JP2006109315A - Image analyzing apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To decide presence/no presence of abnormality in a lens barrel by calculating the number of critical TV resolution stripes. <P>SOLUTION: In an image analyzing apparatus 100, the image of a chart picked up by using the lens barrel 200 is taken in to calculate line width and contrast of a wedge pattern on the chart image, and the number of critical TV resolution stripes is calculated on the basis of these calculation results. The resolution of the lens barrel 200 that has picked up the image of the chart is decided on the basis of the calculated number of critical TV resolution stripes, and presence/no presence of abnormality in the lens barrel 200 is decided. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image analysis apparatus and an image analysis method for analyzing an image and determining the resolution of an image sensor.

  Non-Patent Document 1 discloses the following limit resolution determination algorithm. Based on the detection result, scan the rectangular area until the wedge pattern can no longer be read, and detect the number of wedge-shaped figures for the data of the rectangular area including the wedge pattern composed of multiple wedge-shaped figures. To determine the critical resolution.

Camera Imaging Equipment Industry Standard Standard name: CIPA DC-003-2003 Resolution measurement method for digital cameras (Established on December 17, 2003) 8-P. 19 (issued by the Camera & Imaging Products Association)

  However, in the conventional limit resolution determination algorithm, it is necessary to scan the rectangular area until the wedge pattern cannot be read, which causes a problem that processing takes time.

The image analysis apparatus according to claim 1, wherein a plurality of wedge-shaped figures are spaced apart by a predetermined interval, and a resolution chart formed as a wedge pattern having a bright part and a dark part is imaged by an imaging unit through an optical system, and the image is taken. In the image data, an image analyzing apparatus that calculates the limit TV resolution fringe number by scanning each wedge pattern in the direction of the tip of the wedge-shaped figure, and at least a bright part and a dark part of the wedge pattern at an arbitrary scanning position Line width calculating means for calculating the length of the range including one line as the line width of the wedge pattern, and contrast calculating means for calculating the contrast of the wedge pattern within the line width at an arbitrary scanning position calculated by the line width calculating means And the line width at an arbitrary scanning position calculated by the line width calculating means, and the code calculated by the contrast calculating means at the arbitrary scanning position. Scanning position specifying means for detecting a scanning position in which at least one of the trusts does not satisfy a predetermined condition and specifying a scanning position for calculating the limit TV resolution fringe number based on the detected scanning position. It is characterized by that.
The image analysis apparatus according to claim 2 is the image analysis apparatus according to claim 1, and is specified by a TV resolution fringe number calculating unit that calculates the number of TV resolution fringes at an arbitrary scanning position, and a scanning position specifying unit. It further comprises a limit TV resolution fringe number calculating means for calculating the TV resolution fringe number at the scanned position as the limit TV resolution fringe number.
The image analysis device according to claim 3 is the image analysis device according to claim 1 or 2, wherein the resolution of the optical system is determined based on the limit TV resolution fringe number calculated by the limit TV resolution fringe number calculation means. It further comprises resolution determination means for determining pass / fail.
The image analysis method according to claim 4, wherein a plurality of wedge-shaped figures are spaced apart by a predetermined interval, and a resolution chart formed as a wedge pattern having a bright part and a dark part is imaged by an imaging means through an optical system, and the image is taken. An image analysis method for calculating a limit TV resolution fringe number by scanning each wedge pattern in the direction of the leading edge of a wedge-shaped figure in image data, wherein at least a bright part and a dark part of a wedge pattern at an arbitrary scanning position The length of the range including each one is calculated as the line width of the wedge pattern, the contrast of the wedge pattern within the line width at any scanning position is calculated, the line width calculated at any scanning position, and any scanning A scanning position where at least one of the contrasts calculated at the position does not satisfy a predetermined condition is detected, and based on the detected scanning position. And identifies the scanning position for calculating the limit TV Kaizoshima number Te.
The image analysis method according to claim 5 is the image analysis method according to claim 4, wherein the number of TV resolution fringes at an arbitrary scanning position is calculated, and the number of TV resolution fringes at the specified scanning position is calculated as a limit TV solution. It is calculated as the number of image stripes.
The image analysis method according to claim 6 is characterized in that, in the image analysis method according to claim 4 or 5, the quality of the resolution of the optical system is determined based on the calculated limit TV resolution fringe number. .

