CN102034091B - A light spot recognition method and device for a digital sun sensor - Google Patents

Abstract

The invention provides a light spot identifying method and device for a digital sun sensor. The method comprises the following steps of: obtaining an estimated movement distance of the light spot of a former frame of image of the digital sun sensor in the current frame of image; calculating an actual movement distance of the light spot of the current frame of image relative to the light spot of the former frame of image, comparing the actual movement distance with the estimated movement distance, and marking the light spot of the current frame of image according to a comparison result. In the invention, by identifying the light spot according to the movement track of the light spot, the error generated by distinguishing the light spot with the light spot change rule of the energy cosine effect is overcome, meanwhile, no special narrow belt filter is added for reducing the intensity of the reached earthlight avoiding the effect of the earthlight to the digital sun sensor so that the implementation is simple, and the reliable work of the digital sun sensor is guaranteed; and meanwhile, the device can be realized based on the FPGA (Field Programmable Gata Array), which is beneficial to the integrated design of the digital sun sensor.

Description

A light spot recognition method and device for a digital sun sensor

technical field

The invention relates to image signal processing technology, in particular to a light spot recognition method and device for a digital sun sensor.

Background technique

The sun sensor is an attitude sensor that uses the sun as the reference orientation to measure the angle between the sun's line of sight and a certain body axis or coordinate plane of a spacecraft such as a satellite. The working principle of the digital sun sensor based on the surface frame image sensor and single-hole mask is shown in Figure 1. The sunlight passes through the light-transmitting hole O of the optical mask and is transmitted to the image sensor to form a light spot L. The position of the light spot is different for different incident angles. The digital sun sensor calculates the incident angle β of the sun's rays according to the position of the light spot, so as to determine the attitude of spacecraft such as satellites.

Usually, the earth reflection light is the main interfering stray light that affects the reliable operation of the digital sun sensor. The currently disclosed methods to prevent the earth reflection light from affecting the reliable operation of the digital sun sensor are: to install the digital sun sensor in a special way Or, try to avoid the reflection of the earth from entering the digital sun sensor through spectral filtering, and identify and eliminate the reflection of the earth through the feature of the light spot, so as to overcome the influence of the reflection of the earth to a certain extent.

However, when using a special installation method, when the satellite is in an abnormal attitude, or when encountering strong specular reflections of the earth, such as glacier reflections, the reflected light from the earth will enter the field of view of the digital sun sensor, and If the earth's reflected light is in the normal incident direction and the sun's light is in the oblique incident direction, the light spots formed by the two will not be able to be distinguished according to the light intensity, which will affect the reliable work of the digital sun sensor.

When the intensity of the earth's reflected light reaching the digital sun sensor is reduced by spectral filtering, and the sun's light spot and the earth's light spot formed by the earth's reflected light are distinguished according to the size and intensity of the light spot, so as to ensure the reliable operation of the digital sun sensor and determine the satellite Wait for the status of the spacecraft. However, in this method of spectral filtering, the theoretical basis for distinguishing solar facula and earth facula according to the size and intensity of the facula is: it is believed that the transformation law of facula light intensity satisfies the energy cosine effect, and this assumption not only ignores the The characteristic that the transmittance changes with the change of the incident angle of the sun, and this characteristic will become more obvious as the thickness of the spectral filter increases; the temperature characteristic of the image sensor is also ignored, that is, under the same incident light energy , the intensity of the spot formed on the image sensor will vary with the temperature. Therefore, in this method, according to the change law of the energy cosine effect, there will be a large error in distinguishing the sun spot from the earth spot, and even a misjudgment will occur, which will reduce the reliability of the digital sun sensor.

Contents of the invention

In view of this, the main purpose of the present invention is to provide a kind of spot recognition method that is used for digital sun sensor, can realize the distinction of sun spot and earth spot, thereby avoid the impact of the earth reflection light on digital sun sensor, guarantee The reliable operation of the digital solar sensor and an FPGA implementation device of the method are provided, which is beneficial to the integrated design of the digital solar sensor.

In order to achieve the above object, technical solution of the present invention is achieved in that way:

A method for spot recognition for a digital sun sensor, said method comprising the steps of:

Obtain the estimated moving distance of the light spot of the previous frame image of the digital sun sensor in the current frame image; calculate the actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image; compare the actual moving distance with the estimated moving distance Compare, and mark the light spot of the current frame image according to the comparison result.

