Edge vision enhancement processing method and system in complex environment

Abstract

The invention is suitable for the technical field of image visual processing, and provides an edge visual enhancement processing method and system in a complex environment, and the method comprises the steps: obtaining an initial edge sharpening gain parameter and historical operation data of a target camera when the target camera faces a specified water surface, a water surface fluctuation sample formed after vehicles which belong to the same designated water surface but have different reverse wind forces pass through is screened out from the historical operation data; according to the method, the definition and the contrast ratio of the vehicle edge can still be kept under the complex illumination and environment superposition condition, the stability and the accuracy of traffic flow detection and recognition are remarkably improved, and therefore more reliable data support is provided for intelligent traffic management.

Description

Technical Field

[0001] This invention belongs to the field of image visual processing technology, and in particular relates to a method and system for edge visual enhancement processing in complex environments. Background Technology

[0002] In the field of image vision processing, existing methods for detecting traffic and pedestrian flow in complex environments largely rely on traditional video surveillance cameras combined with image enhancement and target detection technologies. Under normal circumstances, edge sharpening, brightness compensation, noise reduction, and feature extraction can accurately identify the shapes of vehicles or pedestrians under relatively ideal lighting conditions. However, in real urban traffic scenarios, water surfaces formed by road depressions, overflowing fountains, or poor drainage often have specular reflective properties, creating reflections of vehicles as they pass. When external wind forces act on the water surface, the ripples and disturbances caused by passing vehicles combine to cause significant geometric distortion, proportional deviation, and decreased clarity in the reflections. These complex reflection images affected by wind greatly interfere with edge detection and recognition, and existing technologies cannot accurately handle them by simply using fixed sharpening or uniform enhancement.

[0003] Furthermore, at night or under multi-source lighting conditions, the superposition of urban lighting and wind-induced ripples further complicates reflections. Ripples on water caused by strong or headwinds not only alter the shape of the reflection but also amplify the distortion effect of light reflection, resulting in blurred edges and unbalanced brightness and contrast for vehicles in the image. Existing technologies often counteract this interference by increasing sharpening parameters or global contrast, but this method fails to distinguish the specific source of reflection degradation. For example, when wind is the dominant factor, reflection distortion is fundamentally different from decreased water transparency, and existing technologies lack the ability to decompose and quantitatively analyze image deviations under different factors, leading to correction measures that are neither scientific nor targeted.

[0004] Therefore, the core flaw of current technology lies in its inability to accurately model the reflection degradation mechanism by incorporating wind influence, nor can it dynamically distinguish between reflections and environmental factors such as water surface transparency. As a result, when reflections gradually deteriorate over time, existing systems cannot determine whether the degradation is caused by wind disturbance or water surface degradation, thus resorting only to a crude global compensation strategy. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for edge vision enhancement processing in complex environments, aiming to solve the problems mentioned in the background art.

[0006] This invention is implemented as follows: a method for edge visual enhancement processing in complex environments, the method comprising:

[0007] Acquire the initial edge sharpening gain parameters and historical operating data of the target camera when facing the reflection of the specified water surface, and filter out water surface ripple samples formed after vehicles of different headwind strengths pass by the specified water surface from the historical operating data.

[0008] Analyze each water surface ripple sample, analyze the effect of water surface ripples on the vehicle's true reflection image under different headwind forces, and determine the image that best matches the vehicle's specifications and proportions as the reference image, and set the headwind force corresponding to this reference image as the reference headwind force.

[0009] Analyze the local operation data in the historical operation data that are in the same specified water surface and have the same reference headwind force, compare the vehicle images reflected on the specified water surface at different times with the reference image, determine whether there is an image deviation that gradually deteriorates over time, and detect whether the coupling degree between the changing trend of the image deviation and the environmental deterioration trend of the decrease in water surface transparency exceeds the preset threshold.

[0010] When the coupling degree is determined to exceed the preset threshold, the enhancement magnitude is generated based on the image deviation between the most recent vehicle image and the reference image, and the initial edge sharpening gain parameter is corrected based on the enhancement magnitude.

[0011] As a further limitation of the technical solution of the present invention, the designated water surface refers to a water-filled area, road puddles, fountain overflow surface, or similar artificial or natural water body surface with mirror reflection characteristics that is formed by long-term water accumulation, low-lying terrain, fountain overflow, or poor drainage in urban roads or traffic scenarios, and can produce obvious ripples when vehicles pass by.

[0012] The water surface fluctuation samples and local operational data are consistent with the designated water surface in terms of objective background conditions, which include at least the average area, average water depth, boundary morphology and geographical location of the water surface.

[0013] The vehicles selected in the water surface fluctuation samples and local operation data are consistent, meaning that the vehicles have the same appearance size and color.

[0014] As a further limitation of the technical solution of the present invention, the headwind force refers to the wind force acting on the designated water surface in the opposite direction when the direction of the water surface ripples generated after the vehicle passes over the designated water surface is opposite to the direction of the external wind force.

