Satellite-based image transmission method, device and storage medium for unmanned field equipment

Abstract

The present invention discloses a satellite-based image transmission method, device and storage medium for unattended field equipment. The method includes: obtaining ionospheric TEC data and spatial distribution information of ionospheric irregularities corresponding to the observation position of the unattended field equipment through a satellite monitoring system, thereby obtaining the frequency dependence and spatiotemporal variation characteristics of ionospheric delay; obtaining the transmission frequency and transmission power of image transmission of the unattended field equipment according to the frequency dependence of ionospheric delay, and determining the image block size and priority of image monitoring data according to the spatiotemporal variation characteristics of ionospheric delay; obtaining and parsing the image blocks of the unattended field equipment in real time, and fusing the parsed multiple image blocks to reconstruct image data. The present invention realizes image transmission of unattended field equipment by combining the analysis of ionospheric delay data, solving the problems of poor image transmission stability in the prior art.

Description

Technical Field

[0001] The present invention relates to field monitoring, and in particular to a satellite-based field unattended equipment image transmission method, device and storage medium. Background Art

[0002] In satellite monitoring systems, ionospheric delay has a significant impact on real-time data transmission and high-precision positioning. In a high electron density environment, the propagation speed of electromagnetic waves in the ionosphere slows down, resulting in increased signal delay and attenuation. This phenomenon seriously affects the reliability of the data link between the satellite and the ground station, and increases the packet loss rate and bit error rate during data transmission. At the same time, the ionospheric delay effect also introduces additional distance measurement errors, and the GPS positioning accuracy decreases accordingly. In tasks that require real-time response and high-precision positioning, such as forest fire monitoring and maritime search and rescue, the above problems will significantly reduce the efficiency and accuracy of the satellite monitoring system, increase the difficulty of identifying and locating emergencies, and slow down the emergency response speed. Therefore, how to ensure the real-time reliable transmission and high-precision positioning of satellite monitoring data under the influence of ionospheric delay has become a key technical problem that needs to be solved urgently, which is of great significance to improving the adaptability and reliability of satellite monitoring systems in complex ionospheric environments. Summary of the invention

[0003] In order to overcome the deficiencies of the prior art, one of the objects of the present invention is to provide a satellite-based image transmission method for unattended outdoor equipment, which can solve the reliability problem of unattended outdoor equipment in the prior art when performing image transmission.

[0004] A second object of the present invention is to provide a satellite-based image transmission device for unattended outdoor equipment, which can solve the reliability problem of unattended outdoor equipment in the prior art when performing image transmission.

[0005] A third object of the present invention is to provide a computer-readable storage medium that can solve the reliability problem of image transmission in field unattended equipment in the prior art.

[0006] One of the purposes of the present invention is achieved by the following technical solution:

[0007] The satellite-based image transmission method for unmanned field equipment includes:

[0008] Delay analysis step: obtaining ionospheric TEC data and spatial distribution information of ionospheric irregularities corresponding to the observation position of the field unattended equipment through a satellite monitoring system, and deriving frequency dependence and spatiotemporal variation characteristics of ionospheric delay based on the ionospheric TEC data and spatial distribution information of ionospheric irregularities;

[0009] The transmission parameter determination step includes: obtaining ionospheric delay data of different frequency bands according to the frequency dependence of the ionospheric delay, and obtaining the transmission frequency and transmission power of the image transmission of the field unattended device in combination with the device power consumption limit information of the field unattended device, and then sending the transmission frequency and transmission power of the image transmission to the field unattended device;

[0010] The image block setting step includes determining the image block size and the image block priority of the image monitoring data according to the spatiotemporal variation characteristics of the ionospheric delay, and sending the image block size and the image block priority to the field unattended device, so that the field unattended device divides the image in the image transmission according to the image block size and the image block priority to form a plurality of image blocks, and sends the divided image blocks according to the transmission frequency and the transmission power;

[0011] Image reconstruction step: acquiring image blocks of the field unattended device in real time, parsing each image block, and fusing the parsed multiple image blocks to reconstruct the image transmitted by the image.

[0012] Further, the image block setting step of determining the image block size and image block priority of the image monitoring data according to the spatiotemporal variation characteristics of the ionospheric delay specifically includes: firstly, analyzing the spatiotemporal variation characteristics of the ionospheric delay to obtain the variation trend of the ionospheric delay at different time scales; then, performing multi-scale decomposition on the spatiotemporal variation characteristics of the ionospheric delay according to the time-frequency analysis method; finally, obtaining the image block size and image block priority according to the variation trend of the ionospheric delay at different time scales, the time-frequency analysis results and the system preset mapping model of the ionospheric delay and the image block; wherein, the time-frequency analysis results include extracting the delay variation characteristics corresponding to different frequency components;

[0013] In the image block setting step, the field unattended device divides the image in the image transmission according to the image block size and the image block priority to form multiple image blocks, which specifically includes: first, the field unattended device decomposes the image data into multiple image levels according to the pyramid decomposition algorithm, and the image data of each image level is provided with a corresponding image scale and image resolution; then, the image data of each image level is divided into image blocks according to the image block size, and each image block is encoded and the image block priority is set; finally, each image block is compressed by a progressive image compression algorithm, so that the transmitted image data generates a multi-scale, multi-bit stream image data stream, and is sent out according to the transmission frequency and transmission power.

[0014] Furthermore, the image reconstruction step specifically includes: first, acquiring the image data stream of the field unattended equipment in real time and analyzing the image data stream to obtain each image block, and obtaining the corresponding image level according to the encoding of each image block; then, reconstructing the image data of the corresponding image level according to the encoding of the image block, the scale and resolution of the corresponding image level; finally, integrating the image data of multiple image levels according to image enhancement and post-processing technology to construct the final image transmission.

