Image transmission method and system for highway unmanned aerial vehicle inspection

Abstract

The application belongs to the technical field of image transmission, and provides an image transmission method and system for expressway unmanned aerial vehicle inspection, which comprises obtaining state feature data of the expressway in the inspection image, historical occurrence frequency of the corresponding road section abnormal condition and transmission code rate of the image; further, a low code rate transmission feasibility evaluation model is established by fusing the state feature and the historical frequency, the feasibility score of each image is generated, and the low code rate transmission image sequence is screened out; at the same time, a transmission link state evaluation value is generated according to the image transmission code rate; on this basis, an optimization model is established with the low code rate transmission quantity as the independent variable and the link state evaluation value as the dependent variable, and the optimal low code rate transmission quantity is solved; finally, the images in the low code rate transmission sequence are transmitted at a reduced code rate according to the optimal quantity; the application can dynamically adjust the code rate transmission according to the image content and historical data, optimize the bandwidth utilization, and improve the image transmission quality of the key area.

Description

Technical Field

[0001] This invention belongs to the field of image transmission technology, and in particular relates to an image transmission method and system for highway unmanned aerial vehicle (UAV) inspection. Background Technology

[0002] In highway drone inspection operations, real-time transmission of inspection images is a core component of ensuring road safety monitoring. However, the wireless communication link bandwidth between the drone and the ground station has inherent limitations, especially during long-distance flights or in complex terrain or inclement weather, where available bandwidth is significantly reduced. When all inspection images adopt a uniform high bit rate transmission strategy, link congestion is highly likely to occur, leading to increased data transmission delays, higher image frame loss rates, and even communication interruptions, severely compromising the timeliness and reliability of the inspection mission.

[0003] In existing technologies, fixed-rate transmission schemes cannot adapt to dynamically changing link conditions. Adaptive rate adjustment mechanisms based on signal strength only focus on physical layer indicators while ignoring the value of image content. Optimization of image compression algorithms often comes at the cost of visual quality. Although some systems introduce simple priority ranking based on acquisition time or GPS location, such methods are too crude and fail to deeply integrate with the actual needs of highway scenarios. At the bandwidth management level, current practices mostly rely on manually preset static allocation strategies or passive responses based solely on instantaneous link quality, lacking the ability to proactively optimize the transmission process. These technical deficiencies lead to a severe disconnect between rate adjustment decisions and the importance of image content. The image quality of critical abnormal areas (such as road cracks, obstacles, or traffic accident sites) may deteriorate due to inappropriate low-rate transmission, while images of non-critical areas consume excessive bandwidth resources.

[0004] Furthermore, existing methods completely ignore the impact of historical anomaly frequency. For areas with high accident rates and other areas where historical data indicates higher risk, inspection images should receive higher priority for transmission, but the technical solutions fail to reflect these differentiated needs. More significantly, the selection of the number of low-bitrate transmission images lacks scientific basis. It is impossible to quantitatively assess the impact of transmission quantity on link load, and it is also difficult to establish a dynamic balance mechanism between reducing bandwidth pressure and maintaining the integrity of critical information, thus limiting the overall system performance. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an image transmission method and system for unmanned aerial vehicle (UAV) inspection of highways, thus solving the aforementioned problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an image transmission method for unmanned aerial vehicle (UAV) inspection of highways, the method specifically comprising:

[0007] Acquire state characteristic data of highways in UAV inspection images, historical frequency of highway anomalies, and transmission bit rate of UAV inspection images;

[0008] Based on the status characteristic data of highways in UAV inspection images and the historical frequency of abnormal highway conditions, a feasibility assessment model for low bit rate transmission of UAV inspection images is established, and a feasibility score for low bit rate transmission of UAV inspection images is generated.

[0009] Based on the feasibility score of low bit rate transmission of UAV inspection images, a sequence of UAV inspection images that can be transmitted at low bit rate is generated; the sequence of UAV inspection images that can be transmitted at low bit rate refers to the sequence after the UAV inspection images are arranged according to the feasibility score of low bit rate transmission of UAV inspection images.

[0010] Based on the transmission bit rate of the UAV inspection images, a transmission link status assessment value is generated;

[0011] Based on the low-bit-rate transmission sequence of UAV inspection images, a low-bit-rate transmission quantity analysis model is established with the number of low-bit-rate transmissions of UAV inspection images as the independent variable and the transmission link status evaluation value as the dependent variable, to generate the optimal number of low-bit-rate transmissions of UAV inspection images.

[0012] The drone inspection images are transmitted based on the optimal low bit rate transmission quantity.

[0013] Based on the above technical solutions, the present invention also provides the following optional technical solutions:

[0014] Further technical solutions: The specific method for generating the feasibility score for low bit rate transmission of UAV inspection images includes:

[0015] Based on the status feature data of highways in UAV inspection images, a basic score for the feasibility of low bit rate transmission is generated.

[0016] Based on the historical frequency of abnormal conditions on highways, a historical abnormal condition impact coefficient is generated; whereby the historical frequency of abnormal conditions on highways refers to the historical frequency of abnormal conditions occurring at the highway locations corresponding to the UAV inspection images.

[0017] A feasibility assessment model for low bit rate transmission of UAV inspection images is established based on the basic feasibility score for low bit rate transmission and the impact coefficient of historical anomalies, and a feasibility score for low bit rate transmission of UAV inspection images is generated.

[0018] Further technical solutions: The specific method for generating the basic score for the feasibility of low bit rate transmission includes:

[0019] Through the formula: ;

[0020] Generate a basic feasibility score for low bit rate transmission ;

[0021] In the formula, This refers to the basic score for the feasibility of low-bitrate transmission of the i-th drone inspection image. This represents the normalized value of the j-th state feature data of the highway in the i-th UAV inspection image. represents the weight coefficient of the j-th state feature of the highway in the drone inspection image, and s represents the number of state feature data of the highway in the drone inspection image.

[0022] Further technical solution: The specific method for generating the influence coefficient of the historical anomaly includes:

[0023] Through the formula: ;

[0024] Generate historical anomaly impact coefficient ;

[0025] In the formula, This indicates the historical frequency of abnormal situations on highways. This represents the threshold for the frequency of abnormal situations.

[0026] Further technical solution: The specific expression of the feasibility assessment model for low bit rate transmission of UAV inspection images is as follows: ;

[0027] In the expression, This represents the feasibility score for low-bitrate transmission of the i-th drone inspection image. This represents the basic score for the feasibility of low-bitrate transmission of the i-th drone inspection image. This represents the influence coefficient of historical anomalies.

[0028] A further technical solution: The specific method for generating the transmission link state assessment value is as follows:

[0029] Through the formula: ;

[0030] Generate the transmission link state assessment value C;

[0031] In the formula, This represents the estimated bandwidth utilization of the transmission link. This represents the broadband utilization threshold.

