Ice layer bearing capacity intelligent evaluation system and method based on Python image recognition

Abstract

The invention provides an ice layer bearing capacity intelligent evaluation system and method based on Python image recognition, belongs to the technical field of ice layer bearing capacity measurement, and solves the problem that a traditional ice surface crack and weak zone recognition method is low in recognition efficiency. The image acquisition module is used for acquiring high-resolution images of to-be-detected ice layers; the radar module is installed on the crawler trolley and used for collecting radar images of ice layers at different positions and pressure applied to the ice layers by the crawler trolley. The image transmission module is used for receiving the high-resolution image of the ice layer to be detected, the pressure applied to the ice layer by the crawler trolley and the radar images of the ice layers at different positions, and transmitting the images to the data processing module; and the data processing module is used for acquiring an ice layer bearing capacity evaluation result based on the ice layer bearing capacity safety coefficient F.

Description

Technical Field

[0001] The present invention relates to an intelligent evaluation system and method for ice layer bearing capacity based on Python image recognition, and belongs to the technical field of ice layer bearing capacity measurement. Background Art

[0002] When lakes and rivers in cold northern regions freeze over in winter, ice conditions are complex and volatile. Cracks and weak zones in the ice pose a serious threat to people, vehicles, and infrastructure on the surface. For example, ice recreational activities and transportation in cold regions are particularly prone to accidents on cracked ice. Therefore, calculations and analyses of ice cracks, weak zones, and ice thickness are necessary to assess ice safety and stability.

[0003] Traditional methods for identifying cracks and weak zones on the ice surface mostly rely on manual field surveys, which are inefficient and have limited coverage, making them difficult to fully implement on large ice areas. The invention patent with publication number CN 105547839 A proposes an ice bearing capacity testing method and device, which sets a pressure conductor on the upper surface of the test ice layer, and sets a lower pressure distribution plate, a synchronous jack, an upper pressure distribution plate, a pressure gauge and a truss in sequence above the pressure conductor. A plurality of anchor rods are set around the pressure conductor, and an anchor plate is set at the lower end of the anchor rod to hook the lower surface of the ice layer. The upper end of the anchor rod is connected to the truss through a hook. The sum of the contact areas between all anchor plates and the lower surface of the ice layer is greater than the contact area between the pressure conductor and the upper surface of the ice layer. The synchronous jack is raised to load pressure on the ice layer, and the pressure value applied by the synchronous jack is read by the pressure gauge; a deflection tester is set at the center of the lower pressure distribution plate to test the ice layer. The displacement of the ice layer is tested using a complex mechanical device that requires a fixed device for the ice layer depth and a series of sensors to measure it. The range of a single measurement is limited, making it unsuitable for comprehensive assessment of the bearing capacity of large-scale ice layers. Patent publication number CN104568743B proposes a device and method for testing the vertical bonding strength of ice layers on a material surface. First, a conical cylinder in the measuring device is used to prepare ice layers of equal volume and contact area. The ice layers are then lifted and peeled off using a stretching device. The maximum normal force during the peeling process is measured, and the ratio of this value to the contact area between the ice layer and the specimen is used to represent the vertical bonding strength of the ice layer on the material surface. This approach also focuses on calculating the bearing capacity of ice layers over a small area. Furthermore, the two aforementioned patents simulate tests on materials without distinguishing between different ice layers. However, uneven ice thickness distribution and cracks in the ice layer can also affect the overall bearing capacity of the ice layer. Summary of the Invention

[0004] In order to solve the problem of low recognition efficiency of traditional ice surface crack and weak zone identification methods, the present invention proposes an ice layer bearing capacity intelligent evaluation system and method based on Python image recognition.

[0005] The technical solution adopted by the present invention to solve the above problems is: the ice bearing capacity intelligent evaluation system based on Python image recognition proposed by the present invention includes:

[0006] UAV, tracked vehicle, radar module, image acquisition module, image transmission module, data processing module;

[0007] The image acquisition module is set under the drone and is used to collect high-resolution images of the ice layer to be measured;

[0008] The radar module is installed on a tracked vehicle and is used to collect radar images of ice layers at different locations and the pressure applied by the tracked vehicle to the ice layers.