Further, the calculation of the actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image is: calculating the modulus corresponding to the coordinate difference between the light spot centroid of the previous frame image and the light spot centroid of the current frame image, and obtaining the current The actual moving distance of the light spot of the frame image relative to the light spot of the previous frame image.

Further, the step of comparing the actual moving distance with the estimated moving distance, and marking the light spot of the current frame image according to the comparison result is specifically:

Calculate the absolute value of the difference between the actual moving distance and the estimated moving distance. When the absolute value of the difference is less than the preset estimated maximum deviation, the absolute value of the difference is consistent with the estimated maximum deviation. Otherwise, the absolute value of the difference is the same as the estimated maximum deviation. The deviations are inconsistent;

When the two are consistent, and the light spot of the previous frame image is marked as the sun light spot or the earth light spot, then the light spot of the current frame image is marked as the sun light spot or the earth light spot, and if there is another light spot, then it is marked as the earth light spot or sun spots;

When the two are inconsistent, and the light spot of the previous frame image is marked as the sun light spot or the earth light spot, then the light spot of the current frame image is marked as the earth light spot or the sun light spot, and if there is another light spot, then it is marked as the sun light spot or Earth flares.

Wherein, when the light spot of the previous frame image has no light spot mark, the method further includes:

Matching the light spot of the current frame image and the light spot of the previous frame image according to the light spot feature, when there is at least one matching light spot, marking the matched light spot as a hypothetical sun light spot;

Correspondingly, the estimated movement distance of the facula in the next frame image of the previous frame image of the digital sun sensor is specifically: the estimation of the assumed sun facula in the current frame image of the previous frame image of the digital sun sensor Moving distance;

The calculating the actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image is specifically: calculating the actual moving distance of the assumed sun light spot of the current frame image relative to the assumed sun light spot of the previous frame image.

Further, the matching of the light spot of the current frame image and the light spot of the previous frame image according to the feature of the light spot is specifically:

Calculate the absolute value of the difference between the total pixels of the light spot of the previous frame image and the light spot of the current frame image and the maximum gray value of the light spot;

When the absolute value of the difference between the total pixels of the light spot and the absolute value of the difference between the maximum gray value of the light spot is smaller than the preset maximum total pixel difference and the maximum gray value difference, the two match.

Further, comparing the actual moving distance with the estimated moving distance, and marking the light spot of the current frame image according to the comparison result is specifically: calculating the absolute value of the difference between the actual moving distance and the estimated moving distance of the assumed sun spot, when the When the absolute value of the stated difference is less than the preset estimated maximum deviation, it is assumed that the marking is correct; otherwise, it is assumed that the marking is incorrect;

If the label is assumed to be correct, mark the assumed sun spot of the current frame image as a sun spot, if there is another spot, mark it as an earth spot; if the label is incorrect, mark the assumed sun spot of the current frame image as an earth spot, if any Another flare is labeled as a sun flare.

A light spot recognition device for a digital sun sensor, the device includes a light spot recognition and judgment module, which is used to distinguish and identify the sun light spot and the earth light spot according to the moving track of the light spot, and according to the recognition result, the light spot of the current frame image to mark.

Further, the light spot identification and judgment module further includes: a light spot maximum gray value comparison submodule, which is used to calculate the absolute value of the difference between the light spot maximum gray value of the light spot of the previous frame image and the light spot of the current frame image. value, and compare the absolute value of the difference with the preset maximum gray value difference;

The light spot total pixel comparison sub-module is used to calculate the absolute value of the total pixel difference of the light spot between the light spot of the previous frame image and the light spot of the current frame image, and compare the absolute value of the difference with the preset maximum total pixel difference for comparison;

The actual moving distance calculation sub-module is used to calculate the modulus corresponding to the coordinate difference between the spot centroid of the previous frame image stored in the memory of the digital sun sensor and the stored coordinate difference of the spot centroid of the current frame image, to obtain The actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image;

The moving distance comparison sub-module is used to calculate the absolute value of the difference between the actual moving distance obtained by the actual moving distance calculation sub-module and the obtained estimated moving distance, and perform a calculation on the difference between the absolute value of the difference and the preset estimated maximum deviation Compare;

The identification and judgment state machine is used to analyze and judge according to the light spot information of the previous frame image and the current frame image, as well as the output results of the above sub-modules, and complete the identification and marking of the light spot of the current frame image; the read-write address generator uses It is used to read the required light spot information, and store the light spot mark result output by the identification and judgment state machine into the light spot mark memory.

Further, when the device is implemented based on FPGA, the maximum gray value comparison submodule of the light spot is realized by a subtractor and a comparator, and the subtractor is used to calculate the difference between the maximum gray value of the light spot and the current maximum gray value of the light spot of the previous frame image stored in the memory. The absolute value of the difference of the maximum gray value of the light spot of the frame image, the comparator is used to compare the absolute value of the difference with the preset maximum gray value difference, and output the comparison result to the identification and judgment state machine;

The total pixel comparison submodule of the light spot is realized by a subtractor and a comparator. The subtractor is used to calculate the absolute value of the total pixel difference between the total pixels of the light spot of the light spot of the previous frame image stored in the memory and the total pixel difference of the light spot of the light spot of the current frame image. The comparator is used for Comparing the absolute value of the difference with a preset maximum total pixel difference, and outputting the comparison result to the recognition and judgment state machine;

The actual moving distance sub-module is realized by an X subtractor, a Y subtractor and a modulo operator, and the X subtractor is used to calculate the X coordinate of the spot centroid of the previous frame image stored in the memory and the X coordinate of the spot centroid of the current frame image. The difference of the X coordinates; the Y subtractor is used to calculate the difference between the Y coordinate of the spot centroid of the previous frame image stored in the memory and the Y coordinate of the spot centroid of the current frame image; the modulo operator is used for the X subtractor and The coordinate difference value obtained by the Y subtractor is modulo obtained to obtain the actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image;

The moving distance comparison submodule is realized by a subtractor and a comparator, the subtractor is used to calculate the absolute value of the difference between the actual moving distance obtained by the modulo operator and the estimated moving distance obtained, and the comparator is used to compare the absolute value of the difference It is compared with the preset estimated maximum deviation, and the comparison result is output to the recognition and judgment state machine.

The light spot recognition method for the digital sun sensor provided by the present invention distinguishes and identifies the sun light spot and the earth light spot according to the moving track of the light spot, thereby avoiding the influence of the earth's reflected light on the digital sun sensor; The light spot change law of the effect distinguishes the error of the light spot; at the same time, it is not necessary to add a special narrow-band filter to the digital sun sensor to reduce the intensity of the earth's reflection light, which reduces the design difficulty of the digital sun sensor and ensures the digital Reliable work of the solar sensor; in addition, the device can also be realized based on FPGA, which is beneficial to the integrated design of the digital solar sensor.

Description of drawings

Figure 1 is a schematic diagram of the working principle of the digital sun sensor;

Fig. 2 is a schematic flow chart of the realization of the light spot recognition method for the digital sun sensor in the present invention;

Fig. 3 is the composition schematic diagram of the functional module of the spot recognition device that is used for digital sun sensor of the present invention;

FIG. 4 is a schematic diagram of the composition and structure of the light spot recognition device used in the digital sun sensor in FIG. 3 based on FPGA.

Detailed ways

The basic idea of the present invention: obtain the estimated moving distance of the light spot of the previous frame image of the digital sun sensor in the current frame image; calculate the actual moving distance of the light spot of the current frame image relative to the previous frame image, and compare the actual moving distance with the estimated moving distance The distance is compared, and the current frame image spot is marked according to the comparison result.

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail by citing the following embodiments and referring to the accompanying drawings.

Fig. 2 shows the implementation process of the present invention for the spot recognition method of digital sun sensor, as shown in Fig. 2, described method comprises the following steps:

Step S201, the power-on reset signal of the digital sun sensor triggers the initialization of the image sequence variable, and assigns the image sequence variable K to 1;

Wherein, the image sequence variable is used to indicate the frame number of the image currently processed by the digital sun sensor.

Step S202, the digital sun sensor judges whether the processing of the light spot information of the current frame, that is, the Kth frame image, is finished, if so, then execute step S203, otherwise, execute step S202 repeatedly in a certain period or in real time;

Wherein, the processing of the light spot information refers to obtaining the coordinates of the center of mass of the light spot, the number of light spots, the total pixels of the light spots, the maximum gray value of the light spots, etc. of the K-th frame image; the certain period can be set according to the needs of the user.

Step S203, judge whether there is light spot in the image of the Kth frame, if so, execute step S204, otherwise, execute step S216;

In this step, it can be determined whether the Kth frame image has a light spot by judging whether the number of light spots in the Kth frame image obtained in the light spot information processing is greater than 0: when the number of light spots is greater than 0, then the Kth frame image has a light spot ; Otherwise, there is no light spot in the Kth frame image.

Step S204, further judge whether K is equal to 1, if so, execute step S216, otherwise, execute step S205;

Step S205, judging whether the previous frame, that is, the image of the K-1th frame, has light spots, if so, execute step S206, otherwise, execute step S216;

In this step, it can be determined whether there is a light spot in the K-1th frame image by judging whether the number of light spots in the K-1th frame image acquired in the light spot information processing is greater than 0: when the number of light spots is greater than 0, then the K-th The -1 frame image has light spots; otherwise, the K-1th frame image has no light spots.

Step S206, judging whether the light spot of the K-1th frame image has a mark, if so, execute step S207, otherwise, execute step S210;

In this step, when the light spot of the K-1 frame image has a light spot mark, it means that the light spot of the K-1 frame image has been distinguished, and the sun light spot and/or the earth light spot are marked, and step S207 needs to be performed at this time ; When the light spot of the K-1th frame image does not have a light spot mark, it means that the light spot of the K-1th frame image has not been distinguished, and step S210 needs to be executed.

Step S207, obtaining the estimated moving distance ds of the sun spot of the K-1 frame image in the next frame, that is, the K frame image, by the digital sun sensor;

Specifically, in this step, when the K-1th frame image has both sun facula and earth facula, in order to improve processing efficiency, the digital sun sensor can choose one of them for estimation: if the sun facula is estimated, the sun facula is obtained The estimated moving distance ds^ of the light spot; if the earth's light spot is estimated, the estimated moving distance de^ of the earth's light spot is obtained;

In this step, the digital sun sensor estimates the sun facula or the earth's facula, specifically according to the satellite orbit parameters, the sun or the earth's motion parameters, and the position of the marked sun or earth's facula in the K-1 frame image, Estimate its moving distance ds^ or de^ in the next frame, that is, the image of the Kth frame.

Step S208, calculating the actual moving distance d of a certain light spot in the Kth frame image relative to the sun light spot or the earth light spot in the K-1th frame image;

When step S207 is that the solar facula of the K-1 frame image is estimated to obtain ds, then select a light spot from the K frame, and calculate its actual moving distance d relative to the solar facula of the K-1 frame image, Specifically, acquire the spot centroid coordinates of the spot in the K-th frame image obtained in the spot information processing and the spot centroid coordinates of the sun spot in the acquired K-1 frame image, and calculate the acquired K-th frame image and the K-th The coordinate difference of the center of mass coordinates of the light spot of the -1 frame image, and calculate the modulus of the coordinate difference, that is, obtain the actual moving distance d of the light spot relative to the sun light spot of the K-1th frame image;

When step S207 is to estimate the earth light spot of the K-1 frame image to obtain de, then select a light spot from the K frame, and calculate its actual moving distance relative to the corresponding earth light spot in the K-1 frame image d. Specifically, acquire the centroid coordinates of the spot of the spot in the K-th frame image obtained in the spot information processing and the spot centroid coordinates of the spot in the acquired K-1 frame image, and calculate the obtained K-th frame image and the coordinate difference of the light spot centroid coordinates of the K-1th frame image, and calculate the modulus of the coordinate difference value, that is, the actual moving distance d of the light spot relative to the earth’s light spot in the K-1th frame image is obtained.

Step S209, comparing the d and ds^ or de^, and marking the light spot of the Kth frame image according to the comparison result, and then performing step S216;

In this step, when step S207 is to estimate the solar spot of the K-1th frame image to obtain ds^, then calculate the absolute value of the difference between d and ds^, if the absolute value of the difference is less than the preset If the maximum deviation is estimated, the two agree; otherwise, the two disagree. When the comparison result is that the two are consistent, the current spot is marked as a sun spot, and if there is another spot, it is marked as an earth spot; when the comparison result is inconsistent, the current spot is marked as an earth spot, if there is another The other spot is marked as a sun spot.

When step S207 is to estimate the earth flare of the K-1th frame image to obtain de^, then calculate the absolute value of the difference between d and de^, if the absolute value of the difference is less than the preset estimated maximum deviation , the two are consistent; otherwise, the two are inconsistent. When the comparison result is that the two are consistent, mark the current spot as the earth spot, if there is another spot, mark it as the sun spot; when the comparison result is inconsistent, mark the current spot as the sun spot, if there is another The other spot is marked as the Earth spot.

Step S210, matching the light spots of the Kth frame image and the K-1th frame image according to the light spot feature;

In this step, the total pixels of the light spot, the maximum gray value of the light spot, etc. of the light spot of the K-1th frame image are matched with the total pixels of the light spot, the maximum gray value of the light spot, etc. of the light spot of the image of the Kth frame, and the Kth - The total pixels of a certain light spot in one frame image are Pixel(k-1), the maximum gray value of the light spot is Im(k-1), and the total pixels of a certain light spot in the K-th frame image are Pixel(k ), the maximum gray value of the spot is Im(k), if the following conditions are met: |Pixel(k)-Pixel(k-1)|<10; |Im(k)-Im(k-1)|< 15, it can be considered that a certain light spot in the Kth frame image matches a certain light spot in the K-1th frame image, wherein 10 and 15 in the above formula are respectively the preset maximum total pixel difference and The maximum gray value difference, the maximum total pixel difference and the maximum gray value difference can be specifically set according to the pixels and gray levels of the light spot. In this embodiment, the gray level of the light spot is 1 to 255. According to the experiment, the pixels of the light spot 50-100, preferably, the maximum total pixel difference can be set to 10, and the maximum gray value difference can be set to 15.

Step S211, whether there is at least one matching spot, if so, execute step S212, otherwise, execute step S216;

Step S212, assuming that the matching light spot is a sun spot, and marking it as an assumed sun spot;

In this step, the matching light spot can be assumed to be a sun spot, or an earth spot. In this embodiment, the matching spot is marked as an assumed sun spot for illustration; of course, it can also be assumed to be an earth spot, if If the matched spot is marked as a hypothetical earth spot, the subsequent steps are similar to the steps for assuming the sun spot, except that the judgment is made according to the moving track of the earth spot, and will not be repeated here.

Step S213, obtaining the estimated moving distance ds^ of the above-mentioned hypothetical sun spot in the image of the K-1th frame;

Among them, the digital sun sensor estimates the trajectory of the hypothetical sun facula. Specifically, according to the satellite orbit parameters, the sun's motion parameters, and the position of the hypothetical sun facula in the K-1th frame image, it can be estimated that it will be in the next frame, that is, the Kth frame. The moving distance ds^ in the image.

Step S214, calculating the assumed sun facula of the Kth frame image, relative to the actual moving distance d of the assumed sun facula in the K-1th frame image;

Specifically, the spot centroid coordinates of the assumed sun spot of the Kth frame image acquired in the spot information processing and the spot centroid coordinates of the assumed sun spot of the acquired K-1th frame image are obtained, and the acquired Kth frame is calculated. The coordinate difference of the hypothetical solar spot centroid coordinates of the image and the K-1th frame image, and calculate the modulus of the coordinate difference, that is: obtain the K-th frame image of the hypothetical solar spot relative to the K-1th The actual moving distance d of the assumed sun spot of the frame image;

Step S215, calculating the absolute value of the difference between d and ds^, when the absolute value of the difference is less than the preset estimated maximum deviation, the absolute value of the difference is consistent with the estimated maximum deviation; otherwise, the absolute value of the difference is The value does not agree with the estimated maximum deviation. When the comparison result is that the two are consistent, the current assumed sun spot is marked as the sun spot, and if there is another spot, it is marked as the earth spot; when the comparison result is inconsistent, the current assumed sun spot is marked as the earth spot Facula, if there is another facula, it is marked as solar facula, and then step S216 is performed;

Step S216, increase the value of the image sequence variable K by 1, and execute step S202.

Fig. 3 has shown the composition structure of the functional module of the light spot recognition device that the present invention is used for digital sun sensor, and light spot information processing module 11 among the figure and trajectory estimation module 12 are other functional modules of digital sun sensor, memory 14 is the light spot information storage device of the digital sun sensor, which further includes a light spot mark memory 141, a light spot number memory 142, a light spot maximum gray value memory 143, a light spot total pixel memory 144 and a light spot centroid coordinate memory 145; the device The light spot identification and judgment module 13 is included, which is used to distinguish and identify the sun light spot and the earth light spot according to the moving track of the light spot, and mark the light spot of the current frame image according to the identification result.

Further, the light spot identification and judgment module 13 includes a light spot maximum gray value comparison submodule 131, a light spot total pixel comparison submodule 132, an actual moving distance calculation submodule 133, a moving distance comparison submodule 134, an identification and judgment state machine 135, a read-write address generator 136 .

Wherein, the light spot maximum grayscale value comparison submodule 131 is used to calculate the absolute value of the difference between the light spot of the previous frame image and the light spot maximum grayscale value of the light spot of the current frame image, and compare the absolute value of the difference The value is compared with the preset maximum gray value difference;

The light spot total pixel comparison sub-module 132 is used to calculate the absolute value of the total pixel difference of the light spot between the light spot of the previous frame image and the light spot of the current frame image, and compare the difference between the absolute value of the difference and the preset maximum total Pixel difference for comparison;

The actual moving distance calculation sub-module 133 is used to calculate the modulus corresponding to the coordinate difference between the spot centroid of the previous frame image stored in the memory and the stored current frame image's spot centroid, to obtain the current frame image The actual moving distance of the light spot relative to the light spot of the previous frame image;

The moving distance comparison sub-module 134 is used to calculate the absolute value of the difference between the actual moving distance obtained by the actual moving distance calculation sub-module and the estimated moving distance acquired, and calculate the maximum deviation between the absolute value of the difference and the preset estimated maximum Compare;

The identification and judgment state machine 135 is used to analyze and judge according to the light spot information of two consecutive frames of images and the output results of the above sub-modules, and complete the identification and marking of the current frame image light spots.

The read and write address generator 136 is used to generate address signals for reading and writing the spot mark memory 141 , the spot number memory 142 , the maximum gray value memory 143 of the light spot, the total pixel memory 144 of the light spot and the coordinate memory 145 of the spot center of mass.

In addition, the described light spot recognition device for the digital sun sensor can be realized based on Field Programmable Gate Array (Field Programmable Gate Array, FPGA), and because most of the existing digital sun sensors are also realized based on FPGA , so it is more conducive to the integrated design of the device and the digital sun sensor. Fig. 4 shows the composition structure of the light spot recognition device for the digital sun sensor realized by the present invention based on FPGA, wherein the The FPGA chip used by the device can be FPGA chip XCV300 of Xilinx, as shown in FIG. 4 , wherein the spot information processing module 11 , trajectory estimation module 12 and memory 14 are the same as those in FIG. 3 , and will not be repeated here.

By programming the FPGA chip through a high-speed integrated circuit hardware description language (VHDL), the specific realization of the identification and judgment state machine 235 is obtained. The identification and judgment state machine 235 is used to analyze and judge according to the acquired light spot information, Complete the identification and marking of the current frame image spot.

In the same way, the FPGA chip is programmed through VHDL, and the read-write address generator 236 is implemented. The read-write address generator 236 is used to read or save the spot information of the image, including the spot mark, the number of spots, the maximum gray value of the spot, The total pixels of the spot 4 and the coordinates of the center of mass of the spot.

Similarly, the FPGA chip is programmed through VHDL to obtain a subtractor 2311 and a comparator 2312. The subtractor 2311 and the comparator 2312 are specific implementations of the facula maximum gray value comparison sub-module 131 in FIG. 3. The subtractor 2311 is used to calculate the maximum facula The absolute value of the difference between the maximum gray value of the light spot of the light spot of the previous frame image stored in the gray value memory 143 and the maximum gray value of the light spot of the light spot of the current frame image, and the comparator 2312 is used to compare the absolute value of the difference and the maximum gray value of the light spot of the light spot of the current frame image Compared with the preset maximum gray value difference, it can be set that when the absolute value of the difference is smaller than the maximum gray value difference, output 1, otherwise output 0, similarly, it can also be set when the difference When the absolute value is smaller than the maximum gray value difference, output 0, otherwise output 1.

In the same way, the FPGA chip is programmed by VHDL, and the subtractor 2321 and the comparator 2322 are obtained. The subtractor 2321 and the comparator 2322 are the concrete realization of the spot total pixel comparison submodule 132 in FIG. 3: the subtractor 2321 is used for calculating the total pixel memory of the spot 144 stores the absolute value of the difference between the total pixels of the light spot of the previous frame image and the total pixel difference of the light spot of the current frame image, and the comparator 2322 is used to compare the absolute value of the difference with the preset maximum total pixel difference, It can be set that when the absolute value of the difference is smaller than the maximum total pixel difference, 1 is output, otherwise 0 is output. Similarly, it can also be set that when the absolute value of the difference is smaller than the maximum gray value difference, Output 0, otherwise output 1.

Similarly, program the FPGA chip through VHDL to obtain the X subtractor 2331, the Y subtractor 2332 and the modulo operator 2333, and the X subtractor 2331, the Y subtractor 2332 and the modulo operator 2333 are the actual moving distance calculators in Fig. 3 The specific realization of module 133, X subtractor 2331 is used for calculating the difference between the X coordinate of the spot centroid of the previous frame image stored in the spot centroid coordinate memory 145 and the X coordinate of the spot centroid of the current frame image; Y subtractor 2332 is used for Calculate the difference between the Y coordinate of the spot centroid of the previous frame image stored in the spot centroid coordinate memory 145 and the Y coordinate of the spot centroid of the current frame image; the modulo operator 2333 is used to obtain the X subtractor 2331 and the Y subtractor 2332 Calculate the modulo of the coordinate difference of , and obtain the actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image.

Similarly, program the FPGA chip through VHDL to obtain a subtractor 2341 and a comparator 2342, which are the specific implementations of the moving distance comparison sub-module 134 in Figure 3: the subtractor 2341 is used to calculate the modulo operator 2333 to obtain The absolute value of the difference between the actual moving distance and the estimated moving distance obtained by the trajectory estimation module 12, the comparator 2342 is used to compare the absolute value of the difference with the preset estimated maximum deviation, and can be set when the When the absolute value of the difference is smaller than the estimated maximum deviation, 1 is output, otherwise 0 is output. Similarly, when the absolute value of the difference is smaller than the maximum gray value difference, 0 is output, otherwise 1 is output.

The light spot recognition method and device for the digital sun sensor provided by the present invention distinguish and identify the sun light spot and the earth light spot according to the moving track of the light spot, thereby avoiding the influence of the earth's reflected light on the digital sun sensor, not only overcoming the The light spot change law of the energy cosine effect distinguishes the error of the light spot, and at the same time, it is not necessary to add a special narrow-band filter to the digital sun sensor to reduce the intensity of the earth's reflection light, which reduces the design difficulty of the digital sun sensor and ensures The reliable work of the digital sun sensor is ensured; at the same time, the above-mentioned device can also be realized based on FPGA, which is beneficial to the integrated design of the digital sun sensor.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention.

Claims (4)

1. a method for spot recognition for digital sun sensor, is characterized in that, described method comprises the steps: When the light spot of the previous frame image of the digital sun sensor has a light spot mark, the estimated moving distance of the light spot of the previous frame image in the current frame image is obtained; Calculate the actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image; Calculate the absolute value of the difference between the actual moving distance and the estimated moving distance. When the absolute value of the difference is less than the preset estimated maximum deviation, the absolute value of the difference is consistent with the estimated maximum deviation. Otherwise, the absolute value of the difference is the same as the estimated maximum deviation. The deviations are inconsistent; when the two are consistent, and the light spot of the previous frame image is marked as a sun light spot, then the light spot of the current frame image is marked as a sun light spot, and if there is another light spot, it is marked as an earth light spot; When the two are consistent, and the light spot of the previous frame image is marked as the earth light spot, then the light spot of the current frame image is marked as the earth light spot, and if there is another light spot, then it is marked as the sun light spot; when the two are inconsistent, And when the light spot of the previous frame image is marked as the sun light spot, then the light spot of the current frame image is marked as the earth light spot, and if there is another light spot, it is marked as the sun light spot; when the two are inconsistent, and the previous one When the light spot of the frame image is marked as the earth light spot, the light spot of the current frame image is marked as the sun light spot, and if there is another light spot, then it is marked as the earth light spot; When the spot of the previous frame image has no spot mark, calculate the absolute value of the difference between the total pixels of the spot of the spot of the previous frame image and the spot of the current frame image and the maximum gray value of the spot; when the total pixel of the spot is When the absolute value of the difference and the absolute value of the difference between the maximum gray value of the light spot are less than the preset maximum total pixel difference and the maximum gray value difference respectively, then there is a matching light spot, and the matching light spot is marked as hypothetical sun spot; Obtain the estimated moving distance of the assumed sun spot in the current frame image of the previous frame image of the digital sun sensor; Calculate the actual moving distance of the assumed sun facula of the current frame image relative to the assumed sun facula of the previous frame image; Calculate the absolute value of the difference between the actual moving distance of the assumed sun spot and the estimated moving distance. When the absolute value of the difference is less than the preset estimated maximum deviation, it is assumed that the mark is correct; otherwise, the mark is assumed to be incorrect; when the mark is assumed to be correct , mark the assumed sun spot of the current frame image as a sun spot, if there is another spot, mark it as an earth spot; if the label is incorrect, mark the assumed sun spot of the current frame image as an earth spot, if there is another spot, then Marked as sunspots. 2. The method according to claim 1, wherein the calculation of the actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image is: Calculate the modulus corresponding to the coordinate difference between the spot centroid of the previous frame image and the spot centroid of the current frame image, and obtain the actual moving distance of the spot of the current frame image relative to the spot of the previous frame image. 3. A light spot recognition device for a digital sun sensor, characterized in that the device includes a light spot recognition and judgment module for distinguishing and identifying sun light spots and earth light spots according to the moving track of light spots, and according to the recognition result, Mark the spot of the current frame image; where, The spot recognition and judgment module also includes: The maximum gray value comparison submodule of the light spot is used to calculate the absolute value of the difference between the light spot of the previous frame image and the maximum gray value of the light spot of the light spot of the current frame image, and compare the absolute value of the difference with the preset Compare the maximum gray value difference; The light spot total pixel comparison sub-module is used to calculate the absolute value of the total pixel difference of the light spot between the light spot of the previous frame image and the light spot of the current frame image, and compare the absolute value of the difference with the preset maximum total pixel difference for comparison; The actual moving distance calculation sub-module is used to calculate the modulus corresponding to the coordinate difference between the spot centroid of the previous frame image stored in the memory of the digital sun sensor and the stored coordinate difference of the spot centroid of the current frame image, to obtain The actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image; The moving distance comparison sub-module is used to calculate the absolute value of the difference between the actual moving distance obtained by the actual moving distance calculation sub-module and the obtained estimated moving distance, and perform a calculation on the difference between the absolute value of the difference and the preset estimated maximum deviation Compare; The identification and judgment state machine is used to analyze and judge according to the light spot information of the previous frame image and the current frame image, as well as the output results of the above sub-modules, and complete the identification and marking of the light spot of the current frame image; the read-write address generator uses It is used to read the required light spot information, and store the light spot mark result output by the identification and judgment state machine into the light spot mark memory. 4. The device according to claim 3, wherein when the device is implemented based on FPGA, the maximum gray value comparison submodule of the light spot is realized by a subtractor and a comparator, and the subtractor is used to calculate the previous value stored in the memory. The absolute value of the difference between the maximum gray value of the light spot of the frame image and the maximum gray value of the light spot of the current frame image, and the comparator is used to compare the absolute value of the difference with the preset maximum gray value difference, And output the comparison result to the identification and judgment state machine; The total pixel comparison submodule of the light spot is realized by a subtractor and a comparator. The subtractor is used to calculate the absolute value of the total pixel difference between the total pixels of the light spot of the light spot of the previous frame image stored in the memory and the total pixel difference of the light spot of the light spot of the current frame image. The comparator is used for Comparing the absolute value of the difference with a preset maximum total pixel difference, and outputting the comparison result to the recognition and judgment state machine; The actual moving distance sub-module is realized by an X subtractor, a Y subtractor and a modulo operator, and the X subtractor is used to calculate the X coordinate of the spot centroid of the previous frame image stored in the memory and the X coordinate of the spot centroid of the current frame image. The difference of the X coordinates; the Y subtractor is used to calculate the difference between the Y coordinate of the spot centroid of the previous frame image stored in the memory and the Y coordinate of the spot centroid of the current frame image; the modulo operator is used for the X subtractor and The coordinate difference value obtained by the Y subtractor is modulo obtained to obtain the actual moving distance of the light spot of the current frame image relative to the light spot of the previous frame image; The moving distance comparison submodule is realized by a subtractor and a comparator, the subtractor is used to calculate the absolute value of the difference between the actual moving distance obtained by the modulo operator and the estimated moving distance obtained, and the comparator is used to compare the absolute value of the difference It is compared with the preset estimated maximum deviation, and the comparison result is output to the recognition and judgment state machine.

Images