[0032] Further technical solutions: The method for obtaining the estimated bandwidth utilization of the transmission link specifically includes:

[0033] Through the formula: ;

[0034] Estimated bandwidth utilization of the transmission link ;

[0035] In the formula, This represents the transmission bit rate of the k-th UAV inspection image, B represents the total available bandwidth of the transmission link, and n represents the number of UAV inspection images. This represents the fixed consumption item for transmission link bandwidth utilization.

[0036] Further technical solution: The expression for the low bit rate transmission quantity analysis model is specifically as follows: ;

[0037] In the expression, This represents the optimal number of low-bitrate images transmitted during drone inspections. This represents the transmission link status assessment value when the number of low-bit-rate transmissions of UAV inspection images is m, and A represents the set of numbers of UAV inspection images that can be transmitted at low bit rate.

[0038] An image transmission system for highway drone inspection, the system being used to execute the aforementioned image transmission method for highway drone inspection, specifically including:

[0039] The data acquisition unit is used to acquire the status characteristic data of highways in UAV inspection images, the historical frequency of abnormal highway conditions, and the transmission bit rate of UAV inspection images.

[0040] The feasibility analysis unit is used to establish a feasibility assessment model for low bit rate transmission of UAV inspection images based on the status characteristic data of highways in UAV inspection images and the historical frequency of abnormal highway conditions, and to generate a feasibility score for low bit rate transmission of UAV inspection images.

[0041] The sorting unit is used to generate a sequence of UAV inspection images that can be transmitted at low bit rate based on the feasibility score of low bit rate transmission of UAV inspection images. The sequence of UAV inspection images that can be transmitted at low bit rate refers to the sequence after the UAV inspection images are arranged according to the feasibility score of low bit rate transmission of UAV inspection images.

[0042] The transmission link status analysis unit is used to generate a transmission link status evaluation value based on the transmission bit rate of the UAV inspection image.

[0043] The low bit rate transmission quantity analysis unit is used to establish a low bit rate transmission quantity analysis model based on the low bit rate transmission sequence of UAV inspection images, with the number of low bit rate transmissions of UAV inspection images as the independent variable and the transmission link status evaluation value as the dependent variable, to generate the optimal number of low bit rate transmissions of UAV inspection images.

[0044] The image transmission unit is used to transmit UAV inspection images according to the optimal low bit rate transmission quantity of UAV inspection images.

[0045] Further technical solution: The feasibility analysis unit specifically includes:

[0046] The basic analysis module is used to generate a basic score for the feasibility of low bit rate transmission based on the status feature data of highways in UAV inspection images.

[0047] The historical data analysis module is used to generate historical anomaly impact coefficients based on the historical frequency of highway anomalies; where the historical frequency of highway anomalies refers to the historical frequency of anomalies at the highway locations corresponding to the drone inspection images.

[0048] The comprehensive analysis module is used to establish a feasibility assessment model for low bit rate transmission of UAV inspection images based on the basic feasibility score for low bit rate transmission and the impact coefficient of historical anomalies, and to generate a feasibility score for low bit rate transmission of UAV inspection images.

[0049] This invention provides an image transmission method and system for unmanned aerial vehicle (UAV) inspection of highways, which has the following advantages compared with the prior art:

[0050] This invention assesses the feasibility of image transmission and optimizes the amount of low-bitrate transmission. It can dynamically adjust the bitrate transmission based on image content and historical data, optimize bandwidth utilization, and improve the image transmission quality in key areas. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating an image transmission method for highway drone inspection provided by the present invention.

[0052] Figure 2 This is a flowchart illustrating step S20 of the present invention.

[0053] Figure 3 This invention provides a schematic diagram of an image transmission system for highway drone inspection.

[0054] Figure 4 This is a schematic diagram of the feasibility analysis unit provided by the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0056] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0057] Please see Figure 1 An image transmission method for highway unmanned aerial vehicle (UAV) inspection, provided in one embodiment of the present invention, includes the following steps:

[0058] Step S10: Obtain the status feature data of the highway in the UAV inspection image, the historical frequency of abnormal conditions on the highway, and the transmission bit rate of the UAV inspection image;

[0059] Step S20: Based on the status feature data of highways in UAV inspection images and the historical frequency of abnormal highway conditions, establish a feasibility assessment model for low bit rate transmission of UAV inspection images and generate a feasibility score for low bit rate transmission of UAV inspection images.

[0060] Step S30: Generate a sequence of drone inspection images that can be transmitted at low bit rate based on the feasibility score of low bit rate transmission of drone inspection images; the sequence of drone inspection images that can be transmitted at low bit rate refers to the sequence after the drone inspection images are arranged according to the feasibility score of low bit rate transmission of drone inspection images.

[0061] Step S40: Generate a transmission link status assessment value based on the transmission bit rate of the UAV inspection image;

[0062] Step S50: Based on the low bit rate transmission sequence of UAV inspection images, establish a low bit rate transmission quantity analysis model with the number of low bit rate transmissions of UAV inspection images as the independent variable and the transmission link status evaluation value as the dependent variable, and generate the optimal number of low bit rate transmissions of UAV inspection images.

[0063] Step S60: Transmit the UAV inspection images according to the optimal low bit rate transmission quantity of the UAV inspection images;

[0064] Among them, the condition feature data of highways in drone inspection images refers to quantitative information extracted from drone inspection images that reflects the current condition of the highway, such as the length and width of road surface cracks, the depth of potholes, the degree of traffic congestion, and the type of guardrail damage. This data is used to assess the operational status and potential risks of highways;

[0065] The historical frequency of abnormal situations refers to the statistical frequency of traffic accidents, facility damage, or other abnormal events that occur in a specific section or area of ​​a highway over a past period of time. This data reflects the risk level of a specific area and is of reference value for predicting the probability of future abnormal events.

[0066] Transmission rate refers to the amount of digital data transmitted through a communication link per unit of time, usually measured in bits per second (bps). Higher transmission rates typically correspond to higher image quality and lower compression rates, but also consume more bandwidth resources.

[0067] A low-bitrate transmission sequence refers to a sequence obtained by sorting all images to be transmitted based on their feasibility score for low-bitrate transmission of UAV inspection images. In this sequence, images with higher scores are given priority for low-bitrate transmission to ensure that important images can be transmitted at high quality when bandwidth is limited, while the bitrate of secondary images can be reduced. In addition, a low-bitrate transmission feasibility score threshold can be preset when generating this sequence. UAV inspection images with a low-bitrate transmission feasibility score lower than the low-bitrate transmission feasibility score threshold are not included in the low-bitrate transmission sequence. The low-bitrate transmission feasibility score threshold is a preset value used to filter UAV inspection images that can be transmitted at low bitrate. Only when the low-bitrate transmission feasibility score of a UAV inspection image is higher than the threshold is it included in the low-bitrate transmission sequence.