[0015] Furthermore, the image reconstruction step further comprises: a judging step: judging whether each acquired image block is damaged by a cyclic check algorithm, and if so, executing a retransmission step; and judging whether image blocks belonging to the same image level are lost, and if so, executing a retransmission step;

[0016] The retransmission step: when the image block corresponding to the image layer is damaged, start the adaptive retransmission request mechanism to send a retransmission instruction to the field unattended device to reacquire the image block; wherein the retransmission instruction includes the encoding of the image block, the image block priority, the number of image block retransmissions, the time deadline, the image layer to which the image block belongs, and the image transmission corresponding to the image layer to which the image block belongs; the time deadline refers to the image block completing the transmission and decoding within the specified time; wherein, the higher the image block priority, the higher the number of image block retransmissions and the shorter the time deadline.

[0017] Furthermore, in the retransmission step: when the number of retransmissions of an image block reaches a preset number of times or the time deadline reaches a preset time, and the image block is still lost or damaged, an image block adjacent to the image block is obtained based on the encoding of the adaptively lost or damaged image block and is used as a reference image block, and then a weight coefficient is set for the texture direction, edge strength and color features of each reference image block according to a preset weight method, and finally the texture direction, edge strength and color features of the image block are estimated by weighted averaging, and then the corresponding image block is constructed according to the texture direction, edge strength and color features of the image block.

[0018] Furthermore, the step of sending the image data stream according to the transmission frequency and transmission power in the image block setting step further includes: acquiring spatial variation gradient information of ionospheric delay according to ionospheric delay data corresponding to the transmission frequency, and then analyzing and obtaining the variation rate and amplitude of ionospheric delay in different regions, so as to obtain the severity of spatial variation of ionospheric delay and form an ionospheric delay gradient map according to the severity of spatial variation of ionospheric delay, and determining the transmission strategy of image data according to the ionospheric delay gradient map and a preset mapping matching model; and finally sending the image data stream according to the transmission strategy of image data;

[0019] Among them, the transmission strategy of image data follows the following principles: when the degree of change of ionospheric delay in the corresponding area is greater than the preset value, the image data stream is transmitted in sequence according to the resolution of the image level from high to low; when the degree of change of ionospheric delay in the corresponding area is less than the preset value, the image data stream is transmitted in sequence according to the resolution of the image level from low to high.

[0020] Furthermore, it also includes: an abnormality reset step: when an abnormality occurs in the field unattended device, a diagnostic instruction is sent to the field unattended device to start the diagnosis and self-recovery operation of the field unattended device; specifically, by setting an abnormality detection algorithm inside the field unattended device, the abnormality detection algorithm is used to determine in real time whether there is an abnormality in the image transmission process. If so, a reset request is sent to the background service center and then the diagnostic instruction is received to start the diagnosis and self-recovery operation; wherein, the abnormalities in the image transmission process include interruption of the image transmission link, retransmission of image blocks, and failure of retransmission of image blocks.

[0021] Furthermore, the delay analysis step specifically includes: obtaining the observation position of the unattended equipment in the field, and using the GPS occultation observation technology to analyze the phase and delay changes of the GPS signal when it passes through the ionosphere, and inverting to obtain the ionospheric TEC data; and using the observation data of multiple GPS satellites to reconstruct the three-dimensional distribution of the ionospheric electron density using a three-dimensional ionospheric tomography algorithm, and then obtaining the spatial distribution information of the ionospheric irregularities;

[0022] The temporal and spatial variation characteristics of ionospheric delay are extracted by performing time series changes on ionospheric TEC data; and the spatial distribution information of ionospheric irregularities is tracked through the three-dimensional ray tracing method to obtain the ray trajectories and propagation delays of electromagnetic waves of different frequency bands when transmitted in the ionosphere, so as to analyze the frequency dependence of ionospheric delay.

[0023] The second object of the present invention is achieved by adopting the following technical solution:

[0024] A satellite-based image transmission device for unattended outdoor equipment is characterized in that it includes a memory and a processor, wherein the memory stores an unattended outdoor satellite image transmission program that runs on the processor, and the unattended outdoor satellite image transmission program is a computer program. When the processor executes the unattended outdoor satellite image transmission program, it implements the steps of the satellite-based unattended outdoor equipment image transmission method adopted as one of the purposes of the present invention.

[0025] The third object of the present invention is achieved by adopting the following technical solution:

[0026] A computer-readable storage medium stores a field unattended satellite image transmission program, which is a computer program. When the field unattended satellite image transmission program is executed by a processor, the steps of a field unattended device image transmission method based on satellite as one of the purposes of the present invention are implemented.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The present invention determines the optimal transmission frequency and transmission power by accurately analyzing the ionospheric TEC data and the spatial distribution of ionospheric irregularities, thereby improving the adaptability and efficiency of the communication link during image data transmission, reducing the negative impact of ionospheric delay, and realizing low-power transmission while reducing the impact of ionospheric delay on signal quality, extending the battery life of unattended equipment and enhancing the stability of the link; the present invention also dynamically adjusts the image blocking strategy according to the change of ionospheric delay to ensure that key image information can be transmitted preferentially under high-delay conditions, improves the real-time nature of the image and the speed of obtaining important information, and improves the real-time nature and efficiency of image data transmission of unattended equipment in the field, which is more conducive to unmanned monitoring in the field. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flow chart of the satellite-based field unattended equipment image transmission method provided by the present invention. DETAILED DESCRIPTION

[0030] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form a new embodiment.

[0031] Embodiment 1

[0032] The present invention provides a satellite-based image transmission method for unmanned field equipment. Figure 1 As shown, including:

[0033] Step S1, obtaining ionospheric TEC data and spatial distribution information of ionospheric irregularities corresponding to the observation position of the field unattended equipment through a satellite monitoring system, and analyzing the ionospheric TEC data and the spatial distribution information of ionospheric irregularities to obtain the frequency dependence and spatiotemporal variation characteristics of ionospheric delay.