  According to the present invention, the wedge pattern included in the image data of the resolution chart is scanned in the direction of the leading end of the wedge-shaped figure, and at least one of the line width and contrast of the wedge pattern does not satisfy the predetermined reference value. Based on the scanning position, the scanning position for calculating the limit TV resolution fringe number is specified. Thereby, the scanning range of the wedge pattern for specifying the scanning position can be limited, and the scanning position can be specified at high speed.

  FIG. 1 is a block diagram illustrating a configuration example of an embodiment of an image analysis apparatus according to the present embodiment. The image analysis apparatus 100 includes an image capturing I / F 101 for capturing an image captured by a DSC (Digital Still Camera) barrel 200, a control device 102 that performs image processing to be described later, and a monitor that displays the results of the image processing. 103. The image capture I / F 101 is, for example, a memory card slot for capturing an image via a memory card or a USB interface for capturing an image by connecting a DSC barrel 200 via a USB cable.

  The DSC lens barrel 200 is a lens barrel unit that exists alone as a state before being attached to the DSC main body in the DSC manufacturing process, and includes a CCD 201 as an image sensor, a lens 202, and a control device 203. An image of a resolution chart (hereinafter referred to as “chart”) for the electronic still camera shown in FIG. 2 is taken.

  The image analysis apparatus 100 takes in an image of a chart imaged using the lens barrel 200 and executes image processing as follows. For the horizontal and vertical components at the center and four corners on the chart image, that is, for the images in the areas 2a to 2j set on the image, the limit TV resolution fringe number calculation process described later is executed. Then, based on the calculated limit TV resolution fringe number in each region, the resolution of the lens barrel 200 that captures the chart is determined to determine whether the lens barrel 200 is abnormal.

  First, the limit TV resolution fringe number calculation process will be described. FIG. 3 is a diagram schematically showing an enlarged region 2a which is a vertical component on the chart image. The area 2a includes a wedge pattern composed of four black wedge-shaped figures 3a to 3d on a white background, and is indicated by an arrow 3e in the XY coordinate system of the wedge pattern including the four wedges. Thus, the image is scanned at a predetermined interval in the y-axis direction, for example, every pixel. Then, as will be described later, the y coordinate value, the line width, the contrast, and the number of TV resolution fringes at each scanning position are calculated.

  For example, when calculating the y coordinate value, the line width, the contrast, and the number of TV resolution fringes at the scanning position 3f, the following processing is performed. First, the y coordinate value at the scanning position 3f is acquired. The line width at this time is calculated as shown in FIG. That is, a waveform indicating a change in brightness from the wedge-shaped figure 3a side to the 3d side in the range excluding the wedge-shaped figures 3a and 3d on the outer sides of the four wedge-shaped figures 3a to 3d. The distance between the outer sides of the bright peak of the waveform, that is, the length from the point 1 to the point 5 is calculated as the line width 4a at the scanning position 3f. Then, the ratio between the calculated line width 4a at the scanning position 3f and the actual line width at the position corresponding to the scanning position 3f on the chart is calculated. If the ratio is less than a predetermined value, the line width is calculated correctly. Therefore, the contrast is calculated.

The contrast at the scanning position 3f is the average value (brightness MAX) of the points with the highest luminance within the line width 4a, that is, the average value of the luminances at the points 1, 3, and 5, and the point with the lowest luminance within the line width. Based on the average value (luminance MIN), that is, the average value of the luminance at points 2 and 4, it is calculated by the following equation (1).
Contrast = (luminance MAX−luminance MIN) / (luminance MAX + luminance MIN) (1)
If the contrast calculated by equation (1) is equal to or greater than a predetermined reference value, it is determined that the contrast is correctly calculated, and the number of TV resolution fringes at the scanning position 3f is calculated.

