Image transmission control method and system of aviation communication radio station

Abstract

The invention relates to the technical field of image transmission control, and discloses an image transmission control method and system for an aeronautical communication radio station, and the system comprises an image collection and preprocessing module, an intelligent compression and scheduling module, a transmission control and error correction module, and a reception decoding and evaluation module. Image data inside and outside an aircraft are collected in real time and preprocessed, including denoising, enhancement and format standardization, definition and compatibility of the image data are guaranteed, standardized preprocessed image data are generated, uniformity and high efficiency of subsequent transmission processing are guaranteed, and initial quality of image transmission is improved; an intelligent scheduling algorithm is adopted to arrange a transmission sequence according to the flight mission priority and the image content importance, the data compression and transmission sequence is dynamically optimized, different channel conditions are adapted, and the image transmission efficiency and real-time performance are improved.

Description

Technical Field

[0001] This invention relates to the field of image transmission control technology, specifically to an image transmission control method and system for an aviation communication radio. Background Technology

[0002] Aviation communication radios are communication devices installed on aircraft in aviation mobile services. They are mainly used for wireless communication between aircraft and ground base stations or other aircraft; they play a core communication role in ensuring air traffic control and flight safety.

[0003] Currently, in the process of image transmission by aviation communication radio, due to the complex and changeable internal and external environment of aircraft and the unstable state of communication channels, existing technologies are unable to effectively preprocess and adaptively compress the acquired raw image data. There is also a lack of intelligent scheduling mechanisms based on flight mission priority and image content, resulting in low transmission efficiency and the inability to prioritize the protection of critical image data.

[0004] Therefore, a method and system for controlling image transmission in aviation communication radios are proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an image transmission control method and system for aviation communication radios. This solves the problems mentioned in the background section, such as the difficulty in effectively preprocessing and adaptively compressing the acquired raw image data, the lack of an intelligent scheduling mechanism based on flight mission priority and image content, resulting in low transmission efficiency and the inability to prioritize the protection of critical image data.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method and system for controlling image transmission in an aviation communication radio, the method comprising the following steps:

[0007] S1. Real-time acquisition of image data inside and outside the aircraft via image acquisition equipment from aviation communication radio, and generation of raw image dataset;

[0008] S2. Perform image preprocessing operations based on the original image dataset, including denoising, enhancement, and format normalization, and generate preprocessed image data;

[0009] S3. Based on the real-time status parameters of the aviation communication channel, perform adaptive compression coding on the preprocessed image data to generate compressed coded image data.

[0010] S4. An intelligent scheduling algorithm is used to arrange the transmission sequence of the compressed coded image data according to the flight mission priority and the importance of the image content, thereby generating image transmission sequence data;

[0011] S5. The image transmission sequence data is transmitted to the ground control station in segments through the wireless transmission module of the aviation communication radio, and the data packet loss rate is monitored in real time during the transmission process.

[0012] S6. Perform dynamic error correction control processing based on monitoring results, including retransmission mechanism and forward error correction coding adjustment, and generate error-corrected image transmission data;

[0013] S7. At the receiving end, the error-corrected image transmission data is decoded and reconstructed to restore it to visual image data.

[0014] S8. Combine the artificial intelligence model to perform quality assessment on the visual image data and generate an image transmission quality report.

[0015] Preferably, the acquisition of image data of the aircraft's interior and exterior in step S1 includes the following steps:

[0016] S11. Deploy multiple high-definition cameras on the exterior of the aircraft fuselage to collect images of the flight environment, including weather conditions, terrain and obstacle information;

[0017] S12. Install visible light cameras and infrared sensors inside the aircraft cockpit to collect images of the pilot's operating status and the operation of the equipment inside the cockpit.

[0018] S13. Mark the collected image data using timestamps and GPS coordinates;

[0019] S14. Integrate the labeled image data into a structured dataset to generate the original image dataset, whose data format includes JPEG, PNG and RAW formats.

[0020] Preferably, the image preprocessing operation includes the following steps:

[0021] S21. Use Gaussian filtering algorithm to perform noise reduction processing on the original image dataset to reduce random interference during image acquisition;

[0022] S22. Enhance image contrast and brightness through histogram equalization technology to improve image visibility under low light conditions;

[0023] S23. Convert the enhanced image data into the standard H.264 encoding format to ensure compatibility with different aviation communication equipment;

[0024] S24. Perform resolution standardization processing on the converted image data, adjust all images to 1920x1080 pixels, and generate the preprocessed image data.

