A method for on-orbit maintenance of CMOS image sensor single event latch-up combining image edge detection and current monitoring

Abstract

The present application relates to a kind of CMOS image sensor single particle latch in orbit maintenance method of image edge detection and current monitoring combination, the method is that median filter algorithm is used to the image that CMOS image sensor is in orbit real-time acquisition is carried out smooth pretreatment, the gradient intensity and direction of each pixel point of CMOS image sensor are calculated, the gradient intensity of each pixel point is compared with the gradient intensity of adjacent two pixel points along the gradient direction, obtain the image of real edge response, apply double threshold detection to determine the real edge, potential edge and non-edge of pixel point in real edge response image.For potential edge pixel point, as center, whether there is one or more pixel points marked as real edge in its surrounding is detected, finally real edge is morphologically dilated, while CMOS image sensor analog signal circuit module current exceeds 1.2 times normal current value, then it is judged that CMOS image sensor occurs present single particle latch, remote control power off restart.The present application judges whether satellite CMOS detector occurs single particle latch event in orbit, and accuracy is high, and safety is high.

Description

Technical Field

[0001] This invention relates to the field of aerospace technology, specifically to an on-orbit maintenance method for single-event latch of a CMOS image sensor that combines image edge detection and current monitoring. Background Technology

[0002] Single-event latch-up (SETL) refers to the phenomenon where a single particle triggers a parasitic PNPN structure in a CMOS image sensor circuit, causing a large current. SETL leads to an increase in the operating current of the CMOS image sensor, abnormal image acquisition, and device malfunction. On-orbit protection against SETL primarily relies on power supply current limiting. However, if the current threshold is not set appropriately, it can lead to SETL, micro-latch-up, or even burnout of the device. In micro-latch-up, the current increase is not obvious; the current monitoring may not have reached the threshold, but the device is actually latched up. If a power-off restart operation is not performed based on the current threshold, it will seriously affect the safe and reliable operation of the payload in orbit.

[0003] This invention employs a combination of image processing and current monitoring. When single-event latch-up (SUP) image features are detected at the image edge, the current value of the analog signal circuit module of the CMOS image sensor is considered. If the current value exceeds 1.2 times the normal current value, a SUP or micro-latch-up event is determined, and power is immediately cut off and restarted. Compared with the single current threshold method used in previous loads, this invention has higher accuracy, can achieve immunity to SUP, and avoids SUP burn-out, thus ensuring high safety. Summary of the Invention

[0004] The purpose of this invention is to achieve single-event latch-up immunity in orbit by providing an on-orbit maintenance method for single-event latch-up of CMOS image sensors that combines image edge detection and current monitoring. This method first uses a self-adopted median filtering algorithm to smooth and preprocess images acquired in real-time by the CMOS image sensor in orbit. Then, it calculates the gradient intensity and direction of each pixel in the CMOS image sensor and compares the gradient intensity of each pixel with the gradient intensity of two adjacent pixels along the gradient direction to obtain the true edge response image. Dual threshold detection is applied to determine the true edges, potential edges, and non-edges of pixels in the true edge response image. For potential edge pixels, it detects whether there are one or more pixels marked as true edges around the pixel. If so, the potential edge pixel is marked as a true edge. Finally, morphological dilation is performed on the true edges. If a semicircle, small semicircle, or large semicircle appears in the dilated true edge, and the current of the CMOS image sensor's analog signal circuit module exceeds 1.2 times the normal current value, then a single-event latch-up has occurred in the CMOS image sensor, and the power supply is remotely controlled to be shut down and restarted. This invention can determine whether a single-event latch-up event has occurred in a satellite's CMOS detector in orbit with high accuracy and high safety.

[0005] The on-orbit maintenance method for single-event latch of a CMOS image sensor, which combines image edge detection and current monitoring, as described in this invention, is performed according to the following steps:

[0006] a. The median filtering algorithm is used to perform smoothing preprocessing on the images acquired in real time by the CMOS image sensor in orbit;

[0007] b. Calculate the gradient values ​​in the horizontal and vertical directions for each pixel in the smoothed image from step a;

[0008] c. Using the gradient values ​​in the horizontal and vertical directions calculated in step b, calculate the gradient intensity and direction of each pixel in the CMOS image sensor;

[0009] d. For each pixel, compare its gradient strength with the gradient strength of the two adjacent pixels along the gradient direction. If the gradient strength of the pixel is not the maximum value of the two adjacent pixels, set its gradient strength to 0; otherwise, retain the original value, thus obtaining the image of the true edge response.

[0010] e. Apply dual threshold detection to determine the true edges, potential edges, and non-edges of pixels in the true edge response image;

[0011] f. Using the potential edge pixel obtained in step e as the center, detect whether there are one or more pixels marked as real edges around it. If there are real edge pixels in the surrounding 8 neighborhood pixels, then mark the potential edge pixel as a real edge.

