obtained by the following equation.

P = | max (a, b): when min (a, b)> c | min (a, b): when max (a, b) <c | a + b−c: other than the above

The error feedback circuit 402 has a predictive correction value C described later in the state S stored in the memory 404.
The prediction value p is corrected using α to generate a corrected prediction error p ′. That is, if the phase flag R input from the state determination circuit 110 is 0, p ′ = p + Cα, and if 1 the value is p ′ = p−Cα.
And Further, when p ′ is less than 0, p ′ = 0,
When p 'is 255 or more, p' is limited to 255.

The prediction error generation circuit 403 calculates the difference between the pixel value x of the pixel to be encoded and the corrected prediction value p ′, and calculates the prediction error Diff.
Ask for. When the phase flag R is 0, Diff = x−p ′. When the phase flag R is 1, Diff = p′−x.
The prediction error generation circuit 403 adds 255 to Diff when Diff is less than -128, and Di when Diff is 128 or more.
By subtracting 255 from ff, the value of Diff can be expressed within -128 to 127.

The k-parameter generating circuit 406 reads the number of occurrences N of the state S and the cumulative value A of the absolute value of the prediction error in the state S from the memory 404, and obtains min (k | N × 2 ＾ k> = A) is used to determine the value of the parameter k used in Golomb-Rice encoding. Here, min (a | b) means the minimum a that satisfies the condition b.

The parameter updating circuit 405 updates the four parameters N, Cα, A, and Bγ in the state S, and updates N ′, C
Generate α ′, A ′, and Bγ ′. First, Bγ '= Bγ + Dif
f, A ′ = A + | Diff | to obtain Bγ ′ and A ′. Then N
And if N is equal to the threshold Th1, N,
Reduce A and Bγ to 1/2. The threshold value Th1 is for limiting A, Bγ, and N to a certain range, and is set in advance (for example, Th1 = 64). Next, N ′ = N + 1, and N
To update.

FIG. 12 shows a procedure for updating the predicted correction value Cα. After copying the value of Cα to Cα ′, first, in STEP 1201, it is determined whether or not Bγ ′ is −N or less. If Bγ ′ is −N or less, go to STEP 1203; otherwise, go to STEP 1203.
The process proceeds to 1202. In step 1203, the value of Cα ′ is compared with −128.
Subtract 1 from α '(STEP 1204).

Next, N is added to Bγ ′ (STEP 120).
5). If Bγ ′ is −N or less, Bγ ′ = − N + 1 is set (STEPs 1206 and 1207). In STEP1202, B
γ ′ is compared with 0, and if Bγ ′ is 0 or more, STEP 12
08, and if not, the update process ends. In STEP 1208, the value of Cα ′ is compared with 127, and if less than 127, 1 is added to Cα ′ (STE
P1209). Next, N is subtracted from Bγ ′ (STEP 121
0). If Bγ ′ is 0 or more, Bγ ′ = 0 is set (STEPs 1211, 1212). Last updated N ', C
α ′, A ′, Bγ ′ replace the four parameters N, Cα, A, Bγ in the state S stored in the memory 404.

With the above processing, the prediction conversion circuit 102 generates the prediction error Diff and the parameter k, and outputs these to the Golomb-Rice encoding circuit 104.

The Golomb-Rice encoding circuit 104 first
The prediction error Diff output from the prediction conversion circuit 102 is converted into a non-negative integer value V by the following equation.

V = | −2 × Diff−1: when Diff <0 | 2 × Diff: when Diff ≧ 0

Next, this V is converted to Golomb based on the parameter k.
-Rice encoding. The coding procedure for Golomb-Rice coding of the non-negative integer value V based on the parameter k is as follows.

First, the non-negative integer value V is expressed in a binary number.
Next, this is divided into a lower k-bit part and a higher-order remaining bit part. Next, to the lower k-bit part, add the number of '0' to the remaining upper bit part by the number expressed in decimal,
Finally, '1' is added as a code word. As a specific example, when k = 2 and V = 13, the binary representation of V '110
The number represented by the remaining high-order bit part '11', that is, three '0's, is added to the lower two bits'01' of '1' to make it '01000', and '1' is added at the end to form a code word. '010001' is generated.

FIG. 9 shows the correspondence between the non-negative integer values and the code words in the parameters k = 0, 1, and 2 when the above-described encoding process is performed.

