CN118134820A - Self-adaptive optical remote sensing image relative radiation correction method - Google Patents
Self-adaptive optical remote sensing image relative radiation correction method Download PDFInfo
- Publication number
- CN118134820A CN118134820A CN202410558219.7A CN202410558219A CN118134820A CN 118134820 A CN118134820 A CN 118134820A CN 202410558219 A CN202410558219 A CN 202410558219A CN 118134820 A CN118134820 A CN 118134820A
- Authority
- CN
- China
- Prior art keywords
- relative radiation
- radiation correction
- image
- gray value
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005855 radiation Effects 0.000 title claims abstract description 233
- 238000012937 correction Methods 0.000 title claims abstract description 193
- 238000000034 method Methods 0.000 title claims abstract description 60
- 230000003287 optical effect Effects 0.000 title claims abstract description 22
- 239000000523 sample Substances 0.000 claims abstract description 186
- 230000008569 process Effects 0.000 claims abstract description 9
- 238000001514 detection method Methods 0.000 claims abstract description 6
- 230000001186 cumulative effect Effects 0.000 claims description 18
- 238000007619 statistical method Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000012795 verification Methods 0.000 claims description 8
- 238000002310 reflectometry Methods 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 description 12
- 238000003384 imaging method Methods 0.000 description 5
- 230000010354 integration Effects 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000013139 quantization Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000006837 decompression Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The invention provides a self-adaptive optical remote sensing image relative radiation correction method, which comprises the following steps: configuring initial working parameters, wherein the initial working parameters comprise initial relative radiation correction coefficients; acquiring an original image of remote sensing detection, and dividing the original image into strip images; in the process of dividing the strip image, counting the gray value frequency of each line of data of each probe element in the strip image; generating and storing the total frequency of the occurrence of the original gray value of each probe element in the strip image; automatically judging whether the initial relative radiation correction coefficient is suitable for the strip image; if the initial relative radiation correction coefficient is not suitable, automatically rescaling to generate a new relative radiation correction coefficient; and performing relative radiation correction on the band image based on the relative radiation correction coefficient.
Description
Technical Field
The invention relates to the technical field of remote sensing image processing, in particular to a self-adaptive optical remote sensing image relative radiation correction method.
Background
The high resolution optical camera payload is typically composed of a multi-slice TDICMOS or TDICCD splice. The relative radiation correction is based on optical remote sensing image processing, and the relative radiation correction coefficient quantifies the nonlinear relation between the gray value acquired by the probe element and the real radiation source, so that the problems of image quality such as bright and dark stripes, random noise, inter-chip color difference and the like of original images in CMOS and CCD chips caused by inconsistent responses of the probe elements are solved. Along with the extension of the on-orbit time of the load, the probe element can be aged gradually, the adaptability of the relative radiation correction coefficient can be reduced, the image radiation quality is poor, the image radiation quality can be visually distinguished to a certain extent after being accumulated, and even bad wires appear. At present, the lack of effective detection element response change and relative radiation correction coefficient adaptability monitoring means is generally solved by periodically carrying out relative radiation calibration in a period of half a year.
Common radiometric calibration methods are: laboratory calibration, on-orbit calibration, yaw calibration, uniform field calibration, on-orbit statistics and the like. The laboratory calibration method is to simulate various radiation source environments in a laboratory environment, and calculate and acquire the conversion relation between the probe element and the real radiation value. However, due to vibration in the transmitting process, various state changes such as an on-orbit electromagnetic environment and the like, the response of the probe element is not completely consistent with that of a ground laboratory, and the imaging process comprises ground, atmosphere and probe element noise. Thus, the laboratory scaling factor is only valid in the initial stages of the orbit. The on-orbit calibration method supports calibration according to the day and the imaging front/back, can acquire the response data of each probe element in real time, and has the best radiation correction effect. However, the on-orbit satellite calibration method requires that the load has on-orbit calibration capability, has high development cost and is not suitable for all satellites; the on-board calibration method needs to consume satellite energy and occupy satellite imaging time, and the acquisition capacity of satellite monorail data is weakened; the downloading of the on-board calibration data occupies satellite data transmission resources, and the effective image data downloaded to the ground every day is reduced. The yaw calibration method can rapidly acquire the relative radiation correction data of each working gear of the load probe, but the method requires the satellite to have agile maneuvering capability, the load can image around the satellite platform aviation axis by 90 degrees, and the method is not suitable for all satellites; with the continuous change of the response state of the probe element, the load life cycle needs to be corrected by relative radiation once every half year, and aiming at the satellite with the requirement of ultra-stable operation of a satellite platform such as a high-resolution seventh and the like, the relative relation between the load and the space position of other loads such as a star sensor, a GNSS receiver and the like can be destroyed by multiple 90-degree maneuver imaging around a navigation axis, and high-precision geometric calibration needs to be carried out again. Uniform field calibration is the most common calibration method, has good calibration effect and is suitable for common optical satellites. However, the method requires that different uniform ground objects are manually selected as samples for statistics, and the interference of human factors is large; the sample coverage is required to be enough, the calibration period is long, and automation cannot be realized. The on-orbit statistics method is used for obtaining the calibration coefficient by carrying out batch statistics on all data by assuming that the ground object is covered by the image data obtained in a period of load as the uniform ground object, the calibration effect is good, and the method is suitable for common optical satellites, but samples of the method also need to be selected manually, and the interference is large. In addition, batch statistics occupies a large amount of calculation and storage resources, and influences the normal operation of the service system.
Disclosure of Invention
In view of the above problems, the embodiments of the present invention provide a method for correcting relative radiation of an adaptive optical remote sensing image.
