YOLO and thermodynamic diagram mouse disease assessment method, system and equipment and storage medium

Abstract

The invention discloses a YOLO and thermodynamic diagram mouse disease assessment method, system and device and a storage medium, and relates to the technical field of computer vision and intelligent monitoring, and the method comprises the steps: inputting a target region video image into a YOLOv11 target detection model, carrying out the recognition and positioning of a mouse target in the target region video image, and outputting a mouse detection frame set; according to the coordinate data in the detection frame set, constructing a thermodynamic diagram matrix and performing spatial distribution density estimation, generating a thermodynamic diagram of the mouse activity and performing thermodynamic diagram rendering; and based on the mouse detection frame set, calculating a mouse disease safety index and performing visual display. According to the method, the accuracy and the detection efficiency of mouse disease identification are effectively improved, the spatial aggregation characteristics of the mouse activity area can be clearly described, the fine analysis capability of mouse diseases is improved, the interpretability of image expression is enhanced, and quantitative evaluation and graphical expression of mouse disease risk levels are realized.

Description

Technical Field

[0001] This invention relates to the field of computer vision and intelligent monitoring technology, specifically to a YOLO and heat map method, system, device, and storage medium for rodent infestation assessment. Background Technology

[0002] Monitoring and assessment technologies for rodent infestations are gradually evolving from traditional manual capture and mechanical sensing methods towards automation and intelligence. Currently, image recognition technology based on video surveillance is widely used, with deep learning target detection algorithms (such as the YOLO series) becoming an important means of animal target monitoring due to their high detection accuracy and fast processing speed. Simultaneously, spatial data analysis and heatmap visualization are increasingly being applied to environmental risk analysis to reflect the spatial distribution density of target activity.

[0003] However, existing rodent detection technologies generally suffer from the following problems: most methods only reach the target identification stage, failing to further analyze the target's activity trajectory and regional density distribution, and lacking spatial risk representation; they lack unified risk quantification standards, making dynamic risk assessment based on detection results impossible; and the data display format is limited, failing to form an integrated assessment system encompassing identification, analysis, and visualization. Therefore, there is an urgent need for a method that organically combines YOLO detection, heatmap analysis, and safety index calculation to achieve automatic identification, dynamic assessment, and visual display of rodent infestation risks. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that existing image recognition-based rodent infestation detection methods have problems such as low recognition accuracy, inability to obtain spatial distribution information of target activity, lack of unified risk quantification standards, and inability to achieve visual dynamic display.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a YOLO and heatmap method for assessing rodent infestation, comprising: inputting a video image of a target area into a YOLOv11 target detection model; identifying and locating rodent targets in the video image of the target area; outputting a set of rodent detection boxes; constructing a heatmap matrix and estimating its spatial distribution density based on the coordinate data in the set of detection boxes; generating a heatmap of rodent activity and rendering the heatmap; and calculating a rodent infestation safety index based on the set of rodent detection boxes and visualizing it.

[0007] Generating a heatmap of mouse activity involves calculating the heat intensity of each pixel in a fixed-resolution two-dimensional plane using a two-dimensional Gaussian kernel function, based on the center coordinates of the detected mouse target in each frame of the image.

[0008] As a preferred embodiment of the YOLO and heatmap rodent infestation assessment method described in this invention, the step of identifying and locating mouse targets in video images of the target area includes: acquiring video images of the target area captured by a monitoring camera, extracting video image frames, inputting the video image frames into the YOLOv11 target detection model, automatically identifying mouse targets in the video image frames through a one-stage convolutional neural network structure, and outputting a set of mouse detection boxes containing each detected target.

[0009] As a preferred embodiment of the YOLO and heatmap-based rodent infestation assessment method described in this invention, the output includes a set of mouse detection boxes for each detected target. The set of mouse detection boxes output by the YOLOv11 target detection model is represented as follows: , in, This is a set of mouse detection boxes. For the first The x-coordinate of the center point of the mouse detection box For the first The ordinate of the center point of the mouse detection box For the first The width and height of the mouse detection frame. To detect the confidence score, For category labels.

[0010] The center coordinates of the detected mouse target in each frame are recorded to form a time series dataset, as follows: , in, This is a time-series dataset of mouse activity. For the first Frame detection timestamps This represents the total number of targets detected cumulatively.