[0025] Preferably, the adaptive compression coding process includes the following steps:

[0026] S31. Monitor the bandwidth utilization and bit error rate of aviation communication channels in real time and generate channel status indicators;

[0027] S32. Based on the channel state index, dynamically select the compression ratio: when the bandwidth is higher than the threshold, use a lossless compression algorithm; when the bandwidth is lower than the threshold, use a lossy compression algorithm.

[0028] S33. Apply discrete cosine transform to compress image data in the frequency domain to reduce the amount of data.

[0029] S34. Compression efficiency is optimized through entropy coding to generate compressed coded image data, the size of which is adaptive to channel conditions.

[0030] Preferably, the transmission sequence arrangement in S4 includes the following steps:

[0031] S41. Define flight mission priority parameters, including emergency missions, routine missions, and background missions;

[0032] S42. Use convolutional neural networks to analyze image content, identify key objects: runways, other aircraft, and unusual events, and assign importance scores.

[0033] Importance scoring formula: ;

[0034] in Assign importance ratings, This represents the total number of key objects identified in the image. For object indexing, For the first The confidence level of an object's identification. For the first Preset weights for each object type;

[0035] S43. Combining priority parameters and importance scores, a weighted round-robin algorithm is used to arrange the order of image data transmission.

[0036] S44. Allocate more transmission time slots for high-priority image data and generate image transmission sequence data.

[0037] Preferably, the real-time monitoring of data packet loss rate in S5 includes the following steps:

[0038] S51. Divide the image transmission sequence data into fixed-size data packets, and add a sequence number and checksum to each data packet;

[0039] S52. High-speed transmission is achieved via UDP protocol, while ACK confirmation mechanism is enabled to monitor data packet reception status.

[0040] S53. Calculate the data packet loss rate in real time and trigger an alarm when the loss rate exceeds 5%;

[0041] S54. Record transmission timestamps and channel quality indicators, and generate transmission monitoring logs.

[0042] Preferably, the dynamic error correction control process includes the following steps:

[0043] S61. When the data packet loss rate exceeds the threshold, a selective retransmission request is automatically initiated to retransmit only the lost data packets.

[0044] S62. Adjust the redundancy of the forward error correction coding and dynamically adjust the error correction code length according to the channel noise level;

[0045] S63. Reed-Solomon encoding is used to strengthen the retransmitted data;

[0046] S64. Verify the integrity of the corrected data and generate the corrected image transmission data.

[0047] Preferably, the decoding and reconstruction process includes the following steps:

[0048] S71. At the receiving end, the data packets are reassembled in order, and the original sequence is restored based on the sequence number;

[0049] S72. Restore image data by applying inverse discrete cosine transform and decoding algorithm;

[0050] S73. Perform color space conversion and sharpening to improve the visual effect of the image;

[0051] S74 outputs visual image data, supporting real-time display and storage.

[0052] Preferably, generating an image transmission quality report in step S8 includes the following steps:

[0053] S81. Use pre-trained deep learning models to evaluate image quality, including metrics such as peak signal-to-noise ratio and structural similarity index.

[0054] S82. Compare the differences between the images before and after transmission, and calculate the distortion score;

[0055] Distortion rating:

[0056] in Rate the distortion. , , These are the weighting coefficients. Peak signal-to-noise ratio, It is a structural similarity index. A quality score for the output of a deep learning model;

[0057] S83. Generate an image transmission quality report, including the quality level and improvement suggestions.

[0058] Preferably, the system includes an image acquisition and preprocessing module, an intelligent compression and scheduling module, a transmission control and error correction module, and a receiving, decoding, and evaluation module;

[0059] The image acquisition and preprocessing module acquires original images inside and outside the aircraft through a multimodal image acquisition unit, optimizes the images through a noise reduction and enhancement unit, and outputs the preprocessed image data through a format standardization unit.

[0060] The intelligent compression and scheduling module receives the preprocessed image data, obtains real-time communication parameters through the channel state sensing unit, compresses the data using the adaptive coding unit, and outputs image transmission sequence data through the priority scheduling unit.

[0061] The transmission control and error correction module receives the image transmission sequence data, performs segmented transmission through the wireless data transceiver unit, collects data packet loss information using the transmission status monitoring unit, and outputs the corrected image transmission data through the dynamic error correction unit.