[0068] Specifically, firstly, in step S10, it is necessary to acquire the highway status feature data, the historical frequency of highway anomalies, and the transmission bitrate of the drone inspection images from the images. The highway status feature data can be extracted from the drone inspection images using image processing techniques, for example, by manually visually identifying road surface damage and traffic congestion in the images and manually inputting the corresponding feature values. The historical frequency of highway anomalies can be queried from a historical accident record database, for example, by manually counting the number of accidents on a specific road section in the past year. The transmission bitrate of the drone inspection images can be directly provided by the image encoder or transmission module on the drone, for example, by reading the currently set output bitrate of the encoder.

[0069] Further, in step S20, based on the state characteristic data of highways in the UAV inspection images and the historical frequency of highway anomalies, a feasibility assessment model for low-bitrate transmission of UAV inspection images is established, and a feasibility score for low-bitrate transmission of UAV inspection images is generated. Low-bitrate transmission refers to reducing the transmission bitrate of UAV inspection images. For example, a simple weighted summation model can be used, assigning different weights to different state characteristic data (such as the degree of road surface damage and traffic flow), and combining this with the historical frequency of anomalies to calculate a comprehensive score. The higher the score, the more suitable the image is for low-bitrate transmission, because its content may not contain highly urgent or information requiring extremely high precision.

[0070] Based on this, in step S30, a sequence of UAV inspection images that can be transmitted at a low bit rate is generated according to the feasibility score for low bit rate transmission of the UAV inspection images. This sequence refers to the sequence of UAV inspection images arranged according to their feasibility scores. Specifically, all UAV inspection images to be transmitted can be sorted from highest to lowest according to their feasibility scores for low bit rate transmission. Then, a feasibility score threshold is set; for example, all images with scores higher than this threshold are considered suitable for low bit rate transmission and included in the sequence.

[0071] Simultaneously, in step S40, a transmission link status assessment value is generated based on the transmission bitrate of the UAV inspection images. For example, the link load can be roughly estimated by monitoring the ratio between the total bitrate of all images to be transmitted and the total available bandwidth of the transmission link. When the total bitrate approaches or exceeds the available bandwidth, the transmission link status assessment value will increase accordingly, indicating that the link may be in a congested state.

[0072] Subsequently, in step S50, based on the sequence of UAV inspection images that can be transmitted at low bitrates, a low bitrate transmission quantity analysis model is established with the number of UAV inspection images transmitted at low bitrates as the independent variable and the transmission link status evaluation value as the dependent variable, generating the optimal number of UAV inspection images transmitted at low bitrates. For example, a simple lookup table or empirical curve can be constructed, which directly provides a suggested number of low bitrate transmission images based on the current transmission link status evaluation value. The goal of this model is to reduce the link load as much as possible while ensuring transmission quality.

[0073] Finally, in step S60, the UAV inspection images are transmitted according to the optimal number of low-bitrate transmission images. Specifically, images with the optimal number of low-bitrate transmission images are selected from the sequence for low-bitrate transmission, while the remaining images are transmitted at the standard bitrate or a higher bitrate. For example, if the optimal number of low-bitrate transmission images is 5, then the 5 images with the highest scores are selected from the sequence for low-bitrate encoding and transmission, while the remaining images are transmitted at their original bitrate.

[0074] According to the above technical solution, the present invention achieves intelligent dynamic control of UAV inspection image transmission by integrating image content features, historical data and link status. It effectively solves the technical problems existing in the prior art, such as insufficient protection of key information transmission, lack of consideration for high-risk road sections and lack of basis for decision-making on the number of low bit rate transmissions, and improves the real-time performance and reliability of UAV inspection tasks on highways.

[0075] For preferred options, please refer to [link / reference]. Figure 2 The present invention further proposes a method for generating a feasibility score for low bit rate transmission of UAV inspection images, specifically including:

[0076] Step S21: Generate a basic score for the feasibility of low bit rate transmission based on the status feature data of the highway in the UAV inspection images;

[0077] Step S22: Generate historical anomaly impact coefficients based on the historical frequency of highway anomalies; where the historical frequency of highway anomalies refers to the historical frequency of anomalies at the highway locations corresponding to the UAV inspection images.

[0078] Step S23: Establish a feasibility assessment model for low bit rate transmission of UAV inspection images based on the basic feasibility score for low bit rate transmission and the impact coefficient of historical anomalies, and generate a feasibility score for low bit rate transmission of UAV inspection images.

[0079] In step S21, a basic feasibility score for low-bitrate transmission is generated based on the state feature data of the highway in the UAV inspection image. This aims to preliminarily assess the redundancy or information density of the image content itself. For example, image processing algorithms can be used to analyze features such as texture complexity, edge information content, and color distribution uniformity. When the image shows smooth highway conditions, sparse traffic flow, and good weather, the effective information contained in the image may be relatively small, with high redundancy. In this case, a higher basic feasibility score for low-bitrate transmission can be assigned. Conversely, if the image shows cracks, potholes, or foreign objects on the road surface, or traffic congestion and severe weather (such as heavy fog or heavy rain), the image information density is high, and details are crucial. In this case, a lower basic feasibility score for low-bitrate transmission will be assigned. Alternatively, a machine learning model can be trained, using the visual features of the image as input, to directly output a quantified basic feasibility score for low-bitrate transmission.

[0080] In step S22, a historical anomaly impact coefficient is generated based on the historical frequency of highway anomalies. The purpose is to incorporate historical experience data to correct for potential risks associated with low bitrate transmission. The historical frequency of highway anomalies refers to the statistical frequency of past abnormal events such as accidents, congestion, construction, and natural disasters occurring on a specific highway segment covered by the UAV inspection image. For example, a Geographic Information System (GIS) database can be maintained to record historical event data for each highway segment. When a UAV inspection image corresponds to a specific segment, the system queries the historical anomaly frequency for that segment. If a segment has a high historical frequency of anomalies, even if the current image content appears simple, more details may need to be retained due to its high-risk attributes; in this case, the historical anomaly impact coefficient will be higher. Conversely, if a segment has a low historical frequency of anomalies, the impact coefficient will be lower.

[0081] In step S23, a feasibility assessment model for low-bitrate transmission of UAV inspection images is established based on the basic feasibility score for low-bitrate transmission and the impact coefficient of historical anomalies, generating the final feasibility score for low-bitrate transmission of UAV inspection images. This step combines the assessment of the image content itself with the historical risk assessment, forming a more comprehensive and accurate decision-making basis. For example, a multiplicative correction model can be used, multiplying the basic feasibility score for low-bitrate transmission by a correction factor related to the impact coefficient of historical anomalies, or a weighted average method can be used for comprehensive analysis. In this way, even if the image content itself shows low information density, if the corresponding highway section has a high frequency of historical anomalies, the final feasibility score for low-bitrate transmission will be lowered accordingly, thereby avoiding inappropriate low-bitrate transmission in potentially high-risk areas.

