Spatial point target detection image quality feature consistency evaluation method

Abstract

The invention provides a spatial point target detection image quality characteristic consistency evaluation method, which comprises the following steps of: adjusting a rear intercept of a star sensor, and imaging star points at the center, the upper position, the lower position, the left position and the right position of an image surface of a detector to sequentially obtain corresponding star point diagrams; performing image preprocessing, obtaining a star point centroid of each star point diagram by adopting a centroid extraction algorithm, and correspondingly generating an updated star point diagram by taking the star point centroid of each star point diagram in the five-grid diagram as a center; constructing an image evaluation index; establishing a plurality of constraint conditions by utilizing the evaluation indexes; wherein the updated star point diagram simultaneously meets all the constraint conditions, and the back intercept adjustment meets the requirement, otherwise, the back intercept is re-adjusted until all the constraint conditions are met. According to the method, detailed image evaluation indexes are established to restrain the result image of back intercept adjustment, manual errors are eliminated, meanwhile, the calculation amount is greatly reduced, and the method has the remarkable advantages.

Description

Technical Field

[0001] The present invention relates to the technical field of consistency of space point target detection capability, and in particular to a method for evaluating consistency of image quality features of space point target detection. Background Art

[0002] Star sensor is the core component of the spacecraft GNC (guidance, navigation and control subsystem) system. It is a precision instrument developed based on computer vision measurement theory and is mainly used to measure the three-axis attitude of satellites. The components of star sensor include optical and precision structure system, photoelectric detector and signal processing circuit and software. The working principle of star sensor is to take the stable light signal of stars as input, pre-process the star point image within the field of view, obtain the precise position of the star in its body coordinate system, and then obtain the three-axis attitude information of the star sensor body relative to the geocentric inertial coordinate system through the navigation star library and identification matching algorithm stored in the internal memory of the product, so as to obtain the three-axis attitude information of the spacecraft body relative to the geocentric inertial system. Therefore, the star point image obtained by the star sensor is the basis of all subsequent work such as star point extraction, star map recognition, attitude calculation, etc., and also determines the accuracy and sensitivity of the entire star sensor, which is the key factor affecting the accuracy of attitude information solution.

[0003] The back intercept adjustment is an important stage in the development and production process of star sensors. The specific process of this stage is to control the imaging quality of the star sensor for the star point at infinity by adjusting the distance between the lens and the detector. When the imaging of the star point at different positions of the detector is close to the ideal star point image, it is considered that the back intercept meets the use requirements. Therefore, the quality of the back intercept adjustment result directly affects the imaging quality of the star point of the star sensor. The existing method for judging the result of the back intercept adjustment is relatively simple, mainly based on the experience of the technicians to judge whether the image of the star point after adjustment is close to the ideal star point, which introduces a large human error, resulting in large differences in the sensitivity of different products.

[0004] Therefore, it is necessary to establish a unified evaluation method based on the image quality characteristics of point target detection to constrain the image obtained by back-intercept adjustment.

[0005] It should be understood that the above statements merely provide background technology related to the present invention and do not necessarily constitute prior art. Summary of the invention

[0006] The purpose of the present invention is to provide a method for evaluating the consistency of image quality features in spatial point target detection, to establish a unified evaluation method, and to constrain images obtained by back-intercept adjustment.

[0007] In order to achieve the above-mentioned purpose, the present invention provides a method for evaluating the consistency of image quality features of space point target detection, comprising the following steps: S1, adjusting the back intercept of a star sensor, and imaging the star point at the center of the detector image plane, as well as five positions of the top, bottom, left and right through the star sensor with the back intercept adjusted, and obtaining corresponding star point diagrams in sequence to form a five-square diagram; S2, performing image preprocessing on the five-square diagram, using a centroid extraction algorithm to obtain the star point centroid of each star point diagram in the five-square diagram, and generating an updated star point diagram correspondingly with the star point centroid of each star point diagram in the five-square diagram as the center to form a new five-square diagram; S3, constructing an image evaluation index; S4, using the evaluation index to establish a plurality of constraint conditions; S5, the newly generated five-square diagram in S2 satisfies all the constraint conditions at the same time, and the back intercept adjustment meets the requirements, otherwise, turn to S1 to re-adjust the back intercept until all the constraint conditions are met.