[0062] The receiving, decoding, and evaluation module receives the error-corrected image transmission data, restores the visible image through the data decoding and reconstruction unit, generates an image quality report through the transmission quality evaluation unit, and outputs the evaluation results.

[0063] Compared with the prior art, the present invention provides an image transmission control method and system for aviation communication radios, which has the following beneficial effects:

[0064] 1. In this invention, image data from inside and outside the aircraft is collected in real time and preprocessed, including noise reduction, enhancement and format standardization, to ensure the clarity and compatibility of the image data. At the same time, standardized preprocessed image data is generated to ensure the uniformity and efficiency of subsequent transmission processing and improve the initial quality of image transmission.

[0065] 2. In this invention, adaptive compression coding is performed based on the real-time status parameters of the aviation communication channel, and an intelligent scheduling algorithm is used to arrange the transmission sequence according to the flight mission priority and the importance of image content, dynamically optimizing data compression and transmission order, adapting to different channel conditions, and improving the efficiency and real-time performance of image transmission.

[0066] 3. In this invention, by real-time monitoring of data packet loss rate and dynamic error correction control processing, including retransmission mechanism and forward error correction coding adjustment, and at the receiving end performing decoding reconstruction and quality assessment, the integrity and visibility of image data are guaranteed, transmission errors are reduced, and the reliability and overall quality of image transmission are improved. Attached Figure Description

[0067] Figure 1 This is a flowchart of an image transmission control method for an aviation communication radio according to the present invention;

[0068] Figure 2 This is a schematic diagram of the image transmission control system for an aviation communication radio according to the present invention. Detailed Implementation

[0069] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0070] Specific embodiment: A method and system for controlling image transmission in an aviation communication radio, the method comprising the following steps:

[0071] S1. Real-time acquisition of image data inside and outside the aircraft via image acquisition equipment from aviation communication radio, and generation of raw image dataset;

[0072] S2. Perform image preprocessing operations based on the original image dataset, including denoising, enhancement, and format normalization, and generate preprocessed image data;

[0073] S3. Based on the real-time status parameters of the aviation communication channel, perform adaptive compression coding on the preprocessed image data to generate compressed coded image data.

[0074] S4. An intelligent scheduling algorithm is used to arrange the transmission sequence of compressed coded image data according to the flight mission priority and the importance of image content, thereby generating image transmission sequence data;

[0075] S5. The image transmission sequence data is sent to the ground control station in segments through the wireless transmission module of the aviation communication radio, and the data packet loss rate is monitored in real time during the transmission process.

[0076] S6. Perform dynamic error correction control processing based on monitoring results, including retransmission mechanism and forward error correction coding adjustment, and generate error-corrected image transmission data;

[0077] S7. At the receiving end, the error-corrected image transmission data is decoded and reconstructed to restore it to visual image data.

[0078] S8. Combine artificial intelligence models to perform quality assessment on visual image data and generate an image transmission quality report.

[0079] S1 acquires image data of the aircraft's interior and exterior, including the following steps:

[0080] S11. Deploy multiple high-definition cameras on the exterior of the aircraft fuselage to collect images of the flight environment, including weather conditions, terrain and obstacle information;

[0081] S12. Install visible light cameras and infrared sensors inside the aircraft cockpit to collect images of the pilot's operating status and the operation of the equipment inside the cockpit.

[0082] S13. Mark the collected image data using timestamps and GPS coordinates;

[0083] S14. Integrate the labeled image data into a structured dataset to generate the original image dataset, whose data formats include JPEG, PNG and RAW formats.

[0084] Image preprocessing operations include the following steps:

[0085] S21. Use Gaussian filtering algorithm to perform noise reduction processing on the original image dataset to reduce random interference during image acquisition;

[0086] Gaussian filtering is a linear smoothing filter that uses a Gaussian function generated through convolution operations to perform a weighted average on an image, suppressing Gaussian noise. Its core is a two-dimensional Gaussian function, defined as follows: ;

[0087] in These are the weight values ​​of the Gaussian filter kernel at the coordinate points. This represents the horizontal distance of a point within the filter kernel relative to the center point. This represents the vertical distance of a point within the filter kernel relative to the center point. Let $\mathbf{a}$ be the standard deviation of the Gaussian distribution.