An aspect of the present invention provides a method for correcting relative radiation of an adaptive optical remote sensing image, including: configuring initial working parameters, wherein the initial working parameters comprise initial relative radiation correction coefficients; acquiring an original image of remote sensing detection, and dividing the original image into strip images; in the process of dividing the strip image, counting the gray value frequency of each line of data of each probe element in the strip image; generating and storing the total frequency of the occurrence of the original gray value of each probe element in the strip image; automatically determining whether the initial relative radiation correction factor is applicable to the banding image; if the initial relative radiation correction coefficient is not suitable, automatically recalibrating to generate a new relative radiation correction coefficient; and carrying out relative radiation correction on the band image based on the relative radiation correction coefficient.
According to an embodiment of the present invention, the initial operation parameters further include a load continuous operation minimum time interval, and the acquiring the original image of the remote sensing probe and dividing the original image into the strip images includes: obtaining the decompressed and formatted original image according to a data stream or file mode; synchronously analyzing the auxiliary information of the original image to obtain the acquisition time of each line of image data in the original image; when the time difference of the acquisition time of two lines of continuous image data is larger than the shortest time interval of continuous load operation, dividing the two lines of image data with the time difference larger than the shortest time interval of continuous load operation into different strip images.
According to an embodiment of the present invention, the counting the gray value frequency of each line of data of each probe element in the stripe image includes: judging whether the original code stream data corresponding to each line of data of the strip image has error codes or frame loss; if the original code stream data corresponding to the j-th data of the stripe image has error codes or frame loss phenomena, the j-th data does not participate in statistics, and j represents the data line number of the stripe image; carrying out boundary processing on normal probe elements, and calculating gray value frequency P kj (i, g) of the original gray value g of the ith probe element of the jth row in the strip image, wherein i represents the number of the probe element, and g represents the original gray value; when the ith probe element is a bad probe element, recording P kj (i, g) =0; when the original gray value of the load image is greater than or equal to 2 n -1 due to image overexposure or error code of the jth line data of the stripe image, an upper boundary exists in the image corresponding to the jth line data, and P kj (i, g) =0 is recorded; when the error code exists in the j-th data or the reflectivity of the ground object is low, and the original gray value of the load image is smaller than or equal to 0, a lower boundary exists in the image corresponding to the j-th data, and P kj (i, g) =0 is recorded.
According to an embodiment of the present invention, the initial operating parameters further include a minimum number of rows of payload formatted data and a minimum duration of operation of the payload, and the generating and storing the total frequency of occurrence of the original gray value of each of the probe elements in the strip image includes: calculating the total frequency of the occurrence of the original gray value of each probe element in each strip image: if the total line number of the kth stripe image is smaller than the minimum line number of the effective load formatted data or the duration of the kth stripe image is shorter than the minimum working duration of the load, the total frequency of the occurrence of the original gray value of each probe element in the kth stripe image is not counted.
According to an embodiment of the present invention, the automatically determining whether the current relative radiation correction coefficient is applicable to the banding image includes: calculating the total frequency of each original gray value of m strips of the load data in the same wave band, gain and integration level as the kth strip image in the kth-m 1 strips of the k-m1 strip images, wherein m represents the least statistical analysis strip number of the load data, k represents the sequential coding of the strip images, and m1 represents the number of the strip images selected before the kth strip image; matching the load name, the wave band, the gain and the integral series of the kth stripe image, and obtaining the nearest relative radiation coefficient of the kth stripe image; carrying out relative correction on the total frequency of the occurrence of the original gray value of each probe element in the m strip images based on the relative radiation coefficients to obtain the relative radiation correction gray value frequency of each probe element; calculating the relative radiation correction precision between adjacent probe elements based on the relative radiation correction gray value frequency; calculating the relative radiation correction precision of the full view field; if the relative radiation correction accuracy between adjacent probe elements or the full field-of-view relative radiation correction accuracy does not meet the index requirement, the relative radiation coefficient is not suitable for the kth stripe image.
According to an embodiment of the present invention, the calculating the relative radiation correction accuracy between adjacent probe cells based on the relative radiation correction gray value frequency includes: calculating an average gray value of each probe element in the m strip images after relative radiation correction based on the relative radiation correction gray value frequency; based on the average gray value of each probe element after the relative radiation correction, calculating the average gray difference of adjacent probe elements after the relative radiation correction, wherein the bad probe elements do not participate in the calculation of the relative radiation correction precision of the adjacent probe elements; and taking the maximum average gray level difference as the relative radiation correction precision.
According to an embodiment of the present invention, the initial operating parameters further include a bad probe element identification threshold, and the calculating the relative radiation correction accuracy of the adjacent probe elements further includes: if the average gray level difference after the relative radiation correction of the ith probe element and the (i-1) th probe element is larger than the bad probe element identification threshold value and the average gray level difference after the relative radiation correction of the ith probe element and the (i+1) th probe element is larger than the bad probe element identification threshold value, adding the ith probe element into a bad probe element set.
According to an embodiment of the present invention, the calculating the full field-of-view relative radiation correction accuracy includes: calculating the average gray value of the probe cells after the bad probe cells are removed based on the average gray value of each probe cell in the strip image after the relative radiation correction; and calculating the full-view-field relative radiation correction precision based on a preset full-view-field relative radiation correction precision calculation formula and the average gray value of the probe element.
According to an embodiment of the invention, the method further comprises: and if the relative radiation correction precision between adjacent probe elements or the full-view field relative radiation correction precision meets the index requirement and reaches an alarm threshold, alarming and judging that the relative radiation coefficient is not suitable for the kth stripe image.
According to an embodiment of the invention, the automatically rescaling to generate new relative radiation correction factors comprises: automatically calculating the probability distribution of the original gray value and the cumulative probability distribution of the original gray value of each probe element in m strip images, wherein m represents the least statistical analysis strip number of load data; calculating a load comprehensive cumulative probability distribution based on the original gray value cumulative probability distribution; generating a new relative radiation correction coefficient of a kth stripe image according to the original gray value cumulative probability distribution and the load comprehensive cumulative probability distribution of each probe element; and verifying the relative radiation correction coefficient, and storing the relative radiation correction coefficient passing verification according to the load name, the wave band, the gain, the integral series and the calibration time.