[0009] The image transmission module is used to receive high-resolution images of the ice layer to be measured, the pressure applied to the ice layer by the crawler vehicle, and radar images of the ice layer at different positions, and transmit them to the data processing module;

[0010] The data processing module obtains the number of cracks, weak areas and thickness of the ice layer based on the high-resolution image of the ice layer, the pressure applied to the ice layer by the tracked vehicle and the radar image, and obtains the bearing capacity safety factor F of the ice layer based on the number of cracks, weak areas and thickness of the ice layer. Based on the bearing capacity safety factor F of the ice layer, the ice layer bearing capacity evaluation result is obtained.

[0011] Furthermore, the image acquisition module is equipped with a wide-angle, high-resolution camera, which is placed under the drone;

[0012] The crawler trolley is equipped with a load-bearing capacity tester, which measures the pressure applied by the crawler trolley to the ice layer.

[0013] Furthermore, the radar module includes a radar transmitting platform and a radar receiving platform. The radar transmitting platform transmits radar waves to measure the thickness of the ice layer, and the radar receiving platform receives echoes and draws a radar image, wherein the collected radar image is a grayscale image.

[0014] The intelligent evaluation method of ice bearing capacity based on Python image recognition includes:

[0015] Step 1: Use a drone equipped with an image acquisition module to collect high-resolution images of the ice layer to be measured, and transmit the collected high-resolution images to the data processing module through the data transmission module;

[0016] Step 3: Measure the pressure applied to the ice layer by the crawler using a load-bearing capacity tester. Use the crawler to change the radar detection position of the ice layer. Use the radar module to obtain radar images of the ice layer at different positions. Then, transmit the collected radar images and the pressure applied to the ice layer by the crawler to the data processing module via the data transmission module.

[0017] Step 4: The data processing module processes the received radar image in the form of Python code to calculate the thickness of the ice layer. The data processing module processes the received image of the ice layer to be measured at the position corresponding to the radar image in the form of Python code to identify the number of cracks and weak areas in the ice layer. Based on the number of cracks and weak areas in the ice layer and the thickness of the ice layer, the bearing capacity safety factor F of the ice layer is calculated. Based on the bearing capacity safety factor F of the ice layer, the ice layer bearing capacity evaluation result is obtained.

[0018] Furthermore, the calculation of ice thickness in step 4 specifically includes:

[0019] The radar echo waveform characteristics are analyzed to identify the reflected signals from the upper and lower interfaces of the ice layer. The grayscale radar image is converted into a NumPy array. Python code is used to detect the branches of the converted radar image and determine the hierarchical boundaries. The ice thickness is calculated by calculating the difference between adjacent boundaries.

[0020] Furthermore, identifying cracks and weak areas in the ice layer in step 4 specifically includes:

[0021] Step 4.1.1: Use the cv2 library to read the high-resolution image in the specified path and convert the high-resolution image into a grayscale image;

[0022] Step 4.1.2: Binarize the grayscale image by setting a threshold. Detect several white areas in the grayscale image. The white areas are weak areas of the ice layer. Use the contour detection coefficient to obtain the contour information of the white areas and calculate the area of ​​each white area.

[0023] Step 4.1.3: Filter by area to obtain white areas larger than a preset value, and create a mask to block the white areas filtered out;

[0024] Step 4.1.4: Use the Canny edge detection algorithm to extract the edges of the occluded image, detect cracks using the Hough transform line detection algorithm, plot the detected cracks on the read high-resolution image copy, and count the number of cracks;

[0025] Step 4.1.5: The Python code saves the detection result image and visualizes it using the matplotlib library, marking the image to show the number of cracks, the number of white areas, and the proportion of weak areas.

[0026] Furthermore, the calculation formula for the ice layer bearing capacity safety factor F in step 4 is:

[0027]

[0028] In formula (1), F is the ice layer bearing capacity safety factor, P is the load of the crawler on the ice layer, the unit is kPa, P Max is the ultimate bearing capacity of the ice layer, in kPa, n is the number of cracks on the ice surface, A is the pixel area of ​​the weak layer, H is the actual thickness of the ice layer, in cm, k1 and k2 are the correlation coefficients of the influence of the number of cracks and the pixel area of ​​the weak layer on the ultimate bearing capacity of the ice layer, A0 is the reference pixel area of ​​the weak layer of the ice layer, H0 is the reference thickness of the ice layer, in cm;

[0029] If the safety factor F is less than 1, it means that the ice layer load exceeds the ultimate bearing capacity and the ice layer is in an unsafe state. If the safety factor F is close to 1, it means that the ice layer faces a load risk. If the safety factor F is greater than 1, the ice layer is in a complete state.