[0015] As a further limitation of the technical solution of this invention, the steps of analyzing each water surface ripple sample, analyzing the effect of water surface ripples on the actual reflection image of the vehicle under different headwind forces, and determining the image that best matches the vehicle's specifications and proportions as the reference image, and setting the headwind force corresponding to this reference image as the reference headwind force, include:

[0016] Analyze the water surface ripple samples to obtain vehicle reflection images of vehicles traveling in the specified water surface reflection under different headwind conditions;

[0017] Obtain a standard image corresponding to the vehicle, and calculate the morphological matching degree between the reflection image corresponding to each water surface ripple sample and the standard image. The morphological matching degree includes at least the degree of overlap of the outer contour and the consistency of the proportion.

[0018] The reflection image with the highest shape matching degree is determined as the reference image that best matches the vehicle's specifications and proportions, and the headwind force corresponding to this reference image is set as the reference headwind force.

[0019] As a further limitation of the technical solution of the present invention, the image deviation refers to the difference between the vehicle image of the specified water surface reflection at different times and the reference image in at least one of the image features such as geometric contour, scale ratio, edge sharpness and brightness contrast, or the weighted value of the differences of the multiple image features.

[0020] As a further limitation of the technical solution of this embodiment of the invention, when it is determined that the coupling degree exceeds a preset threshold, the step of generating an enhancement magnitude based on the image deviation between the most recent vehicle image and the reference image, and correcting the initial edge sharpening gain parameter based on the enhancement magnitude, includes:

[0021] When an image deviation that gradually deteriorates over time is detected and the coupling between the trend of the image deviation and the trend of environmental degradation with the decrease in water transparency exceeds a preset threshold, the most recent vehicle image is obtained from the local operation data.

[0022] Calculate the image deviation between the most recent vehicle image and the reference image, generate an enhancement magnitude based on the absolute value of the image deviation, and combine the enhancement magnitude with a preset correction magnitude control coefficient to perform positive dynamic correction on the initial edge sharpening gain parameter;

[0023] The modified edge sharpening gain parameters are applied to the real-time image processing flow of the target camera to enhance the clarity and contrast of vehicle edges in the currently and subsequently acquired specified water surface reflections.

[0024] An edge vision enhancement processing system for complex environments, the system comprising:

[0025] The data acquisition module is used to acquire the initial edge sharpening gain parameters and historical operating data of the target camera when facing the reflection of the specified water surface. From the historical operating data, samples of water surface ripples formed after vehicles of different headwind strengths pass by the same specified water surface are selected.

[0026] The reference headwind force determination module is used to analyze each water surface ripple sample, analyze the effect of water surface ripples on the vehicle's real reflection image under different headwind forces, and determine the image that best matches the vehicle's specifications and proportions as the reference image, and set the headwind force corresponding to the reference image as the reference headwind force.

[0027] The image deviation detection module is used to analyze local operation data in historical operation data that belong to the same specified water surface and are consistent with the headwind force. It compares the vehicle images reflected on the specified water surface at different times with the reference image to determine whether there is an image deviation that gradually deteriorates over time. It also detects whether the coupling degree between the trend of image deviation and the trend of environmental deterioration with the decrease in water transparency exceeds a preset threshold.

[0028] The gain correction module is used to generate an enhancement magnitude based on the image deviation between the most recent vehicle image and the reference image when the coupling degree is determined to exceed a preset threshold, and to correct the initial edge sharpening gain parameters based on the enhancement magnitude.

[0029] As a further limitation of the technical solution of the present invention, the designated water surface refers to a water-filled area, road puddles, fountain overflow surface, or similar artificial or natural water body surface with mirror reflection characteristics that is formed by long-term water accumulation, low-lying terrain, fountain overflow, or poor drainage in urban roads or traffic scenarios, and can produce obvious ripples when vehicles pass by.

[0030] The water surface fluctuation samples and local operational data are consistent with the designated water surface in terms of objective background conditions, which include at least the average area, average water depth, boundary morphology and geographical location of the water surface.

[0031] The vehicles selected in the water surface fluctuation samples and local operation data are consistent, meaning that the vehicles have the same appearance size and color.

[0032] As a further limitation of the technical solution of the present invention, the headwind force refers to the wind force acting on the designated water surface in the opposite direction when the direction of the water surface ripples generated after the vehicle passes over the designated water surface is opposite to the direction of the external wind force.

[0033] As a further limitation of the technical solution of this embodiment of the invention, the reference headwind force determination module specifically includes:

[0034] The reflection image acquisition unit is used to analyze each of the water surface ripple samples and acquire reflection images of vehicles traveling in the reflection of the specified water surface under different headwind conditions.

[0035] The morphological matching degree calculation unit is used to acquire a standard image corresponding to the vehicle, and calculate the morphological matching degree between the reflection image corresponding to each water surface wave sample and the standard image, wherein the morphological matching degree includes at least the degree of overlap of the outer contour and the consistency of the proportion.

