JPH05130512A - Picture element defect measuring instrument for solid-state image pickup element - Google Patents

Abstract

PURPOSE:To perform measurement with correlation with the picture element defect visual inspection of a solid-state image pickup element in picture element defect measurement used for the evaluation of the solid-state image pickup element. CONSTITUTION:An A/D converter 2 quantizes the signal of the solid-state image pickup element 1, and accumulates a digital output signal in frame memory 3. A differential signal between a picture element signal from which a random noise is eliminated at an inter-frame averaging part 4 and the shading component of an image detected at a space averaging part 5 is detected at an inter-frame difference part 6. A picture element defect detecting part 8 judges a picture element defective picture element at every picture element base on a histogram analysis at a histogram analyzing part 7, and a picture element position storage part 9 stores the position of the picture element defective picture element and the signal value of the picture element. A picture element defect grouping part 10 recognizes that the picture element is continued if a picture element defect exists in the peripheral picture element of the picture element defective picture element, and performs grouping by judging that it is one group.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a pixel defect measuring device used for evaluating a solid-state image pickup device.

[0002]

2. Description of the Related Art Conventionally, as a method of measuring a pixel defect of a solid-state image pickup device, an oscilloscope has been used to measure the solid-state image pickup device.
One pixel defective pixel is determined as one pixel defect for each scanning line, and the pixel defective pixel for the signal output when the position of the pixel defective pixel and the chart with a constant light amount are imaged by the solid-state image sensor The signal level ratio was measured as the signal level. As a method of visually inspecting pixel defects of the solid-state image sensor, the output signal of the solid-state image sensor is output to a monitor and the pixel defects are visually counted.

[0003]

Conventionally, when measuring a pixel defect of a solid-state image pickup device, one pixel defective pixel is judged to be one pixel defect using an oscilloscope, and the position of the pixel defective pixel is determined. As the signal level was measured as the ratio of the signal value of the pixel defective pixel to the signal output when the chart with a constant light amount was imaged by the solid-state image sensor, spatially continuous pixel defective pixels were detected. It cannot be judged as one lump, and the output signal of the solid-state image pickup device is output to the monitor and the pixel defect visual inspection of the solid-state image pickup device which visually counts the number of pixel defects cannot be taken.

In order to solve the above problems, the present invention provides
If defective pixels are spatially continuous, one group
It is an object of the present invention to provide a pixel defect measuring device for a solid-state image pickup device, which is capable of dealing with the visual inspection by judging each of the groups as one pixel defect.

[0005]

In order to achieve the above object, the present invention is an A / D for quantizing an output signal of a solid-state image pickup device.
A converter, a frame memory for accumulating the output signal of the A / D converter, and a frame memory signal for averaging the signals of the frame memory any number of times to reduce random noise contained in the signal of the solid-state image sensor. An inter-frame averaging unit, a spatial averaging unit that spatially averages the image signals obtained from the inter-frame averaging unit to detect a shading component of an image, and the inter-frame averaging unit. An inter-frame difference section that detects a difference signal between an image signal obtained from an averaging section and an image signal obtained from the spatial averaging section, and an image signal obtained from the inter-frame difference section Pixel defect detection for judging a pixel defective pixel by comparing the image signal obtained from the inter-frame difference unit with a histogram analysis unit for performing a histogram analysis for each pixel with an arbitrary level obtained from the histogram analysis unit Part and the pixel defect A pixel position storage unit that stores a pixel position and a signal value of a pixel, and if a pixel defective pixel exists in the peripheral pixels of the pixel defective pixel, it is recognized as spatially continuous, and one group is used. It is composed of a pixel defect group dividing unit which judges that there is a certain amount, and a signal value input unit which gives an external input a signal value output when a chart of a constant light amount is picked up by the solid-state image pickup device.