[0012] g. Perform morphological dilation on the real edges in steps e and f. If a semicircle, small semicircle or large semicircle appears in the dilated real edge, and the current of the analog signal circuit module of the CMOS image sensor exceeds 1.2 times the normal current value, then it is determined that a single-event latch-up has occurred in the CMOS image sensor.

[0013] j. Remotely control power outage and restart.

[0014] The present invention discloses an on-orbit maintenance method for single-event latch of a CMOS image sensor that combines image edge detection and current monitoring, wherein:

[0015] a. The median filtering algorithm is used to perform smoothing preprocessing on the images acquired in real time by the CMOS image sensor in orbit;

[0016] M(i,j)=median{f(x,y)|(x,y)∈Ω(i,j)} (1)

[0017] Where M(i,j) represents the pixel value at coordinates (i,j) in the output image, f(x,y) represents the pixel value at coordinates (x,y) in the input image, and Ω(i,j) represents the window region centered at coordinates (i,j). The pixel values ​​within the window region are quickly sorted, and the median value is selected as the output value.

[0018] b. Calculate the gradient values ​​in the horizontal and vertical directions for each pixel in the smoothed image after step a using formulas (2) and (3);

[0019] G x =P(x+1, y)-P(x-1, y) (2)

[0020] G y =P(x, y+1)-P(x, y-1) (3)

[0021] P(x, y) represents the pixel value at coordinates (x, y) in the image;

[0022] c. Using the gradient values ​​in the horizontal and vertical directions calculated in step b, calculate the gradient intensity and direction of each pixel of the CMOS image sensor according to formulas (4) and (5);

[0023]

[0024] gradient direction=arctan(grady / gradx) (5)

[0025] d. For each pixel, compare its gradient strength with the gradient strength of the two adjacent pixels along the gradient direction. If the gradient strength of the pixel is not the maximum value of the two adjacent pixels, set its gradient strength to 0; otherwise, keep the original value. This will give you the image of the true edge response.

[0026] e. Apply dual threshold detection to determine the true edges, potential edges, and non-edges of pixels in the true edge response image;

[0027] f. For the potential edge pixel obtained in step e, take it as the center and detect whether there are one or more pixels marked as real edges around it. If there are real edge pixels in the surrounding 8 neighborhood pixels, then mark the potential edge pixel as a real edge.

[0028] g. Perform morphological dilation on the real edges in steps e and f. If a semicircle, small semicircle or large semicircle appears in the dilated real edge, and the current of the analog signal circuit module of the CMOS image sensor exceeds 1.2 times the normal current value, then it is determined that a single-event latch-up has occurred in the CMOS image sensor.

[0029] D(x, y)=max{I(x+i, y+j)(i, j)∈B} (6)

[0030] Where D(x, y) represents the pixel value at coordinates (x, y) in the output image, and I(x+i, y+j) represents the pixel value in the input image corresponding to structuring element B.

[0031] j. Remotely control power outage and restart.

[0032] This invention discloses an on-orbit maintenance method for single-event latch-up (SEL) events in CMOS image sensors that combines image edge detection and current monitoring. The method first uses a self-adopted median filtering algorithm to smooth and preprocess images acquired in real-time by the CMOS image sensor in orbit. Then, it calculates the gradient intensity and direction of each pixel in the CMOS image sensor and compares the gradient intensity of each pixel with the gradient intensity of two adjacent pixels along the gradient direction to obtain the true edge response image. Dual threshold detection is applied to determine the true edges, potential edges, and non-edges of pixels in the true edge response image. For potential edge pixels, the method detects whether there are one or more pixels marked as true edges around the pixel. If so, the potential edge pixel is marked as a true edge. Finally, morphological dilation is performed on the true edges. If a semicircle, small semicircle, or large semicircle appears in the dilated true edge, and the current of the CMOS image sensor's analog signal circuit module exceeds 1.2 times the normal current value, then a SEL event is determined to have occurred in the CMOS image sensor, and the power supply is remotely controlled to be cut off and restarted. This invention can determine whether a single-event latch-up event has occurred in a satellite's CMOS detector in orbit with high accuracy and high safety.

[0033] The present invention provides an on-orbit maintenance method for single-event latch-up (SUP) events in CMOS image sensors that combines image edge detection and current monitoring. This method is applicable to maintenance of any type of CMOS image sensor in an imaging system that experiences an SUP event during on-orbit operation.

[0034] Therefore, this invention is applicable to space payload units that need to perform on-orbit maintenance of CMOS image sensors in the event of a single-event latch-up. Attached Figure Description

[0035] Figure 1 This is the original image collected for this invention;

[0036] Figure 2 This is the original image of the present invention after median filtering.

[0037] Figure 3 This is the image after edge detection following median filtering in this invention;

[0038] Figure 4 This is an image showing the detection of single-event latches after image processing according to the present invention. Detailed Implementation

[0039] The present invention will now be described in further detail with reference to the accompanying drawings.