[0082] This application's solution decomposes the feasibility assessment of low-bitrate transmission of UAV inspection images into a basic feasibility score based on the image content itself and an impact coefficient of historical anomalies based on highway historical anomalies. Furthermore, it combines these two factors to establish a feasibility assessment model for low-bitrate transmission of UAV inspection images. This assessment method ensures that low-bitrate transmission decisions no longer rely solely on the immediate visual information of the image, but comprehensively consider the long-term risk characteristics of the area. Through this dual consideration, the system can more intelligently determine which images can be safely transmitted at low bitrates to save bandwidth, and which images, even if their current content seems unimportant, require higher bitrate transmission to ensure information integrity due to the historically high risk of the area they are located in. This optimizes image transmission efficiency while maintaining inspection quality.

[0083] Through the above technical solution, this application can more comprehensively and accurately assess the feasibility of low-bitrate transmission of UAV inspection images. By introducing the historical frequency of abnormal conditions on highways as a correction factor, it avoids misjudgments that may result from judging solely based on current image content, especially on road sections where visual information is not prominent but historical risk is high, ensuring the complete transmission of critical information. This allows the system to significantly improve the reliability and accuracy of UAV inspections on highways while effectively saving transmission bandwidth, thereby improving overall inspection efficiency and safety.

[0084] Preferably, the present invention further proposes a method for generating the basic score for the feasibility of low bit rate transmission, specifically including:

[0085] Through the formula: ;

[0086] Generate a basic feasibility score for low bit rate transmission ;

[0087] In the formula, This refers to the basic score for the feasibility of low-bitrate transmission of the i-th drone inspection image. This represents the normalized value of the j-th state feature data of the highway in the i-th UAV inspection image. represents the weight coefficient of the j-th state feature of the highway in the UAV inspection image, and s represents the number of state feature data of the highway in the UAV inspection image.

[0088] The low-bitrate transmission feasibility score aims to quantify the inherent transmission priority or importance of the i-th UAV inspection image, without considering historical anomalies. Its value can reflect the level of detail required by the image content; for example, a higher value may indicate that the image content is relatively less critical and more suitable for low-bitrate transmission. This score can be a normalized value between 0 and 1, or a comprehensive index reflecting the image's information density or urgency.

[0089] The normalized value of the j-th state feature data of the highway in the i-th UAV inspection image is a numerical representation of the j-th specific state feature of the highway in the i-th UAV inspection image after standardization. The purpose of normalization is to eliminate the differences in units and numerical ranges between different features, ensuring their comparability in calculation. For example, normalization can use the min-max scaling method to map the feature values ​​to the interval [0,1]; or it can use the Z-score standardization method to make the feature values ​​follow a distribution with a mean of 0 and a standard deviation of 1.

[0090] The weight coefficients of the j-th state feature of a highway in a drone inspection image are used to measure the relative importance of the j-th highway state feature in assessing the feasibility of low-bit-rate image transmission. Different features may have different requirements for image transmission quality; for example, the severity of road surface cracks may be more decisive than the condition of roadside vegetation. These weight coefficients can be set based on expert experience or trained and optimized using machine learning models, combining historical data and transmission results.

[0091] 's' represents the number of highway condition feature data points in the drone inspection image. This parameter indicates the total categories or number of highway condition features considered when evaluating a single drone inspection image. For example, if road surface conditions, traffic flow, and weather conditions are considered simultaneously, then the value of 's' is 3.

[0092] This application's solution quantifies multiple state feature data of highways in UAV inspection images and performs a weighted summation based on their importance to generate a basic score for low-bitrate transmission feasibility. Specifically, for each UAV inspection image, various state feature data of the highway are first identified and extracted. Then, these raw feature data are normalized to eliminate dimensional differences between different features, ensuring fairness in subsequent calculations. Next, a weight coefficient is assigned to each normalized state feature data, reflecting its relative importance in assessing the feasibility of low-bitrate transmission. Finally, all normalized feature data are multiplied by their corresponding weight coefficients, and the products are summed to obtain the basic score for low-bitrate transmission feasibility of the UAV inspection image. This weighted summation-based calculation method allows the basic score to objectively and comprehensively reflect the inherent characteristics of the image content, providing a quantitative and reliable input for the subsequent low-bitrate transmission feasibility assessment model, thereby improving the scientific rigor and accuracy of the entire image transmission decision.

[0093] Through the above technical solution, this application provides a quantitative and objective method for generating a basic score for the feasibility of low bitrate transmission. By normalizing and weighting the various state feature data of highways in UAV inspection images, the inherent characteristics of the image content can be systematically evaluated, thereby overcoming the subjectivity and uncertainty that may exist in traditional methods. This allows the subsequent UAV inspection image low bitrate transmission feasibility assessment model to obtain more accurate and reliable input, thus optimizing the accuracy of low bitrate transmission decisions, ensuring that while guaranteeing the transmission of key information, the bandwidth consumption is effectively reduced, and improving the efficiency and reliability of highway UAV inspection image transmission.

[0094] Preferably, the present invention further proposes a method for generating the influence coefficient of the historical anomaly situation, specifically including:

[0095] Through the formula: ;

[0096] Generate historical anomaly impact coefficient ;

[0097] In the formula, This indicates the historical frequency of abnormal situations on highways. This represents the threshold for the frequency of abnormal situations.

[0098] The historical anomaly impact coefficient quantifies the influence of the historical frequency of highway anomalies on the feasibility of low-bitrate transmission of UAV inspection images. Its calculation formula ensures that the historical anomaly impact coefficient is positive only when the historical frequency of highway anomalies exceeds a preset anomaly frequency threshold, thus negatively impacting the feasibility assessment of low-bitrate transmission. If the historical frequency of highway anomalies does not exceed the preset anomaly frequency threshold, the historical anomaly impact coefficient is zero, indicating that historical anomalies do not negatively affect the feasibility of low-bitrate transmission.

[0099] in, This refers to the statistical frequency of abnormal conditions (such as traffic accidents, road surface damage, and traffic congestion) occurring within a highway area corresponding to the current drone inspection image over a past period. This frequency is an important indicator for assessing the potential risks in the area. This data can be statistically analyzed and calculated from various sources, including historical accident records from traffic management departments, data collected by road sensor networks, and analysis results of past drone inspection images. For example, it can be the ratio of the number of abnormal events occurring on a specific road segment in the past year to the total number of inspections or total operating time for that road segment.

[0100] and This is a preset frequency threshold used to determine whether historical anomalies have reached a level requiring special attention. Its function is to act as a filtering condition, ensuring that only when the frequency of historical anomalies exceeds this threshold is the area considered to have a high historical risk, thus influencing decisions regarding low-bitrate transmission. This threshold can be set based on the highway's grade, traffic flow, safety management regulations, historical data analysis results, and expert experience. For example, it can be set as the average frequency of historical anomalies on a specific road segment or the upper limit specified by a certain safety standard.

