An Automatic Dimming Method Applicable to Manual Rendezvous and Docking

Abstract

The present invention discloses an automatic dimming method suitable for manual rendezvous and docking. This method divides an image into two parts, a central area and a peripheral area, and then separately counts the effective pixels in the central area and the peripheral area. The pixel values in the central area are accumulated and summed using a threshold judgment and weighting method to obtain sum1 and sum2; the pixel values in the peripheral area are accumulated and summed using a threshold judgment method to obtain sum3. The actual pixel average grayscale value Rav of the significant target in the image can be obtained by the accumulated sum of the central area and the peripheral area. Finally, according to the ratio between the actual pixel average grayscale value Rav and the expected grayscale value Eav, the corresponding dimming parameter, that is, the appropriate exposure time EXP, is obtained. i , digital gain ADC i With analog gain PGA i This method is an automatic dimming technology for the specific scenario of manual rendezvous and docking. It can ensure clear imaging of the target during the rendezvous and docking process and ensure the successful completion of the docking mission.

Description

Technical Field

[0001] The present invention belongs to the technical field of space rendezvous and docking, and particularly relates to an automatic dimming method applicable to manual rendezvous and docking. Background Art

[0002] During the in-orbit manual docking process of a spaceship, since there is no atmospheric scattering in outer space, the background is often completely dark. When the camera faces the sun directly, strong sunlight may directly enter the lens; when it is facing away from the sun, the sunlight may illuminate the entire hull, resulting in a large area of high-brightness regions; when far away from the spaceship, the target appears very small, but it is the most important element in the dark background; when approaching the spaceship, the target becomes very large and can even fill most of the image. For these actual situations, the classic dimming method based on the overall average value of the image often cannot achieve good results: when facing the sun, the strong sunlight will make the average value of the image increase, and the algorithm will constantly try to reduce the brightness of the image. In this way, the target information will be submerged in the background; when the spaceship is far away, the target is very small, the background is very dark, and the average value of the image is very small. The algorithm will try to increase the brightness of the image, making the target too bright to be recognized.

[0003] During the spacecraft rendezvous and docking process, it is one of the most important dimming means for astronauts to use a camera based on visual imaging for manual rendezvous and docking.

[0004] However, during the actual rendezvous and docking process, due to the influence of sunlight, large-area bright spots may appear in the field of view. In this case, it is difficult for astronauts to clearly see the specific position of the target, thus bringing great difficulties to manual rendezvous and docking. To improve the impact of this problem, the research on automatic dimming methods is imperative.

[0005] Currently, the automatic dimming methods are mainly based on ideas such as neural networks and image histograms. The method based on neural networks has certain requirements for the processor and memory because a large number of samples are needed for training, so it is not suitable for in-orbit applications. And the method based on image histograms is mainly divided into two types. One is to determine the brightness weighted mean by analyzing the gray histogram; the other is to adjust the exposure by calculating the average gray value of the image.

[0006] However, during the manual rendezvous and docking process, due to the influence of sunlight, sometimes the situation of a bright vehicle and a dark background occurs; sometimes the situation of a bright background and a dark vehicle occurs. Therefore, the traditional automatic dimming methods cannot well handle the manual rendezvous and docking scenario. Summary of the Invention

[0007] The technical problem addressed by this invention is that, given the inability of traditional dimming methods to adapt to the complex scenarios encountered during manual rendezvous and docking, an on-orbit automatic dimming method with strong adaptability and high real-time performance is proposed. This method divides an image into a central region and a peripheral region, performs weighted statistics on the valid pixels in each region, and obtains the dimming parameter ratio coefficients for adjacent frames, ultimately determining the corresponding dimming parameters.

[0008] The technical solution adopted by the present invention is: an automatic dimming method suitable for manual rendezvous and docking, which specifically includes the following steps:

[0009] Step 1: The input image is a grayscale image I;

[0010] Step 2: Divide the input grayscale image I into regions, set the central region, and the image of the central region is M, the first column image of the central region is F, satisfying At the same time, define the area N = IM, C = MF;

[0011] Step 3: Count the number of valid pixels in each area: For the pixels in area N that are greater than the gray threshold Gray L And less than the gray threshold upper limit Gray H ; for pixels in region C that are greater than the grayscale threshold GrayL, a 2-fold weighted statistics is performed; for all pixels in region F, a 2-fold weighted statistics is performed;