[0040] Example

[0041] The present invention discloses an on-orbit maintenance method for a CMOS image sensor using single-event latch that combines image edge detection and current monitoring. The method is specifically operated according to the following steps:

[0042] a. Select an original image with a resolution of 2048×2048 (e.g., Figure 1 The median filtering algorithm is used, with a filter kernel of size 45×45, to perform smoothing preprocessing on the images acquired in real time by the CMOS image sensor in orbit;

[0043] M(i,j)=median{f(x,y)|(x,k)∈Ω(i,j)} (1)

[0044] Where M(i,j) represents the pixel value at coordinates (i,j) in the output image, f(x,y) represents the pixel value at coordinates (x,y) in the input image, and Ω(i,j) represents the window region centered at coordinates (i,j). The pixel values ​​within the 45×45 window region are quickly sorted, and the median value is selected as the output value. Median filtering is then performed by traversing 2048×2048 pixels to obtain the filtered image, as shown below. Figure 2 As shown;

[0045] b. Calculate the gradient values ​​in the horizontal and vertical directions for each pixel in the smoothed image after step a using formulas (2) and (3);

[0046] G x =P(x+1, y)-P(x-1, y) (2)

[0047] G y =P(x,y+1)-P(x,y-1)) (3)

[0048] P(x, y) represents the pixel value at coordinates (x, y) in the image;

[0049] c. Using the gradient values ​​calculated in step b along the horizontal and vertical 2048 pixel directions, calculate the gradient intensity and direction of each pixel of the CMOS image sensor according to formulas (4) and (5);

[0050]

[0051] gradient direction=arc tan(grady / gradx) (5)

[0052] d. For each pixel, compare its gradient strength with the gradient strength of the two adjacent pixels along the gradient direction. If the gradient strength of the pixel is not the maximum value of the two adjacent pixels, set its gradient strength to 0; otherwise, keep the original value. This will give you the image of the true edge response.

[0053] e. Apply dual threshold detection, setting threshold 1 to 20 and threshold 2 to 80 to determine the true edges, potential edges, and non-edges of pixels in the true edge response image;

[0054] f. For the potential edge pixel obtained in step e, using it as the center, detect whether there are one or more pixels marked as real edges around it. If there are real edge pixels within 8 neighboring pixels, then mark the potential edge pixel as a real edge. The result is as follows: Figure 3 As shown;

[0055] g. Morphologically dilate the true edges from steps e and f, setting the dilation kernel shape to a cross shape and a size of 7×7. Small semicircles appear in the dilated true edges. Simultaneously, if the current in the CMOS image sensor analog signal circuit module exceeds 1.2 times the normal current value, then a single-event latch-up occurs in the CMOS image sensor. The result is as follows: Figure 4 As shown;

[0056] D(x, y)=max{I(x+i, y+j)|(i, j)∈B} (6)

[0057] Where D(x, y) represents the pixel value at coordinates (x, y) in the output image, and I(x+i, y+j) represents the pixel value in the input image corresponding to structuring element B.

[0058] j. Remotely control power outage and restart.

[0059] The above description is merely a specific implementation of a single-event latch-up on-orbit maintenance method for a CMOS image sensor that combines image edge detection and current monitoring, provided by the present invention. However, the scope of protection of the present invention is not limited thereto. Any substitutions or additions that can be understood by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of the present invention.

Claims

1. A single-event latch-up on-orbit maintenance method for CMOS image sensors combining image edge detection and current monitoring, characterized in that... The specific steps are as follows: a. The median filtering algorithm is used to perform smoothing preprocessing on the images acquired in real time by the CMOS image sensor in orbit; b. Calculate the gradient values ​​in the horizontal and vertical directions for each pixel in the smoothed image from step a; c. Using the gradient values ​​in the horizontal and vertical directions calculated in step b, calculate the gradient intensity and direction of each pixel in the CMOS image sensor; d. For each pixel, compare its gradient strength with the gradient strength of the two adjacent pixels along the gradient direction. If the gradient strength of the pixel is not the maximum value of the two adjacent pixels, set its gradient strength to 0; otherwise, retain the original value, thus obtaining the image of the true edge response. e. Apply dual threshold detection to determine the true edges, potential edges, and non-edges of pixels in the true edge response image; f. Using the potential edge pixel obtained in step e as the center, detect whether there are one or more pixels marked as real edges around it. If there are real edge pixels in the surrounding 8 neighborhood pixels, then mark the potential edge pixel as a real edge. g. Perform morphological dilation on the real edges in steps e and f. If a semicircle, small semicircle or large semicircle appears in the dilated real edge, and the current of the analog signal circuit module of the CMOS image sensor exceeds 1.2 times the normal current value, then it is determined that a single-event latch-up has occurred in the CMOS image sensor. j. Remotely control power outage and restart.

Images