[0008] Preferably, in S1, the star point diagrams formed by imaging the star points at the upper and lower positions of the detector image plane satisfy central symmetry and are located at the edge of the field of view; and the star point diagrams formed by imaging the star points at the left and right positions of the detector image plane satisfy central symmetry and are located at the edge of the field of view.

[0009] Preferably, taking the centroid of the star points of each star point map in the five-square diagram as the center, a window of n×n pixels is used to divide each star point map in a central windowing manner, and an updated star point map is generated accordingly; wherein n is greater than 5.

[0010] Preferably, the value of n is 9.

[0011] Preferably, in S3, the image evaluation index includes four aspects, namely: grayscale mean, spot size, energy concentration and energy symmetry.

[0012] Preferably, the grayscale mean represents the average value of the star point energy in the five-star diagram, that is, the average value of the sum of the star point energies in the five star point diagrams is taken.

[0013] Preferably, the image spot size represents the number of pixels occupied by the star points in each star point map; wherein, the closer the image spot size is to 3×3 pixels, the better the star point map.

[0014] Preferably, the energy concentration represents the degree of energy concentration of actual star points in the five star point diagrams relative to theoretical star points, that is, the ratio of theoretical star point energy to actual star point energy is calculated; wherein, the closer the ratio is to 1, the more concentrated the actual star point energy is.

[0015] Preferably, the energy symmetry represents the symmetry of the star point energy in the five-square diagram, namely, the star point diagram in which the star points are imaged at the upper and lower positions of the detector image plane, and the star point diagram in which the star points are imaged at the left and right positions; wherein, the star point energy ratio in the two sets of images, namely, the star point diagram in which the star points are imaged at the upper and lower positions of the detector image plane, and the star point diagram in which the star points are imaged at the left and right positions, is calculated respectively, and the closer the obtained ratio is to 1, the better the energy symmetry.

[0016] Preferably, the constraints are: grayscale mean ≥ 1300, the image spot size satisfies the pixel value range of [9, 36], the energy concentration ≥ 0.9, and the energy symmetry satisfies the ratio range of [0.75, 1.25]

[0017] In summary, compared with the prior art, the method for evaluating the consistency of image quality features in spatial point target detection provided by the present invention has at least the following beneficial effects:

[0018] (1) The present invention establishes detailed image evaluation indicators to constrain the result image of the back intercept adjustment, gets rid of human errors, and ensures the consistency of the space point target detection capability of different batches of star sensors;

[0019] (2) The present invention covers the characteristic positions of the star points on the detector image plane by selecting the five-square diagram, and adopts a specific image preprocessing algorithm to reduce the amount of calculation in the subsequent process;

[0020] (3) The design and implementation of the method provided by the present invention are very concise and have strong versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flow chart of the method for evaluating consistency of image quality features for spatial point target detection of the present invention;

[0022] Figure 2 This is an example diagram of the five palace diagram in the present invention;

[0023] Figure 3 Schematic diagram of image evaluation indicators and constraints in the present invention. DETAILED DESCRIPTION

[0024] The following is combined with Figures 1 to 3 , the present invention is further explained by describing a preferred specific embodiment in detail.

[0025] It should be noted that the drawings are in a very simplified form and use non-precise proportions. They are only used to conveniently and clearly assist in explaining the embodiments of the present invention, and are not used to limit the conditions for the implementation of the present invention. Therefore, they have no substantive technical significance. Any structural modification, change in proportional relationship or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical content disclosed by the present invention.

