Satellite data extraction, classification and recombination and camera image combination method

Abstract

The invention discloses a satellite data extraction, classification and recombination and camera image combination method, which comprises the following specific operation steps: (1) acquiring satellite data transmission data, and extracting VCDU data field data from the satellite data transmission data according to a preset data rule; eliminating filling frame data in the VCDU data field data, and extracting effective data field data; (2) classifying the effective data fields according to application process identifiers and frequency bands, and respectively forming S-frequency band effective data field data and X-frequency band effective data field data; (3) the S-frequency-band effective data domain data and the X-frequency-band effective data domain data are classified, recombined, packaged and output effective data of all systems according to application process identifiers and system classifications; wherein each system valid data comprises camera system valid data, and packaging the camera system valid data according to a preset rule and outputting a picture format. The method has the beneficial effects that the data processing service capability and the data processing efficiency are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field related to satellite data processing, and in particular to a method for extracting, classifying, reorganizing and camera grouping satellite data. Background Art

[0002] During the ground development phase of the satellite, data will be transmitted using different devices according to the frequency band. Some devices can only save the original data and cannot extract the valid data domain, which leads to the problem of increased data processing business and reduced processing efficiency. The existing technology can receive and process real-time satellite telemetry data, but it cannot efficiently identify, classify and reorganize the original data of multiple frequency bands and the customized data formats of each system. When encountering data such as camera system image data that requires pixel reorganization according to rules, existing software cannot directly process satellite data, and it is necessary to delete non-image information data from the satellite data before grouping and outputting the image. Summary of the invention

[0003] The present invention aims to overcome the above-mentioned deficiencies in the prior art and provides a method for extracting, classifying and reorganizing satellite data and forming camera images with high processing efficiency.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: A method for extracting, classifying and reorganizing satellite data and camera grouping, the specific operation steps are as follows: (1) Acquire satellite data transmission data, extract VCDU data domain data from the satellite data transmission data according to preset data rules; remove fill frame data in the VCDU data domain data, and extract valid data domain data; (2) classifying the valid data field by frequency band according to the application process identifier to form S-band valid data field data and X-band valid data field data respectively; (3) The S-band valid data domain data and the X-band valid data domain data are reorganized and packaged according to the system classification based on the application process identifier, and the valid data of each system is output; wherein the valid data of each system includes the valid data of the camera system, and the valid data of the camera system is packaged and output in a picture format according to a preset rule.

[0005] The present disclosure provides a satellite data extraction, classification, reorganization and camera grouping method, extracts VCDU data domain data from satellite data transmission data, identifies and classifies system data from VCDU data domain data, and can group and output camera image data according to preset rules, greatly improving data processing business capabilities and data processing efficiency.

[0006] Preferably, in step (1), the satellite telemetry data consists of the data in the VCDU data field and the rest. The rest contains virtual channel identification information, while the data in the VCDU data field contains the E-PDU valid data field data and the header information. The specific operation method for extracting the data in the VCDU data field from the satellite telemetry data is as follows: Arrange the satellite telemetry data matrix according to the transmission frame format, perform format clipping on the arranged matrix data, and the data in the VCDU data field can be obtained without indexing.

[0007] Preferably, in step (1), the preset data rule for obtaining the VCDU data field is customized according to the data format of the output device. By default, it is 1024 bytes in the satellite data format, and the 42nd byte to the 894th byte of each row are extracted; specifically: the satellite data framing format is 1 frame of 1024 bytes, and the data is formed into a matrix with 1024 bytes per row. According to the satellite transmission frame format, the 42nd byte to the 894th byte of each row are extracted, totaling 852 bytes. This part is the data area; then, for these 852 bytes, application process identifier recognition and header recognition are performed according to the preset data rule; after recognition, the filled frame data content is removed, and the data is packed into a new data table according to the corresponding application process identifier.

[0008] Preferably, in step (2), since the satellite transmission frame format contains virtual channel identifiers, and different satellites have different configuration methods for virtual channel identifiers, in this method, a new dictionary is created to store flag_id, and the virtual channel identifier value can be preset according to requirements. The virtual channel identifier recognition differentiates different systems. At the same time, the S-band effective data field data and the X-band effective data field data are formed according to the two data transmission channels of S-band low-speed downlink telemetry and X-band high-speed downlink telemetry in the virtual channel identifier.