[0011] As a preferred embodiment of the YOLO and heatmap-based rodent infestation assessment method described in this invention, the generation of the heatmap of rodent activity includes: on a two-dimensional plane with a fixed resolution, calculating the heat intensity value of each pixel location based on the coordinates of the detected rodent target in each frame using a two-dimensional Gaussian kernel function, thus forming a heatmap of rodent activity represented as follows: , in, Location on the heat map The thermal intensity value indicates the intensity of mouse activity. For smoothing parameters.

[0012] As a preferred embodiment of the YOLO and heatmap rodent infestation assessment method described in this invention, the heatmap rendering includes color mapping based on the numerical range of heat intensity values. , in, To render an image for a heatmap, For color mapping functions, These are the original video image frames.

[0013] As a preferred embodiment of the YOLO and heatmap rodent infestation assessment method described in this invention, the calculation of the rodent infestation safety index includes calculating the rodent infestation safety index based on the total number of mouse detection frames for each camera and the number of active cameras, expressed as follows: , in, For rodent infestation safety index, The total number of mouse detection frames detected by all cameras This represents the number of cameras where mice were detected.

[0014] As a preferred embodiment of the YOLO and heat map rodent infestation assessment method described in this invention, the visualization includes: displaying the rodent infestation safety index as a visual curve; assessing the risk level of each monitoring area based on the heat map rendering image and the rodent infestation safety index analysis results; labeling the risk level of each area in the form of a star rating on the image; and rendering the heat map rendering image, the rodent infestation safety index trend map, and the risk star rating map in real time through the front end.

[0015] Another objective of this invention is to provide a YOLO and heatmap rodent infestation assessment system, which, through a specific system structure comprising a target detection module, a heatmap generation and rendering module, and a rodent infestation safety index calculation and visualization module, solves the problems of low identification accuracy, lack of spatial distribution analysis, and difficulty in quantifying risk levels in current rodent infestation assessment technologies.

[0016] As a preferred embodiment of the YOLO and heatmap rodent infestation assessment system described in this invention, it includes: a target detection module, a heatmap generation and rendering module, and a rodent infestation safety index calculation and visualization module; the target detection module is used to input video images of the target area into the YOLOv11 target detection model, identify and locate the rodent targets in the image, and output a set of rodent detection boxes; the heatmap generation and rendering module is used to construct a heatmap matrix based on the coordinate data in the set of detection boxes, and to estimate the spatial distribution density using a two-dimensional Gaussian kernel function to generate a heatmap of rodent activity, while simultaneously performing color mapping and image overlay for heatmap rendering; the rodent infestation safety index calculation and visualization module is used to calculate the rodent infestation safety index based on the set of rodent detection boxes, count the number of targets and the number of active cameras, and combine the heatmap output to perform risk level assessment and visualization.

[0017] Another object of the present invention is to provide a YOLO and heatmap rodent infestation assessment device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the YOLO and heatmap rodent infestation assessment method.

[0018] Another object of the present invention is to provide a YOLO and heatmap rodent infestation assessment storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the YOLO and heatmap rodent infestation assessment method.

[0019] The beneficial effects of this invention are as follows: The YOLO and heatmap rodent infestation assessment method provided by this invention automatically identifies and locates rodent targets by inputting video images of the target area into the YOLOv11 target detection model, thereby effectively improving the accuracy and efficiency of rodent infestation identification; by extracting coordinate data from the detection box set to construct a heatmap matrix and estimating the spatial distribution density based on a two-dimensional Gaussian kernel function, it can clearly depict the spatial clustering characteristics of rodent activity areas, improving the ability to perform refined analysis of rodent infestations; color mapping rendering of the heatmap makes the rodent activity distribution results intuitive and visual, enhancing the interpretability of image expression; the calculation and visualization of the rodent infestation safety index based on the rodent detection box set realizes the quantitative assessment and graphical expression of rodent infestation risk level, facilitating users to take timely management and response measures. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is an overall flowchart of a YOLO and heat map rodent infestation assessment method provided in Embodiment 1 of the present invention.

[0022] Figure 2 This is a mouse detection information diagram of a YOLO and heatmap mouse infestation assessment method provided in Embodiment 1 of the present invention.

[0023] Figure 3 This is a schematic diagram of heat map generation for a YOLO and heat map rodent infestation assessment method provided in Embodiment 1 of the present invention.

[0024] Figure 4 This is a thermal distribution result diagram of a YOLO and heat map method for assessing rodent infestation provided in Embodiment 1 of the present invention.