The code generated by the Golomb-Rice coding circuit 104 is output to the terminal C 'of the switch 113.

Next, a run termination mode in which the encoding process is performed by the prediction conversion circuit 105, the code prediction conversion circuit 106, and the Golomb-Rice encoding circuit 107 will be described.

The prediction conversion circuit 105 predicts an encoding target pixel value from the peripheral pixel values a and b, and generates a prediction error. FIG.
2 shows a configuration diagram of the prediction conversion circuit 105. In FIG.
01 is a comparator, 502 is a predictor, 503 is a prediction error generation circuit, 504 is a memory, and 505 is a k parameter generation circuit. Although not shown in FIG. 1, pixel value data of peripheral pixels a and b are input from the buffer 115 to the prediction conversion circuit 105. Also, the terminal B of the switch 112
, The encoding target pixel value x is input.

The memory 504 stores three parameters N, A, and Bα corresponding to each state specified by the state number S. N, A, and Bα are N = 1 and A =
4, Bα = 0. Note that N is the number of occurrences of the state S, A is the total of the absolute values of the prediction errors in the state S, and Bα is the number of times a negative prediction error has appeared in the state S.

The operation of the prediction conversion circuit 105 will be described in detail with reference to FIG. First, the comparator 501 has the buffer 115
Is compared with the surrounding pixels a and b of the encoding target pixel passed from
If b <a, 1 is output; otherwise, 0 is output. Hereinafter, the value output from the comparator 501 will be referred to as a phase flag r (1 or 0).

The predictor 502 calculates the predicted value p from the surrounding pixels a and b.
And a state number S are generated. The value of the state number S is 1 when a = b, and 0 when a ≠ b. Also, the predicted value p is always b
And

The prediction error generation circuit 503 generates a prediction error Diff from the pixel value x to be encoded and the prediction value p. Phase flag
When r is 0, Diff = x−p, and when the phase flag r is 1, Diff = p−x.

The k-parameter generating circuit 505 includes a predictor 50
2, the number of occurrences N of the state S and the cumulative value A of the absolute value of the prediction error in the state S are read from the memory 504, and T is obtained by the following equation.

T = | A: When S = 0 | A + N / 2: When S = 1

Golomb-Ri is obtained by min (k | N × 2 ＾ k> = T).
The value of parameter k at the time of ce encoding is obtained and output.

The parameter updating circuit 506 updates the three parameters N, A, and Bα in the state S, and updates N ′, A ′, and B ′.
Generate α ′. First, A ′ is obtained by A ′ = A + | Diff |. Next, the sign of Diff is checked, and if it is positive, Bα '
= Bα, and if negative, Bα ′ = Bα + 1.
That is, Bα indicates the number of times the prediction error Diff is negative. Further, N ′ is obtained by N ′ = N + 1.

The last updated N ', A', and Bα 'replace the three parameters N, A, and Bα in the state S stored in the memory 504.

By the above processing, two parameters Bα and N in the state of the prediction error Diff, the k parameter, the state number S, and the state number S are output.

When the k parameter is 0, the sign prediction conversion circuit 106 calculates the sign (+) of the prediction error Diff from the values of Bα and N.
/ −) Is predicted, and the sign of Diff is changed based on the prediction result.

FIG. 6 shows the configuration of the code prediction conversion circuit 106. In the figure, reference numeral 601 denotes an inversion determination circuit, 602 denotes a sign inversion circuit, and 603 denotes a selector.

The operation of the code prediction conversion circuit 106 will be described in detail with reference to FIG. First, the inversion determination circuit 601 calculates k
In the case of = 0 and Bα> N / 2, a signal '1' instructing sign inversion is output, and in other cases, a signal '0' instructing not to invert sign is output. The sign inversion circuit 602 inverts the sign (+/-) of the Diff corresponding to the encoding target pixel based on the instruction signal (1 or 0), and
iff is output to a signal line 604. When the output value of the inversion determination circuit 601 is “0”, the selector 603
Is selected and output. In the case of “1”, the output value −Diff of the sign inversion circuit 602 is output.

With the above processing, the sign prediction inversion circuit 10
Numeral 6 selects Diff or -Diff for the input value Diff and outputs it.