[0036] The reference headwind force determination unit is used to determine the reflection image with the highest shape matching degree as the reference image that best matches the vehicle body specifications and proportions, and to set the headwind force corresponding to the reference image as the reference headwind force.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] This invention proposes a method for accurately identifying the sources of water reflection degradation and correcting edge sharpening gain in complex environments by introducing the establishment of a reference headwind force and dynamic analysis of image deviation. Compared with existing technologies, the core innovation of this invention lies in that it does not perform global compensation for image degradation in a general way, but instead focuses the correction logic directly on the dynamic improvement of edge sharpening gain by verifying the coupling relationship between image deviation and the decrease in water transparency, thus achieving a scientific closed loop of "compensation based on the amount of deviation".

[0039] Especially in urban traffic scenarios, road flooding and fountain overflows often work together with nighttime city lighting, making vehicle reflections more susceptible to reflection and scattering interference. This invention can maintain the clarity and contrast of vehicle edges under such complex lighting and environmental conditions, significantly improving the stability and accuracy of traffic flow detection and recognition, thereby providing more reliable data support for intelligent traffic management. Attached Figure Description

[0040] Figure 1 A flowchart of the method provided in the embodiments of the present invention;

[0041] Figure 2 This is a flowchart of the step for determining the headwind force in the method provided in the embodiments of the present invention;

[0042] Figure 3 This is a flowchart of the edge sharpening gain parameter correction step in the method provided in the embodiments of the present invention;

[0043] Figure 4 Application architecture diagram of the system provided in the embodiments of the present invention;

[0044] Figure 5 This is a structural block diagram of the reference headwind force determination module in the system provided in the embodiment of the present invention. Detailed Implementation

[0045] 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.

[0046] Figure 1 A flowchart of the method provided by an embodiment of the present invention is shown.

[0047] Specifically, a method for edge vision enhancement in complex environments includes the following steps:

[0048] Step S100: Obtain the initial edge sharpening gain parameters and historical operating data of the target camera when facing the reflection of the specified water surface. Select water surface ripple samples formed by vehicles that are on the same specified water surface but have different headwinds from the historical operating data.

[0049] The designated water surface refers to a water-filled area, puddle, fountain overflow surface, or similar artificial or natural water body surface that is located in urban roads or traffic scenarios and is formed due to long-term water accumulation, low-lying terrain, fountain overflow, or poor drainage, which can produce obvious ripples when vehicles pass by and has mirror reflection characteristics.

[0050] The water surface fluctuation samples and local operational data are consistent with the designated water surface in terms of objective background conditions, which include at least the average area, average water depth, boundary morphology and geographical location of the water surface.

[0051] The vehicles selected in the water surface fluctuation samples and local operation data are consistent, meaning that the vehicles have the same appearance size and color.

[0052] The headwind force refers to the magnitude of the wind force acting on the designated water surface in the opposite direction when the direction of water surface ripples generated after a vehicle passes over the designated water surface is opposite to the direction of external wind force.

[0053] In this embodiment of the invention, the target camera is typically a fixed surveillance camera deployed in urban traffic scenarios. It can be a high-definition optical camera, a low-light night vision camera, or an intelligent video surveillance camera with weak ambient light adaptability. Application scenarios for this type of camera include urban road intersections, low-lying road sections prone to flooding, fountains, or poorly drained traffic hubs. Its function is not only to record vehicle and pedestrian traffic flow, but also to provide raw data support for traffic management and intelligent identification in complex environments.

[0054] The initial edge sharpening gain parameter is an enhancement coefficient set by the camera in the image processing module to improve edge contour sharpness. In traditional technologies, this parameter is usually preset at the factory and remains unchanged during operation, mainly to ensure sufficient sharpness of vehicle and pedestrian edges in the image. However, in scenes with water reflections, water ripples and reflections can interfere with the edge information of the image. If this parameter is not dynamically corrected, the edge information of vehicles or pedestrians may become blurred, thus affecting the accuracy of traffic and pedestrian detection. Therefore, studying the impact of water reflections aims to identify and compensate for deficiencies in camera processing parameters by examining their interference with the real image, thereby making image recognition more reliable.

[0055] The designated water surface has certain characteristics. It must not only be natural or artificial water accumulation within roads or traffic areas, but also meet two conditions: first, it must produce noticeable ripples when vehicles pass over it; second, it must possess specular reflection properties, allowing vehicle images to form clear reflections on the water surface. Only such a water surface can interfere with the footage captured by a camera, thus possessing research value.

[0056] Maintaining consistency between surface ripple samples and local operational data under objective background conditions is crucial to minimizing the influence of irrelevant variables. For example, factors such as average surface area, average depth, boundary morphology, and geographical location directly affect the propagation characteristics of ripples. Without these constraints, differences in water surface environments could obscure the causal relationship between headwinds, ripples, and reflections. However, consistency here does not mean absolute identicality, but rather relative stability within a certain range to ensure data comparability.