[0034] Specifically, this embodiment uses GPS occultation observation technology to analyze the changes in phase and delay of the GPS signal of the satellite monitoring system when it passes through the ionosphere, and then inverts to obtain ionospheric TEC data, and then extracts the spatiotemporal variation characteristics of ionospheric delay by performing time series variation analysis on the ionospheric TEC data. In this embodiment, a time-frequency analysis method such as wavelet transform is used to analyze the time series analysis of the ionospheric TEC data to extract the periodic variation characteristics and trend variation characteristics of the ionospheric TEC data, such as the daily variation characteristics, seasonal variation characteristics and solar activity dependence of the ionospheric TEC data, so as to characterize the spatiotemporal variation characteristics of the ionospheric delay.

[0035] Furthermore, this embodiment obtains observation data of multiple GPS satellites of the satellite monitoring system and uses a three-dimensional ionospheric tomography algorithm to reconstruct the three-dimensional distribution of ionospheric electron density, thereby obtaining the spatial distribution information of ionospheric irregularities. Specifically, the spatial distribution information of ionospheric irregularities is tracked by using a three-dimensional ray tracing method to obtain the impact of electromagnetic waves of different frequencies when transmitted in the ionosphere, so as to obtain the frequency dependence of ionospheric delay. Among them, the three-dimensional ray tracing method includes ray tracing simulation tools, such as HIRT software. Specifically, through the preset ionospheric electron density distribution model, the ray trajectories and propagation delays of electromagnetic waves of different frequencies when propagating in the ionosphere are calculated, and then the frequency dependence of ionospheric delay is analyzed. For example, different frequencies include: GPSL1 frequency (1575.42MHz), L2 frequency (1227.60MHz). In addition, when analyzing the frequency dependence of ionospheric delay, multi-scenario simulations can be performed by setting different parameters such as solar activity conditions and geomagnetic conditions to evaluate the impact of ionospheric environmental changes on ionospheric delay, and then derive the frequency dependence of ionospheric delay.

[0036] Step S2, deriving ionospheric delay data of different frequency bands according to the frequency dependence of ionospheric delay, and combining the device power consumption limit information of the field unattended device to derive the transmission frequency and transmission power of image transmission of the field unattended device, and then sending the transmission frequency and transmission power of image transmission to the field unattended device.

[0037] Specifically, the present invention also determines the transmission frequency and transmission power of the image transmission of the field unattended equipment by comprehensively considering the power consumption limit information of the field unattended equipment and the frequency dependence of the ionospheric delay, so as to select a carrier frequency band with less ionospheric delay influence and better equipment performance, so as to ensure the transmission of monitoring data. That is, this embodiment also comprehensively considers factors such as ionospheric delay data of different frequency bands, receiver sensitivity of the field unattended equipment, antenna gain and interference signals of the field unattended equipment, and preferably obtains a frequency band with small ionospheric delay and strong anti-interference ability as the transmission frequency of the image transmission of the field unattended equipment, such as: GPSL5 frequency (1176.45MHz).

[0038] At the same time, the present invention also adjusts the transmission power of the image transmission to set a suitable transmission power of the image transmission to reduce the influence of the ionospheric delay on the pseudorange. Specifically, by using a suitable transmission power in the frequency band with large ionospheric delay, the influence of the ionospheric delay on the pseudorange can be reduced, and a lower frequency can be used to reduce the influence of the ionospheric delay on the signal carrier phase during image transmission.

[0039] The field unattended equipment in the present invention is internally provided with a satellite communication module, and satellite transmission of monitoring data can be realized through the satellite communication module, so as to remove the restrictions on the installation location of the field unattended equipment and avoid the problem of being unable to communicate with the background service center due to network communication, etc. That is, the present invention realizes data collection and remote control of the field unattended equipment by utilizing satellite communication. Specifically, by utilizing a satellite communication link, a two-way communication between the field unattended equipment and the background service center is established, which can realize the uploading of monitoring data, and can also realize the status monitoring, parameter configuration and software upgrade of the field unattended equipment.

[0040] In addition, the unattended equipment in the field also communicates with the monitoring equipment installed on site through the wireless network to realize data collection. Specifically, the wireless sensor network can be used to realize the data collection and integration between the equipment within the surrounding range of the unattended equipment in the field, thereby improving the efficiency and reliability of data transmission.

[0041] When determining the transmission frequency of image transmission, the present invention first determines the ionospheric delay characteristics of different frequency bands according to the frequency dependence of ionospheric delay, and at the same time obtains the power consumption requirements, power supply mode, and working time of the field unattended equipment to obtain the power consumption limit of the field unattended equipment, such as the average power consumption needs to be controlled within 500mW. Then, the ionospheric delay characteristics and the power consumption limit of the equipment are comprehensively considered to determine the optimal carrier frequency range, such as 1000MHz~1200MHz. Finally, within the above frequency range, the transmission frequency and transmission power of the image transmission are dynamically adjusted according to the changes in the ionospheric delay.

[0042] Specifically, the adjustment rules are based on the following principles: in the frequency band with large ionospheric delay, a lower carrier frequency is used to reduce the impact of ionospheric delay on carrier phase; in the frequency band with large ionospheric delay, a moderate transmission power is used to reduce the impact of ionospheric delay on pseudorange; by optimizing the carrier frequency and transmission power, the impact of ionospheric delay is minimized while meeting the low power consumption requirements of the equipment. For example, when the ionospheric delay is large, a lower carrier frequency such as 1000MHz is selected, and the transmission power is appropriately reduced to 200mW to reduce the impact of ionospheric delay on carrier phase and pseudorange. By optimizing the dynamic adjustment of the carrier frequency within the range of 1000-1200MHz, and moderately controlling the transmission power within the range of 200-500mW, the impact of ionospheric delay can be minimized while meeting the low power consumption requirements of the equipment. In actual use, by continuously monitoring the changes in ionospheric delay and the power consumption status of the equipment, the transmission frequency and transmission power are optimized in real time, which can ensure the reliability and low power consumption of the transmission of monitoring data. For example, the change data of ionospheric delay is collected every 10 minutes, and the equipment power consumption data is collected every 1 minute. The Kalman filter algorithm is used to fuse and predict the data, and the carrier frequency and transmission power are adjusted in real time to ensure the reliability and low power consumption of image transmission.