[0006]

According to the present invention having the above-mentioned structure, the A / D converter quantizes the signal of the solid-state image pickup device, the frame memory stores the digital output signal of the A / D converter, and the inter-frame averaging is performed. The unit takes in the digital output signal of the frame memory any number of times to remove random noise contained in the solid-state image pickup device, and the spatial averaging unit spatially averages the image signals obtained from the inter-frame averaging unit. The shading component of the image is detected, the inter-frame difference unit detects a difference signal between the image signal obtained from the inter-frame averaging unit and the image signal obtained from the spatial averaging unit, and a histogram The analysis unit performs a histogram analysis of the image signal obtained from the inter-frame difference unit, and the pixel defect detection unit obtains the image signal obtained from the inter-frame difference unit from the histogram analysis unit for each pixel. Determined to be a pixel defective pixel, the pixel position storage unit stores the position of the pixel defective pixel and the signal value of the pixel, and the pixel defect group dividing unit is arranged around the pixel defective pixel. If a pixel has a pixel defect, it is recognized as spatially continuous, and it is judged as one group to perform group division.
The signal values of defective pixels of the group are added in the group, and the addition sum is divided by the signal value output when the chart of the constant light amount is imaged by the solid-state image sensor.

As described above, according to the present invention, since the pixel defects are classified into groups, the pixel defects can be accurately measured even when the pixel defects are single or spatially continuous. Since a frame memory is used, pixel defects in one frame area can be measured at high speed.

[0008]

Embodiments of the present invention will be described in detail below with reference to the drawings. (FIG. 1) is a schematic configuration diagram of a pixel defect measuring apparatus for a solid-state image sensor according to an embodiment of the present invention.
The optical system, the drive circuit for driving the solid-state image sensor, the signal processing circuit for obtaining the output of the solid-state image sensor, etc. are omitted.

In FIG. 1, 1 is a solid-state image sensor which is an object to be measured, and 2 is A for quantizing a signal from the solid-state image sensor 1.
A / D converter, 3 is a frame memory for accumulating the digital output signal of the A / D converter 2, and 4 is a solid-state image pickup device for fetching the digital output signal of the frame memory 3 an arbitrary number of times and performing inter-frame averaging. 1 is an averaging unit that reduces random noise included in 1; 5 is a spatial averaging unit that spatially averages the image signals obtained from the inter-frame averaging unit 4 to detect a shading component of the image; Reference numeral 6 is an inter-frame difference section for detecting a difference signal between the image signal obtained from the inter-frame averaging section 4 and the image signal obtained from the spatial averaging section 5, and 7
Is a histogram analysis section for performing a histogram analysis of the image signal obtained from the inter-frame difference section.

A pixel defect detection unit 8 compares the image signal obtained from the inter-frame difference unit 6 with an arbitrary level obtained from the histogram analysis unit 7 to determine a pixel defective pixel.
Is a pixel defect position storage unit for storing the position of the pixel defective pixel detected by the pixel defect detection device 8 and the signal value of the pixel, and 10 is a pixel around the pixel defective pixel stored by the pixel defect position storage unit 9. If there is a defect, it is recognized as spatially continuous, it is judged as one group, the signal values of the pixel defective pixels in the one group are added, and the addition sum is A pixel defect group dividing unit that divides a chart with a constant light amount by a signal value output when an image is captured by a solid-state image sensor, and 11 is a signal value input unit that externally inputs the signal value that divides the addition sum. , 12 are pixel defect measuring devices for a solid-state image sensor, which are configured by using the pixel defect group dividing unit 10 from the above A / D converter 2.

Next, the operation of the pixel defect measuring apparatus for a solid-state image sensor according to the present embodiment is operated to make the solid-state image sensor in a light-shielding state, that is, white defect, which is one kind of pixel defect of the solid-state image sensor, so-called white defect measurement. explain. It should be noted that if a chart of a constant amount of light is imaged by a solid-state image sensor, a black defect or so-called black defect can be measured. In FIG. 1, the solid-state image sensor 1
Is quantized by the A / D converter 2. The digital output signal of the A / D converter 2 is stored in the frame memory 3 and then taken into the inter-frame averaging unit 4, and the image signal is in the state shown by A in FIG. 2). In FIG. 2A, the image signal output from the solid-state image sensor 1 is quantized, and the image signal includes random noise, fixed pattern noise, and a shading component of the image signal. .. The following operations are performed in order to remove random noise and shading components, which are unnecessary for the pixel defect measurement of the solid-state image sensor. The digital output signal of the frame memory 3 is taken into the inter-frame averaging unit 4 for each frame in an arbitrary number of times, the random noise is first removed, and the image signal is output. It becomes a state. Next, in order to detect the shading component included in the output signal of the solid-state image sensor 1, the image signal of the inter-frame averaging unit 4 is taken in by the spatial averaging unit 5.