[0030] The beneficial effects of the present invention are:

[0031] 1. The Python-based image recognition-based intelligent ice bearing capacity assessment system proposed in this paper uses drones to collect ice surface images to detect the number of cracks and weak areas. A small vehicle collects radar images at corresponding locations to detect ice thickness. The system can be remotely controlled, eliminating the need for on-site operators, ensuring accurate test data while reducing risks during the measurement process.

[0032] 2. This invention uses Python code to analyze the collected images and radar images, accurately obtaining information on the number of cracks, weak areas, and ice thickness, providing accurate data support for subsequent calculations of the ice bearing capacity and further for the assessment of the ice bearing capacity, thereby facilitating timely warnings, formulating reasonable ice surface traffic plans, and carrying out ice surface maintenance and management work. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a structural diagram of the ice bearing capacity intelligent evaluation system based on Python image recognition provided by the present invention;

[0034] Figure 2 This is a schematic diagram of the main body of the drone provided by the present invention. Figure 2 In the middle, 1-four sets of propellers, 2-main control board, 3-motor, 4-high-resolution camera;

[0035] Figure 3 This is a schematic diagram of the main body of the crawler vehicle provided by the present invention. Figure 3 In the middle, 5-crawler track, 6-radar equipment loading platform, 7-motor, 8-load capacity tester;

[0036] Figure 4 A schematic diagram of the flow chart of the ice bearing capacity intelligent evaluation method based on Python image recognition provided by the present invention;

[0037] Figure 5 A schematic diagram of the number of ice cracks and weak zone identification results provided by the present invention;

[0038] Figure 6 A schematic diagram of a three-point bending test provided by the present invention;

[0039] Figure 7 A schematic diagram of a uniaxial compression test provided by the present invention;

[0040] Figure 8 This is a schematic diagram of the ice thickness radar identification results provided by the present invention. DETAILED DESCRIPTION

[0041] Specific implementation method 1: Combination Figure 1-3 This embodiment is described as follows. Figure 1 As shown, the structure of the ice bearing capacity intelligent evaluation system based on Python image recognition described in this embodiment includes:

[0042] UAV, tracked vehicle, radar module, image acquisition module, image transmission module, data processing module;

[0043] The structure of the drone is as follows Figure 2 As shown, the drone body mainly includes a motor 3, four sets of propellers 1, a main control board 2, and a remote controller. It can perform test operations at different heights above the ice surface to be tested steadily and vertically, and the flight is controlled by the main control board and the remote controller.

[0044] The image acquisition module is provided with a wide-angle, high-resolution camera 4, which is placed under the drone.

[0045] The radar system includes a radar transmitting platform and a radar receiving platform. The radar system is installed on a tracked vehicle and is used to sense information about weak areas inside the ice layer. The radar detection position of the ice layer is changed by the tracked vehicle, and the radar image is collected by the radar receiving platform. The collected radar image and the pressure applied to the ice layer by the tracked vehicle are transmitted to the data processing module through the data transmission module. The collected radar image is a color image.

[0046] The data processing system uses Python code to batch-identify cracks and weak zones on the ice surface in the collected images. Specifically, it identifies the number of cracks on the ice surface in each image and the ratio of the pixel area of ​​the weak zones to the total pixel area of ​​the image. When detecting ice thickness using radar waves, Python code is also used to classify the collected radar waves into layers, thereby determining the ice thickness. Based on the ice thickness, cracks, and weak zones, the system accurately calculates the ice's bearing capacity safety factor.