According to an embodiment of the present invention, the verifying the relative radiation correction factor further includes: recalculating the relative radiation correction precision of the adjacent probe elements and the relative radiation correction precision of the full field of view by adopting the newly generated relative radiation correction coefficients; if the relative radiation correction precision of the adjacent probe elements and the relative radiation correction precision of the full view field meet the index requirements, verifying the relative radiation correction coefficient; if the relative radiation correction coefficient fails to pass the verification, increasing the value of m or manually selecting any number of conditional images in the first k stripe images, and recalculating the relative radiation correction coefficient.
The above at least one technical scheme adopted in the embodiment of the invention can achieve the following beneficial effects:
The adaptive optical remote sensing image relative radiation correction method provided by the embodiment of the invention utilizes a streaming data processing technology, adopts a small amount of calculation and storage resources, can automatically select samples, performs indiscriminate statistical analysis, realizes quick coverage and statistical analysis of various samples, and removes the influence of human factors; the method introduces a detection method of relative radiation correction precision of adjacent probe elements and relative radiation correction precision of a full-view field of load, carries out statistical analysis on relative radiation correction results in a period of time, and can automatically monitor adaptability of relative radiation correction coefficients; the method can be based on the relative radiation coefficient adaptability monitoring result, can automatically produce relative radiation correction coefficients and deploy, and supports subsequent relative radiation correction.
Drawings
For a more complete understanding of the present invention, and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:
FIG. 1 schematically illustrates a flowchart of a method for correcting relative radiation of an adaptive optics remote sensing image according to an embodiment of the present invention;
Fig. 2 schematically illustrates a detailed flowchart of a method for correcting relative radiation of an adaptive optical remote sensing image according to an embodiment of the present invention.
Detailed Description
Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. It should be understood that the description is only illustrative and is not intended to limit the scope of the invention. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the invention. It may be evident, however, that one or more embodiments may be practiced without these specific details. In addition, in the following description, descriptions of well-known structures and techniques are omitted so as not to unnecessarily obscure the present invention.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The terms "comprises," "comprising," and/or the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It should be noted that the terms used herein should be construed to have meanings consistent with the context of the present specification and should not be construed in an idealized or overly formal manner.
Some of the block diagrams and/or flowchart illustrations are shown in the figures. It will be understood that some blocks of the block diagrams and/or flowchart illustrations, or combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the instructions, when executed by the processor, create means for implementing the functions/acts specified in the block diagrams and/or flowchart.
Thus, the techniques of the present invention may be implemented in hardware and/or software (including firmware, microcode, etc.). Furthermore, the techniques of the present invention may take the form of a computer program product on a computer-readable medium having instructions stored thereon for use by or in connection with an instruction execution system. In the context of the present invention, a computer-readable medium may be any medium that can contain, store, communicate, propagate, or transport the instructions. For example, a computer-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, device, or propagation medium. Specific examples of the computer readable medium include: magnetic storage devices such as magnetic tape or hard disk (HDD); optical storage devices such as compact discs (CD-ROMs); a memory, such as a Random Access Memory (RAM) or a flash memory; and/or a wired/wireless communication link.
Fig. 1 shows a flowchart of a method for correcting relative radiation of an adaptive optical remote sensing image according to an embodiment of the present invention, fig. 2 schematically shows a detailed flowchart of a method for correcting relative radiation of an adaptive optical remote sensing image according to an embodiment of the present invention, and the method for correcting relative radiation of an adaptive optical remote sensing image according to the present invention will be described in detail with reference to fig. 1 and 2.
As shown in FIG. 1, the embodiment of the invention provides a method for correcting relative radiation of a self-adaptive optical remote sensing image, which comprises operations S110-S170.
In operation S110, initial operating parameters including initial relative radiation correction coefficients are configured.
In the embodiment of the invention, the initial working parameters can comprise initial relative radiation correction coefficient and load bad probe element number, bad probe element identification threshold, adjacent probe element relative radiation correction precision and alarm threshold, load full view field relative radiation correction precision and alarm threshold, load continuous working shortest time interval, minimum line number of payload formatted data, minimum working time length of load, minimum statistic analysis strip number of load data and the like.
The initial relative radiation correction coefficient may be a relative radiation correction coefficient obtained by a laboratory, and is denoted as g=r 0 (i, G), which indicates that when the gray value of the original image is G, the gray value of the i-th probe element after the relative radiation correction is G, and is stored in the form of a file according to the load name, the wave band, the gain, the integral series and the laboratory calibration time. Wherein I is the number of the load probe element, and when the load has I probe elements, I is E [1, I ]; g is the gray value of the original image obtained by the load probe element, and g is E [0,2 n -1] when the quantization bit number of the probe element is n.
The initial load bad probe element number set is obtained through experimental calibration, recorded as CCD bad and stored in a file or database form.
The initial bad probe element identification threshold value is estimated according to the load quantization bit number and recorded as bad Default values of 5DN are set and stored in the form of files or databases.
The initial adjacent probe element relative radiation correction precision and the load full-view field relative radiation correction precision are acquired according to the load star-ground integrated index and are respectively recorded as: DN And R DN; simultaneously, relative radiation correction precision and load full-view field relative radiation correction precision abnormal alarm thresholds are set according to satellite-ground integrated indexes and recorded as/>, respectively WDN And R WDN, and/> WDN</> DN And R WDN<RDN; when there is no corresponding index, set/> DN Default 1dn and default 5r DN and stored as a file or database.
Setting the shortest load data acquisition time according to the load design parameters ST And the corresponding formatted data line number/> N . Shortest interval between two adjacent times of data acquisition time/> T And stored in the form of a file or database.
According to the relative radiation correction coefficient histogram matching model, the minimum statistical analysis stripe number m of the load data is set, the default value is 100 stripes, and the load data is stored in a file or database form.
In operation S120, an original image of the remote sensing is acquired and divided into strip images.