[0025] Figure 5 This is a daily change chart of the safety index of a YOLO and heatmap rodent infestation assessment method provided in Embodiment 1 of the present invention.

[0026] Figure 6 This is a weekly chart showing the safety index changes of a YOLO and heatmap rodent infestation assessment method provided in Embodiment 1 of the present invention.

[0027] Figure 7 This is a monthly change graph of the safety index of a YOLO and heatmap rodent infestation assessment method provided in Embodiment 1 of the present invention. Detailed Implementation

[0028] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0029] Example 1, referring to Figures 1-7 As an embodiment of the present invention, a YOLO and heatmap method for assessing rodent infestation is provided, comprising: S1: Input the video image of the target region into the YOLOv11 target detection model 100, identify and locate the mouse target in the video image of the target region, and output a set of mouse detection boxes 101.

[0030] Furthermore, the identification and localization of mouse targets in the target area video image includes acquiring video images of the target area captured by a surveillance camera, extracting video image frames, and inputting the video image frames into the YOLOv11 target detection model 100, such as... Figure 1As shown, a one-stage convolutional neural network structure is used to automatically identify mouse targets in video image frames, and outputs a set of mouse detection boxes 101 containing each detected target.

[0031] It should be noted that video data taken continuously by a surveillance camera at a fixed position for one day was selected as the experimental sample. The camera resolution was 1920×1080 (i.e., 1080P), the video frame rate was 25 frames / second, and the mouse detection box set 101 included information such as the center coordinates, width, height, confidence level, and category label of the mouse target in the image.

[0032] It should also be noted that the YOLOv11 target detection model 100 is used to identify and locate mouse targets in video images captured by surveillance cameras, and automatically generate a set of mouse detection boxes 101. This avoids the tedious manual frame-by-frame annotation, improves detection efficiency and recognition accuracy, and is suitable for real-time rodent infestation analysis in large-scale monitoring scenarios.

[0033] Furthermore, the output includes a set of mouse detection boxes 101 for each detected target. The set of mouse detection boxes 101 output by the YOLOv11 target detection model 100 is represented as follows: , in, The set of mouse detection boxes is 101, such as Figure 2 As shown, For the first The x-coordinate of the center point of the mouse detection box For the first The ordinate of the center point of the mouse detection box For the first The width and height of the mouse detection frame. To detect the confidence score, For category labels.

[0034] The center coordinates of the detected mouse target in each frame are recorded to form a time series dataset, as follows: , in, This is a time-series dataset of mouse activity. For the first The x-coordinate of the center point of the mouse detection box For the first The ordinate of the center point of the mouse detection box For the first Frame detection timestamps This represents the total number of targets detected cumulatively.

[0035] It should be noted that YOLOv11 adopts a unified regression framework, using convolutional neural networks to directly predict the target location and category. The loss function is expressed as: , in, For loss function, The bounding box regression loss measures the deviation between the predicted and ground truth bounding boxes, and uses the generalized intersection-over-union (IoU) ratio. To constrain the detection confidence level, the confidence level is set as the confidence loss. Classification loss measures the accuracy of predicted categories. , , This is the loss weighting coefficient.

[0036] In detail, to remove duplicate mouse detection boxes, the NMS algorithm is used for post-processing. Mouse detection boxes are sorted from high to low confidence levels, and for two candidate boxes... and The intersection-union ratio is expressed as: , set up threshold ,when When that happens, retain those with higher confidence levels.

[0037] It should also be noted that by constructing a standardized mouse detection box structure and binding the coordinates and time information obtained from each frame's detection, a continuously accumulating time-series data is formed. This allows the mouse activity trajectory to no longer be an isolated single-frame result, but rather to be systematically expressed and analyzed over the long term. Based on YOLOv11's unified regression detection mechanism, and through non-maximum suppression and cross-union ratio (CUI) filtering mechanisms, this invention ensures that only the most accurate detection box is retained at each moment, avoiding spatial deviations caused by duplicate counting. This ensures that the high-frequency regions of the heatmap 200 can truly reflect the activity density of the mice, and the statistical basis for calculating the safety index is more rigorous and reliable, collectively forming a complete data link.

[0038] S2: Based on the coordinate data in the detection box set, construct a heatmap matrix and estimate the spatial distribution density to generate a heatmap 200 of mouse activity and render the heatmap.