[0047] The structure of the crawler trolley is as follows Figure 3 As shown, the crawler trolley includes a crawler 5, a radar equipment loading platform 6, an electric motor 7 and a load-bearing capacity tester 8. Under the premise of ensuring safe driving, the crawler trolley is equipped with a radar device to detect the thickness of the ice layer. The load-bearing capacity tester measures the pressure applied to the ice layer by the crawler trolley through a retractable hammer.

[0048] In summary, the intelligent evaluation system for ice bearing capacity based on Python image recognition proposed in this invention collects ice surface images through drones to detect the number of cracks and weak areas, and uses a small car to collect radar images at the corresponding positions to detect the thickness of the ice layer. The system can be remotely controlled and does not require operators to arrive at the site, which ensures the accuracy of test data while reducing the risks in the measurement process.

[0049] Specific implementation method 2: Combination Figure 4-Figure 8 This embodiment is described as follows. Figure 4 As shown, the steps of the ice bearing capacity intelligent evaluation method based on Python image recognition described in this embodiment include:

[0050] S1: Install the radar module on the crawler and place the image acquisition module under the drone;

[0051] S2: By changing the flight altitude of the UAV, the image acquisition module collects high-resolution images of the ice layer to be measured at different altitudes, and transmits the collected high-resolution images to the data processing module through the data acquisition module;

[0052] S3: Measure the pressure applied to the ice layer by the crawler with a retractable hammer, change the radar detection position of the ice layer by the crawler, and transmit the collected radar image and the pressure applied to the ice layer by the crawler to the data processing module through the data transmission module;

[0053] S4: The data processing module uses Python code to analyze and process the received images of the ice layer to be measured, radar waves, and the pressure applied to the ice layer by the tracked vehicle, calculate the thickness and bearing capacity of the ice layer, and identify cracks and weak areas in the ice layer.

[0054] S401: Ice thickness calculation;

[0055] S40101: Radar wave ice thickness measurement utilizes the propagation characteristics of electromagnetic waves in different media. When radar waves are transmitted into the ice, due to the significant differences in electromagnetic properties (such as dielectric constants) between the ice and the air, and between the ice and the underlying medium (such as water or rock), the electromagnetic waves are reflected and refracted at these interfaces. This results in different radar bands in the captured radar image.

[0056] S40102: Identify the reflected signals from the upper and lower interfaces of the ice layer by analyzing the radar echo waveform characteristics. Convert the color image to a grayscale image, and then convert the grayscale image to a NumPy array for calculation.

[0057] S40103: Use the code to detect the peak value of the image, determine the possible layer boundaries, and finally calculate the difference between adjacent boundaries to obtain the layer thickness in pixels. Figure 8 As shown in the figure, after comparing it with the radar band, it can be found that the pixel distance between the first and seventh peaks is the ice thickness. Subsequent manual measurement of ice thickness is required to determine a reliable empirical value for the ice thickness conversion coefficient.

[0058] S402: Identification of cracks and weak areas in the ice;

[0059] This step mainly realizes the detection and analysis of cracks and weak areas in the ice surface image, and integrates and visualizes the relevant data.

[0060] S40201: First, the code uses the cv2 library to read the image in the specified path and convert it to a grayscale image. Then, it binarizes the grayscale image by setting a threshold to detect large white areas:

[0061] S40202: Use the contour detection function to obtain the contour information of the white area, filter out the larger white areas based on their area size, and create a mask to eliminate the interference of the white areas on crack detection in the future. Calculate the proportion of the white areas.

[0062] S40203: In the crack detection section, the Canny edge detection algorithm is used to extract image edges. Then, bitwise operations are performed on the overlapped positions of the weak band and the crack. After eliminating the influence of the weak band on crack identification, the cracks are detected using the Hough transform line detection algorithm. The detected cracks are plotted on a copy of the original image, and the number of cracks is counted.

[0063] S40204: The code integrates some simulation data, including the number of cracks, ultimate strength (measured by beam bending test), temperature (surface temperature of the ice layer to be measured), and the proportion of weak areas (i.e. the proportion of white areas) to build a database. These data are used for subsequent analysis or combined with image detection results for more in-depth research, such as using a multivariate linear regression model. Figure 6 The three-point bending test shown and Figure 7 The uniaxial compression test shown in the figure is used to measure the ultimate bearing capacity of ice blocks. The ultimate bearing capacity data obtained are calculated using the Pearson correlation coefficient method, as shown in formula (1):

[0064]

[0065] In formula (1), k includes the correlation coefficients k1 and k2 of the number of cracks and the pixel area of ​​the weak layer on the ultimate bearing capacity of the ice layer, x i is the number of cracks or the pixel area of ​​the weak layer, y i is the ultimate bearing capacity obtained from the test.