[0026] It should be noted that, in the present invention, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only the elements explicitly listed, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0027] In the present invention, the quality of the back intercept adjustment directly affects the imaging quality of the star sensor star point. Adjusting a suitable back intercept can ensure the consistency of the image quality characteristics of the star sensor imaging. Therefore, Figure 1 As shown, the present invention provides a method for evaluating the consistency of image quality features of spatial point target detection, which specifically comprises the following steps:

[0028] S1, adjusting the back intercept of the star sensor, and imaging the star point at the center, top, bottom, left and right of the detector image plane through the star sensor with the adjusted back intercept, and obtaining the corresponding star point diagrams I1 to I5 in sequence to form a five-palace diagram;

[0029] S2, performing image preprocessing on the five-square diagram, using a centroid extraction algorithm to obtain the centroid of the star points of each star point diagram in the five-square diagram, and taking the centroid of the star points of each star point diagram in the five-square diagram as the center, correspondingly generating updated star point diagrams M1-M5 to form a new five-square diagram;

[0030] S3, construct image evaluation index;

[0031] S4. Using the evaluation index, establish several constraint conditions;

[0032] S5, the five-square diagram newly generated in S2 satisfies all constraints at the same time, and the rear intercept adjustment meets the requirements, otherwise go to S1 to re-adjust the rear intercept until all constraints are met.

[0033] Further, such as Figure 2As shown, in S1, the star point diagrams I2 and I3 formed by imaging the star points at the upper and lower positions of the detector image plane satisfy central symmetry and are located at the edge of the field of view; and the star point diagrams I4 and I5 formed by imaging the star points at the left and right positions of the detector image plane satisfy central symmetry and are located at the edge of the field of view.

[0034] Furthermore, in the S2, the centroid of the star point in each star point image in the five-square diagram is taken as the center, and a window of n×n pixels is used to divide each star point image in a central windowing manner, and an updated star point image is generated accordingly. It should be noted that the imaging range of the star point on the detector image plane is 3×3 pixels to 5×5 pixels, so the value of n is greater than 5, but the larger the value of n, the larger the subsequent calculation amount. In the preferred embodiment of the present invention, the value of n is 9. Among them, the new five-square diagram generated after division can effectively reduce the interference of the background environment compared to the original five-square diagram, and greatly reduce the amount of subsequent calculations.

[0035] Furthermore, it should be noted that the star point diagram is an energy conversion process in which the stellar radiation information is transmitted through the optical system and the grayscale information is output on the image plane. Therefore, the grayscale information on the star point diagram can not only characterize the star point distribution characteristics, but also reflect the star point energy characteristics. Therefore, in the above S3, the image evaluation index includes four aspects, namely: grayscale mean, image spot size, energy concentration and energy symmetry.

[0036] The grayscale mean represents the average value of the star point energy in the five-star diagram, that is, the average value of the sum of the star point energies in the five star point diagrams.

[0037] The image spot size indicates the number of pixels occupied by the star points in each star point map, and is used to determine the reasonable range of the star point energy. In the preferred embodiment of the present invention, it is worth noting that the closer the image spot size is to 3×3 pixels, the better the star point map is.

[0038] Among them, the energy concentration represents the degree of energy concentration of the actual star points in the five star point images relative to the theoretical star points; in the preferred embodiment of the present invention, it is worth noting that the size of the theoretical star point is 3×3 pixels, and a 3×3 pixel window is taken with the centroid of the actual star point as the center, and the ratio of the energy in the window (i.e., the theoretical star point energy) to the actual star point energy is calculated. It can be understood that the closer the ratio is to 1, the more concentrated the star point energy is.

[0039] Among them, the energy symmetry represents the symmetry of the star point energy in the two groups of images, namely, star point diagram I2 and star point diagram I3, in which the star points are imaged at the upper and lower positions of the detector image plane, and star point diagram I4 and star point diagram I5, in the five-square diagram; in the preferred embodiment of the present invention, it is worth noting that the star point energy ratios in the two groups of images, namely, star point diagram I2 and star point diagram I3, and star point diagram I4 and star point diagram I5, are calculated respectively, and the closer the obtained ratio is to 1, the better the energy symmetry is.

[0040] Further, in the S4, if Figure 3 As shown, the constraints are specifically: grayscale mean ≥ 1300, the image spot size satisfies the pixel value range of [9, 36], the energy concentration ≥ 0.9, and the energy symmetry satisfies the ratio range of [0.75, 1.25].