In the embodiment of the invention, the decompression equipment/software outputs the decompressed and formatted original image data in the form of a file/binary stream. The initial operating parameters also comprise the shortest time interval for continuous operation of the load T Dividing the original image into strip images includes operations S121-S123.
In operation S121, the decompressed and formatted original image is acquired in a data stream or file manner.
In operation S122, the original image auxiliary information is analyzed synchronously to obtain the acquisition time T of each line of image data in the original image.
In operation S123, when the time difference between the acquisition times of two consecutive lines of image data is greater than the load consecutive operation minimum time interval, division is performed between the two lines of image data having the time difference greater than the load consecutive operation minimum time interval, and the original image is divided into different strip images.
Recording the j-th row and j-1-th row data acquisition time as T j and T j-1 respectively, if the satellite working time ranges of T j and T j-1 are within, and T j-Tj-1 is more than or equal to T Namely, the j and j-1 th data acquisition time has jump, and the jump time is more than or equal to the shortest continuous load operation time interval/> T And identifying the j-1 th row to load the k-1 th strip image acquisition end row, completing the k-1 th strip segmentation, starting to separate the data of each row of the k-th strip and marking as g k (i, j) when the j-th row is the k-th strip image acquisition start row, and transmitting the load data row by row in a TCP data stream mode to execute step S130.
In operation S130, in the process of dividing the band image, the gray value frequency of each line of data of each probe in the band image is counted.
In the embodiment of the present invention, the j-th line data of the k-th stripe sent in step S120 is received in real time by using the TCP data stream. The statistics of the gray value frequency P kj (i, g) of the j-th line data in the stripe image includes operations S131-S133.
In operation S131, it is determined whether the original code stream data corresponding to each line of the stripe image has an error code or a frame loss phenomenon.
Specifically, whether the error code and the frame loss phenomenon exist in the process of error correction processing, CRC (cyclic redundancy check) and virtual channel separation and channel combination of the data frame of the original code stream corresponding to the j-th data is judged.
In operation S132, if the original code stream data corresponding to the j-th line of the stripe image has an error code or a frame loss phenomenon, the j-th line of data does not participate in statistics, and j represents the data line number of the stripe image.
In operation S133, the normal probe is subjected to boundary processing, and the gray value frequency P kj (i, g) at which the original gray value g of the ith probe in the jth line appears in the stripe image is calculated, i representing the number of the probe, and g representing the original gray value.
In the embodiment of the invention, if the ith probe element is a bad probe element, the ith probe element of the frequency is not involved in statistics, and if the bad probe element is not considered in j rows of data and the original gray value is abnormal, the j rows of data are not involved in statistics. The gray value frequency can be expressed as:
When the ith probe element is a bad probe element, recording P kj (i, g) =0, wherein the ith probe element does not participate in gray value frequency statistics; if the bad probe is not considered in the j-th line data, when the original gray value of the load image is more than or equal to 2 n -1 due to the overexposure or error code of the j-th line data of the stripe image, an upper boundary exists in the image corresponding to the j-th line data, and P kj (i, g) =0 is recorded to participate in gray value frequency statistics; when the j-th data has an error code or the ground object reflectivity is low, and the original gray value of the load image is smaller than or equal to 0, the image corresponding to the j-th data has a lower boundary, and P kj (i, g) =0 is recorded to participate in gray value frequency statistics.
In operation S140, a total frequency of occurrence of the original gray value of each of the probe elements in the strip image is generated and stored.
In the embodiment of the invention, the initial working parameters also comprise the minimum number of lines of the formatted data of the payload N And load shortest operation duration/> st The generating and storing of the total frequency of the occurrence of the original gray value of each probe element in the stripe image includes operations S141 to S144.
In operation S141, a K-th stripe image total acquisition time length K t is calculated:
Kt=TkN-Tk1。
Wherein N is the effective total number of the images of the kth stripe image; t kN is the effective data acquisition time of the last line of the k-th strip image acquisition of the load; t k1 is the load kth stripe image acquisition first line valid data acquisition time.
In operation S142, if the total line number of the kth stripe image is smaller than the minimum line number of the payload formatted data or the duration of the kth stripe image is shorter than the minimum payload duration, i.e. K t < "> st Or N N The method is characterized in that the method comprises the steps that the analysis of the band image data is stopped when the original band image data finally obtained after the kth band image data are transmitted in a star mode and processed by original data is insufficient in one effective statistics unit due to signal interference, equipment failure and the like, and the total frequency of occurrence of the original gray value of each probe element in the kth band image is not counted.
In operation S143, when K t. Gtoreq st Or N is greater than or equal to/> N Counting the total frequency P k of occurrence of the original gray value of each probe element of the original gray image obtained at the kth time, wherein:
。
In operation S144, the total frequency P k at which the original gray value of each of the probe elements of the kth band image data of the load appears is stored in the form of a file according to the load name, the band, the gain, the number of integration stages, and the band image data acquisition time.
The file Size size=i2n 32; If the number of CCD (charge coupled device) probe elements is 10000, the quantization bit number is 11, and when the gray value of a certain probe element is 2 32 -1 at most in a certain strip, each strip can store the gray characteristic of the strip only by 78.2MB, so that the occupied storage size is greatly reduced; meanwhile, the statistics process is synchronous statistics in the output process of decompressed formatted data, and extra computing resources are not required to be occupied.
In operation S150, it is automatically determined whether the initial relative radiation correction coefficient is applicable to the band image.
In the embodiment of the invention, assuming that the ground objects covered by m strips, which are continuously acquired by the same gain and integral stage and have the same load, are uniform ground objects, calculating the relative radiation correction precision of the total adjacent probe elements of the m strips and the relative radiation correction precision of the full field of view to judge whether the current radiation calibration coefficient is suitable for strip data. m represents the least number of statistical analysis strips of the load data. Operation S150 includes operations S151-S156.