Here, the role of the sign inversion will be briefly described. FIG. 13 shows the coding assignment of Golomb-Rice coding when the k parameter is 0 and the k parameter is 2. In FIG. 13, the case where the k parameter is 0 means the case where the values of the peripheral pixels including the coded pixel vary little. Therefore, the integer value V when Golomb-Rice coding is performed
, The coded output and the code length have a relationship as shown in the figure.

On the other hand, when the k parameter is 2, it indicates that the fluctuation of the values of the peripheral pixels including the coded pixel is slightly larger than that in the case of 0. In order to perform coding, the integer value V or the prediction error Diff described later when Golomb-Rice coding is performed, and the coding output or code length have a relationship as shown in the figure.

Here, when the k parameter is 0, there is a characteristic that the code length always differs even though the prediction errors Diff have the same absolute value. On the other hand, when the k parameter is a value other than 0 (for example, 2), the prediction error Di
Those having the same absolute value in ff often have the same code length.

In consideration of this encoding characteristic, sign inversion is performed when the k parameter is 0 and Bα> N / 2 (when the number of times the prediction error in state S is negative exceeds a majority). Otherwise, if the code inversion is not performed in other cases, the one having a higher prediction error Diff (for example, -1 and 1) having the same absolute value and having a higher occurrence frequency can be used as a coded output having a shorter code length. It becomes possible.

That is, if the occurrence frequency of -1 is higher, it is determined that the sign prediction inversion circuit 106 does not invert the sign.
When the frequency of occurrence of 1 is lower (the frequency of occurrence of 1 is higher), encoding is performed by interpreting -1 as 1 and encoding by interpreting 1 as -1 to increase the encoding efficiency. Can be.

On the other hand, the sign inversion processing can be performed even when the k parameter is other than 0, but the sign inversion processing is not performed in this embodiment because it is not so effective.

The above-described sign inversion processing is appropriately switched during the encoding of the image. Needless to say, the decoding side performs the inversion judgment circuit 601 and the sign inversion circuit 602 on the basis of the decoded peripheral pixels. Since it has an equivalent processing unit, the prediction error Diff from any encoding target pixel
You can see whether the interpretation of the sign has changed.

After the sign inversion processing is performed, Golomb
The -Rice coding circuit 107 converts the prediction error Diff 'output from the code prediction conversion circuit 106 into a Golomb-Rice
It encodes and outputs the code to the terminal B ′ of the switch 113. First, Diff is converted into a non-negative integer value V by the following equation.

V = | −2 × (Diff−S) −1: When Diff <0 | 2 × (Diff−S): When Diff ≧ 0

Next, a non-negative integer value V is obtained by the parameter k
Perform mb-Rice encoding. Golomb-Rice coding procedure is Golo
This is as described in the description of the mb-Rice encoding circuit 104.

Finally, the run length counter 108 and Me
The run-length mode in which the encoding process is performed by the lcode encoding circuit 109 will be described.

The run length counter 108 receives the pixel value x to be coded from the terminal A of the switch 112,
, The immediately preceding pixel value a is input, the encoding control signal c is output to the signal line 116, and the same luminance continuous number RL is output to the signal line 117. FIG. 7 shows the configuration of the run length counter 108. In the figure, reference numeral 701 denotes a comparator, and 702 denotes a counter. The comparator 701 compares the encoding target pixel value x with the immediately preceding pixel value a.
If they are not equal, '0' is output as the coding control signal c. The counter 702 has a continuous number RL of the same luminance value internally.
RL is incremented and output when the encoding control signal c output from the comparator 701 is '0', and when the encoding control signal c is '1', After outputting the value, it is reset to zero.

The Melcode encoding circuit 109 is connected to the signal line 116
Is encoded only when the encoding control signal input from is equal to '0'. FIG. 10 shows the procedure of the Melcode encoding.

First, in Step 1001 of FIG.
The run length threshold value RL is determined based on the index (the initial value at the start of encoding is 0) held inside the ode encoding circuit 109.
T is determined. FIG. 11 shows the correspondence between index and RLT.
Next, in STEP 1002, RLT is compared with the run length RL, and RL ≧ R
If LT, the process proceeds to STEP 1003; otherwise, the process proceeds to STEP 1008.

In STEP 1003, “0” is output as a code. Subsequently, in STEP 1004, the value of RL is updated by RL = RL-RLT. In step 1005, the value of the index is incremented. In step 1006, if the index exceeds 31, the index is set to 31 in step 1007 and then in step 10
01, and if the index does not exceed 31, the process returns to STEP 1001 as it is.

