Evaluation system and method for high light performance of detector

Abstract

The invention discloses a detector highlight performance evaluation system and method, and relates to a detector highlight performance evaluation method, which solves the problem that when the incident light energy increases to the point where all detectors are saturated, it is impossible to detect whether the image is abnormal after further increasing the light intensity during the highlight performance detection process of the existing method. Therefore, the highlight performance of the detector is tested before the detector is used to evaluate the imaging effect on track, which can reduce potential losses and improve the observation efficiency on track. The present invention proposes to evaluate the highlight performance of the detector based on a directional light source and a roller with a black and white pattern, and use the roller to change the target moment by moment, so as to judge the boundary of the target light and dark partition, and avoid using an integrating sphere to saturate all pixels under a strong light target, thereby covering up the possible tailing, highlight overflow and other phenomena.

Description

Technical Field

[0001] The present invention relates to a method for evaluating the high light performance of a detector, and in particular to a system and method for evaluating the high light performance of a detector for aerospace applications. Background Art

[0002] Space remote sensing detectors for earth observation may encounter high-brightness reflective targets such as clouds, glass or metal roofs, which may cause excessive charges to accumulate in local areas and local time, resulting in abnormal images. It is possible that the detector cannot immediately return to normal after the high-brightness target disappears, affecting the observation efficiency of non-highlight targets. The traditional radiation calibration method based on the integrating sphere is used to detect the radiation performance of the detector. When the incident light energy increases to the saturation of all detectors, it is impossible to detect whether the image will still be abnormal after further increasing the light intensity. A new method is needed to detect the high-light performance of the detector. Summary of the invention

[0003] The present invention provides a detector high light performance evaluation system and method to solve the problem that when the incident light energy increases to the point that all detectors are saturated during high light performance detection by the existing method, it is impossible to detect whether the image is abnormal after further increasing the light intensity.

[0004] An evaluation system for the high light performance of a detector, the system comprising a lens, an imaging unit, a quick-view ground inspection device and a camera controller; and also comprising a roller and a directional light source group;

[0005] The optical axis of the lens matched with the imaging unit is perpendicular to the central axis of the drum, and the directional light source group is distributed on the side of the imaging unit, and the illuminated area covers the central axis position of the lens relative to the drum;

[0006] The adjustable light output by the directional light source group is irradiated onto a drum that can rotate in both directions. The target pattern on the drum is projected onto the imaging unit of the detector after passing through the lens. The camera controller sets the imaging parameters of the imaging unit and controls the start and end of the video recording via the 422 serial port. The image output by the imaging unit is transmitted to the quick-view ground inspection equipment for analysis and processing to determine the best focusing position and the best line frequency.

[0007] A method for evaluating the high light performance of a detector, which is implemented by the high light performance evaluation system of the detector; the evaluation method includes two steps: obtaining a clear image and performing a photoelectric performance test;

[0008] The specific steps of obtaining a clear image are:

[0009] Step 1: Determine the radius R of the roller according to the length L of the target pattern;

[0010]

[0011] Step 2: Determine the corresponding GSD value according to the highest line frequency, the lowest line frequency, the highest rotation speed and the lowest rotation speed of the drum of the detector, wherein the GSD is the resolution of the image on the drum;

[0012] Step 3: Based on the pixel size a of the detector, the GSD value obtained in step 2 and the distance L of the experimental site lenth , select the focal length f of the lens, and the distance D between the front end of the drum and the center of the lens;

[0013]

[0014] Step 4: Adjust the evaluation system to obtain a clear image; the specific adjustment process is:

[0015] Step 41. Adjust the relative position of the lens and the drum so that the optical axis of the lens passes through the center of the drum;

[0016] Step 42: placing a point light source at the intersection of the drum and the optical axis of the lens, and adjusting the focal length of the lens by moving it up and down to make the image point size generated by the point light source as small as possible;

[0017] Step 43: Place the detector at the position with the smallest image point, move the point light source up and down, and observe whether the generated image point is within the photosensitive surface of the detector to adjust the relative position of the test target pattern and the detector;

[0018] Step 44: Place the lens on a six-dimensional adjustment stand for fine focusing, and determine the optimal focal plane position by judging the number of transition pixels at the edge of the black-and-white strip scene in the same direction as the non-detector scanning direction; that is: δ lm ≤δ li , l i is each focus position, l m is the optimal focusing position, i is an integer, δ lm is the number of transition pixels at the edge of the black and white strip scene in the non-detector scanning direction at the best focal position, δ li is the number of transition pixels at the edge of the black and white strip scene in the non-detector scanning direction at each focus position;