In operation S151, the total frequency P of occurrence of the original gray value of each of m strips of the same band, gain, integration level as the kth strip in the kth-m 1 th strip image to the kth strip image is calculated, m represents the minimum statistical analysis strip number of the load data, k represents the sequential encoding of the strip images, and m1 represents the number of strip images selected before the kth strip image. The calculation formula of the frequency P of each probe element in the k-m 1-k stripe images is as follows:
in operation S152, the relative emissivity which is matched with the load name, band, gain, integral series of the kth stripe image and is most adjacent to the acquisition time of the kth stripe image is used.
In the embodiment of the present invention, in the initial state, the relative emissivity R h (i, g) is set in the initial operating parameters.
In operation S153, the total frequency P of occurrence of the original gray value of each of the m stripe images is relatively corrected based on the relative radiation coefficient R h (i, g), so as to obtain the relative radiation corrected gray value frequency of each of the m stripe images.
The relative radiation corrected gray value frequency for each of the probe cells after the m stripes have undergone relative radiation correction can be expressed as:
，
Where G represents a relative radiation corrected gray value.
In operation S154, the relative radiation correction accuracy between adjacent probe cells is calculated based on the relative radiation correction gray value frequency.
Operation S154 includes S1541-S1543.
In operation S1541, a relative radiation corrected average gray value AVG i of each of the m strip images is calculated based on the relative radiation corrected gray value frequency, wherein:
。
In operation S1542, the average gray-scale difference after the relative radiation correction of the adjacent probe cells is calculated based on the average gray-scale value after the relative radiation correction of each probe cell AVG i, bad probe elements do not participate in the calculation of the relative radiation correction accuracy of adjacent probe elements.
Wherein CCD bad represents the bad set of probe elements.
If there is a probe element i present,AVGi-1≥/> bad And/>AVGi≥/> bad，/> bad And (3) representing a bad probe element identification threshold, namely adding the ith probe element into a bad probe element set if the average gray level difference after the relative radiation correction of the ith probe element and the i-1 th probe element is larger than the bad probe element identification threshold and the average gray level difference after the relative radiation correction of the ith probe element and the i+1 th probe element is larger than the bad probe element identification threshold.
In operation S1543, the maximum average gray-scale difference is takenAVG as relative radiation correction accuracy, wherein:
。
in operation S155, the full field-of-view relative radiation correction accuracy is calculated.
Operation S155 includes S1551 through S1552.
In operation S1551, a probe mean gray value after removing bad probe is calculated based on the average gray value after correcting each probe relative to radiation in the strip imageThe calculation formula can be expressed as:
Wherein Num (CCD bad) is the number of bad probe elements, i.e. the bad probe elements do not participate in calculating the average gray value after the relative radiation correction of m stripes.
In operation S1552, a formula and a probe mean gray value are calculated based on a preset full field-of-view relative radiation correction accuracyThe full field relative radiation correction accuracy G r is calculated.
。
In operation S156, if the relative radiation correction accuracy between the adjacent probe elements or the full field of view relative radiation correction accuracy does not meet the index requirement, the relative radiation coefficient is not suitable for the kth stripe image.
If it isAVG＜/> DN And G r<RDN, i.e. accuracy of relative radiation correction of adjacent probe elements/>The ACG and the full field-of-view relative radiation correction accuracy G r both meet the index requirement, and the relative radiation coefficient nearest to the kth stripe acquisition time is applicable to the kth stripe data.
If it isAVG≥/> DN Or G r≥RDN, i.e. accuracy of relative radiation correction of adjacent probe elements/>The ACG or the full field-of-view relative radiation correction accuracy G r both meet the index requirement, and the relative radiation coefficient nearest to the kth stripe acquisition time is not applicable to the kth stripe data;
In addition, if DN>/>AVG≥/> WDN Or R DN＞Gr≥RWDN, i.e. the relative radiation correction accuracy between adjacent probe elements/>And if the ACG or the full-view-field relative radiation correction precision G r meets the index requirement and reaches the alarm threshold, alarming and judging that the relative radiation coefficient is not suitable for the kth stripe image.
In operation S160, if the initial relative radiation correction coefficient is not suitable, a new relative radiation correction coefficient is automatically rescaled.
Operation S160 includes S161-S164.
In operation S161, an original gray value probability distribution S (i, g) and an original gray value cumulative probability distribution S (i, g) for each of m strip images are automatically calculated, m represents a minimum statistical analysis strip number of the load data, and k represents sequential encoding of the strip images.
In operation S162, a load integrated cumulative probability distribution F (g) is calculated based on the original gray value cumulative probability distribution.
In operation S163, a new relative radiation correction coefficient of the kth band image is generated according to the original gray value cumulative probability distribution S (i, g) and the load integrated cumulative probability distribution F (g) of each probe cell.
In operation S164, the relative radiation correction coefficients are verified, and the verified relative radiation correction coefficients are stored by load name, band, gain, integration progression, and calibration time.
If the verification is not passed, the relative radiation correction coefficient is regenerated again. Recalculating the relative radiation correction precision of the adjacent probe elements and the relative radiation correction precision of the full field of view by adopting the newly generated relative radiation correction coefficients; if the relative radiation correction precision of the adjacent probe elements and the relative radiation correction precision of the full view field meet the index requirements, the relative radiation correction coefficient passes verification; if the relative radiation correction coefficient fails to pass the verification, the value of m is increased or any number of conditional images in the first k stripe images are manually selected, and the relative radiation correction coefficient is recalculated.
Calculating P (i, g) according to the step S151 again by automatically increasing m, and calculating and verifying according to the steps S152-155 based on P (i, g) again to generate a relative radiation correction coefficient; m=m+m，/>M is the step size per increment of m,/>M is more than or equal to 1; alarming and supporting to manually select any one of k strips, calculating P (i, g) according to the step S151, and recalculating and verifying according to the steps S152-155 based on P (i, g) to generate a relative radiation correction coefficient; aiming at S156 relative radiation correction precision abnormal alarming, the method supports manual generation and verification of new relative radiation correction coefficients in advance according to steps S152-S155, simultaneously supports manual selection of any one of k strips, recalculates P (i, g), and recalculates based on P (i, g) to generate the relative radiation correction coefficients.
