JPH0630425A - Solid-state imaging device - Google Patents

Abstract

(57) [Abstract] [Purpose] It is possible to deal with changes in defects due to electrostatic breakdown and changes over time, and it is possible to reliably correct defects that occur in the actual imaging state. Provided is a solid-state imaging device that does not require the above-mentioned distribution form. According to a level difference between pixel signals of a certain pixel and a pixel of the same color adjacent thereto and a level difference between pixel signals of pixels of the same color existing in the vicinity of a certain pixel during normal image pickup. A defect detection circuit 8 for detecting a defective pixel is provided, and an image pickup output of the defective pixel detected by the defect detection circuit 8 is corrected in real time by the defect correction circuit 9.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a solid-state image pickup device,
In particular, it relates to a solid-state imaging device having a function of detecting defective pixels.

[0002]

2. Description of the Related Art In an image pickup device formed of a semiconductor such as a CCD, a defective pixel whose sensitivity is lowered may occur due to a local crystal defect of the semiconductor. In such a case, the defective pixel deteriorates the image quality. It is known to be the cause. In order to eliminate the image quality deterioration due to the defective pixel, defect correction has been conventionally performed in a solid-state imaging device using a CCD or the like.

In the conventional defect correction, defective pixels included in an image pickup device are detected when the image pickup device is shipped and sorted.
The defect data regarding the defective pixel is stored in the ROM,
This ROM is shipped as a pair with the image pickup device, and at the time of normal image pickup, the defective pixel of the image pickup device is specified based on the defect data stored and held in the ROM, and the defect signal is corrected for the defective pixel. I was going to do it.

[0004]

However, in the conventional defect correction, the defect correction is performed by using the defect data stored and held in the ROM at the stage of selecting the shipment of the image pickup device, so that the local crystal of the semiconductor is corrected. Although it is possible to deal with pixel defects due to defects, it is not possible to deal with electrostatic damage when the image sensor is incorporated in the video camera and defect changes due to aging after mounting in the video camera, and shipping selection leaks If it occurs, there is a problem that the defect cannot be corrected.

In addition, a distribution mode in which an image pickup device and a ROM are paired is indispensable, which requires time and effort for physical distribution and manufacturing of a set, and the ROM is attached, which increases the cost and size of the entire set. was there. Furthermore, when detecting a defective pixel having temperature dependency, there is a problem that it takes time and cost to control the temperature of the measuring device.

The present invention can cope with a change in defects due to electrostatic breakdown or a change with time, can reliably correct a defect occurring in an actual image pickup state, and can distribute the image pickup device and the defect data as a pair. An object is to provide a solid-state imaging device that does not require a form.

[0007]

According to the solid-state image pickup device of the present invention, a level difference between pixel signals of a pixel and a pixel of the same color adjacent thereto and each pixel of different color pixels existing around the pixel. The configuration includes a defect detection circuit that detects a defective pixel based on a level difference between signals, and a defect correction circuit that corrects a pixel signal for the defective pixel.

[0008]

With the spatial correlation of each pixel signal taken into consideration, the level difference between each pixel signal of a certain pixel and its adjacent pixels of the same color is detected during normal image pickup, and a judgment is made with respect to a predetermined threshold value. . Further, the defective pixel is detected based on the level difference between the pixel signals of the different-color same-pixels existing in the periphery of the pixel, and the determination is made with respect to a predetermined threshold value. In this way, by adopting the double determination structure, the defective pixel is detected while avoiding erroneous detection due to the edge component of the image. Then, by correcting the pixel signal of the detected defective pixel in real time, the image quality deterioration due to the defective pixel is improved.

[0009]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a solid-state imaging device according to the present invention. In FIG. 1, incident light from a subject is guided by an imaging lens 1 to an imaging surface of a CCD imaging device 3 via an optical filter 2. As the CCD image pickup device 3, for example, a color CCD image pickup device of complementary color checker system color coding is used. This color CC
The D image pickup device 3 employs, for example, an interline transfer system as a charge transfer system.

The reading of signal charges from each pixel of the CCD image pickup device 3 and each drive control of vertical / horizontal transfer are performed by the drive circuit 5 based on various timing signals generated by the timing generation circuit 4. The image pickup output (CCD output) of the CCD image pickup device 3 is sample hold (S / H)
The signal is sampled and held by the circuit 6, digitized by the A / D converter 7, and then supplied to the defect detection circuit 8 and the defect correction circuit 9.