[0057] In the water surface ripple samples and local operational data, the selected vehicles must also be consistent, at least in terms of appearance, size, and color. This is because the shape and color of the vehicle itself significantly affect its appearance in water reflection; if the differences between different vehicles are too large, it cannot be accurately attributed to environmental factors such as headwind or water transparency.

[0058] Headwind force refers to the magnitude of the wind force acting on the water surface in the opposite direction to the external wind direction when a vehicle has passed over a designated water surface. This factor is crucial because the magnitude and direction of the wind directly determine the shape and stability of the water ripples, thus affecting the degree of distortion in the vehicle's reflection on the water surface. Studying the impact of headwind force on water reflections can help us establish a mapping relationship between the true image of the vehicle and the image affected by wind interference, thereby providing a scientific basis for correcting edge sharpening gain parameters.

[0059] Historical operational data primarily comes from long-term data collection during the actual deployment of target cameras, including video frame sequences, environmental parameters, and vehicle information. Specifically, this includes: water reflection images from different periods, camera operating parameters (such as exposure and sharpening factor), environmental data (such as wind speed and direction, and weather conditions), and the external dimensions and colors of passing vehicles. This data forms the basis for subsequent sample selection, reference image determination, image deviation detection, and gain correction.

[0060] Furthermore, the edge vision enhancement processing method in complex environments also includes the following steps:

[0061] Step S200: Analyze each water surface ripple sample, analyze the effect of water surface ripples on the vehicle's true reflection image under different headwind forces, and determine the image that best matches the vehicle's specifications and proportions as the reference image, and set the headwind force corresponding to this reference image as the reference headwind force.

[0062] Specifically, Figure 2 A flowchart illustrating the steps for determining the reference headwind force is shown.

[0063] The process involves analyzing various water surface ripple samples, examining the impact of different headwind forces on the vehicle's true reflection image, and determining the image that best matches the vehicle's specifications and proportions as the reference image. The headwind force corresponding to this reference image is then set as the reference headwind force. The specific steps include:

[0064] Step S201: Analyze each of the water surface ripple samples to obtain vehicle reflection images of vehicles traveling in the reflection of the specified water surface under different headwind conditions;

[0065] Step S202: Obtain a standard image corresponding to the vehicle, and calculate the morphological matching degree between the reflection image corresponding to each water surface ripple sample and the standard image. The morphological matching degree includes at least the degree of overlap of the outline and the consistency of the proportions.

[0066] Step S203: The reflection image with the highest shape matching degree is determined as the reference image that best matches the vehicle body specifications and proportions, and the headwind force corresponding to the reference image is set as the reference headwind force.

[0067] In this embodiment of the invention, in a specific implementation of step S201, analyzing each of the water surface ripple samples and acquiring vehicle reflection images can employ techniques based on image segmentation and motion detection. Specifically, by performing temporal difference analysis or background modeling on video frames, the dynamic changes within the water surface area are extracted. Then, combined with optical flow tracing methods or feature extraction methods from deep convolutional neural networks, the vehicle reflection area in the water reflection is separated. This allows for clear capture of vehicle reflection images under different headwind conditions, providing input for subsequent comparisons.

[0068] In the implementation of step S202, the standard image corresponding to the vehicle can be obtained directly from a camera positioned before the vehicle enters or exits the designated waterway, or obtained through a network search. The morphological matching degree between the standard image and the reflection image can be calculated using edge-feature-based structural similarity (SSIM), Hausdorff distance after contour extraction, proportional consistency metrics, and template matching algorithms. Through these techniques, the degree of consistency between the reflection image and the standard image in terms of shape and proportion can be quantified, ensuring that the measurement of ripple interference effects is scientifically effective.

[0069] In the implementation of step S203, the morphological matching degree of each reflection image is sorted, and the reflection image with the highest matching degree is determined as the reference image. This reference image is closest to the actual size and proportion of the vehicle body and can be used as the sample with the least environmental disturbance. The corresponding headwind force is set as the reference headwind force to calibrate the optimal reflection conditions of the specified water surface under specific wind conditions. Technically, a simple sorting method combined with threshold filtering can be used, or a machine learning model can be introduced to weight multiple matching degree indicators to more accurately select the optimal reflection image.

[0070] Establishing a reference headwind is of great significance. It reflects the condition for minimizing the deviation between the actual vehicle image and its reflection on the water surface in complex environments. In other words, the reference headwind is a calibration point representing the condition under which water surface fluctuations have the least impact on reflection distortion, thus obtaining a reflection image that is closest to the actual size and proportion of the vehicle. This reference provides a benchmark for subsequent analysis of image degradation trends, allowing image deviations to be quantified and tracked under relatively objective conditions, thereby supporting the dynamic correction of edge sharpening gain parameters.