[0009] Preferably, corresponding indexes are set and fields are defined in the application process identifier, where the fields are used to represent the data belonging frequency band and the data belonging system; corresponding indexes are set and fields are defined in the header, where the fields are used to represent the data valid length; corresponding indexes are set and fields are defined in the valid data of the camera system, where the fields are used to represent the camera manufacturer, the camera bit depth size, the picture size, the pixel recombination rule, and the picture auxiliary data format.

[0010] Preferably, in step (3), the E-PDU valid data field data in the 852-byte data area is 836 bytes of data and the header information is 16 bytes of data; in each frame of the 836-byte data field, the first packet of data will have a header identification word, which is matched with the packet length of the corresponding header subsystem after identification, and the data is extracted according to the length to output this packet of data; the system classification and reorganization is specifically as follows: confirming the system to which the data belongs according to the application process identifier, and after obtaining the reorganized 836-byte data table, reconfirming the system header to which the data belongs, and after confirming the match, extracting the corresponding length data according to the packet length information and storing it in a new data table, and after processing the matrix data in sequence, obtaining the final data.

[0011] Preferably, in step (3), in addition to the above-mentioned application system identifier, the camera system also has two rows of auxiliary data. After obtaining the 836-byte camera system data table according to the above method, the camera auxiliary data content is identified, including the camera identifier and the camera bit depth identifier. Different cameras calculate the image size according to the bit depth. First, the camera image size is preset in the packet length information. If the bit depth is 8 bits, it is directly output according to the preset size. If the bit depth is 10 bits or 14 bits, the preset size is enlarged by 2 times according to the pixel reorganization rule and output.

[0012] The beneficial effects of the present invention are: extracting VCDU data domain data from satellite data transmission data, identifying and classifying system data from the VCDU data domain data, and being able to group and output camera image data according to preset rules, thereby greatly improving data processing business capabilities and data processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a flow chart of the method of the present invention; Figure 2 This is a schematic diagram of the arrangement of visible light camera sensors. DETAILED DESCRIPTION

[0014] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0015] like Figure 1 In the embodiment described above, a method for extracting, classifying and reorganizing satellite data and forming camera images has the following specific steps: (1) Acquire satellite data transmission data, extract VCDU data domain data from the satellite data transmission data according to preset data rules; remove fill frame data in the VCDU data domain data, and extract valid data domain data; The satellite telemetry data consists of the data in the VCDU data field and the rest. The rest contains virtual channel identification information, and the data in the VCDU data field contains the E-PDU valid data field data and the header information. The specific operation method for extracting the data in the VCDU data field from the satellite telemetry data is as follows: Arrange the satellite telemetry data matrix according to the transmission frame format, perform format clipping on the arranged matrix data, and the data in the VCDU data field can be obtained without indexing.

[0016] The general format of the satellite transmission frame is as follows:

[0017] The virtual signal identifier is used to distinguish different systems such as cameras, infrared earth sensors, digital sun sensors, etc., as follows:

[0018] The content format of the data area is as follows. Each frame of the E-PDU data field has 836 bytes of valid data. This method is to extract this part of the content for splicing and output. The data type can be further identified and distinguished according to the application process identifier.

[0019] The application process identifier is shown as follows:

[0020] The preset data rule for obtaining the VCDU data field is customized according to the data format of the output device. By default, it is 1024 bytes in the satellite data format, and the 42nd byte to the 894th byte of each row are extracted; the output data of different devices may be different, and the corresponding length rule and the starting rule of the data field can be configured.

[0021] Specifically: The satellite data framing format is 1 frame of 1024 bytes. The data is formed into a matrix with 1024 bytes per row. According to the satellite transmission frame format, the 42nd byte to the 894th byte of each row are extracted, a total of 852 bytes. This part is the data area; then, for these 852 bytes, the application process identifier and the header are identified according to the preset data rule; after identification, the filled frame data content is removed, and the data is packed into a new data table according to the corresponding application process identifier. Among them, the virtual signal identifier will identify the data type, including four categories: subsystem 1, 2, 3S telemetry data, subsystem 1, 2, 3X telemetry data, satellite real-time telemetry, and satellite delayed telemetry; then, it is further subdivided through the application process identifier, mainly for satellite delayed telemetry, identifying subsystem 1, 2, 3 delayed telemetry.