The defect detection circuit 8 is a feature of the present invention, and is provided for detecting defective pixels based on the CCD output during normal image pickup. The operation and the like will be described later. On the other hand, the defect correction device 9 provides the pixel signals of the defective pixels detected by the defect detection circuit 8 in real time, for example, the pixel signal of the pixel two lines before, the pixel signal of the pixel two pixels before in the same line. The defect correction is performed by so-called average value interpolation in which the average value of the signal and the pixel signal of the pixel after two pixels is replaced. The defect-corrected CCD output is subjected to various kinds of signal processing in the signal processing circuit 10 to become a video output.

An example of the spatial arrangement of pixel signals obtained from the CCD image pickup device 3 is shown in FIG. In this example,
A pixel signal array consisting of four types of color information of A, B, C, and D is taken up, and the signal output waveform is shown in FIG. As shown in FIG. 3, pixel signals A _k1 , B _k2 , A _k3 , ... _Are sequentially obtained on the k line. In the next l line, C _l1 , D
_l2 , C _l3 , ... Pixel signals different from the k-th line are sequentially obtained, and in the next m-th line, A _m1 , B
A pixel signal is obtained again with the same combination of colors as _m2 , A _m3 , ... And the k-th line.

When an image of a subject with little change in brightness is taken,
The following equation is locally established between pixel signals.

## EQU1 ## A _{k1 ≈A k3 ≈A k5} ≈ ..., A _{k1 ≈A m1} , A _{k3 ≈A m3} , ...
… B _k2 ≈ B _k4 ≈ B _k6 ≈ ……, B _k2 ≈ B _m2 , B _k4 ≈ B _m4 ,…
_{... C l1 ≒ C l3 ≒ C} l5 ≒ ......, C l1 ≒ C n1, C l3 ≒ C n3, ...
_{... D l2 ≒ D l4 ≒ D} l6 ≒ ......, D l2 ≒ D n2, D l4 ≒ D n4, ...
That is, the output level difference between adjacent pixels of the same color is small.

However, when a certain pixel has a defect, _assuming that the defective pixel is a pixel corresponding to B _m4 (shown by hatching in FIG. 2), the signal output waveform of each pixel is as shown in FIG. In the defective pixel portion, the pixel signal B
Each pixel signal B _m2 , B _m6 , B _{k4 of} the same color pixel adjacent to _m4 ,
There will be a level difference with B _p4 . That is,
Generally, as a property of CCD output of the CCD image pickup device 1, between a certain pixel and a neighboring pixel which obtains the same color information as that of the pixel except when a subject whose image brightness is extremely changed is imaged. Has a strong signal correlation, on the contrary,
It can be said that the image defect portion of the CCD image pickup device 1 represented by the point defect has a very low correlation with the neighboring pixels.

The present invention is characterized in that the solid-state image pickup device is equipped with the defect detection circuit 8 in order to detect the defective pixel based on the spatial correlation of each pixel signal. I am trying. The operation of the defect detection circuit 8 will be described below. Now, in the spatial arrangement of the pixel signals shown in FIG. 2, the pixel corresponding to B _m4 (hatched portion) is the defective pixel to be determined. In this case, the adjacent pixels of the same color for which the correlation of the pixel signal B _m4 should be determined are pixel signals B _m2 ,
There are four pixels of B _m6 , B _k4 , and B _p4 . First, the output level difference between the pixel signals B _m2 , B _m6 , B _k4 , and B _p4 of the adjacent pixels and the pixel signal B _m4 of the defective pixel is detected including the positive / negative polarities.

In detecting the level difference between the adjacent pixels, the level difference between the pixel signals B _k4 and B _p4 of the adjacent pixels in the vertical direction is 4 lines (k, l, m,
By storing and holding the pixel signals for n) in four line memories (first synchronization circuits) for each line, the pixel signals B _k4 and B _p4 are synchronized and stored in the line memory of the k-th line. Level difference between the pixel signal B _k4 and the pixel signal B _m4 stored in the m-th line memory, and the pixel signal B _m4 stored in the m-th line memory and the p-th pixel signal B _p4 Is performed by each subtracter (first level difference detection circuit).