[0071] Furthermore, the edge vision enhancement processing method in complex environments also includes the following steps:

[0072] Step S300: Analyze the local operation data in the historical operation data that belongs to the same specified water surface and has the same reference headwind force, compare the vehicle images of the specified water surface reflection at different times with the reference image, determine whether there is an image deviation that gradually deteriorates over time between the two, and detect whether the coupling degree between the changing trend of the image deviation and the environmental deterioration trend of the decrease in water surface transparency exceeds a preset threshold.

[0073] The image deviation refers to the difference between vehicle images of a specified water surface reflection at different times and reference images in at least one of the following image features: geometric contour, scale ratio, edge sharpness, and brightness contrast, or the weighted value of the differences in the multiple image features.

[0074] In this embodiment of the invention, image deviation is an important indicator for quantifying the difference between a vehicle's reflection and a reference image. Specifically, it reflects the degree of difference between the vehicle's appearance in the water reflection and the actual image. Image deviation is not limited to a single dimension but is composed of multiple features, including the degree of change in geometric contours, deviation in scale ratio, decrease in edge sharpness, and imbalance in brightness contrast. These features can be extracted using image processing algorithms. For example, contour matching can be achieved using edge detection and shape similarity measurement, scale ratio can be achieved through keypoint detection and affine transformation comparison, edge sharpness can be quantified using gradient intensity or spectral energy distribution, and brightness contrast can be calculated using grayscale histograms or local contrast statistics. In practical applications, the differences in different features can be analyzed individually or weighted to form a comprehensive image deviation index.

[0075] If, under the same wind conditions, the image deviation shows a gradual deterioration trend, this indicates that the distortion of the reflected image is not due to wind disturbance, but rather to other environmental factors. In this invention, a typical such factor is the decrease in water transparency. Reduced transparency alters the propagation characteristics of light on and within the water surface, causing blurring, distortion, or reduced brightness in vehicle reflections, thus leading to a gradual increase in image deviation. Therefore, the continuous deterioration of image deviation under constant wind conditions can be considered an indirect reflection of water surface environmental degradation.

[0076] The trend of environmental degradation due to decreased water transparency can be inferred from optical parameters in historical operational data, such as the brightness distribution of the water surface area, the sharpness of reflection edges, or the scattering characteristics of the water body. The trend of image deviation is calculated from the periodic differences between the vehicle reflection and the reference image. The coupling degree between the two can be determined by correlation coefficient, fit curve consistency, or trend similarity index. When the coupling degree exceeds a preset threshold, it indicates that the trend of image deviation and decreased water transparency is highly consistent, meaning that the degradation of the reflection is mainly caused by the deterioration of water transparency.

[0077] In step S300, by comparing the image deviation between the vehicle reflection and the reference image at different times, and combining this with a coupling analysis of the environmental degradation trend of decreasing water transparency, the source of the image deviation can be clearly determined. When the verification results show that the image deviation is indeed mainly due to changes in the water environment, rather than other complex interference factors, it is reasonable to directly adjust and increase the initial edge sharpening gain parameter to compensate for the reflection degradation. This approach is the core of this research, and its value lies in accurately tracing the source and locking onto a single environmental variable as the basis for correction, thereby avoiding the introduction of unnecessary complex compensation methods during image processing.

[0078] In other words, if it can be confirmed that the image deviation originates from a decrease in water transparency, then the entire system does not need to consider other potentially uncertain and difficult-to-quantify external influences (such as random lighting interference, differences in vehicle reflection colors, etc.). This avoids the multi-directional dispersion of compensation logic, reduces model complexity and computational redundancy, ensures a clear and targeted correction path, and avoids correction distortion problems caused by the coupling of multiple factors. Through this logic, the system can directly focus on the single correction method of increasing the edge sharpening gain parameter, so that the clarity and contrast of vehicle edges in the reflection are reliably enhanced, thereby improving the stability and scientific validity of subsequent traffic flow detection and recognition.

[0079] Furthermore, the edge vision enhancement processing method in complex environments also includes the following steps:

[0080] Step S400: When it is determined that the coupling degree exceeds the preset threshold, an enhancement magnitude is generated based on the image deviation between the most recent vehicle image and the reference image, and the initial edge sharpening gain parameter is corrected based on the enhancement magnitude.

[0081] Specifically, Figure 3 A flowchart of the edge sharpening gain parameter correction steps is shown.

[0082] Specifically, when the coupling degree exceeds a preset threshold, an enhancement magnitude is generated based on the image deviation between the most recent vehicle image and the reference image, and the initial edge sharpening gain parameter is corrected based on the enhancement magnitude. This includes the following steps:

[0083] Step S401: When an image deviation that gradually deteriorates over time is detected and the coupling degree between the image deviation change trend and the environmental deterioration change trend of the water surface transparency decreases exceeds a preset threshold, the most recent vehicle image is obtained from the local operation data.

[0084] Step S402: Calculate the image deviation between the most recent vehicle image and the reference image, generate an enhancement magnitude based on the absolute value of the image deviation, and combine the enhancement magnitude with a preset correction magnitude control coefficient to perform positive dynamic correction on the initial edge sharpening gain parameter;

[0085] Step S403: The corrected edge sharpening gain parameters are applied to the real-time image processing flow of the target camera to enhance the clarity and contrast of vehicle edges in the currently and subsequently acquired specified water surface reflections.