[0022] Set corresponding indexes and define fields in the application process identifier, where the fields are used to characterize the frequency band to which the data belongs and the system to which the data belongs; set corresponding indexes and define fields in the packet header, where the fields are used to characterize the valid length of the data; set corresponding indexes and define fields in the valid data of the camera system, where the fields are used to characterize the manufacturer of the camera, the bit depth size of the camera, the picture size, the pixel recombination rule, and the picture auxiliary data format.

[0023] (2)Classify the valid data domain by frequency band according to the application process identifier, and respectively form the S-band valid data domain data and the X-band valid data domain data; Since the satellite transmission frame format contains virtual channel identifiers, and different satellites have different configuration methods for virtual channel identifiers, this method creates a dictionary to store flag_id, and can preset the virtual channel identifier value according to requirements. The virtual channel identifier is used to identify and distinguish different systems. At the same time, the S-band valid data domain data and the X-band valid data domain data are formed according to the two data transmission channels of S-band low-speed downlink data transmission and X-band high-speed downlink data transmission in the virtual channel identifier.

[0024] (3)Classify and reorganize the S-band valid data domain data and the X-band valid data domain data according to the application process identifier by system and pack and output the valid data of each system; the valid data of each system includes the valid data of the camera system, and the valid data of the camera system is packed and output in the picture format according to the preset rules; In the 852-byte data area, the E-PDU valid data domain data is 836 bytes of data and the packet header information is 16 bytes of data; in each frame of 836-byte data domain, for the first packet of data, there will be a packet header identification word. After identification, match the packet length of the corresponding packet sub-system of the packet header, and extract the data according to this length and output this packet of data; for the E-PDU valid data domain, the length of each frame is different for different systems, and the packet headers are different; this method creates a dictionary to store the FRAME_HEAD packet header and the FRAME_LEN packet length information; the system classification and reorganization is specifically: confirm the system to which the data belongs according to the application process identifier, and after obtaining the reorganized 836-byte data table, reconfirm the packet header of the system to which the data belongs. After confirmation and matching, extract the data of the corresponding length according to the packet length information and store it in a new data table. After processing all the matrix data in turn, the final data is obtained.

[0025] In addition to the above application system identifier, there are two rows of auxiliary data in the camera system. After obtaining the 836-byte camera system data table according to the above method, identify the content of the camera auxiliary data, including the camera identifier and the camera bit depth identifier; different cameras calculate the picture size according to the bit depth. First, preset the camera picture size in the packet length information. For 8-bit bit depth, it is directly output according to the preset size. If it is 10-bit or 14-bit bit depth, the preset size is enlarged by 2 times according to the pixel recombination rule and output.

[0026]

[0027] It should be noted that the part of creating a new dictionary is implemented by software methods. The virtual channel identifiers in the original frame format of satellite transmission are stored in the vflag dictionary; the application process identifiers in the satellite VCDU data area are stored in the appid dictionary; FRAME_HEAD and FRAME_LEN are also dictionaries established according to the protocol, storing the packet header and packet length data; the internal content of the above dictionaries can be changed according to different data protocols, that is, preset rules; the actual satellite transmission of original data may contain data of multiple systems, that is, different identifier data. Therefore, a flag_id determination is added at the software level. After the matching of each identifier is completed, the current subsystem is recorded in flag_id, and at the same time, the data valid flag is set to valid, allowing the software to perform data output and the next round of matching queries; if there is no matching item for the identifier, flag_id is set to NA, and the data valid flag is set to invalid, indicating that there is no valid data in the remaining data, and the matching query is stopped.