On the other hand, the level difference between the pixel signals B _m2 and B _m6 of adjacent pixels in the horizontal direction is 2 by the latch circuit (second synchronization circuit) of the pixel signal B _m4 with respect to the pixel signal B _m6 . The pixel signals B _m6 and B _m4 are synchronized at the same time by latching for a period corresponding to the pixel pitch, and the pixel signal B _m2 with respect to the pixel signal B _m4 is further 2 pixels by a latch circuit (second synchronization circuit). The pixel signals B _m4 and B _m2 are synchronized by latching for a period corresponding to the pitch, and the level difference between the pixel signal B _m6 and the pixel signal B _m4 and the level between the pixel signal B _m4 and the pixel signal B _m2 are respectively set. It is performed by using a subtracter (second level difference detection circuit).

Next, each of the four detected level differences is compared with a preset predetermined threshold value V _th by a comparator, and the comparison result is "L" when each level difference is less than the threshold value V _th.
When the level is _{equal to} or higher than the threshold value V _th , it is defined as “H” level.
Then, if there are n "H" levels in the determination circuit, the pixel of the pixel signal B _m4 is determined to be a defective pixel. Here, since n has four adjacent pixels, n can be arbitrarily set within the range of n = 1 to 4. However, when n is small, FIG.
As shown in, when the edge component of the image is input,
If it is erroneously recognized as a defective pixel, and if n is large, the defective pixel is not detected when the edge component of the image is input to the defective pixel portion, as shown in FIG. 5B. Problem arises.

In order to identify the edge component and the defective pixel of this image, the level difference between the pixel signals of the adjacent pixels of the same color in another combination existing around the determination pixel is determined. In FIG. 2, A _k3 and A _m3 , A are provided around the determination pixel (B _m4 ).
_k5 and A _m5, the combination of the pixel signals, such as D _l2 and D _l4, D _n2 and D _n4 corresponds to it. The level difference determination in the combination of these respective pixels is also performed in the same manner as the level difference determination between the pixel signal B _m4 of the determination pixel and the pixel signals B _m2 , B _m6 , B _k4 and B _p4 of the adjacent pixels. .

Here, a case will be described as an example where it is determined in the first level determination that the level difference between the pixel signals B _m2 and B _m4 is not less than the threshold value V _th (comparison result of the comparator is “H” level). At this time, the pixel signal D _l2 and D _l4, if the level difference between the pixel signal D _n2 and D _n4 is not less than the threshold value V _th,
As shown in FIG. 6, since the pattern in which the luminance of the pixel signals B _m2 and B _m4 varies across the pixels, that is, the edge component of the image is input, the levels of the pixel signals B _m2 and B _m4 are input. It is highly possible that the difference is equal to or greater than the threshold value V _th, and the determination result in this case is set to the “L” level.

On the contrary, the pixel signals D _l2 and D _l4 and the pixel signal D
_{When the} level difference between _n2 and D _n4 is less than the threshold value V _th (the comparison result of the comparator is “L” level), the pixel of the pixel signal B _m4 is a defective pixel, and therefore the pixel signals B _m2 and B _m4. There is a high possibility that the level difference of 2 is equal to or greater than the threshold value V _th, and the determination result in this case is set to the “H” level. Similarly, in the first level determination, when the level difference between the pixel signals B _k4 and B _m4 is equal to or _larger than the threshold value V _th , the combination of the pixel signals A _k3 and A _m3 and the pixel signals A _k5 and A _m5 is _added again. H "level /
The "L" level is judged.

In the final defect determination, when there are n (n = 1 to 4) "H" levels, it is determined that the pixel is a defective pixel by the pixel signals B _m2 , B _m6 , B _{k4 of} adjacent pixels. This is similar to the case of determining the level difference from B _p4 . The processing procedure of the series of defect detection described above is shown in the flowchart of FIG. By sequentially applying this algorithm to all pixels, point defects in the screen are detected.

In this manner, the level difference between the pixel signals of the adjacent pixels of the same color of the defect determination pixel is detected and determined, and the level difference between the pixel signals of the adjacent pixels of the same color of different combinations existing around the pixel is detected. By detecting and determining
The defective pixel can be detected while avoiding erroneous detection due to the edge component of the image.