[0086] In this embodiment of the invention, selecting the most recent vehicle image as the reference object in step S401 is reasonable. This is because the degradation of the water surface environment and its transparency is usually dynamic and cumulative, and the most recent vehicle image best reflects the actual quality of the reflection under the current water surface conditions. If an earlier vehicle image is selected, the correction may be biased due to lag differences in environmental conditions, reducing the real-time nature and specificity of the compensation. Therefore, adjusting parameters based on the latest image ensures that the adjustment result remains consistent with the current environmental state, thereby more accurately improving the visual effect of the reflection.

[0087] In step S402, the absolute value of the image deviation between the vehicle image and the reference image is used to generate the enhancement magnitude. This ensures the compensation amount is always positive, avoiding uncertainty introduced by the sign. Image deviation may manifest as a decrease in image features (negative value) or an abnormal increase (positive value) under certain conditions. However, in the edge sharpening correction logic, we only care about the absolute magnitude of the deviation, i.e., the magnitude of the deviation from the true reference. For example, when the comprehensive difference is −0.2, it means that the image sharpness or contour fidelity has decreased by 20%. In this case, the absolute value of 0.2 represents the deviation magnitude. The system can then generate a proportional enhancement magnitude according to the principle of "how much decrease, how much gain increase," to compensate for the initial edge sharpening gain parameters. The calculation of the comprehensive difference can be based on the weighted difference of multiple image features, such as: geometric contour deviation × weight 1 + scale ratio deviation × weight 2 + edge sharpness deviation × weight 3 + brightness contrast deviation × weight 4, finally yielding a unified difference index. This processing integrates the changes in multi-dimensional features into a single operable parameter, facilitating dynamic correction.

[0088] Between steps S402 and S403, it can be further explained that: during the subsequent long-term operation, the increase is not only generated based on the most recent image deviation, but also needs to be combined with the slope of the image degradation trend for upward processing, so as to reflect the cumulative effect brought about by the continuous decrease in water surface transparency, thereby ensuring that the correction parameter gradually increases with the dynamic aggravation of long-term environmental degradation.

[0089] In step S403, the corrected edge sharpening gain parameter is applied to the real-time image processing flow of the target camera. Specifically, the camera's built-in image signal processing module (ISP) calls the new edge sharpening gain parameter to enhance the current input image frame in real time, focusing on improving the gradient intensity and contrast of the edge regions. Simultaneously, the system can dynamically call this corrected parameter in subsequent consecutive frames via hardware or software to achieve continuous image quality enhancement, ensuring that even under consistent wind conditions but decreased water transparency, the edges of vehicles in the reflection maintain high clarity and recognizability. This application method not only works on single frames but can also be extended to long-running continuous frame processing to ensure the stability of traffic flow detection and recognition.

[0090] The overall beneficial effect of this invention lies in its proposal of an edge visual enhancement method based on environmental degradation tracing and reference condition calibration. In complex traffic environments, especially in scenes with water reflections, traditional methods often struggle to distinguish image quality degradation caused by wind disturbances and decreased water transparency. This invention, by establishing a reference headwind force and introducing dynamic analysis of image bias, can scientifically identify the source of image degradation and directly implement corrective measures to enhance edge sharpening gain parameters. This unidirectional and precise compensation logic avoids interference from irrelevant factors, ensuring the targeted and efficient nature of image enhancement.

[0091] In terms of application prospects, this invention can be widely used in urban intelligent traffic monitoring systems, especially in complex environments with water reflections, such as low-lying areas prone to water accumulation, near fountains, or where road drainage is poor. Through dynamic compensation, the edge clarity and contrast of vehicle reflections can be maintained under different environmental conditions, thereby improving the accuracy of vehicle and pedestrian flow detection. Furthermore, the concept of this method can be extended to other scenarios with environmental degradation interference, such as nighttime reflective environments and light scattering interference in rainy or foggy weather, providing a general compensation framework for edge vision enhancement, and possessing high application value and promotion potential.

[0092] Furthermore, Figure 4 An application architecture diagram of the system provided in an embodiment of the present invention is shown.

[0093] In another preferred embodiment of the present invention, an edge vision enhancement processing system for complex environments includes:

[0094] The data acquisition module 100 is used to acquire the initial edge sharpening gain parameters and historical operating data of the target camera when facing the reflection of the specified water surface, and to filter out water surface ripple samples formed by vehicles that are on the same specified water surface but have different headwinds from the historical operating data.

[0095] The designated water surface refers to a water-filled area, puddle, fountain overflow surface, or similar artificial or natural water body surface that is located in urban roads or traffic scenarios and is formed due to long-term water accumulation, low-lying terrain, fountain overflow, or poor drainage, which can produce obvious ripples when vehicles pass by and has mirror reflection characteristics.