[0043] Step S3, determining the image block size and image block priority of the image transmitted by the field unattended device according to the spatiotemporal variation characteristics of the ionospheric delay and sending them to the field unattended device, so that the field unattended device divides the image data of each transmitted image into multiple image blocks according to the image block size and image block priority and sends the divided image blocks according to the transmission frequency and transmission power.

[0044] Specifically, since image data is generally large, in order to ensure the stability of image transmission during data transmission, the present embodiment also divides the image data into image blocks by image segmentation, and then reassembles the received image blocks at the receiving end to form image data. In this way, even if an image block is lost or damaged due to external factors, only the image block needs to be retransmitted, and there is no need to transmit the entire image data, thereby saving resources and improving the transmission efficiency of image data.

[0045] Specifically, for the image block size, this embodiment is also set according to the spatiotemporal variation characteristics of the ionospheric delay. That is, first, the variation trend of the ionospheric delay on different time scales is obtained according to the spatiotemporal variation characteristics of the ionospheric delay, and the spatiotemporal variation characteristics of the ionospheric delay are multi-scale decomposed according to the time-frequency analysis method, and finally the image block size and image block priority are obtained according to the variation trend of the ionospheric delay on different time scales, the time-frequency analysis results and the system preset ionospheric delay and image block mapping model. Among them, the time-frequency analysis results include the delay variation characteristics corresponding to the different frequency components obtained by extraction. This embodiment solves the problem of incomplete data transmission when the ionospheric delay is large by setting the image block size and the image block priority. Specifically, when the image blocks are transmitted, they are transmitted in sequence according to the priority of the image blocks, such as ensuring that some image blocks with key features are transmitted first to ensure the data integrity of the transmitted image.

[0046] When the field unattended device receives the image block size and image block priority issued by the background service center, it divides the transmitted image data into image blocks according to the image block size and image block priority, sets a priority for each image block, and then sends the divided image blocks to the background service center through the satellite communication module according to the transmission frequency and transmission power to realize data analysis and further remote monitoring.

[0047] In addition, this embodiment also needs to encode the divided image blocks to encode the positions of the image blocks in the corresponding image data for subsequent reconstruction of the transmitted image data. Specifically, the image blocks can be encoded or sorted according to the segmentation order by setting the segmentation order.

[0048] More preferably, the present invention further layers the transmitted image data before performing image segmentation. That is, the field unattended device decomposes the transmitted image data into multiple image levels according to the pyramid decomposition algorithm, and the image data of each image level is provided with an image scale and an image resolution. Specifically, the image level may include a base layer, an enhancement layer, etc., and the image size and image resolution of each image level are different. Specifically, the original high-resolution image is decomposed into multiple levels through algorithms such as wavelet transform or Gaussian pyramid to generate image subbands of different scales, forming an image pyramid structure, in which each layer represents a specific scale and resolution.

[0049] Then the image data of each image layer is divided into image blocks according to the image block size and image block priority, and the priority of each image block is set and each image block is encoded. In this way, when the ionospheric delay increases, the size, number and priority of the image block are dynamically adjusted, and corresponding priorities are set for different image layers. For example, a higher priority is assigned to the image blocks of the base layer to ensure the transmission quality of the image of the base layer and the integrity of the image data.

[0050] Finally, a progressive image compression algorithm is used to compress each image block, so that the image data of the transmitted image generates a multi-scale, multi-bitstream image data stream and is sent out according to the transmission frequency and transmission power. Specifically, according to the scale and resolution of each layer of the image pyramid, appropriate coding parameters are selected, and a progressive image compression algorithm such as SPIHT is used to compress multiple image blocks of each layer of sub-band to generate a multi-scale, multi-bitrate image data stream, thereby realizing progressive compression and transmission of image data.

[0051] That is, this embodiment first solves the problem of incomplete data transmission when the ionospheric delay is large by combining layered transmission with image segmentation; at the same time, smaller image segments are used for data transmission to avoid the problem of incomplete image transmission caused by an overly large image. At the same time, a suitable priority is set for each image segment to better ensure the clarity of the transmitted image while ensuring the data integrity of the transmitted image.

[0052] Furthermore, in this embodiment, the setting of the image block size and the image block priority is obtained according to the mapping model of ionospheric delay and image block constructed in the system. The mapping model specifically analyzes the change trend of ionospheric delay on different time scales according to the spatiotemporal change characteristics of ionospheric delay, and the time scale can be: hours, days, weeks, months, etc., so as to determine the long-term change mode and short-term change mode of ionospheric delay; then, a time-frequency analysis method such as wavelet transform or empirical mode decomposition is used to perform multi-scale decomposition on the ionospheric delay time series, and the delay change characteristics corresponding to different frequency components are extracted, such as: low-frequency components reflect long-term trends, and high-frequency components reflect short-term fluctuations; finally, the long-term change mode and short-term change mode of ionospheric delay data are combined with the time-frequency analysis results, and used as the basis for dynamically adjusting the image block size and priority, so as to realize the setting of the image block size and priority. Furthermore, this embodiment constructs a nonlinear mapping model of ionospheric delay data and image block data by using a machine learning algorithm, such as a neural network or a support vector machine, so that when the ionospheric delay is obtained, it is substituted into the nonlinear mapping model to obtain the corresponding image block size and priority. The image block data includes the image block size and the image block priority. Furthermore, the model is constructed by using the historical delay data and the corresponding image block data as training samples, and the nonlinear mapping model is trained according to the training samples to construct a mapping relationship between the ionospheric delay data and the image block data, so as to realize the dynamic mapping of the ionospheric delay to the image block data.