In operation S170, a relative radiation correction is performed on the band image based on the relative radiation correction coefficient.
According to the actual demands of users, whether the current relative radiation correction coefficient meets the precision index or not, a new calibration coefficient can be generated by manual driving according to the method of the step S150. The kth imaging data G k (i, j) is subjected to relative radiation correction using a relative radiation correction coefficient which is matched with the load name, the wave band, the gain, the integral series and is nearest to the acquisition time, and a corrected image G k (i, j) is acquired, wherein:
。
The adaptive optical remote sensing image relative radiation correction method provided by the embodiment of the invention utilizes a streaming data processing technology, adopts a small amount of calculation and storage resources, can automatically select samples, performs indiscriminate statistical analysis, realizes quick coverage and statistical analysis of various samples, and removes the influence of human factors. Wherein automatically picking samples comprises:
In operation S120, when sample data to be statistically analyzed is extracted, j-1 time of the j-th row is determined to be valid by analyzing the load data of each row subjected to decompression formatting processing, and T j-Tj-1 is not less than T Starting a task to be counted and analyzed once;
In operation S130, performing boundary processing and statistics on the sample data, and when the i-th probe is a bad probe, the probe does not participate in statistical analysis; when the original gray value of the load image is more than or equal to 2 n -1 due to image overexposure or error code of the jth line of data of the sample, the line of image data has an upper boundary and does not participate in statistical analysis; when the j-th line has error codes or the reflectivity of the ground object is low, and the original gray value of the load image is smaller than or equal to 0, the line of image data has a lower boundary and does not participate in statistical analysis;
In operation S140, it is determined whether the sample data subjected to the boundary processing is a valid sample: the valid sample acquisition time and the valid data acquisition time are respectively greater than the payload shortest data acquisition time st And least number of valid data lines/> N Counting the total frequency P k of each original grey value of the sample, otherwise, terminating the subsequent statistical analysis task; and storing the total frequency P k of each original gray value of each probe element of the kth time of the load strip sample data according to the load name, the wave band, the gain, the integral series and the strip data acquisition time in the form of a file.
The adaptive optical remote sensing image relative radiation correction method provided by the embodiment of the invention introduces the detection method of the relative radiation correction precision of adjacent probe elements and the relative radiation correction precision of the full field of view of the load, carries out statistical analysis on the relative radiation correction result in a period of time, can automatically monitor the adaptability of the relative radiation correction coefficient, and comprises the following steps:
in operation S151, the k-th frequency data acquisition time is taken as an end reference, and a total frequency P of each original gray value of each probe element in a continuous period of time is calculated;
in operations S152 to 153, obtaining a relative radiation correction coefficient R h (i, g) nearest to the k-th frequency data obtaining time, and calculating a total frequency C of the corresponding probe after the relative radiation correction according to the total frequency P of the original gray value of each probe;
In operation S1541, an average gray value AVG i after the relative radiation correction of each probe element is calculated;
In operation S1542, the average gray-scale difference after the adjacent probe element relative radiation correction is calculated AVG i, when/>AVGi-1≥ bad And/>AVGi≥/> bad I.e. the gray level difference between the ith probe element and the i-1 th probe element as well as between the ith probe element and the (i+1) th probe element is larger than the bad probe element identification threshold value, and the ith probe element is the bad probe element;
According to S156, when adjacent probe cells are relatively corrected for radiation AVG＜/> DN Based on the relative radiation correction coefficient R h (i, g), the relative radiation correction precision of m adjacent probe elements of the k-m 1-k strip images, which are acquired by the same working parameters as the k-th strip image, is not satisfied, which indicates that R h (i, g) is not suitable for the k-th strip data and needs to carry out relative radiation calibration again;
According to S156, when the full field-of-view relative radiation accuracy G r≥RDN, that is, the full field-of-view relative radiation correction accuracy of m stripe acquired by the same working parameters as the kth stripe image in the kth-m 1-k stripe images based on the relative radiation correction coefficient R h (i, G) does not meet the requirement, it is indicated that R h (i, G) is not suitable for the kth stripe data, and the relative radiation calibration needs to be performed again.
The adaptive optical remote sensing image relative radiation correction method provided by the embodiment of the invention can be based on the relative radiation coefficient adaptive monitoring result, can automatically produce relative radiation correction coefficients and deploy, supports subsequent relative radiation correction, and comprises the following steps:
In operation S161, an original gray value probability distribution S (i, g) and an accumulated probability distribution S (i, g) for each of the probe cells are automatically calculated;
in operation S162, a load comprehensive cumulative probability distribution F (g) is calculated from the raw gray value probability distribution of each probe cell;
in operation S163, a histogram matching algorithm is adopted to generate a new relative radiation scaling coefficient according to the original gray value cumulative probability distribution S (i, g) and the load comprehensive cumulative probability distribution F (g) of each probe element;
In operation S164, automatically verifying the relative radiation correction system and storing the relative radiation correction coefficient according to the load name, the band, the gain, the integration series, and the calibration time;
In operation S170, the relative radiation correction coefficient is automatically matched to perform relative radiation correction.
Those skilled in the art will appreciate that the features recited in the various embodiments of the invention can be combined in a variety of combinations and/or combinations, even if such combinations or combinations are not explicitly recited in the present invention. In particular, the features recited in the various embodiments of the invention can be combined and/or combined in various ways without departing from the spirit and teachings of the invention. All such combinations and/or combinations fall within the scope of the invention.
While the invention has been shown and described with reference to certain exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended embodiments and equivalents thereof. Thus, the scope of the invention should not be limited to the embodiments described above, but should be determined not only by the appended embodiments, but also by equivalents of the appended embodiments.