[0096] The water surface fluctuation samples and local operational data are consistent with the designated water surface in terms of objective background conditions, which include at least the average area, average water depth, boundary morphology and geographical location of the water surface.

[0097] The vehicles selected in the water surface fluctuation samples and local operation data are consistent, meaning that the vehicles have the same appearance size and color.

[0098] The headwind force refers to the magnitude of the wind force acting on the designated water surface in the opposite direction when the direction of water surface ripples generated after a vehicle passes over the designated water surface is opposite to the direction of external wind force.

[0099] Furthermore, the edge vision enhancement processing system for complex environments also includes:

[0100] The reference headwind force determination module 200 is used to analyze each water surface ripple sample, analyze the effect of water surface ripples on the vehicle's real reflection image under different headwind forces, and determine the image that best matches the vehicle's specifications and proportions as the reference image, and set the headwind force corresponding to the reference image as the reference headwind force.

[0101] Specifically, Figure 5 The diagram shows a structural block diagram of the reference headwind force determination module 200 in the system provided in an embodiment of the present invention.

[0102] In a preferred embodiment provided by the present invention, the reference headwind force determination module 200 specifically includes:

[0103] The reflection image acquisition unit 201 is used to analyze each of the water surface ripple samples and acquire reflection images of vehicles traveling in the reflection of the specified water surface under different headwind conditions.

[0104] The morphological matching degree calculation unit 202 is used to acquire a standard image corresponding to the vehicle, and calculate the morphological matching degree between the reflection image corresponding to each water surface wave sample and the standard image, wherein the morphological matching degree includes at least the degree of overlap of the outline and the consistency of the proportions.

[0105] The reference headwind force determination unit 203 is used to determine the reflection image with the highest shape matching degree as the reference image that best matches the vehicle body specifications and proportions, and set the headwind force corresponding to the reference image as the reference headwind force.

[0106] Furthermore, the edge vision enhancement processing system for complex environments also includes:

[0107] The image deviation detection module 300 is used to analyze local operation data in historical operation data that belong to the same specified water surface and have the same reference headwind force. It compares the vehicle images reflected on the specified water surface at different times with the reference image to determine whether there is an image deviation that gradually deteriorates over time. It also detects whether the coupling degree between the changing trend of the image deviation and the environmental deterioration trend of the water surface transparency decreases exceeds a preset threshold.

[0108] Furthermore, the edge vision enhancement processing system for complex environments also includes:

[0109] The gain correction module 400 is used to generate an enhancement magnitude based on the image deviation between the most recent vehicle image and the reference image when the coupling degree is determined to exceed a preset threshold, and to correct the initial edge sharpening gain parameters based on the enhancement magnitude.

[0110] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0111] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0112] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0113] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for edge visual enhancement processing in complex environments, characterized in that, The method includes: Acquire the initial edge sharpening gain parameters and historical operating data of the target camera when facing the reflection of the specified water surface, and filter out water surface ripple samples formed after vehicles of different headwind strengths pass by the specified water surface from the historical operating data. Analyze each water surface ripple sample, analyze the effect of water surface ripples on the vehicle's true reflection image under different headwind forces, and determine the image that best matches the vehicle's specifications and proportions as the reference image, and set the headwind force corresponding to this reference image as the reference headwind force. Analyze the local operation data in the historical operation data that are in the same specified water surface and have the same reference headwind force, compare the vehicle images reflected on the specified water surface at different times with the reference image, determine whether there is an image deviation that gradually deteriorates over time, and detect whether the coupling degree between the changing trend of the image deviation and the environmental deterioration trend of the decrease in water surface transparency exceeds the preset threshold. When the coupling degree is determined to exceed the preset threshold, the enhancement magnitude is generated based on the image deviation between the most recent vehicle image and the reference image, and the initial edge sharpening gain parameter is corrected based on the enhancement magnitude.

2. The edge visual enhancement processing method in complex environments according to claim 1, characterized in that, The designated water surface refers to a water-filled area, puddle, fountain overflow surface, or similar artificial or natural water body surface that is located in urban roads or traffic scenarios and is formed due to long-term water accumulation, low-lying terrain, fountain overflow, or poor drainage, which can produce obvious ripples when vehicles pass by and has mirror reflection characteristics. The water surface fluctuation samples and local operational data are consistent with the designated water surface in terms of objective background conditions, which include at least the average area, average water depth, boundary morphology and geographical location of the water surface. The vehicles selected in the water surface fluctuation samples and local operation data are consistent, meaning that the vehicles have the same appearance size and color.

3. The edge visual enhancement processing method in complex environments according to claim 2, characterized in that, The headwind force refers to the magnitude of the wind force acting on the designated water surface in the opposite direction when the direction of water surface ripples generated after a vehicle passes over the designated water surface is opposite to the direction of external wind force.