The number of TV resolution stripes at this time is calculated as follows by interpolating the line width 4a at the scanning position 3f described above. FIG. 5 is an enlarged view of a portion including the chart areas 2 a and 2 b captured by the lens barrel 200. At this time, assuming that the actual line width at the symbol a on the chart is A, the actual line width at the symbol b is B, and the symbol c, that is, the line width 4a on the chart image at the scanning position 3f is C, the TV resolution stripes The number is calculated by the following equation (2).
Number of TV resolution stripes = 600 + ((C−A) / (B−A)) × 400 (2)

  As a result of scanning the image in the y-axis direction and calculating the y coordinate value, the line width, the contrast, and the number of TV resolution fringes at each scanning position, the line width at the certain scanning position, for example, the scanning position 3h, and When at least one of the contrasts does not satisfy the above-described condition for determining that the contrast is correct, it is determined that the scanning position exceeds the limit TV resolution of the CCD 201 mounted on the lens barrel 200 that captured the chart. Then, the calculation result of the TV resolution fringe number at the scanning position 3g scanned immediately before is displayed on the monitor 103 as the limit TV resolution fringe number.

  For example, a matrix diagram as shown in FIG. 6 is created in the memory space of the control apparatus 102, and each item in the matrix diagram is filled when each scanning position is scanned. That is, the y coordinate value of the scanning position during scanning is acquired and stored, and the line width and contrast are calculated. Then, it is determined by ○ X whether or not the line width and contrast at the scanning position satisfy the above-described conditions. When both satisfy the condition, that is, when it is ◯, the number of TV resolution stripes is calculated and the next scanning position is scanned. On the other hand, when one of the line width and the contrast is X as in the scanning position 3h, the scanning position returns to the scanning position 3g scanned immediately before, and the number of TV resolution stripes “EEE” at this time is calculated. The number of limit TV resolution stripes.

  In addition, although the case where the above-described processing is performed on the region 2a that is a vertical component on the chart image has been described, the above-described processing is performed on other vertical components, thereby calculating the limit TV resolution fringe number. Is possible. Further, for the region that is a horizontal component (for example, the region 2b), as shown in FIG. 7, the image is scanned in the x-axis direction as indicated by an arrow 7e, and the length from point 1 to point 5 is defined as the line width. By executing the processing described above, it is possible to calculate the number of limit TV resolution fringes.

  If the limit TV resolution fringe number calculation process is executed for all the areas 2a to 2j, and all the limit TV resolution fringe numbers calculated in each area are equal to or greater than a predetermined reference value, the chart image is displayed. The CCD 201 mounted on the imaged lens barrel 200 is determined to satisfy the required resolution, and the lens barrel 200 is determined to have no abnormality. On the other hand, if the limit TV resolution fringe number in any region is less than a predetermined reference value, the lens barrel 200 is determined to be abnormal. Then, the determination result is displayed on the monitor 103. Note that a predetermined reference value to be compared with the limit TV resolution fringe number in order to determine the presence or absence of the lens barrel 200 is set in advance for each camera model in which the lens barrel 200 is mounted.

  FIG. 8 is a flowchart showing processing of the image analysis apparatus 100 in the present embodiment. The process shown in FIG. 8 shows the process executed for any one of the areas 2a to 2j for the sake of simplification, and it is assumed that the chart image is captured via the image capture I / F 101. It is executed by the control device 102 as a program to be activated when it is determined. In step S10, data of the first scanning position on the chart image is read, and the process proceeds to step S20.

  In step S20, the line width at the current scanning position is calculated, and the process proceeds to step S30. In step S30, a ratio between the calculated line width and the actual line width at a position corresponding to the current scanning position on the chart is calculated, and whether or not a condition that the ratio is less than a predetermined value is satisfied. to decide. If the condition is not satisfied, the process proceeds to step S80 described later. If the condition is satisfied, the process proceeds to step S40.

  In step S40, the contrast at the current scanning position is calculated, and the process proceeds to step S50. In step S50, it is determined whether or not a condition that the calculated contrast is equal to or greater than a predetermined reference value is satisfied. If the condition is not satisfied, the process proceeds to step S80 described later. If the condition is satisfied, the process proceeds to step S60. In step S60, the number of TV resolution stripes is calculated by the above-described equation (2), and the process proceeds to step S70. In step S70, the data of the next scanning position on the chart image is read, and the process returns to step S20.