On the other hand, if the processing has proceeded to step 1008, “1” is first output as a code in STEP 1008. Next, in STEP 1009, RL is expressed in binary notation,
The lower RLT bit is output as a code. STEP101
At 0, the value of index is decremented.
It is determined whether ex has become 0 or less. If the index has become 0 or less, STEP 1012 sets index to 0, and the processing ends. If the index does not become 0 or less, the processing ends.

As described above, a code for the run length RL is generated and output.

The above processing is repeated until the last pixel input from the input unit 100, and the encoding processing is performed. As a result, a code sequence for the input image is output to the signal line 114.

For a description of the above-mentioned Melcode encoding method, see “Transactions of the Institute of Electronics, Information and Communication Engineers, '77 / 12 Vo.
1. J60-A No. 12 ", and the details are omitted.

The three encoding modes described above are:
It is used in accordance with the switching of the encoding mode shown in FIG.

According to the above-described encoding method, lossless encoding can be performed without losing the information amount of a color or monochrome multivalued image.

In the lossless encoding shown in the present embodiment, the sign inversion processing shown in FIG.
For entropy coding such as Golomb-Rice coding in which the prediction error Diff has the same absolute value but a different code length, this encoding property is taken into consideration, and the pixel before the pixel to be coded is considered. The entropy is determined based on whether the positive or negative sign of the prediction error Diff has occurred before.
By controlling whether or not the sign of the prediction error Diff is inverted before coding (whether or not to change the interpretation of the sign of the prediction error Diff), the frequency of occurrence of the prediction error Diff having the same absolute value is high. The code length can be shorter, and efficient entropy coding can be performed.

The sign inversion processing of the above embodiment is
A major feature is that the inversion process is performed based on the mere number of occurrences of which one of the positive sign and the negative sign of the prediction error Diff has occurred many times before the pixel before the encoding target pixel. For example, compared to the method of performing the inversion processing based on whether the cumulative value of the prediction error Diff up to the previous pixel of the encoding target pixel is a positive sign or a negative sign, However, processing is easy. Furthermore, when a large positive or negative value occurs in the prediction error Diff due to image noise or the like, the above-mentioned accumulated value is biased to one of the positive and negative sides, and the optimum sign inversion cannot be performed. According to this, it is not necessary to be affected by such noise.

(Modification) The present invention is not limited to the above embodiment. For example, the present invention can be applied to a case where the previous value prediction is simply used as the prediction method of the pixel value to be coded. Several prediction methods may be prepared and switched as appropriate. In addition, Golomb-
Although Rice code and Melcode are used, these may be changed to another entropy coding method such as Huffman code.
In the above-described embodiment, the sign inversion processing is performed in the run end mode. However, the present invention is not limited to this, and a coding mode in which a predicted value represented by the normal mode is entropy-coded. Alternatively, the present invention can be applied to a single encoding process, and these modifications are also included in the present invention.

The present invention can be applied to a system composed of a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), or to an apparatus composed of one device (for example, a copying machine or a facsimile machine). You may.

Also, in order to operate various devices to realize the functions of the above-described embodiment, a device connected to the various devices or a computer in a system is provided with software for realizing the functions of the above-described embodiment. The present invention also includes a program code supplied and executed by operating a computer (CPU or MPU) of the system or apparatus according to a stored program to operate the various devices.

In this case, the program code of the software implements the functions of the above-described embodiment. The program code itself and means for supplying the program code to a computer, for example, the program code The storage medium storing the information constitutes the present invention.

As a storage medium for storing such a program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, magnetic tape, nonvolatile memory card, ROM or the like can be used.

When the computer executes the supplied program code, not only the functions of the above-described embodiment are realized, but also the OS (operating system) running on the computer or another program. Needless to say, even when the functions of the above-described embodiments are realized in cooperation with application software and the like, such program codes are included in the embodiments of the present invention.