4. The edge visual enhancement processing method in complex environments according to claim 3, characterized in that, The steps of analyzing various water surface ripple samples, examining the impact of water surface ripples on the vehicle's true reflection image under different headwind forces, and determining the image that best matches the vehicle's specifications and proportions as the reference image, and setting the headwind force corresponding to this reference image as the reference headwind force, include: Analyze the water surface ripple samples to obtain vehicle reflection images of vehicles traveling in the specified water surface reflection under different headwind conditions; Obtain a standard image corresponding to the vehicle, and calculate the morphological matching degree between the reflection image corresponding to each water surface ripple sample and the standard image. The morphological matching degree includes at least the degree of overlap of the outer contour and the consistency of the proportion. The reflection image with the highest shape matching degree is determined as the reference image that best matches the vehicle's specifications and proportions, and the headwind force corresponding to this reference image is set as the reference headwind force.

5. The edge visual enhancement processing method in complex environments according to claim 4, characterized in that, The image deviation refers to the difference between vehicle images of a specified water surface reflection at different times and reference images in at least one of the following image features: geometric contour, scale ratio, edge sharpness, and brightness contrast, or the weighted value of the differences in the multiple image features.

6. The edge visual enhancement processing method in complex environments according to claim 5, characterized in that, When the coupling degree is determined to exceed a preset threshold, the steps of generating an enhancement magnitude based on the image deviation between the most recent vehicle image and the reference image, and correcting the initial edge sharpening gain parameters based on the enhancement magnitude, include: When an image deviation that gradually deteriorates over time is detected and the coupling between the trend of the image deviation and the trend of environmental degradation with the decrease in water transparency exceeds a preset threshold, the most recent vehicle image is obtained from the local operation data. Calculate the image deviation between the most recent vehicle image and the reference image, generate an enhancement magnitude based on the absolute value of the image deviation, and combine the enhancement magnitude with a preset correction magnitude control coefficient to perform positive dynamic correction on the initial edge sharpening gain parameter; The modified edge sharpening gain parameters are applied to the real-time image processing flow of the target camera to enhance the clarity and contrast of vehicle edges in the currently and subsequently acquired specified water surface reflections.

7. An edge vision enhancement processing system for complex environments, characterized in that, The system includes: The data acquisition module is used to acquire the initial edge sharpening gain parameters and historical operating data of the target camera when facing the reflection of the specified water surface. From the historical operating data, samples of water surface ripples formed after vehicles of different headwind strengths pass by the same specified water surface are selected. The reference headwind force determination module is used to analyze each water surface ripple sample, analyze the effect of water surface ripples on the vehicle's real reflection image under different headwind forces, and determine the image that best matches the vehicle's specifications and proportions as the reference image, and set the headwind force corresponding to the reference image as the reference headwind force. The image deviation detection module is used to analyze local operation data in historical operation data that belong to the same specified water surface and are consistent with the headwind force. It compares the vehicle images reflected on the specified water surface at different times with the reference image to determine whether there is an image deviation that gradually deteriorates over time. It also detects whether the coupling degree between the trend of image deviation and the trend of environmental deterioration with the decrease in water transparency exceeds a preset threshold. The gain correction module is used to generate an enhancement magnitude based on the image deviation between the most recent vehicle image and the reference image when the coupling degree is determined to exceed a preset threshold, and to correct the initial edge sharpening gain parameters based on the enhancement magnitude.

8. The edge vision enhancement processing system in complex environments according to claim 7, characterized in that, The designated water surface refers to a water-filled area, puddle, fountain overflow surface, or similar artificial or natural water body surface that is located in urban roads or traffic scenarios and is formed due to long-term water accumulation, low-lying terrain, fountain overflow, or poor drainage, which can produce obvious ripples when vehicles pass by and has mirror reflection characteristics. The water surface fluctuation samples and local operational data are consistent with the designated water surface in terms of objective background conditions, which include at least the average area, average water depth, boundary morphology and geographical location of the water surface. The vehicles selected in the water surface fluctuation samples and local operation data are consistent, meaning that the vehicles have the same appearance size and color.

9. The edge vision enhancement processing system in complex environments according to claim 8, characterized in that, The headwind force refers to the magnitude of the wind force acting on the designated water surface in the opposite direction when the direction of water surface ripples generated after a vehicle passes over the designated water surface is opposite to the direction of external wind force.

10. The edge vision enhancement processing system in complex environments according to claim 9, characterized in that, The reference headwind force determination module specifically includes: The reflection image acquisition unit is used to analyze each of the water surface ripple samples and acquire reflection images of vehicles traveling in the reflection of the specified water surface under different headwind conditions. The morphological matching degree calculation unit is used to acquire a standard image corresponding to the vehicle, and calculate the morphological matching degree between the reflection image corresponding to each water surface wave sample and the standard image, wherein the morphological matching degree includes at least the degree of overlap of the outer contour and the consistency of the proportion. The reference headwind force determination unit is used to determine the reflection image with the highest shape matching degree as the reference image that best matches the vehicle body specifications and proportions, and to set the headwind force corresponding to the reference image as the reference headwind force.

Images