[0066] S40205: Finally, the code saves the detection result image and uses the matplotlib library to visualize the detection result image. Key information such as the number of cracks, the number of white areas, and the proportion of weak areas are displayed in the image title, making it easy to visually observe the analysis results of the ice surface image. The number of ice cracks and the identification results of weak zones are shown in the figure below. Figure 5 As shown, the overall code function provided by this embodiment is relatively complete, which can effectively realize the preliminary detection analysis and result display tasks of ice surface images, and provide basic data and intuitive basis for further research on ice surface conditions.

[0067] In summary, this embodiment uses Python code to accurately extract the characteristic information of cracks and weak zones, accurately measure the thickness of the ice layer, and provide accurate data support for the evaluation of the bearing capacity of the ice layer.

[0068] S403: Ice bearing capacity calculation;

[0069] Combining the characteristic information of cracks and weak zones calculated in S401 and S402 and the precise data of ice thickness, this embodiment calculates the safety factor F of the ice layer bearing capacity based on these data. Specifically, see formula (2). If the safety factor is less than 1, it means that the ice layer load exceeds the ultimate bearing capacity and the ice layer is in an unsafe state. If the safety factor approaches 1, it means that the ice layer faces a bearing risk. If the safety factor is greater than 1, the ice layer is in a safe state.

[0070] The calculation formula of ice bearing capacity safety factor F is:

[0071]

[0072] In formula (2), F is the safety factor of ice layer bearing capacity, P is the load of the crawler on the ice layer, the unit is kPa, P Max is the ultimate bearing capacity of the ice layer, in kPa, n is the number of cracks on the ice layer surface, A is the pixel area of ​​the weak layer, H is the actual thickness of the ice layer, in cm, k1 and k2 are the correlation coefficients of the influence of the number of cracks and the pixel area of ​​the weak layer on the ultimate bearing capacity of the ice layer, A0 is the reference pixel area of ​​the weak layer of the ice layer, and H0 is the reference thickness of the ice layer, in cm.

[0073] In summary, this protection implementation utilizes the characteristic information of cracks and weak zones calculated above and the precise data of ice thickness to provide accurate data support for ice bearing capacity calculation, and facilitates timely warning, formulation of reasonable ice surface traffic planning, and ice surface maintenance and management.

[0074] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any technician familiar with the present profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical content disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement of the above embodiments made according to the technical essence of the present invention, within the spirit and principles of the present invention, without departing from the content of the technical solution of the present invention, shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. The ice bearing capacity intelligent evaluation system based on Python image recognition is characterized by: The ice layer bearing capacity intelligent evaluation system based on Python image recognition includes: UAV, tracked vehicle, radar module, image acquisition module, image transmission module, data processing module; The image acquisition module is set under the drone and is used to collect high-resolution images of the ice layer to be measured; The radar module is installed on a tracked vehicle and is used to collect radar images of ice layers at different locations and the pressure applied by the tracked vehicle to the ice layers. The image transmission module is used to receive high-resolution images of the ice layer to be measured, the pressure applied to the ice layer by the crawler vehicle, and radar images of the ice layer at different positions, and transmit them to the data processing module; The data processing module obtains the number of cracks, weak areas and thickness of the ice layer based on the high-resolution image of the ice layer, the pressure applied to the ice layer by the tracked vehicle and the radar image, and obtains the bearing capacity safety factor F of the ice layer based on the number of cracks, weak areas and thickness of the ice layer. Based on the bearing capacity safety factor F of the ice layer, the ice layer bearing capacity evaluation result is obtained.

2. The ice layer bearing capacity intelligent evaluation system based on Python image recognition according to claim 1 is characterized in that: The image acquisition module is equipped with a wide-angle, high-resolution camera and is placed under the drone; The crawler trolley is equipped with a load-bearing capacity tester, which measures the pressure applied by the crawler trolley to the ice layer.