[0039] Furthermore, generating the heatmap of mouse activity 200 includes, on a two-dimensional plane with a fixed resolution, calculating the heat intensity value of each pixel location based on the coordinates of the detected mouse target in each frame using a two-dimensional Gaussian kernel function, forming the heatmap of mouse activity 200 as follows: , in, Location on the heat map The thermal intensity value indicates the intensity of mouse activity. For the first The x-coordinate of the center point of the mouse detection box For the first The ordinate of the center point of the mouse detection box The total number of targets detected cumulatively. For smoothing parameters.

[0040] It should be noted that the mouse center coordinates detected in each frame are mapped to a two-dimensional plane (1920×1080) with the same size as the video frame, and a heatmap matrix is ​​defined. Initialize it to a zero matrix. For the detected mouse coordinates in each frame ( In the thermal matrix, with that point as the center, a two-dimensional Gaussian function is used to accumulate heat, such as... Figure 3 As shown, after all frames have been processed, the thermal matrix... This records the spatial activity density distribution of the rats throughout the entire video period.

[0041] It should also be noted that by mapping the mouse center coordinates detected in each frame to a two-dimensional plane with the same size as the video frame, and then performing density superposition using a two-dimensional Gaussian kernel centered on each detection point on this plane, the thermal matrix can record the spatial activity intensity distribution of the mouse throughout the entire monitoring period. This processing method can form a continuous and smooth density region representation in space based on activity frequency, no longer limited to single-point recording.

[0042] Furthermore, the heatmap rendering process includes color mapping based on the numerical range of heat intensity values, as follows: , in, To render an image for a heatmap, For color mapping functions, These are the original video image frames. Location on the heat map The thermal intensity value.

[0043] It should be noted that, according to The data range is color-mapped (e.g., green-to-red gradient). Darker colors indicate more frequent mouse activity or longer dwell times. Green represents areas with low or almost no mouse activity, yellow to orange represents areas with moderate activity, and red represents high-risk areas where mice frequently pass through or linger for extended periods. Color mapping is performed by a color mapping function, generating a color image with the same dimensions as the original video frame. After obtaining the color image, it is overlaid onto the original video frame. The overlay method is pixel-level, preserving the structure and background information of the original video frame. The color layer of the heatmap 200 is semi-transparently overlaid on the original image, allowing users to simultaneously see the monitored scene and the distribution of mouse activity density. The resulting image is the heatmap rendering image, which retains the resolution and content of the original video while accurately representing the density areas of mouse activity in space.

[0044] It should also be noted that, through rendering the image, the heatmap 200 clearly shows the activity areas of the rats. The red areas represent high-risk areas where rats appear most frequently or stay for the longest time, while the green areas represent low-risk areas where there is less activity or no activity. The heatmap 200 effect can provide intuitive and effective support for subsequent risk assessment and control measures.

[0045] S3: Based on the set of 101 mouse detection boxes, calculate the rodent infestation safety index and visualize it.

[0046] Furthermore, the calculation of the rodent infestation safety index includes calculating the rodent infestation safety index based on the total number of mouse detection frames for each camera and the number of active cameras, expressed as follows: , in, For rodent infestation safety index, This represents the total number of mouse detection frames detected by all cameras. This represents the number of cameras where mice were detected.

[0047] It should be noted that after completing the statistics on the number of rats detected and the recording of camera activity, the next step is to calculate the rat infestation safety index. This involves summarizing the number of rat target frames detected by all cameras within a selected time period (e.g., one day) and generating a total count, thus calculating the number of cameras that had rats present during that period. These two statistical measures are then used to convert the detection count and camera activity count into safety index evaluation factors.

[0048] Specifically, in the actual calculation process, the two statistics are subjected to cube root processing to reduce the impact of excessively rapid growth in magnitude on the index results, making the index calculation comparable for monitoring areas of different sizes. The two processed parameters are then multiplied to obtain the risk quantification value for that period. By using a fixed percentage level as a benchmark, this risk quantification value is subtracted from the percentage level to finally obtain the rodent infestation safety index in the range of 0 to 100, where a higher index indicates a relatively safe environment, and a lower index indicates a higher risk level.

[0049] It should also be noted that the safety index is further mapped to a star rating. A preset 10-level rating range is used, with the safety index corresponding to 10 stars to 0 stars from high to low, realizing the conversion from absolute values ​​to intuitive levels.

[0050] Furthermore, based on the rodent infestation safety index, the weekly and monthly rodent infestation safety indices are obtained by averaging over a time scale and expressed as follows: , , in, The safety index for rat infestation. The safety index for the lunar rat infestation. This represents the number of days in the current month.