[0019] Step 4 and 5: Adjust the scanning direction and line frequency of the detector to achieve a fine match between the line frequency of the detector and the rotation speed of the drum; determine the optimal line frequency value by judging the number of transition pixels at the edge of the black and white strip scene in the same scanning direction; that is: δ fm ≤δ fi , where f i is the frequency value of each line, f m is the optimal line frequency value; δ fiis the number of transition pixels at the edge of the black and white strip scene in the detector scanning direction at each focus position, δ fm is the number of transition pixels at the edge of the black and white strip scene in the detector scanning direction at the best focal position;

[0020] The process of conducting photoelectric performance testing is:

[0021] Step 1: Paste a white image with high reflectivity on the roller, and test the maximum full well charge number of the detector based on the photon transfer curve. When the incident light energy increases, the DN value of the detector increases, and the variance value of the image also increases accordingly. When the variance of the image reaches the critical point where it does not increase, the incident light energy A corresponds to the maximum full well of the detector.

[0022] Step 2: Use three images, black and white staggered horizontal stripes, white blocks in a completely black image, and black blocks in a completely white image, to perform imaging, and increase the incident light energy. When a black block or a black bar image begins to have brighter bright stripes or bright spots, that is, the uniformity index of the dark image deviates by more than 20% compared with the previous one, it is considered that highlight overflow occurs. At this time, the corresponding incident light energy is the critical value B of highlight overflow; the anti-halation ability is = B / A;

[0023] Step 3: Use black and white staggered horizontal stripes, white blocks in a completely black image, and black blocks in a completely white image to perform imaging, increase the incident light energy, and when the black and white boundary begins to blur, it is considered that tailing occurs. At this time, the corresponding incident light energy is the critical value C of highlight overflow; the anti-tailing ability is = C / A;

[0024] Step 4: Use three types of images, namely black and white alternating horizontal stripes, white blocks in a completely black image, and black blocks in a completely white image, to perform imaging, and increase the incident light energy. When the white block image has not reached saturation and bright lines or bright spots begin to appear in the black block image, it is considered that a reverse charge integral overflow phenomenon has occurred. At this time, the corresponding incident light energy is the critical value D of the reverse charge integral overflow; the anti-tailing capability is =D / A.

[0025] Beneficial effects of the present invention:

[0026] 1. The present invention evaluates the high light performance of the detector based on a directional light source and a roller with a black and white pattern. The roller is used to change the target moment by moment, so that the boundary of the target's bright and dark partitions can be judged, avoiding the situation where all pixels of the integrating sphere enter a saturated state under a strong light target, thus covering up the possible tailing, highlight overflow and other phenomena;

[0027] 2. The present invention can determine the limit threshold of charge transfer for an unblurred image by judging the clarity of the text image, and accordingly set reasonable imaging parameters to avoid excessive accumulated charge, which results in reduced transfer efficiency and blurred images.

[0028] 3. The present invention can identify the threshold of detector highlight overflow by judging the uniformity of the dark area image behind the bright block image, and set reasonable imaging parameters accordingly. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a principle block diagram of the evaluation system of the detector high light performance of the present invention;

[0030] Figure 2 Schematic diagram of drum imaging. DETAILED DESCRIPTION

[0031] Specific implementation method 1. Combination Figure 1 and Figure 2 To illustrate this embodiment, a detector high light performance evaluation system is as follows: Figure 1 As shown, it includes a drum, a directional light source group, a lens, an imaging unit, a quick-view ground detector and a camera controller. The adjustable light output by the directional light source group is irradiated onto the drum that can rotate in both directions. The target pattern on the drum is projected onto the detector of the imaging unit after passing through the lens. The camera controller sets the imaging parameters of the imaging unit and controls the start and end of the video recording through the 422 serial port. The image output by the imaging unit is sent to the quick-view ground detector for analysis and processing.