3. The ice bearing capacity intelligent evaluation system based on Python image recognition according to claim 1 is characterized in that: The radar module includes a radar transmitting platform and a radar receiving platform. The radar transmitting platform transmits radar waves to measure the thickness of the ice layer, and the radar receiving platform receives the echo and draws a radar image. The collected radar image is a grayscale image.

4. The method for intelligent evaluation of ice bearing capacity based on Python image recognition is applied to the intelligent evaluation system for ice bearing capacity based on Python image recognition according to any one of claims 1 to 3, characterized in that: include: Step 1: A high-resolution image of the ice layer to be measured is collected by a drone equipped with an image acquisition module, and the collected high-resolution image is transmitted to the data processing module through the data transmission module; Step 3: Measure the pressure applied to the ice layer by the crawler using a load-bearing capacity tester. Use the crawler to change the radar detection position of the ice layer. Use the radar module to obtain radar images of the ice layer at different positions. Then, transmit the collected radar images and the pressure applied to the ice layer by the crawler to the data processing module via the data transmission module. Step 4: The data processing module processes the received radar image in the form of Python code to calculate the thickness of the ice layer. The data processing module processes the received image of the ice layer to be measured at the position corresponding to the radar image in the form of Python code to identify the number of cracks and weak areas in the ice layer. Based on the number of cracks and weak areas in the ice layer and the thickness of the ice layer, the bearing capacity safety factor F of the ice layer is calculated. Based on the bearing capacity safety factor F of the ice layer, the ice layer bearing capacity evaluation result is obtained.

5. The method for intelligent evaluation of ice bearing capacity based on Python image recognition according to claim 4 is characterized in that: The calculation of ice thickness in step 4 specifically includes: The radar echo waveform characteristics are analyzed to identify the reflected signals from the upper and lower interfaces of the ice layer. The grayscale radar image is converted into a NumPy array. Python code is used to detect the branches of the converted radar image and determine the hierarchical boundaries. The ice thickness is calculated by calculating the difference between adjacent boundaries.

6. The method for intelligent evaluation of ice bearing capacity based on Python image recognition according to claim 4 is characterized in that: Step 4 involves identifying cracks and weak areas in the ice: Step 4.1.1: Use the cv2 library to read the high-resolution image in the specified path and convert the high-resolution image into a grayscale image; Step 4.1.2: Binarize the grayscale image by setting a threshold. Detect several white areas in the grayscale image. The white areas are weak areas of the ice layer. Use the contour detection coefficient to obtain the contour information of the white areas and calculate the area of ​​each white area. Step 4.1.3: Filter by area to obtain white areas larger than a preset value, and create a mask to block the white areas filtered out; Step 4.1.4: Use the Canny edge detection algorithm to extract the edges of the occluded image, detect cracks using the Hough transform line detection algorithm, plot the detected cracks on the read high-resolution image copy, and count the number of cracks; Step 4.1.5: The Python code saves the detection result image and visualizes it using the matplotlib library, marking the image to show the number of cracks, the number of white areas, and the proportion of weak areas.

7. The method for intelligent evaluation of ice bearing capacity based on Python image recognition according to claim 4 is characterized in that: The calculation formula for the ice layer bearing capacity safety factor F in step 4 is: In formula (1), F is the safety factor of ice layer bearing capacity, P is the load of the crawler on the ice layer, the unit is kPa, P Max is the ultimate bearing capacity of the ice layer, in kPa, n is the number of cracks on the ice surface, A is the pixel area of ​​the weak layer, H is the actual thickness of the ice layer, in cm, k1 and k2 are the correlation coefficients of the influence of the number of cracks and the pixel area of ​​the weak layer on the ultimate bearing capacity of the ice layer, A0 is the reference pixel area of ​​the weak layer of the ice layer, H0 is the reference thickness of the ice layer, in cm; If the safety factor F is less than 1, it means that the ice layer load exceeds the ultimate bearing capacity and the ice layer is in an unsafe state. If the safety factor F is close to 1, it means that the ice layer faces a load risk. If the safety factor F is greater than 1, the ice layer is in a complete state.

Images