[0053] Furthermore, the present embodiment also compresses each image block. Specifically, the present embodiment compresses the image blocks using a progressive image compression algorithm. For example, the SPIHT algorithm is first used for adaptive quantization, and the quantization step size is dynamically adjusted according to the change of ionospheric delay. For delay-sensitive image areas, such as edges and textures, a smaller quantization step size is used to reduce the impact of delay on image quality; then the CABAC algorithm (Context-Based AdaptiveBinary Arithmetic Coding) is used for adaptive entropy coding, and the coding table is dynamically adjusted according to the statistical characteristics of different image areas to improve coding efficiency. Specifically, for example, by using a GPS receiver with a sampling frequency of 1Hz to continuously collect 24 hours of ionospheric delay data, by analyzing the change trend of the delay on different time scales, such as hours, days, weeks, months, etc., the long-term change mode and short-term change mode of the delay are determined. For example, by analyzing the delay data for 30 consecutive days, it is found that the delay peaks from 14:00 to 18:00 every day, showing obvious daily variation characteristics; and on a time scale of one month, the delay shows a trend of gradually increasing with the increase of solar activity. Then, the wavelet transform is used to perform multi-scale decomposition of the ionospheric delay time series, and the Daubechies wavelet function is selected. The decomposition layer number is 5, and the delay change characteristics corresponding to different frequency components are extracted. Among them, the low-frequency component reflects the long-term trend, and the high-frequency component reflects the short-term fluctuation. Through analysis, it is found that the high-frequency component with a scale of 1 shows obvious random fluctuation characteristics, while the low-frequency component with a scale of 5 reflects the long-term change trend of the delay. The long-term and short-term change patterns are combined with the time-frequency analysis results as the basis for dynamically adjusting the image block size and priority.

[0054] When the delay shows obvious daily variation characteristics, the size and number of image blocks can be dynamically adjusted according to the peak time period of the delay. A smaller image block size, such as 256 pixels × 256 pixels, is used for image data in the period from 14:00 to 18:00, and a higher priority is assigned to ensure the quality of image transmission. In addition, this embodiment also uses a three-layer neural network to establish a nonlinear mapping model between ionospheric delay and image block size and priority. The relevant delay data of the ionosphere for 30 consecutive days and the corresponding optimal image block and priority are used as training samples, and the model is trained multiple times to achieve dynamic mapping of ionospheric delay to image block data. When optimizing image block data, a genetic algorithm is used for search, and the chromosome encoding length is 20, of which the first 10 bits represent the image block size, ranging from 64 to 512, and the last 10 bits represent the image block priority, ranging from 1 to 10.

[0055] More preferably, sending the image data stream according to the transmission frequency and transmission power in step S3 also includes: acquiring spatial variation gradient information of ionospheric delay according to ionospheric delay data, and then analyzing and obtaining the variation rate and amplitude of ionospheric delay in different regions to obtain the severity of spatial variation of ionospheric delay, and forming an ionospheric delay gradient map according to the severity of spatial variation of ionospheric delay, determining the transmission strategy of image data according to the ionospheric delay gradient map and a preset mapping matching model; and finally sending the image data stream according to the transmission strategy of image data.

[0056] That is, according to the obtained ionospheric delay gradient map, the image data can be divided into different areas, including areas where the ionospheric delay changes gently and areas where the ionospheric delay changes dramatically. In this way, the corresponding transmission strategy is selected according to the ionospheric delay changes in the image data. Specifically, in areas where the ionospheric changes gently, image data at the image level with high resolution can be preferentially transmitted to provide more detailed information of the image data; in areas where the ionospheric changes dramatically, image data at the image level with low resolution can be preferentially transmitted to ensure the continuity of image transmission.

[0057] More specifically, the present invention also uses a support vector regression (SVR) model to establish a mapping model between ionospheric delay gradient and image scale and resolution. Specifically, the ionospheric delay gradient data obtained by historical calculation and the corresponding optimal image scale and resolution are used as training samples, and the hyperparameters of SVR, such as kernel function type, regularization parameter, etc., are optimized through cross-validation and grid search to improve the fitting performance and generalization ability of the model. Through the mapping model obtained by training, the optimal image level and resolution of image data transmitted in different regions are dynamically determined to realize an adaptive image transmission strategy. During the image data transmission process, the change of ionospheric delay can also be monitored in real time, so as to dynamically adjust the scale and resolution of image data transmission according to the amplitude and rate of ionospheric delay change, that is, select the image block of the corresponding image level for transmission. That is, when the ionospheric delay changes gradually, the transmission strategy is adjusted to the first transmission mode, that is, the scale and resolution of image data of the high-resolution image level are preferentially transmitted to improve the image quality; when the ionospheric delay changes drastically, it is switched to the second transmission mode in time, and the image data of the low-resolution image level is preferentially transmitted to ensure the continuity and recognizability of the image.

[0058] Step S4, acquiring image blocks of the field unattended equipment in real time, parsing each image block, and fusing the parsed image blocks to reconstruct the image for image transmission.

[0059] That is, the background service center as the receiving end analyzes and reconstructs the received image blocks to construct the image data. More specifically, after receiving the image blocks, the background service center first uses a progressive image decompression method to decompress the image blocks, and analyzes the image blocks in real time to obtain the image data of the image blocks, the encoding of the image blocks, the image level to which the image blocks belong, the priority of the image blocks, and other data, and then reconstructs all image blocks belonging to the same image level to obtain the image data of the corresponding image level, and finally merges all the image data of all image levels to obtain the final image data, thereby realizing the transmission of image data.

[0060] In the background service center, it adopts progressive image decompression algorithms corresponding to unattended equipment in the field, such as SPIHT, to decode the received image data stream in real time, derive each image layer according to the size and resolution of each image layer in the image data stream, and then gradually restore the different scales and resolutions of the image to achieve progressive reconstruction of the image.

[0061] Furthermore, due to external or internal factors, the received image blocks may be lost or damaged. In order to ensure the integrity of the image data, this embodiment also retransmits the lost or damaged image blocks. That is, by setting a retransmission mechanism, the lost or damaged image blocks are restored, and the image enhancement and post-processing technology are used to improve the quality and details of the reconstructed image. According to the self-diagnosis and self-recovery capabilities of the field unattended equipment, transmission anomalies are discovered and handled in a timely manner.