Claims (11)
1. The method for correcting the relative radiation of the self-adaptive optical remote sensing image is characterized by comprising the following steps of:
configuring initial working parameters, wherein the initial working parameters comprise initial relative radiation correction coefficients;
Acquiring an original image of remote sensing detection, and dividing the original image into strip images;
In the process of dividing the strip image, counting the gray value frequency of each line of data of each probe element in the strip image;
Generating and storing the total frequency of the occurrence of the original gray value of each probe element in the strip image;
automatically determining whether the initial relative radiation correction factor is applicable to the banding image;
If the initial relative radiation correction coefficient is not suitable, automatically recalibrating to generate a new relative radiation correction coefficient;
And carrying out relative radiation correction on the band image based on the relative radiation correction coefficient.
2. The method of claim 1, wherein the initial operating parameters further comprise a load continuous operation for a minimum time interval, wherein the acquiring the original image of the remote sensing and dividing the original image into the strip images comprises:
obtaining the decompressed and formatted original image according to a data stream or file mode;
synchronously analyzing the auxiliary information of the original image to obtain the acquisition time of each line of image data in the original image;
When the time difference of the acquisition time of two lines of continuous image data is larger than the shortest time interval of continuous load operation, dividing the two lines of image data with the time difference larger than the shortest time interval of continuous load operation into different strip images.
3. The method of claim 1, wherein said counting the gray value frequency of each line of data for each probe element in the strip image comprises:
Judging whether the original code stream data corresponding to each line of data of the strip image has error codes or frame loss;
if the original code stream data corresponding to the j-th data of the stripe image has error codes or frame loss phenomena, the j-th data does not participate in statistics, and j represents the data line number of the stripe image;
Carrying out boundary processing on normal probe elements, and calculating gray value frequency P kj (i, g) of the original gray value g of the ith probe element of the jth row in the strip image, wherein i represents the number of the probe element, and g represents the original gray value;
when the ith probe element is a bad probe element, recording P kj (i, g) =0; when the original gray value of the load image is greater than or equal to 2 n -1 due to image overexposure or error code of the jth line data of the stripe image, an upper boundary exists in the image corresponding to the jth line data, and P kj (i, g) =0 is recorded; when the error code exists in the j-th data or the reflectivity of the ground object is low, and the original gray value of the load image is smaller than or equal to 0, a lower boundary exists in the image corresponding to the j-th data, and P kj (i, g) =0 is recorded.
4. The method of claim 1, wherein the initial operating parameters further comprise a minimum number of rows of payload formatting data and a minimum length of payload operation, and wherein generating and saving the total frequency of occurrence of the original gray value for each of the probe elements in the strip image comprises:
calculating the total frequency of the occurrence of the original gray value of each probe element in each strip image:
If the total line number of the kth stripe image is smaller than the minimum line number of the effective load formatted data or the duration of the kth stripe image is shorter than the minimum working duration of the load, the total frequency of the occurrence of the original gray value of each probe element in the kth stripe image is not counted.
5. The method of claim 1, wherein automatically determining whether a current relative radiation correction factor is applicable to the banding image comprises:
Calculating the total frequency of occurrence of the original gray value of each probe element of m strips obtained by the same working parameters as the kth strip image in the kth-m 1 strip images to the kth strip image, wherein m represents the least statistical analysis strip number of load data, k represents the sequential coding of the strip images, and m1 represents the number of the strip images selected before the kth strip image;
Matching the load name, the wave band, the gain and the integral series of the kth stripe image, and obtaining the nearest relative radiation coefficient of the kth stripe image;
carrying out relative correction on the total frequency of the occurrence of the original gray value of each probe element in the m strip images based on the relative radiation coefficients to obtain the relative radiation correction gray value frequency of each probe element;
calculating the relative radiation correction precision between adjacent probe elements based on the relative radiation correction gray value frequency;
Calculating the relative radiation correction precision of the full view field;
if the relative radiation correction accuracy between adjacent probe elements or the full field-of-view relative radiation correction accuracy does not meet the index requirement, the relative radiation coefficient is not suitable for the kth stripe image.
6. The method of claim 5, wherein said calculating relative radiation correction accuracy between adjacent probe elements based on said relative radiation correction gray value frequency comprises:
Calculating an average gray value of each probe element in the m strip images after relative radiation correction based on the relative radiation correction gray value frequency;
Based on the average gray value of each probe element after the relative radiation correction, calculating the average gray difference of adjacent probe elements after the relative radiation correction, wherein the bad probe elements do not participate in the calculation of the relative radiation correction precision of the adjacent probe elements;
and taking the maximum average gray level difference as the relative radiation correction precision.
7. The method of claim 6, wherein the initial operating parameters further comprise bad-probe-element identification thresholds, and wherein calculating the relative radiation correction accuracy of adjacent probe elements further comprises:
If the average gray level difference after the relative radiation correction of the ith probe element and the (i-1) th probe element is larger than the bad probe element identification threshold value and the average gray level difference after the relative radiation correction of the ith probe element and the (i+1) th probe element is larger than the bad probe element identification threshold value, adding the ith probe element into a bad probe element set.
8. The method of claim 5, wherein the calculating the full field of view relative radiation correction accuracy comprises:
Calculating the average gray value of the probe cells after the bad probe cells are removed based on the average gray value of each probe cell in the strip image after the relative radiation correction;
And calculating the full-view-field relative radiation correction precision based on a preset full-view-field relative radiation correction precision calculation formula and the average gray value of the probe element.
9. The method of claim 5, wherein the method further comprises:
And if the relative radiation correction precision between adjacent probe elements or the full-view field relative radiation correction precision meets the index requirement and reaches an alarm threshold, alarming and judging that the relative radiation coefficient is not suitable for the kth stripe image.