[0028] In this embodiment, the satellite data transmission data is directly subjected to matrix recombination and cropping according to the preset data rules to obtain the VCDU data field, where the preset data rules include the original length of each frame of data transmission data and the length of each frame of VCDU data field. Extract the transmission frame data from the satellite data transmission data, including: classifying the VCDU data field data according to the application process identifier in the VCDU data field according to the preset system; among them, the application process identification index includes the data frequency band attribution and the data system attribution; construct the index of each system; among them, the index includes the packet header flag of each system data packet and the data packet length. After identifying the camera system from the application process identification index, identify the picture parameters according to the camera index and extract the picture data; among them, the camera index includes the camera type and the corresponding pixel recombination rule, the picture size, and the picture bit depth. According to the parameters obtained from the camera index, group the extracted picture data and output the picture data using CV2.

[0029] The implementation example is as follows: Perform matrix recombination on the satellite data transmission data in 1000-byte blocks, with the VCDU data field being the 0th to 500th bytes, and crop and retain the 0 - 500th bytes for each row of 1000 bytes.

[0030] Set the data valid flag to True, index the data from the application process identifier library, first index the data of each system in the X frequency band, and then index the data of each system in the S frequency band; if not found, set the data valid flag to False, indicating that this portion of data is all invalid padding frames. This flag is added in the code to determine whether the search for data of a certain system is completed.

[0031] Set the packet header flag to NA, index the S-band data from the S packet header library, index the X-band data from the X packet header library, and set the packet header flag to the system name after successful recognition.

[0032] If it is recognized as a camera payload system, the data needs to be grouped into images; analyze the image parameters in the data according to the camera index, extract a certain length of data according to the photo size combined with the bit depth information, and save it as image data. Taking the format of Camera 1 as an example: The sizes of different cameras vary. Configure them in one-to-one correspondence according to the packet header in advance, and then analyze the auxiliary data after the packet header to identify the bit depth and camera mode. There is a windowing mode in the camera mode. In the windowing mode, the image size is extracted as X*Y, and in other modes, it is extracted as the default size of 9000*7000. The size of a complete full-frame image data of the visible light unit of the camera is 9000×7002 pixels, including 2 rows of auxiliary data rows and 7000 rows of image data rows. At the same time, the visible light unit can achieve windowed imaging with a fixed size. The size of a complete windowed image data is 3000×5002 pixels, including 2 rows of auxiliary data rows and 5000 rows of image data rows. Whether it is a full-frame or windowed image, the first row of auxiliary data row contains 167 pixels of valid auxiliary data, and the remaining part is filled with AAH; the second row of auxiliary data row contains 87 pixels of valid auxiliary data, and the remaining part is filled with 00H.

[0033] The quantization bit number is selectable as 8bit / 12bit (default 8bit). When the pixel bit width is greater than 8bit, the auxiliary data is filled in the lower 8bit, and the high bit is filled with 0. For 8bit quantization bit number, the image space-ground transmission data type is int8, and each pixel occupies one channel; for 12bit quantization bit number, the image space-ground transmission data type is int16, and each pixel's image data is 12bit + 4bit of high-bit filled 0 data = 2Byte, occupying 2 channels.

[0034] Arrange and combine the image data according to the pixel rules in the camera index, use CV2 to perform image optimization such as color adjustment and white balance, and finally output the image.

[0035] The data of different cameras has special arrangement rules. Taking Camera 2 as an example: When actually outputting, 7000 lines of data for each focal plane (a total of 14000 lines of data for two focal planes) are packed and transmitted together with 1 line of attitude and orbit auxiliary data. Among them, the first line is the attitude and orbit auxiliary data, the second to 7001st lines are the data of focal plane 1, and the 7002nd to 14001st lines are the data of focal plane 2. (Each line of auxiliary data is 94 bytes, arranged as 1, 2, 3......92, 93, 94 bytes, and the image data is 9344 bytes, arranged as 1, 2, 3......9342, 9343, 9344 pixels).

[0036] Using COLOR_BG2RGB2 in OPENCV to convert the raw grayscale to RGB color and output a color picture; according to the optical imaging system, the visible light camera imaging needs to be processed by left-right mirroring (in the cross-track direction) to obtain the actual image. The output of the visible light camera is in Bayer format, and the layout of the area array filter is as Figure 2 shown, and the image interpolation calculation can be referred to the filter arrangement for processing.