[0032] In this embodiment, the illumination light source of the drum is a directional light source; the optical axis of the lens matched with the imaging unit is perpendicular to the central axis of the drum; the directional light source group is distributed on the side of the imaging unit, and the area illuminated by it covers the central axis position of the lens relative to the drum;

[0033] Four patterns can be selected on the roller, including a highly reflective white image, alternating black and white stripes, white blocks in a completely black image, and black blocks in a completely white image.

[0034] Specific implementation method 2: Combination Figure 1 and Figure 2 The present embodiment describes a method for evaluating the high light performance of a detector, which is implemented by using the detector high light performance evaluation system described in Specific Implementation 1. The method is implemented by first adjusting the evaluation system to obtain a clear image, and then performing a photoelectric performance test.

[0035] The steps for adjusting the evaluation system are:

[0036] Step 1: Determine the radius R of the roller according to the length L of the target pattern;

[0037]

[0038] Step 2: Determine the corresponding GSD value according to the highest line frequency, the lowest line frequency of the detector, the highest speed and the lowest speed of the drum;

[0039] First, determine the line frequency range of the detector's vertical charge transfer based on the specific application, so that the line frequency variation range and the rotation speed adjustment range can match. When changing the line frequency during the experiment, only the rotation speed needs to be adjusted while maintaining a constant focal length.

[0040] Drum imaging requires that the linear velocity of the rolling rotation matches the transfer speed of the charge corresponding to the scene on the detector.

[0041] The linear velocity of the front end surface of the roller is: V=Rw; where R is the radius of the roller and w is the rotation speed of the roller.

[0042] Then the charge moving speed on the focal plane is: Where f is the focal length of the lens, and D is the distance between the front end surface of the drum and the center of the lens.

[0043] The line frequency of the corresponding detector is: Where a is the pixel size of the detector and GSD is the resolution of the image on the drum.

[0044] In this embodiment, the minimum line frequency f of the detector is 行频_min , the detector's maximum line frequency f 行频_max , the maximum speed of the drum w max , minimum drum speed w min The constraints are as follows:

[0045]

[0046] Step 3: Based on the calculated GSD value, the pixel size a of the detector, and the distance L of the experimental site lenth To select the focal length f of the lens and the distance D from the drum;

[0047]

[0048] Step 4: Adjust the evaluation system

[0049] (1) Adjust the relative position of the lens and the drum so that the optical axis of the lens passes through the center of the drum;

[0050] (2) Place a point light source at the intersection of the drum and the optical axis of the lens, move it up and down, and adjust the focal length of the lens to make the image point size generated by the point light source as small as possible;

[0051] (3) Place the detector at the position with the smallest image point, move the point light source up and down, and see whether the generated image point is within the photosensitive surface of the detector, so as to adjust the relative position of the test pattern and the detector;

[0052] (4) Place the lens on a six-dimensional adjustment frame for fine focusing, and determine the optimal focal position by judging the number of transition pixels at the edge of the black and white strip scene in the same direction as the non-scanning direction; lm ≤δ li ( i The value of is an integer, l i Represents each focus position, l m represents the best focus position)

[0053] (5) Adjust the scanning direction and line frequency of the detector to achieve a fine match between the line frequency of the detector and the speed of the drum. The optimal line frequency value is determined by judging the number of transition pixels at the edge of the black and white strip scene in the same scanning direction; fm ≤δ fi ( i The value of f is an integer. i Represents the frequency value of each line, f m represents the best line frequency value)

[0054] The judgment is further assisted by judging the deviation of the number of pixels occupied by the circular target in the scanning direction and the direction perpendicular to the scanning direction. The value with the smallest difference in the number of pixels occupied in the two directions is the best line frequency value. fm ≤β fi ( i The value of f is an integer. i Represents the frequency value of each line, f m represents the best line frequency value)

[0055] The process of conducting photoelectric performance testing is:

[0056] First, a white image with high reflectivity is pasted on the drum, and the maximum full well charge number of the detector is tested based on the photon transfer curve: as the incident light energy increases, the DN value of the detector increases, and the variance value of the image also increases accordingly; the incident light energy A at the critical point where the variance of the image begins to stop increasing corresponds to the maximum full well of the detector.

[0057] Three types of images, black and white alternating horizontal stripes, white blocks in a completely black image, and black blocks in a completely white image, are used for imaging. The incident light energy is increased. When brighter bright stripes or bright spots begin to appear in the black block or black bar image, that is, the uniformity index of the dark image deviates by more than 20% compared with the previous one, it is considered that blooming has occurred. At this time, the corresponding incident light energy is the critical value B of blooming; the anti-halo ability is = B / A.