Further, the supplied program code is stored in a memory provided in a function expansion board of the computer or a function expansion unit connected to the computer, and then stored in the function expansion board or the function storage unit based on the instruction of the program code. It is needless to say that the present invention also includes a case where a provided CPU or the like performs a part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

[0093]

As described above, according to the present invention, a predicted value is generated based on the peripheral pixels of the encoding target pixel, and the prediction error between the predicted value and the pixel value of the encoding target pixel is entropy-encoded. Predictive encoding mode, a coding apparatus having a run-length encoding mode for run-length encoding the pixel value without obtaining a prediction error, the interpretation of the sign of the prediction error in the prediction encoding mode, Since there is a sign inversion processing means for inverting so that the code amount after encoding is reduced, the one having a higher occurrence frequency in the prediction error Diff of the same absolute value can have a shorter code length, and an efficient entropy code Is possible.

[Brief description of the drawings]

FIG. 1 shows a block diagram according to a first embodiment of the present invention.

FIG. 2 illustrates a positional relationship between a pixel to be encoded and peripheral pixels.

FIG. 3 illustrates a method of quantizing a difference value between peripheral pixels.

4 shows a configuration of a prediction conversion circuit 102. FIG.

5 shows a configuration of a prediction conversion circuit 105. FIG.

FIG. 6 shows a configuration of a sign prediction inversion circuit 106.

FIG. 7 shows a configuration of a run length counter 108.

FIG. 8 shows the control of the encoding mode by the mode selector 111.

FIG. 9: Golomb-Rice in k parameters 0, 1, and 2
The example of a code | symbol was shown.

FIG. 10 shows a Melcode encoding procedure.

FIG. 11 shows RLT values with respect to index values in run-length encoding.

FIG. 12 shows a procedure for updating a predicted correction value C.

FIG. 13: When the k parameter is 0 and when the k parameter is 2
This shows how coding is allocated in Golomb-Rice coding in the case of.

[Explanation of symbols]

 Reference Signs List 100 input unit 101, 114, 116, 117 signal line 102, 105 prediction conversion circuit 104, 107 Golomb-Rice coding circuit 106 code prediction conversion circuit 108 run-length counter 109 Melcode coding circuit 110 state discrimination circuit 111 mode selector 112 , 113 switch 115 buffer for storing image data sequence

Claims (20)