[0062] When there are lost or damaged image blocks, the image blocks are retransmitted by starting an adaptive retransmission request mechanism. More specifically, the present embodiment adopts an adaptive retransmission request mechanism to determine the retransmission parameters of the image blocks and to make a retransmission request for the lost image blocks. The retransmission parameters include the image block size, the image block priority, the image layer to which the image blocks belong, the time deadline, the retransmission parameters and the time interval. According to the above retransmission parameters, the lost image blocks or damaged image blocks can be retransmitted to ensure the integrity of the image data.

[0063] More specifically, the time deadline means that the image block must be transmitted and decoded within the specified time. After the time deadline is exceeded, the real-time and effectiveness of the image block transmission will be greatly reduced. Specifically, when the ionospheric delay increases and causes the image block to be lost, the high-priority image blocks are retransmitted first, and the number of retransmissions and time intervals are dynamically adjusted. That is, for the image blocks of the high-priority image base layer, a shorter retransmission time interval, such as 200 milliseconds, and a larger number of retransmissions, such as 3 times, are set to ensure the basic quality of the image; for the image blocks of the low-priority image enhancement layer, a longer retransmission time interval, such as 500 milliseconds, and a smaller number of retransmissions, such as 2 times, are set to reduce the retransmission overhead to ensure the reliable transmission of key image blocks.

[0064] Furthermore, when the number of image block retransmissions reaches a preset number of times or the time deadline reaches a preset time, if the image block is still lost or damaged, it is considered that the image block retransmission fails. When the image block retransmission fails, an adaptive weighted interpolation algorithm is used to perform error concealment and fault tolerance processing to correct the image block. Specifically, first, the lost image block is determined according to the number of the lost image block and the image level to which the image block belongs, and then the adjacent image block of the image block is selected and used as the reference image block at the position of the image block in the image level; then, the texture direction, edge strength and color characteristics of the reference image block are analyzed and extracted, and then the corresponding weight coefficient is set according to the texture direction, edge strength and color characteristics of each reference image block, and the texture direction, edge strength and color characteristics of each reference image block are weighted averaged by weighted averaging, and then the pixel value of the image block is obtained, that is, the image data of the lost image block is obtained. When setting the weight coefficient, an adaptive model based on gradient direction and edge strength is used to give greater weight to reference image blocks with high texture and edge strength similarity to reduce image distortion.

[0065] When constructing the image hierarchy based on the image blocks of each image level, image enhancement and post-processing techniques are also used to improve the visual quality of the reconstructed image. For example: the contrast and dynamic range of the image are improved through adaptive contrast enhancement algorithms, such as histogram equalization or Retinex (image enhancement method) algorithm; among them, histogram equalization enhances the overall contrast of the image by stretching the grayscale distribution of the image; the Retinex algorithm achieves adaptive adjustment of brightness and contrast by estimating the illumination and reflectivity of the image. For another example, an adaptive sharpening filter algorithm, such as bilateral filtering, is used to enhance the edge and texture details of the image; among them, bilateral filtering combines the weights of the spatial domain and the grayscale domain to smooth the image noise and improve the clarity of the image while maintaining the edge.

[0066] Specifically, at the receiving end, the SPIHT (image compression algorithm based on wavelet transform) algorithm is used to decompress and reconstruct the received image bit stream. The SPIHT algorithm achieves progressive transmission of images by encoding the spatial position and amplitude of wavelet coefficients. During the decoding process, the image base layer with low frequency resolution is first decoded, and then the detail layer with high frequency resolution is gradually decoded until the target bit rate or the parameters of reconstructed image quality are reached.

[0067] In addition, when receiving each image block, the background service center of this embodiment also performs integrity check on the image block to determine whether the image block is complete, that is, to determine whether the image block is damaged. If damaged, the retransmission request mechanism is activated. Similarly, when reconstructing the image data of each image layer, it is also necessary to determine whether there is a lost image block. If so, the retransmission request mechanism is activated.

[0068] Furthermore, a 32-bit CRC check code is used to perform integrity check on the received image data. By calculating whether the CRC value of the received data matches the transmitted CRC value, it is determined whether there is an error in the image block. When a CRC check failure is detected, an adaptive retransmission request (ARQ) is immediately sent to the sender to request retransmission of the erroneous image block. For wireless transmission channels, the timeout time and number of retransmissions are dynamically adjusted according to the channel delay jitter and packet loss rate. For example, in a channel with a packet loss rate of 5%, the retransmission timeout time of the base layer image block is set to 200 milliseconds and the number of retransmissions is set to 3 times; while for the enhanced layer image block, the timeout time is extended to 500 milliseconds and the number of retransmissions is reduced to 2 times. Through the adaptive retransmission strategy, while ensuring the reliable transmission of important image blocks, the redundant retransmission overhead is reduced. For image blocks that cannot be recovered through retransmission, an adaptive weighted interpolation algorithm is used for error concealment. First, according to the position of the lost image block, select 8 adjacent 4x4 image blocks as reference blocks; then calculate the gradient difference between each reference block and the lost block in the horizontal, vertical and diagonal directions, and obtain the weight coefficient in each direction after normalization; finally, use the weighted average of the 8 reference blocks to fill the lost block. The calculation formula of the weight coefficient is: W i =exp(-abs(G i -lost) / (2*σ^2)), where: W i is the weight of the i-th reference block, G iis the gradient value of the reference block in a certain direction, lost is the estimated gradient value of the lost block in this direction, and σ is the standard deviation of the gradient difference. In the image enhancement and post-processing stage, the histogram equalization algorithm is also used to stretch the grayscale distribution of the image to [0,255] to improve the overall contrast of the image. At the same time, the bilateral filtering algorithm is used to denoise and sharpen the image. Bilateral filtering combines spatial domain Gaussian filtering and grayscale domain Gaussian filtering to smooth image noise while keeping edge and texture details from being overly blurred. The standard deviation of the spatial domain Gaussian filter is set to 1.5, and the standard deviation of the grayscale domain Gaussian filter is set to 25.