[0088] First, set the standard deviation value according to the required smoothness and determine a convolution kernel of size 5×5; then, discretize the above Gaussian function, calculate the weight value of each position in the convolution kernel, and ensure that the sum of all weights is 1 to keep the overall brightness of the image unchanged; finally, perform a sliding window convolution operation on the preprocessed image data with this convolution kernel, and replace the value of the center pixel with the weighted average of the pixels in the kernel, thereby completing the noise reduction process.

[0089] S22. Enhance image contrast and brightness through histogram equalization technology to improve image visibility under low light conditions;

[0090] First, calculate the gray-level histogram of the input image and count the probability of each gray level; then, calculate its cumulative distribution function as the transformation function. ;

[0091] in The transformed image grayscale value, For transformation function, For the first part of the original image grayscale value, gray levels in the original image The probability of occurrence;

[0092] Finally, this transformation function is used to map each pixel gray value in the original image to a new gray value, making the histogram of the output image approximately uniformly distributed, thereby expanding the dynamic range of the image and improving the visibility of image details under low light conditions.

[0093] S23. Convert the enhanced image data into the standard H.264 encoding format to ensure compatibility with different aviation communication equipment;

[0094] S24. Perform resolution standardization on the converted image data, adjusting all images to 1920x1080 pixels to generate preprocessed image data.

[0095] All images after format conversion are uniformly adjusted to a standard resolution of 1920x1080 pixels using a bilinear interpolation image scaling algorithm. Bilinear interpolation calculates the value of the target pixel by performing linear interpolation in both the horizontal and vertical directions, which maintains the smoothness of the scaled image to a certain extent and reduces jagged edges. This process ensures the consistency of image size requirements in subsequent compression, transmission and decoding stages, avoiding processing errors and compatibility issues caused by resolution differences.

[0096] Adaptive compression encoding processing includes the following steps:

[0097] S31. Monitor the bandwidth utilization and bit error rate of aviation communication channels in real time and generate channel status indicators;

[0098] S32. Based on channel state indicators, dynamically select the compression ratio: when the bandwidth is higher than the threshold, use a lossless compression algorithm; when the bandwidth is lower than the threshold, use a lossy compression algorithm.

[0099] The system presets a bandwidth threshold. When the channel bandwidth monitored in real time is higher than the threshold, PNG and FLIF lossless compression algorithms are used to ensure that the image information is completely restored after compression and decoding without any loss. When the channel bandwidth is lower than the threshold, transmission efficiency is prioritized, and a lossy compression algorithm based on the H.264 standard is used to reduce the amount of data by discarding some visual information that is not sensitive to the human eye, so that the image data can still be transmitted in a basic manner under poor channel conditions.

[0100] S33. Apply discrete cosine transform to compress image data in the frequency domain to reduce the amount of data.

[0101] The formula for two-dimensional DCT transformation is as follows: ; ;

[0102] in Let be the coefficient values ​​at position (u,v) in the transformed frequency domain. This represents the pixel value at position (x, y) in the original image spatial domain. The size of the image patch. The index of the horizontal frequency component in the frequency domain and 0 ≤ u <N, The index of the vertical frequency component in the frequency domain and 0 ≤ v <N, x is the horizontal pixel index in the spatial domain and 0 ≤ x <N, The vertical pixel index in the spatial domain and 0 ≤ y <N, These are the normalization coefficients;

[0103] First, the preprocessed image is divided into multiple 8×8 small pixel blocks. Then, the above two-dimensional DCT formula is applied to each pixel block to convert it into a frequency domain coefficient matrix composed of DC and AC components. After the transformation, the low-frequency coefficients in the upper left corner of the matrix carry the main information of the image, while the high-frequency coefficients in the lower right corner represent the detail information. Subsequently, a quantization operation is performed, that is, the DCT coefficients are divided by the corresponding values ​​in a predetermined quantization matrix and rounded. This step will reduce the value of the high-frequency coefficients or even bring them to zero, thereby achieving data compression.

[0104] S34. Compression efficiency is optimized by entropy coding to generate compressed coded image data, the size of which is adaptive to channel conditions.

[0105] After DCT and quantization, the generated coefficient matrix is ​​zigzagged to convert the two-dimensional matrix into a one-dimensional sequence, making consecutive zero-value coefficients appear more frequently. Then, entropy coding techniques such as arithmetic coding are used to compress the sequence. Entropy coding is a lossless compression technique that assigns short codewords to symbols with high probability of occurrence and long codewords to symbols with low probability of occurrence, thereby further reducing the amount of data and generating the final compressed coded image data that is adaptive to channel conditions.