[0012] Step 4: Perform comprehensive statistics on the valid pixels in regions N, C, and F that meet the requirements of step 3 to obtain the cumulative grayscale sum of valid pixels and the number of valid pixels num;

[0013] Step 5: Calculate the average grayscale value Rav of the valid pixels in the image:

[0014]

[0015] Step 6: According to the effective pixel average grayscale value Rav and the expected grayscale value Eav, the dimming parameter ratio coefficient ratio is obtained:

[0016]

[0017] Step 7: Calculate the ratio of the nearest 8 frames i , where i = 0, 1, 2, 3, 4, 5, 6, 7; thus the corrected comprehensive dimming parameter ratio coefficient ratio′ is obtained:

[0018]

[0019] Step 8: Calculate the total gain value G of the previous frame imagei-1 :

[0020] G i-1 =ADC i-1 ×PGA i-1 ×EXP i-1 ;

[0021] Among them, ADC i-1 、PGA i-1 、EXP i-1 are the digital gain, analog gain, and exposure time of the previous frame of image respectively;

[0022] According to the comprehensive dimming parameter ratio coefficient ratio′, the total gain value G required to be set for the current frame image is obtained. i :

[0023] G i =G i-1 ×ratio′;

[0024] Step 9: According to the total gain value G of the current frame image i , calculate the dimming parameters required for the current frame, that is, the exposure time EXP i , analog gain PGA i , digital gain ADC i ; Among them, the priority adjustment is the exposure time EXP i :

[0025]

[0026] Among them, EXP min is the minimum exposure time, EXP max is the maximum exposure time;

[0027] ADC def The default initial digital gain, PGA def The default initial analog gain; when automatic dimming is enabled, the ADC def =ADC i-1 , PGA def =PGA i-1 .

[0028] Furthermore, the size of the central area image is 256×256.

[0029] Furthermore, the gray threshold lower limit GrayL=16, the gray threshold upper limit Gray H =1000.

[0030] Furthermore, the expected grayscale value Eav=144.

[0031] Further, the default initial digital gain of the ADC def = 44, and the default initial analog gain of the PGA def = 1.

[0032] Further, the minimum exposure time EXP min = 1510 and the maximum exposure time EXP max = 201500.

[0033] Further, in step 9, if the adjusted exposure time EXP i is within the range between the maximum exposure time EXP max and less than the minimum exposure time EXP min , then no further adjustment is made to the current frame; if the adjusted exposure time EXP i has exceeded the maximum exposure time EXP max or is less than the minimum exposure time EXP min , then the digital gain of the ADC i and the analog gain of the PGA i need to be further adjusted.

[0034] Further, when the adjusted exposure time is greater than the maximum exposure time EXP max , the digital gain of the ADC i needs to be adjusted: At this time, if it satisfies that the analog gain of the PGA i-1 = 0 and the digital gain of the ADC i-1 < 46, or satisfies that the PGA i-1 = 1 and the digital gain of the ADC i-1 < 61, then the digital gain of the current frame image ADC i = ADC i-1 + 1; if these two conditions are not met, then the analog gain value of the PGA i = PGA i-1 + 1.

[0035] Further, when the adjusted exposure time is less than the minimum exposure time EXP min , the analog gain of the PGA i needs to be adjusted: At this time, if it satisfies that the PGA i-1 > 0 and the digital gain of the ADC i-1 = 46, or satisfies that the PGA i-1 > 1 and the digital gain of the ADC i-1 = 61, then the analog gain of the current frame image PGA i = PGA i-1 - 1; if these two conditions are not met, then the digital gain of the ADC i = ADC i-1 - 1.

[0036] Compared with the prior art, the significant advantages of the present invention are as follows: (1) First, it has higher applicability to the manual rendezvous and docking scenario. The image is divided into a central region and a peripheral region, and different weight values are given according to the threshold judgment, which can effectively adjust the light for the complex environment during the rendezvous and docking process; (2) High real-time performance. Mainly, the gray values are statistically analyzed, and then the corresponding light adjustment parameters are calculated according to the proportional relationship, with low calculation complexity. Description of the Drawings

[0037] Figure 1 is a schematic diagram of the work flow;

[0038] Figure 2 is a diagram of image region segmentation;

[0039] Figure 3 is a schematic diagram before light adjustment;

[0040] Figure 4 is a schematic diagram after light adjustment. Detailed Embodiment

[0041] The present invention will be further described below in conjunction with the drawings and the detailed embodiment.