The spatial averaging unit 5 performs spatial averaging, that is, 2
Since the three-dimensional low-pass filter is multiplied by the image signal, the fixed pattern noise included in the image signal is removed, and the image signal output by the spatial averaging unit 5 is (FIG. 2) C. It will be in the state shown. Further, in order to detect the fixed pattern noise of the solid-state image pickup device, the spatial averaging unit 5 converts the image signal output from the inter-frame averaging unit 4 from the image signal.
The inter-frame difference of the image signal output in (3) is performed by the inter-frame difference unit 6, and the image signal is in the state shown by D in FIG. Due to the fact that pixel pattern is included in the fixed pattern noise due to a well-known fact in pixel defect detection,
Generally, as a method of detecting a pixel defect, a histogram analysis of fixed pattern noise is performed and a pixel pattern having a large absolute value level of fixed pattern noise is detected as a pixel defect. Therefore, the histogram analysis unit 7 performs the histogram analysis of the image signal output by the inter-frame difference unit 6.

The result of the histogram analysis is shown in FIG. 3 and the tendency of the magnitude of the level of the fixed pattern noise can be grasped. In the present embodiment, in the histogram analysis result of (FIG. 3), a pixel whose level is larger than a is treated as a pixel defect. Since the pixel defect detection unit 8 detects one of the image signals output by the inter-frame difference unit 6 that has a level higher than a, the output image signal is only a white defect (see FIG. 2E). The state is indicated by. The pixel signal ((FIG. 2) E) output from the pixel defect detection unit 8 represents a white defect, but even if it is spatially continuous, one group
Is not recognized.

Therefore, grouping is performed using the pixel defect position storage unit 9 and the pixel defect group dividing unit 10. The pixel defect position storage unit 9 stores the position and signal value of the white defective pixel in one frame area indicated by E (FIG. 2).
The stored contents are fetched from the pixel defect position storage unit 9 into a pixel defect group dividing unit 10 having a memory for one frame area, and the pixel defect group dividing unit 10 groups the white defective pixels. Division and the signal values of the white defect pixels of one group divided by the group are added, and the sum of the signal value and the signal value output when the chart of a constant light amount is imaged by the solid-state image sensor Calculate the ratio. The signal value output when the chart of the constant light amount is imaged by the solid-state image sensor is externally input by the signal value input unit 11. The operation of the pixel defect group dividing unit 10 will be described with reference to (FIG. 4) to (FIG. 6).

First, the peripheral pixels for recognizing whether the target white defective pixel is spatially continuous with the peripheral white defective pixel will be described. In FIG. 4, a lattice represents a pixel unit, S indicated by diagonal lines represents a target white defective pixel, and a lattice region surrounded by a dotted line represents peripheral pixels. next,
Group division for recognizing one group if it is spatially continuous using the peripheral pixels will be described with reference to FIGS. 5 and 6.

FIG. 5 shows the operation of group division by an algorithm, and FIG. 6 shows the state of white defective pixels in which the algorithm of FIG. 5 divides white defective pixels into groups. .. In (FIG. 5), the peripheral pixel recognition I observes the peripheral pixels described in (FIG. 4), determines whether there are white defective pixels in the peripheral pixels of S, and after the determination, the white defective pixels in the same group. Label them so that they can have a common recognition.
In FIG. 5, the new labeling K is k1, k2 when there is no white defect pixel in the surrounding pixels described in (FIG. 4) and when the noted S is not labeled.
Labeling with a coefficient k as described below, the sub-labeling J operates when there is a white defective pixel in the peripheral pixels described in (FIG. 4), and the small labeling J is labeled in the S and already in the peripheral pixels. Replace the label of the coefficient with k with the smallest label of the coefficient associated with k in the labeling in the surrounding pixels. The 1-frame end L is determined by checking the pixel position of S to see if the labeling described above is completed within the 1-frame area.

The algorithm operation will be described using an embodiment. (FIG. 6) A has no white defective pixels in the peripheral pixels, so a new labeling K is performed by the peripheral pixel recognition I in (FIG. 5).
It is determined that the process proceeds to step (4), and the new label K is passed (FIG. 6) to become B, and the label S is given to S. (FIG. 6) C has a white defective pixel in its peripheral pixels, so it proceeds to the small labeling J, passes through the small labeling J and becomes (FIG. 6) D, and the label of k1 has already been given in the peripheral pixels. Therefore, a label k1 is given to the S.