  Next, processing when the line width or contrast does not satisfy the condition will be described. In this case, the process proceeds to step S80, and the number of TV resolution fringes at the immediately preceding scanning position is set as the limit TV resolution fringe number. Thereafter, the process proceeds to step S90, and it is determined whether or not the limit TV resolution fringe number is equal to or greater than a predetermined value. If it is determined that the limit TV resolution fringe number is equal to or greater than the predetermined value, the process proceeds to step S100, where it is determined that the lens barrel 200 that has captured the chart image has no abnormality, and the process proceeds to step S120.

  On the other hand, if it is determined that the limit TV resolution fringe number is less than the predetermined value, the process proceeds to step S110, where it is determined that the lens barrel 200 that has captured the chart image is abnormal, and the process proceeds to step S120. In step S120, the limit TV resolution fringe number and the presence or absence of the lens barrel 200 are displayed on the monitor 103, and the process ends.

According to the present embodiment described above, the following operational effects can be obtained.
(1) The limit TV resolution fringe number is calculated based on the chart image captured by the lens barrel 200, and the resolution of the lens barrel 200 is determined based on the calculation result to determine whether there is an abnormality. . Thus, conventionally, the resolution of the lens barrel 200 is visually determined by a human by projection resolution. However, by automatically performing this, the determination time can be shortened and the determination accuracy can be improved.
(2) When calculating the limit TV resolution fringe number, when either the line width or contrast at the scanning position during scanning does not satisfy the standard, the TV resolution fringe number at the immediately preceding scanning position is calculated as the limit TV resolution. The number of image stripes was determined. As a result, the scanning range of the wedge pattern can be limited, and the processing can be performed at high speed.

In addition, it can also deform | transform as follows.
(1) In the above-described embodiment, the case where the number of wedge-shaped figures on the chart image is four has been described, but the present invention can be applied to the case of three or less and the case of five or more.

(2) In the above-described embodiment, when calculating the line width, a waveform indicating a change in brightness is calculated in a range excluding the wedge-shaped figures on both outer sides, and outside the peak of the bright part of the waveform. The length between them was calculated as the line width at the scanning position. However, a range including at least one bright peak and one dark peak in the waveform may be calculated as the line width. For example, the length between the outer sides of the peak of the bright part of the waveform may be calculated as the line width at the scanning position without removing the wedge-shaped figures on the outer sides.

(3) In the embodiment described above, when calculating the limit TV resolution fringe number, the line width and the contrast are calculated at the time of scanning each scanning position as shown in the matrix diagram of FIG. It is determined whether or not the result of the calculation satisfies the condition, and if it satisfies the condition, the number of TV resolution stripes is calculated and the next scanning position is scanned. If either one of the line width and the contrast does not satisfy the condition, return to the previous scanning position and calculate the number of TV resolution fringes at the scanning position as the limit TV resolution fringe number. did.

  However, the present invention is not limited to this. For example, only the line width is calculated when each scanning position is scanned, and when the calculated line width does not satisfy the condition, the contrast is calculated by returning to the previous scanning position. If the contrast satisfies the condition, the number of TV resolution fringes at this scanning position may be calculated and used as the limit TV resolution fringe number. When the contrast calculated at this scanning position does not satisfy the condition, the same processing may be performed by returning to the previous scanning position and calculating the contrast. This eliminates the need to calculate the line width and contrast at all scanning positions, thereby further speeding up the processing.

(4) In the embodiment described above, the example in which the contrast in the matrix diagram shown in FIG. 6 is determined to be ◯ when the calculation result according to the expression (1) satisfies the condition has been described. However, the present invention is not limited to this. For example, the maximum values of the respective contrasts of points 1, 3, and 5 shown in FIG. 3 at an arbitrary scanning position and the background outside the wedge-shaped figures on the outer sides are in the white area. 6 is calculated when the contrast at each point on the arbitrary scanning position is calculated, at least one of these contrasts is equal to or less than a predetermined reference value, and the calculation result by the expression (1) satisfies the condition. The contrast in the matrix diagram shown may be determined as ◯.