[0041] In summary, the present invention provides a method for evaluating the consistency of image quality features in space point target detection, establishes detailed image evaluation indicators to constrain the result image of the back intercept adjustment, gets rid of human errors, and ensures the consistency of space point target detection capabilities of star sensors in different batches; covers the characteristic positions of star points on the detector image plane by selecting the five-square diagram, and uses a specific image preprocessing algorithm to reduce the amount of calculation in the subsequent process; the method design and implementation are very concise and have strong versatility.

[0042] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be appreciated that the above description should not be considered as a limitation of the present invention. After reading the above content, it will be apparent to those skilled in the art that various modifications and substitutions of the present invention will occur. Therefore, the protection scope of the present invention should be limited by the appended claims.

Claims

1. A method for evaluating the consistency of image quality features in spatial point target detection, characterized in that: The following steps are involved: S1, adjusting the back intercept of the star sensor, and imaging the star point at the center, top, bottom, left and right of the detector image plane through the star sensor with the adjusted back intercept, and obtaining the corresponding star point diagrams in turn to form a five-square diagram; S2, performing image preprocessing on the five-square diagram, using a centroid extraction algorithm to obtain the centroid of the star points of each star point diagram in the five-square diagram, and generating an updated star point diagram correspondingly with the centroid of the star points of each star point diagram in the five-square diagram as the center to form a new five-square diagram; S3, construct image evaluation index; S4. Using the evaluation index, establish several constraint conditions; S5, the five-square diagram newly generated in S2 satisfies all constraints at the same time, and the rear intercept adjustment meets the requirements, otherwise go to S1 to re-adjust the rear intercept until all constraints are met.

2. The method for evaluating consistency of image quality features of spatial point target detection according to claim 1, characterized in that: In the above S1, the star point diagrams formed by imaging the star points at the upper and lower positions of the detector image plane satisfy central symmetry and are located at the edge of the field of view; and the star point diagrams formed by imaging the star points at the left and right positions of the detector image plane satisfy central symmetry and are located at the edge of the field of view.

3. The method for evaluating consistency of image quality features of spatial point target detection according to claim 1, characterized in that: Taking the centroid of each star point map in the five-square diagram as the center, a window of n×n pixels is used to divide each star point map in a central windowing manner, and an updated star point map is generated accordingly; The value of n is greater than 5.

4. The method for evaluating consistency of image quality features of spatial point target detection according to claim 3, characterized in that: The value of n is 9.

5. The method for evaluating consistency of image quality features of spatial point target detection according to claim 1, characterized in that: In the above-mentioned S3, the image evaluation index includes four aspects, namely: grayscale mean, spot size, energy concentration and energy symmetry.

6. The method for evaluating consistency of image quality features of spatial point target detection according to claim 5, characterized in that: The grayscale mean represents the average value of the star point energy in the five-star diagram, that is, the average value of the sum of the star point energies in the five star point diagrams.

7. The method for evaluating the consistency of image quality features of spatial point target detection according to claim 5, characterized in that: The image spot size represents the number of pixels occupied by the star points in each star point image; The closer the image spot size is to 3×3 pixels, the better the star point image is.

8. The method for evaluating the consistency of image quality features of spatial point target detection according to claim 5, characterized in that: The energy concentration indicates the degree of energy concentration of the actual star points relative to the theoretical star points in the five star point diagrams, that is, the ratio of the theoretical star point energy to the actual star point energy; The closer the ratio is to 1, the more concentrated the actual star point energy is.

9. The method for evaluating consistency of image quality features of spatial point target detection according to claim 5, characterized in that: The energy symmetry refers to the symmetry of the star point energy in the star point diagram of the upper and lower positions of the detector image plane and the star point diagram of the left and right positions in the five-square diagram; Among them, the star point diagrams of the star points imaged at the upper and lower positions of the detector image plane, as well as the star point diagrams of the star point images at the left and right positions, are calculated respectively. The energy ratio of the star points in these two sets of images is calculated. The closer the ratio is to 1, the better the energy symmetry is.

10. The method for evaluating consistency of image quality features of spatial point target detection according to claim 1, characterized in that: The constraints are as follows: grayscale mean ≥ 1300, the image spot size satisfies the pixel value range of [9, 36], the energy concentration ≥ 0.9, and the energy symmetry satisfies the ratio range of [0.75, 1.25].

Images