[0106] The transmission sequence arrangement in S4 includes the following steps:

[0107] S41. Define flight mission priority parameters, including emergency missions, routine missions, and background missions;

[0108] S42. Use convolutional neural networks to analyze image content, identify key objects: runways, other aircraft, and unusual events, and assign importance scores.

[0109] The compressed images are analyzed using a convolutional neural network model pre-trained on an aerial imagery dataset. This CNN model is able to identify key objects in the images: runways, other aircraft, clouds, terrain, and unusual events. The model ultimately outputs an importance score for each image, based on the type of object identified and its criticality to flight safety.

[0110] Importance scoring formula: ;

[0111] in Assign importance ratings, This represents the total number of key objects identified in the image. For object indexing, For the first The confidence level of an object's identification. For the first Preset weights for each object type;

[0112] S43. Combining priority parameters and importance scores, a weighted round-robin algorithm is used to arrange the order of image data transmission.

[0113] Calculate a comprehensive weight for each image data packet: ;

[0114] in To calculate the overall transmission weight of the image data packets, These are priority parameters defined based on flight missions. The importance score is derived from the analysis of image content by a convolutional neural network. This is the task priority weighting coefficient. This refers to the weighting coefficient for content importance.

[0115] The transmission scheduler sorts each data packet according to its weight, prioritizes the transmission of data packets with higher weights, and allocates more transmission time slots to high-priority image data, thereby generating the final image transmission sequence data.

[0116] S44. Allocate more transmission time slots for high-priority image data and generate image transmission sequence data.

[0117] Real-time monitoring of packet loss rate in S5 includes the following steps:

[0118] S51. Divide the image transmission sequence data into fixed-size data packets, and add a sequence number and checksum to each data packet;

[0119] S52. High-speed transmission is achieved via UDP protocol, while ACK confirmation mechanism is enabled to monitor data packet reception status.

[0120] To balance transmission speed and reliability, the UDP protocol is used for high-speed transmission of image data packets. At the same time, a selective ACK confirmation mechanism is implemented at the application layer. The sender marks each data packet with a sequence number and sends it. After successfully receiving the data packet, the receiver returns an ACK confirmation signal to the sender, which contains the sequence number of the successfully received data packet. The sender tracks the data packet reception status by monitoring the ACK signal.

[0121] S53. Calculate the data packet loss rate in real time and trigger an alarm when the loss rate exceeds 5%; ;

[0122] in For the calculated packet loss rate, This refers to the total number of data packets sent by the sender within a specific time period. The total number of acknowledgment data packets received by the sender within a specific time period;

[0123] S54. Record transmission timestamps and channel quality indicators, and generate transmission monitoring logs.

[0124] Dynamic error correction control processing includes the following steps:

[0125] S61. When the data packet loss rate exceeds the threshold, a selective retransmission request is automatically initiated to retransmit only the lost data packets.

[0126] S62. Adjust the redundancy of the forward error correction coding and dynamically adjust the error correction code length according to the channel noise level;

[0127] The redundancy of FEC is adjusted by the length of the error-correcting code and the coding rate: ;

[0128] in For coding rate, It is the length of the original data. This is the total length after encoding;

[0129] The system dynamically adjusts the coding scheme based on the channel noise level monitored in real time. When the noise level is high and the packet loss rate is high, a lower coding rate and a longer Reed-Solomon code are used to enhance error correction capability. Conversely, when the channel conditions are good, a higher coding rate is used to reduce redundancy overhead.

[0130] S63. Reed-Solomon encoding is used to strengthen the retransmitted data;

[0131] For data packets that need to be retransmitted, Reed-Solomon encoding is performed before retransmission; RS A code can encode a raw data block of k symbols into a codeword of n symbols, and can correct up to [number missing] errors. The system will divide the retransmitted data into blocks, apply RS encoding to each block, generate a hardened data packet containing redundancy check information, and then transmit it, thereby improving the probability of successful reception of retransmitted data under poor channel conditions.

[0132] S64. Verify the integrity of the corrected data and generate the corrected image transmission data.