[0051] It should also be noted that by statistically analyzing the number of mouse detection frames and the number of cameras showing mice within a uniform time period for all cameras, and combining this with a percentage-based mapping method to generate a rodent infestation safety index ranging from 0 to 100, comparable safety assessment results can be obtained for monitoring areas of different sizes and camera densities. This avoids evaluation bias caused by a single detection count, ensuring that risk assessment no longer relies on the anomaly count of a particular camera. By averaging the daily safety index over time, weekly and monthly safety indices are obtained, expanding the description of rodent infestation risk from an instantaneous state to a trend-based expression, facilitating the identification of long-term patterns and periodic changes.

[0052] Furthermore, visualization can be achieved by overlaying heat maps onto the original surveillance video to visually reflect the mouse's activity paths and hotspot areas. Figure 4 As shown, the rodent infestation safety index is displayed as a visual curve, such as... Figures 5-7 As shown, based on the heat map rendering image and the analysis results of the rodent infestation safety index, the risk level of each monitoring area is assessed, and the risk level of each area is labeled on the image in the form of a star rating. The heat map rendering image, the rodent infestation safety index trend map and the risk star rating map are rendered in real time through the front end.

[0053] It should be noted that the coordinates of the mice in each frame of the image are processed to generate a heatmap. In the heatmap, the color of different areas represents the intensity (frequency and duration) of mouse activity. The color gradually changes from green (less activity) to red (more activity). Through the front-end interface, users can clearly view the distribution of mouse activity in real time or historical time periods. The Safety Index (RSI) reflects the safety status of the environment in different time periods through time series analysis of mouse activity data. The Safety Index is updated in real time and displayed in the form of a line graph. Users can view the changing trend of the rodent infestation over a certain period of time through charts. For example, users can view the changes in the rodent infestation safety index over the past week or month to understand changes in environmental risk.

[0054] Based on the results of heatmap and safety index calculations, the system automatically assesses the rodent infestation risk level of each monitored area. The risk level is mapped to a star rating (e.g., 10 stars represent low risk, and 0 stars represent extremely high risk). The risk star rating for each area is displayed as a label on the image, allowing users to quickly identify high-risk areas and implement appropriate control measures. The interface uses a front-end visualization library for real-time rendering and supports area filtering and time-based query functions. Users can flexibly filter and query through the interface: users can select a specific area to view its heatmap and safety index. The system will display data based on the user-selected area, and users can choose different time periods to view the rodent activity heatmap and safety index trends over a specific time period. For example, users can choose to view the rodent activity heatmap and safety index for the past day, week, or month.

[0055] It should also be noted that the target coordinates of each frame of mouse activity image are processed to generate a heatmap, with different areas varying in color from green to red to visually represent the density of mouse activity in the corresponding area. Users can view the distribution of mouse activity and the changes in the Safety Index (RSI) over real-time or historical periods on the front-end interface, and understand the time trend of the risk index through line graphs. Based on the heatmap and the calculation results of the safety index, a risk level star rating is automatically generated for each monitored area, and the rating is marked on the visualization layer in the form of a star rating. This allows a large amount of high-dimensional monitoring data to be presented in a unified symbolic way, enabling users to quickly identify high-risk areas and make action decisions.

[0056] Example 2, an embodiment of the present invention, provides a YOLO and heatmap rodent infestation assessment system, including a target detection module, a heatmap generation and rendering module, and a rodent infestation safety index calculation and visualization module.

[0057] Among them, the target detection module is used to input the video image of the target area into the YOLOv11 target detection model 100, identify and locate the mouse target in the image, and output a set of mouse detection boxes 101.

[0058] The heatmap generation and rendering module is used to construct a heatmap matrix based on the coordinate data in the detection box set, and to estimate the spatial distribution density using a two-dimensional Gaussian kernel function to generate a heatmap of mouse activity. At the same time, it performs color mapping and image overlay to render the heatmap.

[0059] The rodent infestation safety index calculation and visualization module is used to calculate the rodent infestation safety index based on the set of 101 rodent detection boxes, count the number of targets and the number of active cameras, and combine the output of heat map to conduct risk level assessment and visualization display.

[0060] This embodiment also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements a YOLO and heat map rodent infestation assessment system as proposed in the above embodiment.

[0061] This embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a YOLO and heatmap rodent infestation assessment system as described in the above embodiment.