[0042] Combined with Figure 1 , the automatic light adjustment method of the present invention applicable to manual rendezvous and docking is as follows:

[0043] Step 1: The input image is a grayscale image I, the current frame image number is i, and the previous frame image number is i - 1;

[0044] Step 2: As Figure 2 shown, the input grayscale image I is divided into regions. The image of the central region 256×256 is M, the first column image of the central region is F, and it satisfies Therefore, the regions N = I - M and C = M - F can be defined;

[0045] Step 3: Count the number of valid pixels in each region. For the pixels in region N that are greater than the lower gray threshold Gray L and less than the upper gray threshold Gray H , the statistics are performed to obtain the sum of pixel gray values as sum1 and the number of valid pixels as num1; for the pixels in region C that are greater than Gray L , 2-fold weighted statistics are performed to obtain the sum of pixel gray values as sum2 and the number of valid pixels as num2; for all pixels in region F, 2-fold weighted statistics are performed to obtain the sum of pixel gray values as sum3 and the number of valid pixels as num3;

[0046] Step 4: Conduct a comprehensive statistics on the valid pixels in regions N, C, and F that meet the requirements of Step 3 to obtain the cumulative sum sum of the valid pixels and the number num of the valid pixels;

[0047] where sum = sum1 + sum2 + sum3;

[0048] num = num1 + num2 + num3;

[0049] Step 5: Calculate the average gray value Rav of the valid pixels in the calculated image:

[0050]

[0051] Step 6: Obtain the dimming parameter ratio coefficient ratio based on the average gray value Rav of the valid pixels and the expected gray value Eav:

[0052]

[0053] Step 7: Calculate the ratio coefficient ratio obtained from the nearest 8 frames of images i , where i = 0, 1, 2, 3, 4, 5, 6, 7. The corrected comprehensive dimming parameter ratio coefficient ratio′ can be obtained:

[0054]

[0055] Step 8: Calculate the total gain value G of the previous frame of image i-1 :

[0056] G i-1 = ADC i-1 ×PGA i-1 ×EXP i-1 ;

[0057] According to the comprehensive dimming parameter ratio coefficient ratio′, the total gain value G that needs to be set for the current frame of image can be obtained i :

[0058] G i = G i-1 ×ratio′;

[0059] Step 9: Calculate the dimming parameters that need to be configured for the current frame based on the total gain value G i , that is, the exposure time EXP i , the analog gain PGA i , and the digital gain ADC i .

[0060] The exposure time EXP is preferentially configured i :

[0061]

[0062] When the exposure time has reached the maximum exposure time EXP max , the exposure time is maintained unchanged, and the digital gain ADC is further adjusted i . At this time, if the analog gain PGA i-1 = 0 and the digital gain ADC i-1 < 46, or if PGA i-1 = 1 and the digital gain ADC i-1 < 61, then the digital gain ADC of the current frame image i = ADC i-1 + 1; if these two conditions are not met, the analog gain value PGA is adjusted i = PGA i-1 + 1.

[0063] When the exposure time has reached the minimum exposure time EXP min , the analog gain PGA needs to be adjusted i . At this time, if PGA i-1 > 0 and the digital gain ADC i-1 = 46, or if PGA i-1 > 1 and the digital gain ADC i-1 = 61, then the analog gain PGA of the current frame image i = PGA i-1 - 1; if these two conditions are not met, the digital gain ADC is adjusted i = ADC i-1 - 1.

[0064] Figure 3 It is a schematic diagram before dimming, Figure 4 and it is a schematic diagram after dimming by the method of the present invention. It can be seen that before dimming, due to the direct sunlight, the spaceship itself shows a high-brightness saturation phenomenon, and there are ghosts of the spaceship around it; after adopting the automatic dimming method of the present invention, not only do the ghosts disappear, but at the same time, the target spaceship also restores a clear visible outline and is no longer in a high-brightness saturation state, which can prove that the dimming method described in the present invention has a good dimming effect.

[0065] The parts not elaborated in detail in the present invention belong to the well-known technology in the art.