[0133] Decoding and reconstruction processing includes the following steps:

[0134] S71. At the receiving end, the data packets are reassembled in order, and the original sequence is restored based on the sequence number;

[0135] S72. Restore image data by applying inverse discrete cosine transform and decoding algorithm;

[0136] For each data block after sequential reassembly, entropy decoding is performed first, followed by inverse quantization, and finally, inverse discrete cosine transform is applied to restore the frequency domain coefficients back to the spatial domain pixel values; the formula for IDCT is the inverse process of DCT: ;

[0137] in For the reconstructed image spatial domain The pixel value of the location, For the frequency domain after inverse quantization The coefficient value of the location, The size of the image patch. This is the index of the horizontal frequency components in the frequency domain. This is the index of the vertical frequency component in the frequency domain. For the horizontal pixel index in the spatial domain This is the vertical pixel index in the spatial domain. These are the normalization coefficients;

[0138] S73. Perform color space conversion and sharpening to improve the visual effect of the image; the original image underwent color space conversion before compression, so the inverse conversion is performed here to restore the RGB format visual image; in order to improve the image clarity that may be lost due to compression and transmission, the Laplacian operator image sharpening filter is used for post-processing to enhance the image edge details and improve the visual effect.

[0139] S74 outputs visual image data, supporting real-time display and storage.

[0140] Generating an image transmission quality report in S8 includes the following steps:

[0141] S81. Use pre-trained deep learning models to evaluate image quality, including metrics such as peak signal-to-noise ratio and structural similarity index.

[0142] A pre-trained deep learning model—a proprietary model trained on the ResNet architecture—is used to evaluate the decoded and reconstructed image. This model outputs a quality score consistent with human subjective perception, with evaluation metrics including peak signal-to-noise ratio (PSNR) and structural similarity index. PSNR is calculated based on mean squared error, using the following formula: ; ;

[0143] in Peak signal-to-noise ratio, This represents the maximum possible pixel value for a given image pixel. The mean square error between the original image and the reconstructed image. For the original image, To reconstruct the image, and For the dimensions of the image, For the original image at position Pixel value at that location, To reconstruct the image at location Pixel value at;

[0144] S82. Compare the differences between the images before and after transmission, and calculate the distortion score;

[0145] The system aligns and performs pixel-level comparisons between the image finally received and decoded by the ground station and the original image that has been preprocessed at the aircraft end but has not yet been transmitted. By calculating PSNR and SSIM indices and combining them with the comprehensive score given by the deep learning model, a comprehensive distortion score is calculated, which quantitatively reflects the impact of the entire transmission process on image quality.

[0146] Distortion rating:

[0147] in Rate the distortion. , , These are the weighting coefficients. Peak signal-to-noise ratio, It is a structural similarity index. A quality score for the output of a deep learning model; ;

[0148] in The mean of the original image and the reconstructed image. The standard deviations of the original image and the reconstructed image are given. The covariance of the original image and the reconstructed image. It is a constant;

[0149] S83. Generate an image transmission quality report, including the quality level and improvement suggestions.

[0150] The system includes an image acquisition and preprocessing module, an intelligent compression and scheduling module, a transmission control and error correction module, and a receiving, decoding, and evaluation module.

[0151] The image acquisition and preprocessing module acquires raw images inside and outside the aircraft through the multimodal image acquisition unit, optimizes the images through the noise reduction and enhancement unit, and outputs the preprocessed image data through the format standardization unit.

[0152] The intelligent compression and scheduling module receives preprocessed image data, obtains real-time communication parameters through the channel state sensing unit, compresses data using the adaptive coding unit, and outputs image transmission sequence data through the priority scheduling unit.

[0153] The transmission control and error correction module receives image transmission sequence data, performs segmented transmission through the wireless data transceiver unit, collects data packet loss information through the transmission status monitoring unit, and outputs the corrected image transmission data through the dynamic error correction unit.

[0154] The receiving, decoding, and evaluation module receives the image transmission data after error correction, restores the visible image through the data decoding and reconstruction unit, generates an image quality report through the transmission quality evaluation unit, and outputs the evaluation results.

[0155] The operation steps of this method and system are as follows:

[0156] First, image data of the flight environment, pilot operation status, and equipment operation are collected in real time using image acquisition equipment on the aircraft, including high-definition cameras on the exterior of the fuselage and visible light cameras and infrared sensors inside the cabin. This data is then labeled with timestamps and GPS coordinates to generate a structured raw image dataset. Next, the raw image dataset is preprocessed. Gaussian filtering is used for noise reduction, histogram equalization is used to enhance contrast and brightness, and all images are uniformly converted to the standard H.264 encoding format and a resolution of 1920x1080 pixels to generate preprocessed image data.