  The correspondence between the constituent elements of the claims and the embodiment will be described. The control device 102 corresponds to a line width calculation unit, a contrast calculation unit, a scanning position specifying unit, a TV resolution fringe number calculation unit, a limit TV resolution fringe number calculation unit, and a resolution determination unit, and the lens barrel 200 corresponds to an imaging unit. To do. Note that the present invention is not limited to the configurations in the above-described embodiments as long as the characteristic functions of the present invention are not impaired.

It is a block diagram which shows the structural example of one Embodiment of an image-analysis apparatus. It is a figure which shows the chart imaged using the lens-barrel 200. FIG. It is the figure which expanded the area | region of the vertical component on a chart image. It is a figure which shows the waveform which shows the change of the brightness from the wedge-shaped figure 3a side to 3d side. It is the figure which expanded the part containing the area | regions 2a and 2b of a chart. It is a figure which shows the specific example of the matrix figure which memorize | stores the y coordinate value of the scanning position in scanning, the determination result of line width and contrast, and the number of TV resolution fringes. It is the figure which expanded the area | region of the horizontal component on a chart image. 3 is a flowchart showing processing of the image analysis apparatus 100. FIG.

Explanation of symbols

100 Image Analysis Device 101 Image Import I / F
102 control device 103 monitor 200 lens barrel 201 CCD
202 Lens 203 Control device

Claims (6)

A resolution chart formed as a wedge pattern having a bright portion and a dark portion with a plurality of wedge-shaped figures spaced apart by a predetermined interval is imaged by an imaging means through an optical system, and each wedge pattern is wedge-shaped in the captured image data. An image analyzer that calculates the number of limit TV resolution fringes by scanning in the direction of the tip of a figure,
A line width calculating means for calculating a length of a range including at least one bright part and dark part of the wedge pattern at an arbitrary scanning position as a line width of the wedge pattern;
Contrast calculation means for calculating the contrast of the wedge pattern within the line width at the arbitrary scanning position calculated by the line width calculation means;
A scanning position in which at least one of the line width at an arbitrary scanning position calculated by the line width calculating unit and the contrast calculated by the contrast calculating unit at the arbitrary scanning position does not satisfy a predetermined condition is detected. An image analyzing apparatus comprising: a scanning position specifying means for specifying a scanning position for calculating the limit TV resolution fringe number based on the detected scanning position. The image analysis apparatus according to claim 1,
TV resolution fringe number calculating means for calculating the TV resolution fringe number at the arbitrary scanning position;
An image analysis apparatus, further comprising: a limit TV resolution fringe number calculating unit that calculates the number of TV resolution fringes at the scanning position specified by the scanning position specifying unit as the limit TV resolution fringe number. In the image analysis device according to claim 1 or 2,
An image analyzing apparatus, further comprising: a resolution determining unit that determines whether the resolution of the optical system is good or not based on the limit TV resolution fringe number calculated by the limit TV resolution fringe number calculating unit. A resolution chart formed as a wedge pattern having a bright portion and a dark portion with a plurality of wedge-shaped figures spaced apart by a predetermined interval is imaged by an imaging means through an optical system, and each wedge pattern is wedge-shaped in the captured image data. An image analysis method for calculating the number of limit TV resolution fringes by scanning in the direction of the tip of a figure,
Calculating the length of a range including at least one bright part and dark part of the wedge pattern at an arbitrary scanning position as the line width of the wedge pattern;
Calculating the contrast of the wedge pattern within the line width at the arbitrary scanning position;
A scanning position where at least one of the line width calculated at the arbitrary scanning position and the contrast calculated at the arbitrary scanning position does not satisfy a predetermined condition is detected, and a limit TV solution is detected based on the detected scanning position. An image analysis method characterized by specifying a scanning position for calculating the number of image stripes. The image analysis method according to claim 4,
Calculate the number of TV resolution stripes at the arbitrary scanning position,
An image analysis method characterized in that the number of TV resolution fringes at the specified scanning position is calculated as a limit TV resolution fringe number. In the image analysis method according to claim 4 or 5,
An image analysis method comprising: determining whether the optical system has good resolution based on the calculated limit TV resolution fringe number.

Images