[0101] This application's solution introduces an abnormal situation occurrence frequency threshold and uses a formula to generate a historical abnormal situation impact coefficient. The formula first calculates the difference between the historical occurrence frequency of highway abnormal situations and the abnormal situation occurrence frequency threshold. A max function ensures that the difference is positive only when the historical occurrence frequency exceeds the preset threshold; otherwise, it is zero. Subsequently, this positive difference is normalized by dividing by the threshold, resulting in a quantified historical abnormal situation impact coefficient. When the historical occurrence frequency does not exceed the threshold, the impact coefficient is set to zero, indicating a low risk of historical abnormal situations in the area and no negative impact on the feasibility assessment of low-bitrate transmission. Conversely, when the historical occurrence frequency is significantly higher than the threshold, the historical abnormal situation impact coefficient will be a large positive value, reflecting a higher historical risk in the area. This calculated historical abnormal situation impact coefficient is then integrated into the feasibility assessment model for low-bitrate transmission of UAV inspection images, for example, by weighting or multiplying it with the basic feasibility score for low-bitrate transmission, to correct the final feasibility score. This mechanism enables feasibility assessments of low bitrate transmission to more accurately reflect the actual impact of historical anomalies on transmission risks, avoiding overly conservative judgments on low-risk areas while ensuring sufficient attention to high-risk areas.

[0102] Through the above technical solution, this application can accurately generate historical anomaly impact coefficients based on the historical frequency and threshold values ​​of highway anomalies. This solution, by introducing a threshold mechanism, effectively distinguishes different levels of historical anomaly risk, avoiding overreaction to minor historical anomalies below the threshold. This makes the feasibility assessment of low-bitrate transmission of UAV inspection images more reasonable and accurate. This helps optimize the allocation of transmission resources, maximizing transmission efficiency while ensuring the quality and security of inspection image transmission. Especially in areas with low historical anomaly frequency, low-bitrate transmission can be used with greater confidence, while in areas with high historical anomaly frequency, the feasibility of low-bitrate transmission will be correspondingly reduced, thus ensuring the reliable transmission of critical information.

[0103] Preferably, the present invention further proposes the following expression for the feasibility assessment model of low bit rate transmission of UAV inspection images: ;

[0104] In the expression, This represents the feasibility score for low-bitrate transmission of the i-th drone inspection image. This represents the basic score for the feasibility of low-bitrate transmission of the i-th drone inspection image. This represents the influence coefficient of historical anomalies;

[0105] The feasibility assessment model for low-bitrate transmission of UAV inspection images provides a quantitative framework for comprehensively considering multiple factors to determine the overall feasibility of low-bitrate transmission of specific UAV inspection images. This model can employ various mathematical forms, such as linear combinations, weighted averages, or nonlinear functions, to adapt to different application scenarios and data characteristics.

[0106] The feasibility score for low-bitrate transmission of the i-th UAV inspection image represents the final comprehensive evaluation result of low-bitrate transmission of that image. This score is a key basis for subsequent transmission decisions; a higher score generally means that the image is more suitable for low-bitrate transmission. The score can be a normalized value between 0 and 1, or a level value reflecting relative feasibility.

[0107] This application's solution achieves a quantitative fusion of the basic score for the feasibility of low-bitrate transmission of UAV inspection images and the influence coefficient of historical anomalies by introducing a mathematical expression. The core of this expression lies in applying the influence of historical anomalies as a multiplicative correction factor to the basic score. Specifically, a higher historical anomaly influence coefficient indicates a higher frequency of historical anomalies in the highway area. The value will decrease accordingly, thus lowering the feasibility score for low-bitrate transmission of drone inspection images. Conversely, a lower historical anomaly impact coefficient indicates lower historical risk in the area. The value will be close to 1, making the feasibility score for low-bitrate transmission of drone inspection images closer to the baseline feasibility score for low-bitrate transmission of drone inspection images. This design ensures that even if the visual content of the current drone inspection image (composed of...) is close to 1, the feasibility score for low-bitrate transmission of drone inspection images will be closer to the baseline feasibility score for low-bitrate transmission of drone inspection images. While a region may appear suitable for low-bitrate transmission, the system will automatically lower its priority or feasibility for low-bitrate transmission if the region has a historically high risk of anomalies. This avoids missing potentially critical information due to over-compression. In this way, the model can comprehensively consider the real-time content characteristics of the image and the historical risks of the region, providing a more comprehensive and prudent assessment of the feasibility of low-bitrate transmission, effectively solving the problem of how to scientifically quantify and integrate different assessment dimensions.

[0108] Through the above technical solution, this application provides a clear and quantitative feasibility assessment model for low-bitrate transmission of UAV inspection images. This model effectively solves the problem of how to scientifically, reasonably, and quantitatively integrate different assessment dimensions by incorporating the influence coefficient of historical anomalies as a multiplicative correction factor into the basic feasibility score for low-bitrate transmission. This integration method ensures that the final feasibility score for low-bitrate transmission of UAV inspection images not only considers the real-time content characteristics of the image itself but also fully considers the historical risk background of the corresponding highway area. Therefore, even if the current image content seems non-urgent, if the area has a high frequency of historical anomalies, the model can promptly reduce the feasibility of low-bitrate transmission, thereby avoiding the loss of potentially critical information due to blindly reducing the bitrate and significantly improving the accuracy, reliability, and security of UAV inspection image transmission decisions.

[0109] Preferably, the present invention further proposes a specific method for generating the transmission link state evaluation value as follows:

[0110] Through the formula: ;

[0111] Generate the transmission link state assessment value C;

[0112] In the formula, This represents the estimated bandwidth utilization of the transmission link. This represents the broadband utilization threshold;

[0113] The transmission link status assessment value is used to quantify the current congestion level or stress level of the transmission link. Its concept is to provide a standardized, comparable metric to reflect the performance of the transmission link under current transmission load. This assessment value can be a dimensionless numerical value, whose magnitude is directly related to the health status of the transmission link. For example, this value can represent the degree of link overload or the gap between link resources and demand.

[0114] The estimated bandwidth utilization of a transmission link represents the proportion of bandwidth expected to be used under current or predicted transmission tasks. It is a prediction or estimate of the actual load on the transmission link. This utilization can be obtained by monitoring the ratio of the current amount of data transmitted to the total available bandwidth in real time, or by estimating the amount of data to be transmitted over a future period. For example, it can be estimated by statistically analyzing the average data throughput over a period of time, or by predicting the total bitrate of the images to be transmitted.