[0157] Based on this, the system dynamically selects lossless and lossy compression algorithms according to the aviation communication channel status parameters obtained through real-time monitoring. It then applies discrete cosine transform and entropy coding to adaptively compress and encode the preprocessed image data, generating compressed coded image data with a data volume adapted to the current channel conditions. Next, an intelligent scheduling algorithm is employed. Based on predefined flight mission priorities and importance scores obtained through convolutional neural network analysis of image content, a weighted round-robin algorithm is used to arrange the compressed image data into a transmission sequence, generating image transmission sequence data to ensure that high-priority images are transmitted first.

[0158] During the transmission phase, the image transmission sequence data is segmented into data packets with sequence numbers and checksums via the wireless transmission module of the aviation communication radio. These packets are then sent to the ground control station in segments using the UDP protocol and an ACK confirmation mechanism. Simultaneously, the data packet loss rate is monitored in real time. If the loss rate exceeds a threshold, dynamic error correction control is immediately initiated. Lost data packets are retransmitted through a selective retransmission request mechanism, and the redundancy of the forward error correction coding is dynamically adjusted. Reed-Solomon coding is used to strengthen the data, generating error-corrected image transmission data.

[0159] Finally, at the receiving end, the error-corrected image transmission data packets are sequentially reassembled, and the inverse discrete cosine transform and decoding algorithm are applied for decoding and reconstruction. After color space conversion and sharpening, the images are restored to high-quality visual image data. Then, combined with a pre-trained deep learning model, the quality of the restored images is evaluated, and the peak signal-to-noise ratio, structural similarity index, and distortion score are calculated. Finally, a comprehensive image transmission quality report is generated, completing the entire image transmission control process.

[0160] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0161] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for controlling image transmission in an aviation communication radio, characterized in that: The method includes the following steps: S1. Real-time acquisition of image data inside and outside the aircraft via image acquisition equipment from aviation communication radio, and generation of raw image dataset; S2. Perform image preprocessing operations based on the original image dataset, including denoising, enhancement, and format normalization, and generate preprocessed image data; S3. Based on the real-time status parameters of the aviation communication channel, perform adaptive compression coding on the preprocessed image data to generate compressed coded image data. S4. An intelligent scheduling algorithm is used to arrange the transmission sequence of the compressed coded image data according to the flight mission priority and the importance of the image content, thereby generating image transmission sequence data; S5. The image transmission sequence data is transmitted to the ground control station in segments through the wireless transmission module of the aviation communication radio, and the data packet loss rate is monitored in real time during the transmission process. S6. Perform dynamic error correction control processing based on monitoring results, including retransmission mechanism and forward error correction coding adjustment, and generate error-corrected image transmission data; S7. At the receiving end, the error-corrected image transmission data is decoded and reconstructed to restore it to visual image data. S8. Combine the artificial intelligence model to perform quality assessment on the visual image data and generate an image transmission quality report.

2. The image transmission control method for an aviation communication radio according to claim 1, characterized in that: The acquisition of image data of the aircraft's interior and exterior in S1 includes the following steps: S11. Deploy multiple high-definition cameras on the exterior of the aircraft fuselage to collect images of the flight environment, including weather conditions, terrain and obstacle information; S12. Install visible light cameras and infrared sensors inside the aircraft cockpit to collect images of the pilot's operating status and the operation of the equipment inside the cockpit. S13. Mark the collected image data using timestamps and GPS coordinates; S14. Integrate the labeled image data into a structured dataset to generate the original image dataset, whose data format includes JPEG, PNG and RAW formats.

3. The image transmission control method for an aviation communication radio according to claim 1, characterized in that: The image preprocessing operation includes the following steps: S21. Use Gaussian filtering algorithm to perform noise reduction processing on the original image dataset to reduce random interference during image acquisition; S22. Enhance image contrast and brightness through histogram equalization technology to improve image visibility under low light conditions; S23. Convert the enhanced image data into the standard H.264 encoding format to ensure compatibility with different aviation communication equipment; S24. Perform resolution standardization processing on the converted image data, adjust all images to 1920x1080 pixels, and generate the preprocessed image data.