Similarly to (C) of FIG. 6, (E) of FIG. 6 becomes (F) of FIG. 6 through the small labeling J, and the label of S is given to the S, and the label of k2 already given is given. Replaces the label k1. The algorithm operation is performed until the S reaches the final pixel in one frame area, and the algorithm operation end confirmation is 1
This is performed by determining the pixel position of S at the frame end L. By the algorithm operation, labeling of white defective pixels, that is, grouping is completed.

Next, in order to calculate the ratio of the group of one white defective pixel divided into groups to the signal output when the chart of a constant light quantity is imaged by the solid-state image pickup device, the group is calculated. When the pixels are divided into groups, those having the same label k given to the white defective pixels are added, and the addition sum is divided by the external input value given from the signal value input unit 11.

The characteristic of the measurement value of the white defective pixel is that the position and the signal level are required. However, the first position labeled as k in the cluster of the one white defective pixel is the position. The quotient obtained by the division is given as a signal level. This structure can be easily realized by using a computer, and accurate and highly reliable measurement can be realized.

[0021]

As described above, according to the present invention, the following effects can be obtained. (A) If the pixel defects are spatially continuous, it is judged as one pixel defective pixel, and therefore it is compatible with the conventional visual inspection for pixel defects of the solid-state image sensor. (B) 1 frame at high speed because frame memory is used
Pixel defects in the pixel area can be measured.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a pixel defect measuring device for a solid-state image sensor according to an embodiment of the present invention.

FIG. 2 is a schematic diagram showing a pixel signal indicating an operation of the pixel defect measuring device of the solid-state image sensor.

FIG. 3 is an output signal waveform diagram of a histogram analysis unit in the pixel defect measuring device of the solid-state image sensor.

FIG. 4 is a diagram illustrating peripheral pixels and a target pixel used for grouping.

FIG. 5 is a flowchart illustrating a grouping operation.

FIG. 6 is a diagram for explaining the labeling operation.

[Explanation of symbols]

 1 Solid-state image sensor 2 A / D converter 3 Frame memory 4 Inter-frame averaging unit 5 Spatial averaging unit 6 Inter-frame difference unit 7 Histogram analysis unit 8 Pixel defect detection unit 9 Pixel defect position storage unit 10 Pixel defect grouping unit 11 Signal value input unit 12 Pixel defect measuring device for solid-state imaging device

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Ryuichiro Kuga 1006 Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd.

Claims (2)

[Claims] 1. An A / D for quantizing a signal of a solid-state image sensor.
The calculation includes a converter, a frame memory for accumulating the digital output signal of the A / D converter, and an arithmetic processing unit for arithmetically processing the digital output signal of the frame memory to calculate a pixel defect. The processing unit averages the digital output signal of the frame memory between frames at an arbitrary number of times to reduce random noise contained in the signal of the solid-state image pickup device, and an inter-frame averaging means. Spatial averaging means for spatially averaging the image signals obtained from the averaging means to detect the shading component of the image,
An inter-frame difference means for detecting a difference signal between the image signal obtained by the inter-frame averaging means and the image signal obtained by the spatial averaging means, and the inter-frame difference means. Histogram analysis means for performing a histogram analysis of the image signal, and the image signal obtained by the inter-frame difference means is compared pixel by pixel with an arbitrary level obtained by the histogram analysis means to determine a pixel defective pixel. Pixel defect detection means, pixel defect position storage means for storing the pixel position of the pixel defective pixel and the signal value of the pixel, and spatially continuous if a pixel defect exists in the peripheral pixels of the pixel defective pixel. A pixel defect group dividing means for recognizing that there is one group, and a signal value given by an external input as a signal value output when a chart of a constant light amount is imaged by the solid-state image sensor Entering Pixel defect measuring apparatus of the solid-state imaging device which comprises a means. 2. A peripheral pixel of the pixel defective pixel includes a pixel one pixel before the pixel defective pixel, a pixel one scanning line before the pixel defective pixel, and a pixel one scanning line before the pixel defective pixel. The pixel defect measuring device for a solid-state image sensor according to claim 1, wherein the pixel is a pixel one pixel before and a pixel one pixel after the pixel one scanning line before the pixel defective pixel.