[0058] Three types of images, namely black and white alternating horizontal stripes, white blocks in a completely black image, and black blocks in a completely white image, are used for imaging. The incident light energy is increased. When the black and white boundary begins to blur (for one pixel at the black and white pixel boundary), it is considered that tailing has occurred. At this time, the corresponding incident light energy is the critical value C of blooming; the anti-tailing ability is = C / A.

[0059] Three images, black and white staggered horizontal stripes, white blocks in all-black images, and black blocks in all-white images, are used for imaging. The incident light energy is increased. When the white block image has not reached saturation and bright lines or bright spots begin to appear in the black block image, it is considered that the reverse charge integral overflow phenomenon has occurred. At this time, the corresponding incident light energy is the critical value D of the reverse charge integral overflow; the anti-tailing ability is = D / A

[0060] In this embodiment, the difference between reverse charge integral overflow and blooming is that the former occurs when the image in the bright area is not saturated, and the latter occurs after the image in the bright area is saturated. Tailing and blooming, the former may occur when the image in the bright area is not saturated; the latter occurs after the image in the bright area is saturated. The difference between reverse charge integral overflow and tailing is that the latter is caused by incomplete charge transfer, showing that the residual DN value in the dark area corresponds to the distribution of the bright area; the former is relatively random, and is usually manifested as bright lines or bright spots appearing in some areas, but not in other areas.

[0061] In this embodiment, the directional light source group adopts an LED array; the roller uses a customized bidirectional adjustable speed roller; the lens adopts a shelf product; the imaging unit adopts an imaging unit based on a linear array TDICMOS detector; the camera controller adopts a laptop computer + moxa box; the quick-view ground inspection equipment adopts a customized product of Wuhan Zhuomu Company.

Claims

1. A system for evaluating the high light performance of a detector, the system comprising a lens, an imaging unit, a quick-view ground inspection device and a camera controller; the system is characterized by: Also included are a drum and a directional light source set; The optical axis of the lens matched with the imaging unit is perpendicular to the central axis of the drum, and the directional light source group is distributed on the side of the imaging unit, and the illuminated area covers the central axis position of the lens relative to the drum; The adjustable light output by the directional light source group is irradiated onto a drum that can rotate in both directions. The target pattern on the drum is projected onto the imaging unit of the detector after passing through the lens. The camera controller sets the imaging parameters of the imaging unit and controls the start and end of the video recording via the 422 serial port. The image output by the imaging unit is transmitted to the quick-view ground inspection equipment for analysis and processing to determine the best focusing position and the best line frequency.

2. The detector high light performance evaluation system according to claim 1, characterized in that: Four patterns are selected on the roller, namely, a white image with high reflectivity, black and white alternating horizontal stripes, a white block in a completely black image, or a black block in a completely white image.