4. The image transmission control method for an aviation communication radio according to claim 1, characterized in that: The adaptive compression coding process includes the following steps: S31. Monitor the bandwidth utilization and bit error rate of aviation communication channels in real time and generate channel status indicators; S32. Based on the channel state index, dynamically select the compression ratio: when the bandwidth is higher than the threshold, use a lossless compression algorithm; when the bandwidth is lower than the threshold, use a lossy compression algorithm. S33. Apply discrete cosine transform to compress image data in the frequency domain to reduce the amount of data. S34. Compression efficiency is optimized through entropy coding to generate compressed coded image data, the size of which is adaptive to channel conditions.

5. The image transmission control method for an aviation communication radio according to claim 1, characterized in that: The transmission sequence arrangement in S4 includes the following steps: S41. Define flight mission priority parameters, including emergency missions, routine missions, and background missions; S42. Use convolutional neural networks to analyze image content, identify key objects: runways, other aircraft, and unusual events, and assign importance scores. Importance scoring formula: ； in Assign importance ratings, This represents the total number of key objects identified in the image. For object indexing, For the first The confidence level of an object's identification. For the first Preset weights for each object type; S43. Combining priority parameters and importance scores, a weighted round-robin algorithm is used to arrange the order of image data transmission. S44. Allocate more transmission time slots for high-priority image data and generate image transmission sequence data.

6. The image transmission control method for an aviation communication radio according to claim 1, characterized in that: The real-time monitoring of data packet loss rate in S5 includes the following steps: S51. Divide the image transmission sequence data into fixed-size data packets, and add a sequence number and checksum to each data packet; S52. High-speed transmission is achieved via UDP protocol, while ACK confirmation mechanism is enabled to monitor data packet reception status. S53. Calculate the packet loss rate in real time and trigger an alarm when the loss rate exceeds 5%; S54. Record transmission timestamps and channel quality indicators, and generate transmission monitoring logs.

7. The image transmission control method for an aviation communication radio according to claim 1, characterized in that: The dynamic error correction control process includes the following steps: S61. When the data packet loss rate exceeds the threshold, a selective retransmission request is automatically initiated to retransmit only the lost data packets. S62. Adjust the redundancy of the forward error correction coding and dynamically adjust the error correction code length according to the channel noise level; S63. Reed-Solomon encoding is used to strengthen the retransmitted data; S64. Verify the integrity of the corrected data and generate the corrected image transmission data.

8. The image transmission control method for an aviation communication radio according to claim 1, characterized in that: The decoding and reconstruction process includes the following steps: S71. At the receiving end, the data packets are reassembled in order, and the original sequence is restored based on the sequence number; S72. Restore image data by applying inverse discrete cosine transform and decoding algorithm; S73. Perform color space conversion and sharpening to improve the visual effect of the image; S74 outputs visual image data, supporting real-time display and storage.

9. The image transmission control method for an aviation communication radio according to claim 1, characterized in that: The process of generating an image transmission quality report in S8 includes the following steps: S81. Use pre-trained deep learning models to evaluate image quality, including metrics such as peak signal-to-noise ratio and structural similarity index. S82. Compare the differences between the images before and after transmission, and calculate the distortion score; Distortion rating: ； in Rate the distortion. , , These are the weighting coefficients. Peak signal-to-noise ratio, It is a structural similarity index. A quality score for the output of a deep learning model; S83. Generate an image transmission quality report, including the quality level and improvement suggestions.

10. An image transmission control system for an aviation communication radio, used to implement the image transmission control method for an aviation communication radio as described in any one of claims 1-9, characterized in that: The system includes an image acquisition and preprocessing module, an intelligent compression and scheduling module, a transmission control and error correction module, and a receiving, decoding, and evaluation module. The image acquisition and preprocessing module acquires original images inside and outside the aircraft through a multimodal image acquisition unit, optimizes the images through a noise reduction and enhancement unit, and outputs the preprocessed image data through a format standardization unit. The intelligent compression and scheduling module receives the preprocessed image data, obtains real-time communication parameters through the channel state sensing unit, compresses the data using the adaptive coding unit, and outputs image transmission sequence data through the priority scheduling unit. The transmission control and error correction module receives the image transmission sequence data, performs segmented transmission through the wireless data transceiver unit, collects data packet loss information using the transmission status monitoring unit, and outputs the corrected image transmission data through the dynamic error correction unit. The receiving, decoding, and evaluation module receives the error-corrected image transmission data, restores the visible image through the data decoding and reconstruction unit, generates an image quality report through the transmission quality evaluation unit, and outputs the evaluation results.

Images