[0069] In addition, during the reception and reconstruction of the transmitted image, the support vector machine (SVM) algorithm is also used for anomaly detection. Specifically, by extracting eight statistical features such as the mean, variance, and entropy of the received image data, an SVM anomaly detection model is constructed. The RBF kernel function is used to select the optimal model parameters through cross-validation, such as the penalty coefficient C=10 and the kernel function parameter γ=0.1. When the feature vector of the received image data deviates from the support vector of the normal model, it is judged as an anomaly and the corresponding recovery mechanism is triggered. Among them, the training samples of the anomaly detection model include 500 normal images and 100 abnormal images. Through 10-fold cross-validation, the anomaly detection accuracy of the model reaches more than 95%.

[0070] In addition, this embodiment also adopts a remote software upgrade solution based on the MQTT protocol to realize online upgrades and fault repair of unattended equipment in the field. Among them, the unattended equipment in the field, as an MQTT client, subscribes to the firmware upgrade topic, receives upgrade tasks and firmware packages; the upgrade process is divided into multiple stages, each stage transmits a part of the firmware package, and the size of the entire firmware package is 500KB. During the firmware upgrade process, the AES-128 encryption algorithm is also used to encrypt the firmware package to ensure the security of transmission. After receiving the firmware package, the unattended equipment in the field confirms the integrity and correctness of the package through MD5 verification, and stores the firmware package in an external Flash. After the upgrade is completed, the device restarts, loads the new firmware, and completes the software upgrade.

[0071] Preferably, the present invention also utilizes the self-diagnosis and self-recovery capabilities of the unattended equipment in the field to monitor abnormal conditions in real time during image transmission and reconstruction. Specifically, by embedding anomaly detection algorithms such as isolation forests or support vector machines in the unattended equipment in the field, the abnormal state in the image transmission process is analyzed, and abnormal states such as transmission link interruption, data packet loss, decoding errors, etc. are judged in real time. Once an abnormal state is detected, the corresponding recovery mechanism is automatically triggered, such as re-establishing the transmission link, requesting retransmission of lost data, resetting the decoder state, etc. Among them, the anomaly detection algorithm specifically constructs a normal state model by training normal data and abnormal data. Once it is detected that new data deviates from the normal state model, the new data is considered to be abnormal, that is, the unattended equipment in the field is in an abnormal state at this time.

[0072] Preferably, this embodiment can also warn of ionospheric delay. That is, according to the ionospheric delay data, solar activity, and historical observation data of geomagnetic disturbances corresponding to the observation position of the field unattended equipment, the correlation strength and hysteresis effect of ionospheric delay with solar activity and geomagnetic disturbances are quantified through cross-correlation analysis and cross-spectrum analysis, and then the correlation analysis results of ionospheric delay with solar activity and geomagnetic disturbances are obtained. Then, based on the correlation analysis results of ionospheric delay with solar activity and geomagnetic disturbances, machine learning algorithms such as extreme gradient boosting and long short-term memory networks are used to construct a multivariate ionospheric delay prediction model. When constructing the model, the solar activity index (such as F10.7, sunspot number), geomagnetic disturbance index (such as Kp, Dst) and ionospheric key parameters (such as foF2, TEC data) are used as input features, and ionospheric delay is used as the output target for model training; and in the model training process, feature engineering and model integration technology are used to optimize model hyperparameters to improve prediction accuracy and stability. The key ionospheric parameters corresponding to the observation position of the field unattended equipment, as well as the data of solar activity, geomagnetic disturbance and historical observation data of ionospheric delay are input into the multivariate ionospheric delay prediction model constructed above to estimate the trend of ionospheric delay changes in the future, thereby forming ionospheric delay warning and forecast information. In this way, according to the ionospheric delay warning and forecast information, the transmission frequency, transmission power and transmission parameters of the image transmission of the field unattended equipment are dynamically estimated.

[0073] The above-mentioned embodiments are only preferred embodiments of the present invention, and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technical users in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. A satellite-based image transmission method for unattended outdoor equipment, characterized in that: include: Delay analysis step: obtaining ionospheric TEC data and spatial distribution information of ionospheric irregularities corresponding to the observation position of the field unattended equipment through a satellite monitoring system, and deriving frequency dependence and spatiotemporal variation characteristics of ionospheric delay based on the ionospheric TEC data and spatial distribution information of ionospheric irregularities; The transmission parameter determination step includes: obtaining ionospheric delay data of different frequency bands according to the frequency dependence of the ionospheric delay, and obtaining the transmission frequency and transmission power of the image transmission of the field unattended device in combination with the device power consumption limit information of the field unattended device, and then sending the transmission frequency and transmission power of the image transmission to the field unattended device; The image block setting step includes determining the image block size and the image block priority of the image monitoring data according to the spatiotemporal variation characteristics of the ionospheric delay, and sending the image block size and the image block priority to the field unattended device, so that the field unattended device divides the image in the image transmission according to the image block size and the image block priority to form a plurality of image blocks, and sends the divided image blocks according to the transmission frequency and the transmission power; The image block setting step of determining the image block size and image block priority of the image monitoring data according to the spatiotemporal variation characteristics of the ionospheric delay specifically includes: firstly, analyzing the spatiotemporal variation characteristics of the ionospheric delay to obtain the variation trend of the ionospheric delay at different time scales; then, performing multi-scale decomposition on the spatiotemporal variation characteristics of the ionospheric delay according to the time-frequency analysis method; finally, obtaining the image block size and image separation priority according to the variation trend of the ionospheric delay at different time scales, the time-frequency analysis results and the system preset mapping model of the ionospheric delay and the image block; wherein the time-frequency analysis results include extracting the delay variation characteristics corresponding to different frequency components; In the image block setting step, the field unattended device divides the image in the image transmission according to the image block size and the image block priority to form a plurality of image blocks, which specifically includes: firstly, the field unattended device decomposes the image data into a plurality of image levels according to the pyramid decomposition algorithm, and the image data of each image level is provided with a corresponding image scale and image resolution; then, the image data of each image level is divided into image blocks according to the image block size, and each image block is encoded and the image block priority is set; finally, each image block is compressed by using a progressive image compression algorithm, so that the transmitted image data generates a multi-scale, multi-bit stream image data stream, and is sent out according to the transmission frequency and transmission power; Image reconstruction step: acquiring image blocks of the field unattended device in real time, parsing each image block, and fusing the parsed multiple image blocks to reconstruct the image transmitted by the image.