[0115] A bandwidth utilization threshold is a preset reference point used to determine whether a transmission link is operating normally or ideally. When the estimated bandwidth utilization of a transmission link exceeds this threshold, it usually means that the link may face the risk of congestion or performance degradation. This threshold can be set according to factors such as the actual application scenario, network equipment performance, and quality of service requirements. For example, it can be set as a percentage of the total link bandwidth, such as 80% or 90%, to leave a certain margin.

[0116] The above calculation formula provides a specific method for converting the estimated bandwidth utilization and bandwidth utilization threshold of a transmission link into a transmission link status assessment value. The core of this formula is that when the estimated bandwidth utilization does not exceed the bandwidth utilization threshold, the transmission link status assessment value is 0, indicating that the link status is good or has not reached the warning level. Once the estimated bandwidth utilization exceeds the bandwidth utilization threshold, the transmission link status assessment value is calculated based on the proportion of the excess, thus intuitively reflecting the degree of link overload. This calculation method can effectively distinguish between the normal operation state and the potential congestion state of the link and quantify the severity of congestion.

[0117] This application's solution introduces a method for generating transmission link status assessment values, enabling quantitative evaluation of the real-time status of transmission links. Specifically, the method first obtains the estimated bandwidth utilization rate of the transmission link, which reflects the current or upcoming transmission tasks' occupation of link resources. Simultaneously, a bandwidth utilization threshold is set, representing the maximum acceptable load of the link under normal operating conditions. By comparing the estimated bandwidth utilization rate with the bandwidth utilization threshold and calculating using the aforementioned formula, a transmission link status assessment value can be generated. When the estimated bandwidth utilization rate does not exceed the bandwidth utilization threshold, the transmission link status assessment value is zero, indicating that the transmission link is in a healthy state and no additional low-bit-rate transmission measures are required. Once the estimated bandwidth utilization rate exceeds the bandwidth utilization threshold, the transmission link status assessment value will be positive, and its magnitude directly reflects the degree of link overload. This quantitative assessment mechanism provides accurate input for subsequent low-bit-rate transmission decisions, allowing the system to dynamically adjust the amount of low-bit-rate transmission based on the actual pressure on the transmission link. This effectively avoids link congestion while ensuring image transmission quality, ensuring timely return of inspection images.

[0118] Through the above technical solution, this application provides an effective method for quantitatively assessing the status of transmission links. By introducing the estimated bandwidth utilization and bandwidth utilization threshold of the transmission link, and using a specific formula to calculate the transmission link status assessment value, the real-time load and potential congestion risk of the transmission link can be accurately reflected. This quantitative assessment avoids subjective judgment or simple binary judgment, enabling the system to perceive link pressure more precisely, thus providing an accurate and reliable basis for subsequent low-bit-rate transmission decisions. This helps the system dynamically optimize transmission strategies while ensuring image transmission quality, effectively avoiding link congestion, ensuring stable and efficient backhaul of highway UAV inspection images, and solving the technical problem of how to accurately assess the status of transmission links to guide low-bit-rate transmission decisions.

[0119] Preferably, the present invention further proposes a method for obtaining the estimated bandwidth utilization of the transmission link, specifically including:

[0120] Through the formula: ;

[0121] Estimated bandwidth utilization of the transmission link ;

[0122] In the formula, This represents the transmission bit rate of the k-th UAV inspection image, B represents the total available bandwidth of the transmission link, and n represents the number of UAV inspection images. This represents the fixed consumption item for transmission link bandwidth utilization;

[0123] The transmission bitrate of the kth UAV inspection image refers to the data transmission rate required for a single UAV inspection image during transmission. This bitrate can be preset or dynamically adjusted according to parameters such as the image's encoding format, resolution, frame rate, and compression ratio. It can also be evaluated through an analysis model of the complexity of the image content.

[0124] The total available bandwidth of a transmission link refers to the maximum data transmission capacity that the transmission link can provide under ideal conditions. This bandwidth can be determined by network device configuration information, service level agreements (SLAs) of communication service providers, or actual link performance tests, or it can be obtained by analyzing the physical characteristics of the wireless communication channel.

[0125] The number of drone inspection images refers to the total number of drone inspection images that are currently waiting to be processed or transmitted. This number can be obtained by using the counter of the image acquisition module, the length of the image buffer queue, or by counting the image files acquired within a specific time window.

[0126] The fixed consumption item of transmission link bandwidth utilization refers to the relatively stable bandwidth occupancy rate of the transmission link due to other necessary communications (such as flight control commands, telemetry data, system heartbeat packets, etc. sent from the ground station to the UAV) in addition to the transmission of UAV inspection images itself. This fixed consumption item can be determined by long-term statistics and measurement of non-image data communication traffic, or it can be preset according to system design specifications or experience values ​​and calibrated regularly to adapt to the actual operating environment.

[0127] The proposed solution uses the aforementioned formula to sum the transmission bitrates of all UAV inspection images to be transmitted, obtaining the total bandwidth requirement for image transmission. This total bandwidth requirement is then calculated as a ratio to the total available bandwidth of the transmission link, thus yielding the bandwidth utilization rate for the image transmission portion. Furthermore, a fixed consumption term for the transmission link bandwidth utilization rate is added. This fixed consumption term considers the bandwidth required for non-image data transmission, such as flight control commands sent from the ground station to the UAV, ensuring a comprehensive assessment of the total load on the transmission link. In this way, the system obtains a comprehensive and accurate estimated bandwidth utilization rate for the transmission link. This estimated bandwidth utilization rate is then used to generate a transmission link status assessment value, providing accurate input for subsequent low-bitrate transmission quantity analysis models. This allows the system to make more reasonable and efficient low-bitrate transmission decisions based on a clear understanding of the actual load on the transmission link.

[0128] Through the above technical solution, this application can accurately calculate the estimated bandwidth utilization of the transmission link. This calculation method comprehensively considers the transmission bit rate of all UAV inspection images to be transmitted, the total available bandwidth of the transmission link, and the fixed consumption items of the transmission link bandwidth utilization, thus making the assessment of the transmission link load more comprehensive and accurate. This effectively avoids transmission link overload or bandwidth resource waste caused by inaccurate estimation, provides reliable basic data for the generation of subsequent transmission link status assessment values, and thus ensures the scientific validity and effectiveness of low bit rate transmission decisions for UAV inspection images, improving the stability and efficiency of highway UAV inspection image transmission.

[0129] Preferably, the present invention further proposes the following expression for the low bit rate transmission quantity analysis model: ;

[0130] In the expression, This represents the optimal number of low-bitrate images transmitted during drone inspections. This represents the transmission link status evaluation value when the number of low-bit-rate transmissions of UAV inspection images is m, and A represents the set of numbers of UAV inspection images that can be transmitted at low bit rate.