3. A method for evaluating the high light performance of a detector, characterized by: The evaluation method is implemented by the evaluation system for the high light performance of the detector according to claim 1 or 2; the evaluation method has two steps of obtaining a clear image and performing a photoelectric performance test; the specific steps of obtaining a clear image are: Step 1: Determine the radius R of the roller according to the length L of the target pattern; Step 2: Determine the corresponding GSD value according to the highest line frequency, the lowest line frequency, the highest rotation speed and the lowest rotation speed of the drum of the detector, wherein the GSD is the resolution of the image on the drum; Step 3: Based on the pixel size a of the detector, the GSD value obtained in step 2 and the distance L of the experimental site lenth , select the focal length f of the lens, and the distance D between the front end of the drum and the center of the lens; Step 4: Adjust the evaluation system to obtain a clear image; the specific adjustment process is: Step 41. Adjust the relative position of the lens and the drum so that the optical axis of the lens passes through the center of the drum; Step 42: placing a point light source at the intersection of the drum and the optical axis of the lens, and adjusting the focal length of the lens by moving it up and down to make the image point size generated by the point light source as small as possible; Step 43: Place the detector at the position with the smallest image point, move the point light source up and down, and observe whether the generated image point is within the photosensitive surface of the detector to adjust the relative position of the test target pattern and the detector; Step 44: Place the lens on a six-dimensional adjustment stand for fine focusing, and determine the optimal focal plane position by judging the number of transition pixels at the edge of the black-and-white strip scene in the same direction as the non-detector scanning direction; that is: δ lm ≤δ li , l i is each focus position, l m is the optimal focusing position, i is an integer, δ lm is the number of transition pixels at the edge of the black and white strip scene in the non-detector scanning direction at the best focal position, δ li is the number of transition pixels at the edge of the black and white strip scene in the non-detector scanning direction at each focus position; Step 4 and 5: Adjust the scanning direction and line frequency of the detector to achieve a fine match between the line frequency of the detector and the rotation speed of the drum; determine the optimal line frequency value by judging the number of transition pixels at the edge of the black and white strip scene in the same scanning direction; that is: δ fm ≤δ fi , where f i is the frequency value of each line, f m is the optimal line frequency value; δ fi is the number of transition pixels at the edge of the black and white strip scene in the detector scanning direction at each focus position, δ fm is the number of transition pixels at the edge of the black and white strip scene in the detector scanning direction at the best focal position; The process of conducting photoelectric performance testing is: Step 1: Paste a white image with high reflectivity on the roller, and test the maximum full well charge number of the detector based on the photon transfer curve. When the incident light energy increases, the DN value of the detector increases, and the variance value of the image also increases accordingly. When the variance of the image reaches the critical point where it does not increase, the incident light energy A corresponds to the maximum full well of the detector. Step 2: Use three images, black and white staggered horizontal stripes, white blocks in a completely black image, and black blocks in a completely white image, to perform imaging, and increase the incident light energy. When a black block or a black bar image begins to have brighter bright stripes or bright spots, that is, the uniformity index of the dark image deviates by more than 20% compared with the previous one, it is considered that highlight overflow occurs. At this time, the corresponding incident light energy is the critical value B of highlight overflow; the anti-halation ability is = B / A; Step 3: Use black and white staggered horizontal stripes, white blocks in a completely black image, and black blocks in a completely white image to perform imaging, increase the incident light energy, and when the black and white boundary begins to blur, it is considered that tailing occurs. At this time, the corresponding incident light energy is the critical value C of highlight overflow; the anti-tailing ability is = C / A; Step 4: Use three types of images, namely black and white alternating horizontal stripes, white blocks in a completely black image, and black blocks in a completely white image, to perform imaging, and increase the incident light energy. When the white block image has not reached saturation and bright lines or bright spots begin to appear in the black block image, it is considered that a reverse charge integral overflow phenomenon has occurred. At this time, the corresponding incident light energy is the critical value D of the reverse charge integral overflow; the anti-tailing capability is =D / A.

4. The method for evaluating the high light performance of a detector according to claim 3, characterized in that: In step 2, the specific process of determining the resolution GSD value of the corresponding image on the drum is: Firstly, the line frequency range of the detector's vertical charge transfer is determined so that the range of the line frequency changes matches the range of the drum's rotation speed. In the process of changing the line frequency range, only the drum's rotation speed needs to be adjusted to maintain a constant focal length. Secondly, the roller imaging requirement is that the linear velocity of the rolling rotation matches the transfer velocity of the charge corresponding to the scene on the detector; Then, set the linear velocity of the front end of the roller V = Rw, w is the rotation speed of the roller; then the charge moving speed V of the focal plane 焦面 The formula is: In the formula, f is the focal length of the lens, and D is the distance between the front end of the drum and the center of the lens; Finally, the line frequency f of the corresponding detector 行频 for:

5. The method for evaluating the high light performance of a detector according to claim 3, characterized in that: The minimum line frequency f of the detector 行频_min , the detector's maximum line frequency f 行频_max , the maximum speed of the drum w max , minimum drum speed w min The constraint relationship is:

6. The method for evaluating the high light performance of a detector according to claim 3, characterized in that: In step 4 and 5, further auxiliary judgment is made by judging the deviation of the number of pixels occupied by the circular target in the scanning direction and the direction perpendicular to the scanning direction, and comparing the deviation values ​​of the number of pixels occupied in the two directions. When the deviation value reaches the minimum value, it corresponds to the optimal line frequency value; that is: β fm ≤β fi , where f i is the frequency value of each line, f m is the optimal line frequency value; β fi is the deviation of the number of pixels occupied in both directions at each line frequency, β fm It is the deviation value of the number of pixels occupied in two directions at the optimal line frequency.

Images