[0062] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0063] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0064] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0065] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0066] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A YOLO and heatmap method for assessing rodent infestation, characterized in that, include: The target region video image is input into the YOLOv11 target detection model to identify and locate the mouse target in the target region video image and output a set of mouse detection boxes; Based on the coordinate data in the detection box set, a heat map matrix is ​​constructed and the spatial distribution density is estimated to generate a heat map of mouse activity and render the heat map. Based on a set of mouse detection boxes, a rodent infestation safety index is calculated and visualized. Generating a heatmap of mouse activity involves calculating the heat intensity of each pixel in a fixed-resolution two-dimensional plane using a two-dimensional Gaussian kernel function, based on the center coordinates of the detected mouse target in each frame of the image.

2. The YOLO and heatmap method for assessing rodent infestation as described in claim 1, characterized in that: The process of identifying and locating mouse targets in video images of the target area includes acquiring video images of the target area captured by a surveillance camera, extracting video image frames, inputting the video image frames into the YOLOv11 target detection model, automatically identifying mouse targets in the video image frames through a one-stage convolutional neural network structure, and outputting a set of mouse detection boxes containing each detected target.

3. The YOLO and heatmap method for assessing rodent infestation as described in claim 2, characterized in that: The output includes a set of mouse detection boxes for each detected target. The set of mouse detection boxes output by the YOLOv11 target detection model is represented as follows: ， in, This is a set of mouse detection boxes. For the first The x-coordinate of the center point of the mouse detection box For the first The ordinate of the center point of the mouse detection box For the first The width and height of the mouse detection frame. To detect the confidence score, Category labels; The center coordinates of the detected mouse target in each frame are recorded to form a time series dataset, as follows: ， in, This is a time-series dataset of mouse activity. For the first Frame detection timestamps This represents the total number of targets detected cumulatively.

4. The YOLO and heatmap method for assessing rodent infestation as described in claim 2 or 3, characterized in that: The generation of the heatmap of mouse activity includes, on a two-dimensional plane with a fixed resolution, calculating the heat intensity value of each pixel location based on the coordinates of the detected mouse target in each frame using a two-dimensional Gaussian kernel function, forming a heatmap of mouse activity as follows: ， in, Location on the heat map The thermal intensity value indicates the intensity of mouse activity. For smoothing parameters.

5. The YOLO and heatmap method for assessing rodent infestation as described in claim 4, characterized in that: The heatmap rendering process includes color mapping based on the numerical range of heat intensity values, as follows: ， in, To render an image for a heatmap, For color mapping functions, These are the original video image frames.

6. The YOLO and heatmap method for assessing rodent infestation as described in any one of claims 2, 3, and 5, characterized in that: The calculation of the rodent infestation safety index includes the calculation of the rodent infestation safety index based on the total number of rodent detection frames of each camera and the number of active cameras, expressed as follows: ， in, For rodent infestation safety index, This represents the total number of mouse detection frames detected by all cameras. This represents the number of cameras where mice were detected.

7. The YOLO and heatmap method for assessing rodent infestation as described in claim 6, characterized in that: The visualization includes displaying the rodent infestation safety index as a visual curve, assessing the risk level of each monitored area based on the heat map rendering image and the rodent infestation safety index analysis results, and annotating the risk level of each area in the form of a star rating on the image. The heat map rendering image, the rodent infestation safety index trend map, and the risk star rating map are rendered in real time through the front end.

8. A YOLO and heatmap rodent infestation assessment system, employing the YOLO and heatmap rodent infestation assessment method as described in any one of claims 1 to 7, characterized in that: Includes a target detection module, a heatmap generation and rendering module, and a rodent infestation safety index calculation and visualization module; The target detection module is used to input video images of the target area into the YOLOv11 target detection model, identify and locate mouse targets in the image, and output a set of mouse detection boxes; The heatmap generation and rendering module is used to construct a heatmap matrix based on the coordinate data in the detection box set, and to estimate the spatial distribution density through a two-dimensional Gaussian kernel function to generate a heatmap of mouse activity. At the same time, it performs color mapping and image overlay to render the heatmap. The rodent infestation safety index calculation and visualization module is used to calculate the rodent infestation safety index based on the set of rodent detection boxes, count the number of targets and the number of active cameras, and combine the output of heat map to conduct risk level assessment and visualization display.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the YOLO and heatmap rodent infestation assessment method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the YOLO and heatmap rodent infestation assessment method as described in any one of claims 1 to 7.

Images