[0131] This low-bitrate transmission quantity analysis model aims to quantify the relationship between the number of low-bitrate transmitted images and the transmission link status, thereby guiding the system to make optimal decisions. Its role is to provide a mathematical framework that allows the system to dynamically adjust the number of low-bitrate transmitted images based on the actual situation of the current transmission link, achieving optimal resource utilization. This model can be constructed based on various mathematical methods, such as regression analysis, machine learning algorithms (e.g., support vector machines, neural networks), or optimization algorithms. In one implementation, the model can be a pre-trained predictive model that predicts the transmission link status evaluation value based on the input low-bitrate transmission quantity. In another implementation, the model can be a real-time computational model that finds the optimal solution through iteration or search.

[0132] in, This refers to the number of possible low-bitrate transmissions that minimizes the transmission link state evaluation value. It is the core output of the system's decision-making process, directly determining how many images will ultimately be transmitted at a low bitrate. Determining the optimal number of low-bitrate transmissions can employ optimization algorithms such as exhaustive search, binary search, and gradient descent. For example, it can be calculated by iterating through every possible number of low-bitrate transmissions *m* in set A. Then select to make The smallest m as .

[0133] This indicates the load or congestion level of the transmission link when m drone inspection images are transmitted at a low bit rate. The smaller the value, the less pressure on the transmission link and the better the transmission status. The calculation relies on the estimated bandwidth utilization of the transmission link, which in turn is related to the transmission bitrate of all images (including images transmitted at low bitrates and those transmitted at non-low bitrates), the total available bandwidth of the transmission link, fixed consumption items, and other factors. Therefore, This is a dynamically changing quantity based on the number of low-bitrate transmissions of UAV inspection images, reflecting the link performance at a specific number of low-bitrate transmissions. Set A contains all the possible numbers of images that can be transmitted at low bitrate. The upper limit of this set is typically determined by the length of the sequence of UAV inspection images suitable for low-bitrate transmission, i.e., the maximum number of images that can be determined to be suitable for low-bitrate transmission.

[0134] The elements of set A can be any integer between 0 and the sequence length, or filtered according to actual needs. For example, if there are 100 images in the sequence that can be transmitted at a low bit rate, then A can be a set of integers from 0 to 100.

[0135] This application's solution, by introducing the aforementioned low-bit-rate transmission quantity analysis model, aims to accurately determine the optimal number of images capable of low-bit-rate transmission under current transmission link conditions. The core of this model lies in using the number of low-bit-rate transmission images as the independent variable and the transmission link state evaluation value as the dependent variable, finding the optimal solution through the aforementioned mathematical expression. Specifically, the system first generates a sequence of UAV inspection images suitable for low-bit-rate transmission based on the highway state characteristic data and the historical frequency of highway anomalies in the UAV inspection images. This sequence contains all images suitable for low-bit-rate transmission. Simultaneously, the system generates a transmission link state evaluation value based on the transmission bit rate of the UAV inspection images. Based on this, the model iteratively or computationally evaluates the change in the transmission link state evaluation value when different numbers m of images are transmitted at low bit rates. By minimizing... That is, to find the number of low-bit-rate transmissions m that minimizes the pressure on the transmission link, thereby determining the optimal number of low-bit-rate transmissions. This method avoids the problems of blindly transmitting at low bit rates or failing to fully utilize the advantages of low bit rate transmission. It enables the image transmission strategy to be dynamically optimized based on the actual link conditions and the importance of the image, thereby effectively reducing the burden on the transmission link while ensuring transmission quality.

[0136] Through the above technical solution, this application can accurately determine the optimal number of low-bitrate transmissions of UAV inspection images under the current transmission link conditions. This allows the system to avoid transmission link overload or resource waste caused by blindly selecting a low-bitrate transmission quantity. By minimizing the transmission link state evaluation value, this solution ensures that the utilization efficiency of the transmission link is optimized to the maximum extent while guaranteeing image transmission quality, effectively reducing the risk of transmission link congestion and improving the stability and reliability of highway UAV inspection image transmission. This dynamically optimized transmission strategy enables the system to intelligently adjust its transmission behavior according to the actual transmission environment and image characteristics, thereby improving the overall execution efficiency of the inspection task and the timeliness of data return.

[0137] Please see Figure 3 In another embodiment, the present invention also proposes an image transmission system for highway drone inspection, which is used to execute the above-described image transmission method for highway drone inspection, specifically including:

[0138] The data acquisition unit 10 is used to acquire the status feature data of the highway in the UAV inspection images, the historical frequency of abnormal conditions of the highway, and the transmission bit rate of the UAV inspection images.

[0139] Feasibility analysis unit 20 is used to establish a feasibility assessment model for low bit rate transmission of UAV inspection images based on the status characteristic data of highways in UAV inspection images and the historical frequency of abnormal highway conditions, and to generate a feasibility score for low bit rate transmission of UAV inspection images.

[0140] The sorting unit 30 is used to generate a sequence of UAV inspection images that can be transmitted at low bit rate based on the feasibility score of low bit rate transmission of UAV inspection images; the sequence of UAV inspection images that can be transmitted at low bit rate refers to the sequence after the UAV inspection images are arranged according to the feasibility score of low bit rate transmission of UAV inspection images.

[0141] The transmission link status analysis unit 40 is used to generate a transmission link status evaluation value based on the transmission bit rate of the UAV inspection image.

[0142] The low bit rate transmission quantity analysis unit 50 is used to establish a low bit rate transmission quantity analysis model based on the low bit rate transmission sequence of UAV inspection images, with the number of low bit rate transmissions of UAV inspection images as the independent variable and the transmission link status evaluation value as the dependent variable, to generate the optimal number of low bit rate transmissions of UAV inspection images.

[0143] The image transmission unit 60 is used to transmit the UAV inspection images according to the optimal low bit rate transmission quantity of the UAV inspection images.

[0144] For preferred options, please refer to [link / reference]. Figure 4 The present invention further proposes that the feasibility analysis unit 20 specifically includes:

[0145] Basic analysis module 21 is used to generate a basic score for the feasibility of low bit rate transmission based on the status feature data of highways in UAV inspection images.

[0146] The historical data analysis module 22 is used to generate historical abnormality impact coefficients based on the historical frequency of occurrence of abnormal conditions on highways; wherein, the historical frequency of occurrence of abnormal conditions on highways refers to the historical frequency of occurrence of abnormal conditions at the highway locations corresponding to the UAV inspection images.

[0147] The comprehensive analysis module 23 is used to establish a feasibility assessment model for low bit rate transmission of UAV inspection images based on the basic feasibility score for low bit rate transmission and the influence coefficient of historical anomalies, and to generate a feasibility score for low bit rate transmission of UAV inspection images.