2. The satellite-based field unattended equipment image transmission method according to claim 1, characterized in that: The image reconstruction step specifically includes: first, acquiring the image data stream of the field unattended equipment in real time and analyzing the image data stream to obtain each image block, and obtaining the corresponding image level according to the encoding of each image block; then, reconstructing the image data of the corresponding image level according to the encoding of the image block, the scale and resolution of the corresponding image level; finally, integrating the image data of multiple image levels according to image enhancement and post-processing technology to construct the final image transmission.

3. The satellite-based field unattended equipment image transmission method according to claim 2, characterized in that: The image reconstruction step further includes: a judging step: judging whether each acquired image block is damaged by a cyclic check algorithm, and if so, executing a retransmission step; and judging whether image blocks belonging to the same image level are lost, and if so, executing a retransmission step; The retransmission step: when the image block corresponding to the image layer is damaged, start the adaptive retransmission request mechanism to send a retransmission instruction to the field unattended device to reacquire the image block; wherein the retransmission instruction includes the encoding of the image block, the image block priority, the number of image block retransmissions, the time deadline, the image layer to which the image block belongs, and the image transmission corresponding to the image layer to which the image block belongs; the time deadline refers to the image block completing the transmission and decoding within the specified time; wherein, the higher the image block priority, the higher the number of image block retransmissions and the shorter the time deadline.

4. The satellite-based field unattended equipment image transmission method according to claim 3, characterized in that: In the retransmission step: when the number of retransmissions of an image block reaches a preset number of times or the time deadline reaches a preset time, and the image block is still lost or damaged, an image block adjacent to the image block is obtained based on the encoding of the adaptively lost or damaged image block and used as a reference image block, and then a weight coefficient is set for the texture direction, edge strength and color characteristics of each reference image block according to a preset weight method, and finally the texture direction, edge strength and color characteristics of the image block are estimated by weighted averaging, and then the corresponding image block is constructed according to the texture direction, edge strength and color characteristics of the image block.

5. The satellite-based field unattended equipment image transmission method according to claim 1, characterized in that: The step of sending the image data stream according to the transmission frequency and transmission power in the image block setting step further includes: acquiring spatial variation gradient information of ionospheric delay according to ionospheric delay data corresponding to the transmission frequency, and then analyzing and obtaining the variation rate and amplitude of ionospheric delay in different regions, so as to obtain the spatial variation severity of ionospheric delay and form an ionospheric delay gradient map according to the spatial variation severity of ionospheric delay, and determining the transmission strategy of image data according to the ionospheric delay gradient map and a preset mapping matching model; and finally sending the image data stream according to the transmission strategy of image data; Among them, the transmission strategy of image data follows the following principles: when the degree of change of ionospheric delay in the corresponding area is greater than the preset value, the image data stream is transmitted in sequence according to the resolution of the image level from high to low; when the degree of change of ionospheric delay in the corresponding area is less than the preset value, the image data stream is transmitted in sequence according to the resolution of the image level from low to high.

6. The satellite-based field unattended equipment image transmission method according to claim 1, characterized in that: Also includes: Abnormal reset step: when an abnormality occurs in the field unattended device, a diagnostic instruction is sent to the field unattended device to start the diagnosis and self-recovery operation of the field unattended device; specifically, by setting an abnormality detection algorithm inside the field unattended device, the abnormality detection algorithm is used to determine in real time whether there is an abnormality in the image transmission process. If so, a reset request is sent to the background service center and then the diagnostic instruction is received to start the diagnosis and self-recovery operation; wherein, the abnormalities in the image transmission process include interruption of the image transmission link, retransmission of image blocks, and failure of retransmission of image blocks.

7. The satellite-based field unattended equipment image transmission method according to claim 1, characterized in that: The delay analysis step specifically includes: obtaining the observation position of the unattended equipment in the field, and using the GPS occultation observation technology to analyze the phase and delay changes of the GPS signal when it passes through the ionosphere, and inverting to obtain the ionospheric TEC data; and using the observation data of multiple GPS satellites to reconstruct the three-dimensional distribution of the ionospheric electron density using a three-dimensional ionospheric tomography algorithm, thereby obtaining the spatial distribution information of the ionospheric irregularities; The temporal and spatial variation characteristics of ionospheric delay are extracted by performing time series changes on ionospheric TEC data; and the spatial distribution information of ionospheric irregularities is tracked through the three-dimensional ray tracing method to obtain the ray trajectories and propagation delays of electromagnetic waves of different frequency bands when transmitted in the ionosphere, so as to analyze the frequency dependence of ionospheric delay.

8. Satellite-based field unattended equipment image transmission device, characterized in that: It includes a memory and a processor, and is characterized in that the memory stores a field unattended satellite image transmission program running on the processor, and the field unattended satellite image transmission program is a computer program. When the processor executes the field unattended satellite image transmission program, it implements the steps of the satellite-based field unattended equipment image transmission method as described in any one of claims 1-7.

9. A computer-readable storage medium storing a field unattended satellite image transmission program, characterized in that: The field unattended satellite image transmission program is a computer program. When the field unattended satellite image transmission program is executed by a processor, the steps of the satellite-based field unattended device image transmission method as described in any one of claims 1 to 7 are implemented.

Images