In the above embodiment, when the level difference determination data is n or more, the pixel is regarded as a defective pixel. However, which of n = 1 to 4 is selected depends on the defect detection accuracy. And is optional. In addition, it is possible to increase the number of adjacent pixels of the same color in different combinations in the periphery in order to improve the accuracy of identifying the defect and the edge component.

Further, in the above embodiment, a pixel (B _m4 )
4 adjacent pixels of the same color (B _k4 , B _m2 , B _m6 , B _p4 ) were used as the comparison target in order to determine as the defective pixel, the pixel for the pixel signal B _p4 was excluded from the comparison target.
It is also possible to make a comparison using three pixels. According to this, although the detection accuracy is deteriorated, the circuit scale can be almost halved.

The present invention also includes four types of A, B, C and D.
Not only the application to the color solid-state image pickup device that handles the color information of the species, but also the color solid-state image pickup device of other color coding, A,
It is also applicable to a monochrome solid-state imaging device in which B, C, and D all have the same color information. Further, to simplify the circuit, it is also within the scope of the present invention to select the number of adjacent pixels of the same color for level comparison.

[0027]

As described above, according to the present invention,
By detecting the defective pixel based on the level difference between the pixel signals of one pixel and the adjacent same-color pixels and the level difference between the pixel signals of the different-color same-pixels existing in the vicinity of the pixel signal, the defective pixel is detected. Since defect detection can be performed at any time based on image information obtained from the element, the following effects can be obtained. Since defect detection can be performed in a normal imaging state,
It is also possible to deal with electrostatic breakdown when the image sensor is incorporated in the video camera and defect changes due to aging after mounting in the video camera. Since initial defect information is not required, shipping sorting is not required, and time and cost including temperature control of the measuring device can be reduced. Since initial defect information is not required and defect correction is performed in real time, a non-volatile memory that stores and retains defect data is not required, which reduces the cost and size of the entire set and saves labor in logistics and manufacturing. It is possible to plan. Since there is no need to perform shipping sorting, there is no sorting leak, and it is possible to reliably correct defects that occur in the actual imaging state.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a spatial arrangement of pixel signals.

FIG. 3 is a waveform diagram of a signal output in a pixel signal array including four types of color information.

FIG. 4 is a waveform diagram of a signal output when a defective pixel is included.

FIG. 5 is a diagram (part 1) showing a relationship between a defective pixel and an edge image.

FIG. 6 is a diagram (part 2) showing the relationship between a defective pixel and an edge image.

FIG. 7 is a flowchart showing a defect detection processing procedure.

[Explanation of Codes] 3 CCD image sensor 4 Timing generation circuit 6 S / H (sample hold) circuit 8 Defect detection circuit 9 Defect correction circuit

Claims (3)

[Claims] 1. A defective pixel is detected based on a level difference between pixel signals of a pixel and a pixel of the same color adjacent to the pixel and a level difference between pixel signals of pixels of the same color of different types existing around the pixel. And a defect correction circuit that corrects a pixel signal for the defective pixel. 2. The defect detection circuit includes a first synchronization circuit that synchronizes a pixel signal of the one pixel with pixel signals of a pixel two lines before and a pixel two lines after the first synchronization circuit. A second synchronization circuit that synchronizes the pixel signal of the pixel with each pixel signal of the pixel two pixels before and two pixels after the pixel signal; and the pixel signal of each pixel signal synchronized by the first synchronization circuit. A first level difference detection circuit that obtains a level difference, a second level difference detection circuit that obtains a level difference between the pixel signals synchronized by the second synchronization circuit, and the first level difference detection circuit A first comparator that compares the level difference detected in step 1 with a predetermined threshold value; a second comparator that compares the level difference detected by the second level difference detection circuit with a predetermined threshold value; Defective pixel based on each comparison output of the second comparator The solid-state imaging device according to claim 1, further comprising a determination circuit for determining. 3. The defect correction circuit outputs a pixel signal of the defective pixel to a pixel signal of a pixel two lines before the defective pixel, a pixel signal of a pixel two pixels before the same line, and a pixel two pixels after the defective pixel. 2. The solid-state imaging device according to claim 1, wherein the solid-state imaging device replaces the average value of the pixel signal of