[Claims] 1. A prediction coding mode for generating a prediction value based on neighboring pixels of an encoding target pixel, and entropy encoding a prediction error between the prediction value and a pixel value of the encoding target pixel, and obtaining a prediction error. A run-length coding mode for run-length coding a pixel value, and interprets the sign of the prediction error in the prediction coding mode so that the code amount after coding is reduced. An encoding device, comprising: sign inversion processing means for inverting. 2. The encoding apparatus according to claim 1, wherein the run-length encoding mode is performed when a value calculated using a plurality of peripheral pixels is a predetermined value. 3. The predictive encoding mode has a first predictive encoding mode and a second predictive encoding mode for encoding the pixel to be encoded as a multi-valued image, and the second predictive encoding mode The encoding apparatus according to claim 1, wherein the encoding of the mode is performed on a pixel immediately after the pixel encoded in the run-length encoding mode. 4. The encoding apparatus according to claim 3, wherein said first predictive encoding mode generates a predicted value based on a plurality of said peripheral pixels. 5. The encoding in the predictive encoding mode includes Go
The encoding device according to claim 1, wherein lomb-Rice encoding is used. 6. The encoding apparatus according to claim 1, wherein the run-length encoding is Melcode encoding. 7. The method according to claim 1, wherein the inversion of the interpretation of the positive / negative sign by the sign inversion processing means inverts the positive / negative sign of the prediction error before performing encoding in the prediction encoding mode. An encoding device according to claim 1. 8. The inversion of the interpretation of the positive / negative sign by the sign inversion processing means, as for the prediction error occurring up to the pixel preceding the pixel to be encoded, whichever of the positive prediction error or the negative prediction error occurs more frequently. The encoding apparatus according to claim 1, wherein the encoding is performed based on information indicating whether the encoding is performed. 9. The encoding apparatus according to claim 3, wherein the first and second predictive encoding modes and the run-length encoding mode can be switched for each encoding target pixel. 10. The image processing apparatus according to claim 1, wherein the encoding apparatus encodes a color image and switches a plurality of components constituting the color image for each pixel. 11. The apparatus according to claim 1, wherein the encoding apparatus encodes a color image and switches a plurality of components constituting the color image for each line composed of one line or a plurality of lines. Image processing device. 12. The image processing apparatus according to claim 1, wherein the encoding apparatus encodes a color image and switches a plurality of components constituting the color image for each screen. 13. Generating a predicted value based on surrounding pixels,
An encoding method capable of switching between a prediction encoding mode for entropy encoding a prediction error between a prediction value and a pixel value of a pixel to be encoded and a run-length encoding mode for run-length encoding a pixel value without obtaining a prediction error And a sign inversion step of inverting the interpretation of the sign of the prediction error in the prediction encoding mode so that the code amount after encoding is reduced so as to be reduced. 14. A method for generating a predicted value based on surrounding pixels,
A coding program capable of switching between a prediction coding mode for entropy coding of a prediction error between a prediction value and a pixel value of a pixel to be coded, and a run-length coding mode for run-length coding a pixel value without obtaining a prediction error. And a sign inversion step of inverting the interpretation of the sign of the prediction error in the prediction encoding mode so that the code amount after encoding is reduced so that the code amount is reduced. A storage medium characterized by having. 15. A first mode and a second mode in which a prediction value is generated based on neighboring pixels of an encoding target pixel, and a Golomb-Rice encoding is performed on a prediction error between the prediction value and a pixel value of the encoding target pixel. A third mode for performing Melcode encoding of the pixel value of the pixel to be encoded, wherein the mode switching control for each pixel can be controlled in accordance with a predetermined condition. When encoding, the second
The value of the k parameter used for Golomb-Rice encoding in the predictive coding mode of, and which of the positive prediction error and the negative An encoding apparatus, comprising: sign inversion processing means for inverting the sign of a prediction error corresponding to the encoding target pixel based on information indicating whether the encoding error has occurred. 16. A first mode and a second mode in which a predicted value is generated based on peripheral pixels of a pixel to be encoded and a prediction error between the predicted value and a pixel value of the pixel to be encoded is Golomb-Rice encoded; A third mode for performing Melcode encoding of the pixel value of the pixel to be encoded, and a mode switching control for each pixel according to a predetermined condition, wherein the second predictive encoding mode is When performing, the value of the k parameter used for Golomb-Rice encoding in the second prediction encoding mode, and the positive prediction error and the negative prediction for the pixel before the pixel to be encoded in the second prediction encoding mode An encoding method, comprising: a sign inversion processing step of inverting the sign of the prediction error corresponding to the encoding target pixel based on information indicating which of the errors has occurred more frequently. 17. A first mode and a second mode in which a prediction value is generated based on peripheral pixels of an encoding target pixel, and a Golomb-Rice encoding of a prediction error between a prediction value and a pixel value of the encoding target pixel is performed. A third mode for performing Melcode encoding of the pixel value of the encoding target pixel, and a storage medium storing an encoding program capable of mode switching control for each pixel in accordance with predetermined conditions in a state readable by a computer When performing the second predictive encoding mode, the value of the k parameter used for Golomb-Rice encoding in the second predictive encoding mode, and the value of the pixel to be encoded in the second predictive encoding mode A sign inversion processing step of inverting the sign of the prediction error corresponding to the encoding target pixel based on information indicating which of the positive prediction error and the negative prediction error with respect to the previous pixel has occurred more frequently. Storage medium characterized by having the program. 18. A plurality of encoding modes including a prediction encoding mode for generating a prediction value based on a peripheral pixel of the encoding target pixel and entropy encoding a prediction error between the prediction value and a pixel value of the encoding target pixel. And a sign inversion processing unit for inverting interpretation of positive and negative signs of a prediction error in the prediction encoding mode so as to reduce a code amount after encoding. An encoding device characterized by the above-mentioned. 19. A plurality of encoding modes including a prediction encoding mode for generating a prediction value based on peripheral pixels of an encoding target pixel and entropy encoding a prediction error between the prediction value and a pixel value of the encoding target pixel. Is a coding method that can be selectively used for each pixel, comprising a sign inversion step of inverting the interpretation of the positive / negative sign of the prediction error in the prediction encoding mode so that the amount of code after encoding is reduced. An encoding method characterized by the following. 20. A plurality of encoding modes including a prediction encoding mode for generating a prediction value based on a peripheral pixel of the encoding target pixel and entropy encoding a prediction error between the prediction value and a pixel value of the encoding target pixel. A computer-readable storage medium storing an encoding program that can be selectively used for each pixel, wherein the interpretation of the positive and negative signs of the prediction error in the prediction encoding mode is reduced in code amount after encoding. A storage medium having a sign inversion step for inverting as much as possible in the program.