Claims

1. An automatic dimming method applicable to manual rendezvous and docking, characterized in that, The method includes the following steps: Step 1: The input image is a grayscale image I; Step 2: Divide the input grayscale image I into regions, set the central region, and the image of the central region is M, and the first column image of the central region is F, satisfying Meanwhile, define region N = I - M, C = M - F; Step 3: Count the number of valid pixels in each region: For the pixels in region N that are greater than the lower gray level threshold Gray L and less than the upper gray level threshold Gray H count them; For the pixels in region C that are greater than the lower gray level threshold Gray L count them with a 2-fold weighting; For all pixels in region F, count them with a 2-fold weighting; Step 4: For the valid pixels in regions N, C, and F that meet the requirements of Step 3, perform comprehensive statistics to obtain the sum of the grayscale values of the valid pixels, sum, and the number of valid pixels, num; Step 5: Calculate the average grayscale value Rav of the valid pixels in the image: Step 6: Based on the average grayscale value Rav of the valid pixels and the expected grayscale value Eav, obtain the dimming parameter proportionality coefficient ratio: Step 7: Calculate the proportionality coefficient ratio obtained from the eight nearest frames of images i , where i = 0, 1, 2, 3, 4, 5, 6, 7; thus, the corrected proportionality coefficient ratio' of the comprehensive dimming parameter is obtained: Step 8: Calculate the total gain value G of the previous frame image i-1 : G i-1 = ADC i-1 × PGA i-1 × EXP i-1 , Among them, ADC i-1 , PGA i-1 , EXP i-1 are respectively the digital gain, analog gain, and exposure time of the previous frame image; Obtain the total gain value G to be set for the current frame image according to the comprehensive dimming parameter proportionality coefficient ratio′ i : G i = G i-1 × ratio'; Step 9: According to the total gain value G of the current frame image i , calculate the dimming parameters to be configured for the current frame, namely the exposure time EXP i , the analog gain PGA i , and the digital gain ADC i ; among them, the exposure time EXP is preferentially adjusted i : where EXP min is the minimum exposure time, and EXP max is the maximum exposure time; ADC def is the default initial digital gain, and PGA def is the default initial analog gain; after automatic dimming is started, ADC def = ADC i-1 , and PGA def = PGA i-1 .

2. The automatic dimming method applicable to manual rendezvous and docking according to claim 1, wherein: The size of the central region image is 256×256.

3. The automatic dimming method applicable to manual rendezvous and docking according to claim 1, characterized in that: The lower limit of the gray scale threshold Gray L = 16, and the upper limit of the gray scale threshold Gray H = 1000.

4. An automatic dimming method applicable to manual rendezvous and docking according to claim 1, characterized in that: The expected grayscale value Eav = 144.

5. The automatic dimming method applicable to manual rendezvous and docking according to claim 1, characterized in that: Default initial digital gain ADC def = 44, default initial analog gain PGA def = 1.

6. The automatic dimming method applicable to manual rendezvous and docking according to claim 1, characterized in that: Minimum exposure time EXP min = 1510 and maximum exposure time EXP max = 201500.

7. An automatic dimming method applicable to manual rendezvous and docking according to claim 1, characterized in that: In step 9, if the adjusted exposure time EXP i is within the maximum exposure time EXP max and less than the minimum exposure time EXP min range, then the current frame is not further adjusted; If the adjusted exposure time EXP i has been greater than the maximum exposure time EXP max or less than the minimum exposure time EXP min , then the digital gain ADC i and the analog gain PGA i .

8. An automatic dimming method applicable to manual rendezvous and docking according to claim 7, characterized in that: When the adjusted exposure time is greater than the maximum exposure time EXP max , the digital gain ADC needs to be adjusted i : At this time, if the analog gain PGA i-1 = 0 and the digital gain ADC i-1 < 46, or if PGA i-1 = 1 and the digital gain ADC i-1 < 61, then the digital gain ADC of the current frame image i = ADC i-1 + 1; if these two conditions are not met, then adjust the analog gain value PGA i = PGA i-1 + 1.

9. An automatic dimming method applicable to manual rendezvous and docking according to claim 7, characterized in that: When the adjusted exposure time is less than the minimum exposure time EXP min , it is necessary to adjust the analog gain PGA i : At this time, if PGA i-1 > 0 and the digital gain ADC i-1 = 46, or PGA i-1 > 1 and the digital gain ADC i-1 = 61, then the analog gain PGA of the current frame image i = PGA i-1 - 1; If these two conditions are not met, then adjust the digital gain of the ADC i = ADC i-1 - 1.

Images