[0148] 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. An image transmission method for unmanned aerial vehicle (UAV) inspection of highways, characterized in that, The method specifically includes: Acquire state characteristic data of highways in UAV inspection images, historical frequency of highway anomalies, and transmission bit rate of UAV inspection images; Based on the status characteristic data of highways in UAV inspection images and the historical frequency of abnormal highway conditions, a feasibility assessment model for low bit rate transmission of UAV inspection images is established, and a feasibility score for low bit rate transmission of UAV inspection images is generated. Based on the feasibility score of low bit rate transmission of UAV inspection images, a sequence of UAV inspection images that can be transmitted at low bit rate is generated; the sequence of UAV inspection images that can be transmitted at low bit rate refers to the sequence after the UAV inspection images are arranged according to the feasibility score of low bit rate transmission of UAV inspection images. Based on the transmission bit rate of the UAV inspection images, a transmission link status assessment value is generated; Based on the low-bit-rate transmission sequence of UAV inspection images, a low-bit-rate transmission quantity analysis model is established with the number of low-bit-rate transmissions of UAV inspection images as the independent variable and the transmission link status evaluation value as the dependent variable, to generate the optimal number of low-bit-rate transmissions of UAV inspection images. The drone inspection images are transmitted according to the optimal number of low-bitrate transmissions. Specifically, the optimal number of low-bitrate transmission images are selected from the sequence of images that can be transmitted at low bitrate, while the remaining images are transmitted at the standard bitrate. The specific method for generating the feasibility score for low bit rate transmission of UAV inspection images includes: Based on the status feature data of highways in UAV inspection images, a basic score for the feasibility of low bit rate transmission is generated. Based on the historical frequency of abnormal conditions on highways, a historical abnormal condition impact coefficient is generated; whereby the historical frequency of abnormal conditions on highways refers to the historical frequency of abnormal conditions occurring at the highway locations corresponding to the UAV inspection images. A feasibility assessment model for low bit rate transmission of UAV inspection images is established based on the basic feasibility score for low bit rate transmission and the impact coefficient of historical anomalies, and a feasibility score for low bit rate transmission of UAV inspection images is generated.

2. The image transmission method for highway unmanned aerial vehicle (UAV) inspection according to claim 1, characterized in that, The specific methods for generating the basic score for the feasibility of low bit rate transmission include: Through the formula: ; Generate a basic feasibility score for low bit rate transmission ; In the formula, This refers to the basic score for the feasibility of low-bitrate transmission of the i-th drone inspection image. This represents the normalized value of the j-th state feature data of the highway in the i-th UAV inspection image. represents the weight coefficient of the j-th state feature of the highway in the drone inspection image, and s represents the number of state feature data of the highway in the drone inspection image.

3. The image transmission method for highway unmanned aerial vehicle (UAV) inspection according to claim 1, characterized in that, The specific methods for generating the impact coefficient of the historical anomalies include: Through the formula: ; Generate historical anomaly impact coefficient ; In the formula, This indicates the historical frequency of abnormal situations on highways. This represents the threshold for the frequency of abnormal situations.

4. The image transmission method for highway unmanned aerial vehicle (UAV) inspection according to claim 1, characterized in that, The specific expression for the feasibility assessment model of low bit rate transmission of UAV inspection images is as follows: ; In the expression, This represents the feasibility score for low-bitrate transmission of the i-th drone inspection image. This represents the basic score for the feasibility of low-bitrate transmission of the i-th drone inspection image. This represents the influence coefficient of historical anomalies.

5. The image transmission method for highway unmanned aerial vehicle (UAV) inspection according to claim 1, characterized in that, The specific method for generating the transmission link state evaluation value is as follows: Through the formula: ; Generate the transmission link state assessment value C; In the formula, This represents the estimated bandwidth utilization of the transmission link. This represents the broadband utilization threshold.

6. The image transmission method for highway unmanned aerial vehicle (UAV) inspection according to claim 1, characterized in that, The methods for obtaining the estimated bandwidth utilization of the transmission link specifically include: Through the formula: ; Estimated bandwidth utilization of the transmission link ; In the formula, This represents the transmission bit rate of the k-th UAV inspection image, B represents the total available bandwidth of the transmission link, and n represents the number of UAV inspection images. This represents the fixed consumption item for transmission link bandwidth utilization.

7. The image transmission method for highway unmanned aerial vehicle (UAV) inspection according to claim 1, characterized in that, The specific expression for the low bit rate transmission quantity analysis model is as follows: ; In the expression, This represents the optimal number of low-bitrate images transmitted during drone inspections. This represents the transmission link status assessment value when the number of low-bit-rate transmissions of UAV inspection images is m, and A represents the set of numbers of UAV inspection images that can be transmitted at low bit rate.

8. An image transmission system for unmanned aerial vehicle (UAV) inspection of highways, characterized in that, The system is used to perform the image transmission method for highway unmanned aerial vehicle (UAV) inspection as described in any one of claims 1-7, specifically including: The data acquisition unit is used to acquire the status characteristic data of highways in UAV inspection images, the historical frequency of abnormal highway conditions, and the transmission bit rate of UAV inspection images. The feasibility analysis unit is used to establish a feasibility assessment model for low bit rate transmission of UAV inspection images based on the status characteristic data of highways in UAV inspection images and the historical frequency of abnormal highway conditions, and to generate a feasibility score for low bit rate transmission of UAV inspection images. The sorting unit is used to generate a sequence of UAV inspection images that can be transmitted at low bit rate based on the feasibility score of low bit rate transmission of UAV inspection images. The sequence of UAV inspection images that can be transmitted at low bit rate refers to the sequence after the UAV inspection images are arranged according to the feasibility score of low bit rate transmission of UAV inspection images. The transmission link status analysis unit is used to generate a transmission link status evaluation value based on the transmission bit rate of the UAV inspection image. The low bit rate transmission quantity analysis unit is used to establish a low bit rate transmission quantity analysis model based on the low bit rate transmission sequence of UAV inspection images, with the number of low bit rate transmissions of UAV inspection images as the independent variable and the transmission link status evaluation value as the dependent variable, to generate the optimal number of low bit rate transmissions of UAV inspection images. The image transmission unit is used to transmit UAV inspection images according to the optimal number of low-bit-rate transmission images; specifically, it selects the images with the optimal number of low-bit-rate transmission images from the sequence of images that can be transmitted at low bit rate, while the remaining images are transmitted at the standard bit rate. The feasibility analysis unit specifically includes: The basic analysis module is used to generate a basic score for the feasibility of low bit rate transmission based on the status feature data of highways in UAV inspection images. The historical data analysis module is used to generate historical anomaly impact coefficients based on the historical frequency of highway anomalies; where the historical frequency of highway anomalies refers to the historical frequency of anomalies at the highway locations corresponding to the drone inspection images. The comprehensive analysis module is used to establish a feasibility assessment model for low bit rate transmission of UAV inspection images based on the basic feasibility score for low bit rate transmission and the impact coefficient of historical anomalies, and to generate a feasibility score for low bit rate transmission of UAV inspection images.

Images