Claims

1. A method for extracting, classifying and reorganizing satellite data and camera grouping, characterized in that: The specific steps are as follows: (1) Acquire satellite data transmission data, extract VCDU data domain data from the satellite data transmission data according to preset data rules; remove fill frame data in the VCDU data domain data, and extract valid data domain data; (2) classifying the valid data field by frequency band according to the application process identifier to form S-band valid data field data and X-band valid data field data respectively; (3) reorganizing and packaging the S-band valid data domain data and the X-band valid data domain data according to the application process identifier and system classification, and outputting the valid data of each system; The valid data of each system includes the valid data of the camera system, and the valid data of the camera system is packaged and output in a picture format according to preset rules.

2. The method for extracting, classifying, reorganizing and camera grouping satellite data according to claim 1, characterized in that: In step (1), the satellite data transmission data consists of VCDU data domain data and the rest, the rest includes virtual channel identification information, and the VCDU data domain data includes E-PDU valid data domain data and packet header information; the specific operation method for extracting the VCDU data domain data from the satellite data transmission data is: arranging the satellite data transmission data in a matrix according to a transmission frame format, formatting the matrix arrangement data, and obtaining the VCDU data domain data without indexing.

3. A method for extracting, classifying and reorganizing satellite data and grouping images with cameras according to claim 2, characterized in that, in step (1), the preset data rule for obtaining the VCDU data domain is customized according to the data format of the output device, and the satellite data format is 1024 bytes by default, and the 42nd byte to the 894th byte are extracted from each row; specifically: the satellite data framing format is 1 frame of 1024 bytes, and the data is organized into a matrix of 1024 bytes per row, and according to the satellite transmission frame format, the 42nd byte to the 894th byte are extracted from each row, a total of 852 bytes, which is the data area; then, application process identifier identification and packet header identification are performed on these 852 bytes according to the preset data rules; after identification, the fill frame data content is removed, and the data is packaged into a new data table according to the corresponding application process identifier.

4. The method for extracting, classifying, reorganizing and camera grouping satellite data according to claim 3, characterized in that: In step (2), since the satellite transmission frame format includes a virtual channel identifier, and different satellites have different virtual channel identifier configuration methods, this method creates a new dictionary to store flag_id, and the virtual channel identifier value can be preset according to requirements. The virtual channel identifier identifies and distinguishes different systems. At the same time, the virtual channel identifier is composed of S-band effective data domain data and X-band effective data domain data according to the two data transmission channels of S-band low-speed downlink data transmission and X-band high-speed downlink data transmission.

5. The method for extracting, classifying, reorganizing and camera grouping satellite data according to claim 3, characterized in that: The application process identifier sets a corresponding index and defines a field, wherein the field is used to characterize the frequency band to which the data belongs and the system to which the data belongs; sets a corresponding index and defines a field in the packet header, wherein the field is used to characterize the effective length of the data; sets a corresponding index and defines a field in the camera system effective data, wherein the field is used to characterize the camera manufacturer, camera bit depth, image size, pixel reorganization rules, and image auxiliary data format.

6. The method for extracting, classifying, reorganizing and camera grouping satellite data according to claim 5, characterized in that: In step (3), the E-PDU valid data field data in the 852-byte data area is 836 bytes of data and the header information is 16 bytes of data; in each frame of the 836-byte data field, the first packet of data will have a header identification word, which is matched with the packet length of the corresponding header subsystem after identification, and the data is extracted according to the length and output for this packet of data; the system classification and reorganization is specifically as follows: confirm the system to which the data belongs according to the application process identifier, and after obtaining the reorganized 836-byte data table, confirm the system header to which the data belongs for the second time. After confirming the match, extract the corresponding length data according to the packet length information and store it in a new data table. After processing the matrix data in sequence, the final data is obtained.

7. The method for extracting, classifying, reorganizing and camera grouping satellite data according to claim 6, characterized in that: In step (3), in addition to the above-mentioned application system identifier, the camera system also has two lines of auxiliary data. After obtaining the 836-byte camera system data table according to the above method, the camera auxiliary data content is identified, including the camera identifier and the camera bit depth identifier. Different cameras calculate the image size according to the bit depth. First, the camera image size is preset in the packet length information. If the bit depth is 8 bits, it is directly output according to the preset size. If the bit depth is 10 bits or 14 bits, the preset size is enlarged by 2 times according to the pixel reorganization rule and output.

Images