10. The method of claim 1, wherein the automatically rescaling to generate new relative radiation correction factors comprises:
Automatically calculating the probability distribution of the original gray value and the cumulative probability distribution of the original gray value of each probe element in m strip images, wherein m represents the least statistical analysis strip number of load data, and k represents the sequential coding of the strip images;
calculating a load comprehensive cumulative probability distribution based on the original gray value cumulative probability distribution;
generating a new relative radiation correction coefficient of a kth stripe image according to the original gray value cumulative probability distribution and the load comprehensive cumulative probability distribution of each probe element;
and verifying the relative radiation correction coefficient, and storing the relative radiation correction coefficient passing verification according to the load name, the wave band, the gain, the integral series and the calibration time.
11. The method of claim 10, wherein said validating said relative radiation correction factor further comprises:
Recalculating the relative radiation correction precision of the adjacent probe elements and the relative radiation correction precision of the full field of view by adopting the newly generated relative radiation correction coefficients;
If the relative radiation correction precision of the adjacent probe elements and the relative radiation correction precision of the full view field meet the index requirements, verifying the relative radiation correction coefficient;
If the relative radiation correction coefficient fails to pass the verification, increasing the value of m or manually selecting any number of conditional images in the first k stripe images, and recalculating the relative radiation correction coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410558219.7A CN118134820A (en) | 2024-05-08 | 2024-05-08 | Self-adaptive optical remote sensing image relative radiation correction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410558219.7A CN118134820A (en) | 2024-05-08 | 2024-05-08 | Self-adaptive optical remote sensing image relative radiation correction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118134820A true CN118134820A (en) | 2024-06-04 |
Family
ID=91238121
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410558219.7A Pending CN118134820A (en) | 2024-05-08 | 2024-05-08 | Self-adaptive optical remote sensing image relative radiation correction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118134820A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140146935A1 (en) * | 2011-04-01 | 2014-05-29 | Siemens Aktiengesellschaft | Method and device for correcting artefacts during x-ray imagery, especially computer tomography, with a moving modulator field |
| CN104050643A (en) * | 2014-06-18 | 2014-09-17 | 华中师范大学 | Remote-sensing image relative correction method and system integrating geometry and radiation |
| CN113469899A (en) * | 2021-06-04 | 2021-10-01 | 中国资源卫星应用中心 | Optical remote sensing satellite relative radiation correction method based on radiant energy reconstruction |
| CN116071640A (en) * | 2023-02-17 | 2023-05-05 | 自然资源部国土卫星遥感应用中心 | Hyperspectral satellite remote sensing image radiation quality evaluation method based on noise factors |
-
2024
- 2024-05-08 CN CN202410558219.7A patent/CN118134820A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140146935A1 (en) * | 2011-04-01 | 2014-05-29 | Siemens Aktiengesellschaft | Method and device for correcting artefacts during x-ray imagery, especially computer tomography, with a moving modulator field |
| CN104050643A (en) * | 2014-06-18 | 2014-09-17 | 华中师范大学 | Remote-sensing image relative correction method and system integrating geometry and radiation |
| CN113469899A (en) * | 2021-06-04 | 2021-10-01 | 中国资源卫星应用中心 | Optical remote sensing satellite relative radiation correction method based on radiant energy reconstruction |
| CN116071640A (en) * | 2023-02-17 | 2023-05-05 | 自然资源部国土卫星遥感应用中心 | Hyperspectral satellite remote sensing image radiation quality evaluation method based on noise factors |
Non-Patent Citations (1)
| Title |
|---|
| 许和鱼 等: "基于太阳漫反射板线阵CCD相对辐射定标方法研究", 光学学报, 11 December 2019 (2019-12-11), pages 1 - 15 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109359270B (en) | Threshold model establishing method for integrity risk monitoring of Beidou foundation enhancement system | |
| CN112016674B (en) | Knowledge distillation-based convolutional neural network quantification method | |
| US20030176974A1 (en) | Data management for seismic acquisition using variable compression ratio as a function of background noise | |
| US6625318B1 (en) | Robust sequential approach in detecting defective pixels within an image sensor | |
| US8866645B2 (en) | Method and apparatus for compression of generalized sensor data | |
| CN109525337B (en) | WiFi fingerprint acquisition method, device, storage medium and equipment | |
| US20030103676A1 (en) | Data compression method and recording medium with data compression program recorded therein | |
| CN113723338A (en) | Sensor abnormality detection method, sensor abnormality detection device, and computer-readable storage medium | |
| US20120082393A1 (en) | Attitude estimation with compressive sampling of starfield data | |
| Lesaffre et al. | Objective determination of snow-grain characteristics from images | |
| CN113642447B (en) | Monitoring image vehicle detection method and system based on convolutional neural network cascade | |
| CN118134820A (en) | Self-adaptive optical remote sensing image relative radiation correction method | |
| CN112101313B (en) | Machine room robot inspection method and system | |
| CN112906883A (en) | Hybrid precision quantization strategy determination method and system for deep neural network | |
| CN111951611A (en) | ADS-B weak signal detection device and method based on multi-feature fusion | |
| CN111581407B (en) | Method, device and medium for constructing global geographic positioning reference image database | |
| CN115601293A (en) | Object detection method and device, electronic equipment and readable storage medium | |
| US7392161B2 (en) | Identifying a state of a system using an artificial neural network generated model | |
| CN112700007A (en) | Training method, forecasting method and device of ionosphere electron content forecasting model | |
| Qian et al. | Quantitative evaluation of hyperspectral data compressed by near lossless onboard compression techniques | |
| US20040022317A1 (en) | Digital video processing method and apparatus thereof | |
| CN103268593B (en) | The separation method of signal and noise in a kind of target in hyperspectral remotely sensed image | |
| JP4214477B2 (en) | Signal restoring method, program therefor, recording medium, signal restoring device, and receiving station and receiving system including the same | |
| CN112184570A (en) | Method for detecting response state change of probe element based on relative radiation correction coefficient | |
| CN116935227B (en) | Crop yield analysis method, system and medium based on satellite remote sensing image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |