Deep foundation pit slope soil displacement monitoring method and device and storage medium
Through the real-time monitoring method of high-resolution cameras and environmental sensing data combined with YOLO model, the problem of regular inspection errors and insufficient coverage of traditional monitoring methods is solved, real-time, automated and accurate displacement monitoring of deep foundation pit slopes is achieved, and the improvement of improvement of the improvement of the Construction safety and data accuracy.
Patent Information
- Application Number
- CN202410770865.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-14
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-06-14
AI Technical Summary
In existing deep foundation pit projects, traditional monitoring methods require manual regular inspection, which has errors and is difficult to achieve real-time monitoring and early warning. In addition, physical measurement tools can only monitor at fixed points, making it difficult to fully cover the displacement of foundation pit slopes.
Use high-resolution cameras to obtain real-time high-definition images and environmental sensing data, pre-process and recognition analysis through the trained and improved YOLO model, combine visual information and environmental sensing data for fusion calculation, and real-time monitoring and analysis of deep-based Displacement of pit slope.
Improve the automation level and accuracy of monitoring, real-time fine-grained monitoring of deep foundation pit slopes, timely identify soil cracks and displacement risks, improve construction safety, reduce costs, and enhance the accuracy of monitoring data. Sex and reliability.
Smart Images

Figure CN118980316A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of deep foundation pit engineering, and in particular to a method, device and storage medium for monitoring the displacement of soil on a slope of a deep foundation pit. Background Art
[0002] In the field of deep foundation pit engineering, monitoring the displacement and stability of the foundation pit is a crucial task to ensure construction safety and structural integrity.
[0003] At present, most of the traditional monitoring methods are used to monitor the displacement of foundation pits. The traditional monitoring method is to use physical measurement tools (such as inclinometers, displacement meters and other equipment) to monitor the displacement of foundation pits. Although these physical measurement tools can provide accurate displacement data, they have some limitations, such as the following limitations:
[0004] (1) Traditional monitoring methods often require regular manual inspections, which is not only time-consuming but may also lead to errors due to human factors.
[0005] (2) Physical measurement tools can usually only monitor at a few fixed points, making it difficult to achieve comprehensive coverage. In addition, due to delays in data collection and processing, it is difficult to achieve real-time monitoring and early warning. Summary of the invention
[0006] The present application provides a method, device and storage medium for monitoring soil displacement on the slope of a deep foundation pit to solve the problems existing in the related technologies. The technical solution is as follows:
[0007] In a first aspect, an embodiment of the present application provides a method for monitoring soil displacement of a deep foundation pit slope, comprising:
[0008] Acquire a real-time high-definition image covering a target deep foundation pit slope and environmental sensing data of the target deep foundation pit slope;
[0009] Preprocessing the real-time high-definition image to obtain a processed real-time high-definition image;
[0010] Calling the trained improved YOLO model to perform recognition analysis on the processed real-time high-definition image to obtain visual information of the target deep foundation pit slope, wherein the visual information includes displacement recognition data;
[0011] The visual information and the environmental sensor data are fused and calculated to obtain the final displacement data of the target deep foundation pit slope.
[0012] In one embodiment, obtaining a real-time high-definition image covering a target deep foundation pit slope and environmental sensing data of the target deep foundation pit slope includes:
[0013] Using a high-resolution camera to regularly photograph the target deep foundation pit slope to obtain the real-time high-definition image, the shooting range of the high-resolution camera covering the target deep foundation pit slope;
[0014] The environmental sensor is used to collect the relevant data of the target deep foundation pit slope to obtain the environmental sensing data.
[0015] In one implementation, preprocessing the real-time high-definition image to obtain a processed real-time high-definition image includes:
[0016] De-noising, adjusting brightness and contrast, cropping and scaling the real-time high-definition image to obtain a real-time high-definition image after preliminary processing;
[0017] The real-time high-definition image after the preliminary processing is normalized to obtain the processed real-time high-definition image.
[0018] In one embodiment, the trained improved YOLO model is obtained through the following training process:
[0019] Acquire an image sample set, wherein the image sample set includes a plurality of deep foundation pit slope images;
[0020] Preprocessing each deep foundation pit slope image in the image sample set to obtain each processed deep foundation pit slope image;
[0021] Marking each processed deep foundation pit slope image, marking key features in each deep foundation pit slope image, and obtaining a marked image sample set, wherein the key features include soil cracks and soil displacement;
[0022] A preset YOLO model is trained according to the labeled image sample set, and during the training process, a GPU is used to accelerate the training process, and the network structure and loss function of the preset YOLO model are continuously adjusted to obtain the trained improved YOLO model.
[0023] In one embodiment, adjusting the network structure of the preset YOLO model includes: adjusting the anchor box aspect ratio and anchor box size of the preset YOLO model, the number of convolutional layers of the preset YOLO model, the filter size of the preset YOLO model, and the image input size of the preset YOLO model.
[0024] In one embodiment, the adjustment of the loss function of the preset YOLO model includes: adjusting the position loss of the loss function, the size loss of the loss function, the confidence loss of the loss function, and the category probability loss of the loss function.
[0025] In one embodiment, fusing the visual information and the environmental sensor data to obtain the final displacement data of the target deep foundation pit slope includes:
[0026] The displacement recognition data in the visual information and the displacement sensing data in the environmental sensing data are substituted into a specified formula for fusion calculation to obtain the final displacement data.
[0027] In a second aspect, the embodiment of the present application further provides a deep foundation pit slope soil displacement monitoring device, comprising:
[0028] A processing unit, used to obtain a real-time high-definition image covering a target deep foundation pit slope and environmental sensing data of the target deep foundation pit slope; pre-process the real-time high-definition image to obtain a processed real-time high-definition image;
[0029] The monitoring unit is used to call the trained improved YOLO model to perform recognition and analysis on the processed real-time high-definition image to obtain visual information of the target deep foundation pit slope, wherein the visual information includes displacement recognition data; and fuse the visual information with the environmental sensor data to obtain the final displacement data of the target deep foundation pit slope.
[0030] In one embodiment, when the processing unit is used to obtain a real-time high-definition image covering a target deep foundation pit slope and environmental sensor data of the target deep foundation pit slope, it is specifically used to:
[0031] Using a high-resolution camera to regularly photograph the target deep foundation pit slope to obtain the real-time high-definition image, the shooting range of the high-resolution camera covering the target deep foundation pit slope;
[0032] The environmental sensor is used to collect the relevant data of the target deep foundation pit slope to obtain the environmental sensing data.
[0033] In one implementation, when the processing unit is used to pre-process the real-time high-definition image to obtain the processed real-time high-definition image, it is specifically used to:
[0034] De-noising, adjusting brightness and contrast, cropping and scaling the real-time high-definition image to obtain a real-time high-definition image after preliminary processing;
[0035] The real-time high-definition image after the preliminary processing is normalized to obtain the processed real-time high-definition image.
[0036] In one implementation, the trained improved YOLO model is obtained by the monitoring unit through the following training process:
[0037] Acquire an image sample set, wherein the image sample set includes a plurality of deep foundation pit slope images;
[0038] Preprocessing each deep foundation pit slope image in the image sample set to obtain each processed deep foundation pit slope image;
[0039] Marking each processed deep foundation pit slope image, marking key features in each deep foundation pit slope image, and obtaining a marked image sample set, wherein the key features include soil cracks and soil displacement;
[0040] A preset YOLO model is trained according to the labeled image sample set, and during the training process, a GPU is used to accelerate the training process, and the network structure and loss function of the preset YOLO model are continuously adjusted to obtain the trained improved YOLO model.
[0041] In one embodiment, adjusting the network structure of the preset YOLO model includes: adjusting the anchor box aspect ratio and anchor box size of the preset YOLO model, the number of convolutional layers of the preset YOLO model, the filter size of the preset YOLO model, and the image input size of the preset YOLO model.
[0042] In one embodiment, the adjustment of the loss function of the preset YOLO model includes: adjusting the position loss of the loss function, the size loss of the loss function, the confidence loss of the loss function, and the category probability loss of the loss function.
[0043] In one embodiment, when the monitoring unit is used to fuse the visual information and the environmental sensor data to obtain the final displacement data of the target deep foundation pit slope, it is specifically used to:
[0044] The displacement recognition data in the visual information and the displacement sensing data in the environmental sensing data are substituted into a specified formula for fusion calculation to obtain the final displacement data.
[0045] In a third aspect, an embodiment of the present application further provides a computer device, comprising: a memory and a processor, wherein the memory stores instructions, and the instructions are loaded and executed by the processor to implement a method in any one of the above-mentioned aspects, wherein the memory and the processor communicate with each other through an internal connection path.
[0046] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. When the computer program runs on a computer, the method in any one of the above-mentioned embodiments is implemented.
[0047] The advantages or beneficial effects of the above technical solution include at least:
[0048] 1. Improve the level of monitoring automation: By using a well-trained deep learning model (such as the trained improved YOLO model) to automatically analyze the real-time high-definition images of the real-time target deep foundation pit slope, reduce manual intervention and improve the level of automation of the monitoring process.
[0049] 2. Expand monitoring scope and accuracy: Obtain real-time high-definition images covering the target deep foundation pit slope. For example, high-resolution cameras can be used to cover a wider area. Combined with the high-precision recognition function of the trained improved YOLO model, fine-grained monitoring of the entire deep foundation pit area can be achieved.
[0050] 3. Realize real-time data processing and early warning: The trained and improved YOLO model can process image data in real time, promptly identify soil cracks, soil displacement and other safety risks on the slopes of deep foundation pits, and quickly issue early warnings, greatly improving response speed and timeliness.
[0051] 4. Improve safety: Real-time and comprehensive monitoring can promptly detect potential risks, so that corresponding safety measures can be taken to reduce the probability of accidents and protect the safety of workers and equipment.
[0052] 5. Reduce costs: Automated monitoring reduces dependence on manpower. At the same time, rapid identification and response to potential problems reduces the cost of long-term maintenance and emergency treatment.
[0053] 6. Enhance the accuracy and reliability of monitoring data: By combining multi-source data through data fusion technology, the comprehensiveness and accuracy of monitoring data are improved, making engineering management decisions more scientific and effective.
[0054] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present application will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings represent the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in the present application and should not be regarded as limiting the scope of the present application.
[0056] Figure 1 A schematic diagram of a flow chart of a method for monitoring soil displacement of a deep foundation pit slope provided in an embodiment of the present application;
[0057] Figure 2 An example of a field image of a deep foundation pit slope provided in an embodiment of the present application;
[0058] Figure 3 Another example of a field image of a deep foundation pit slope provided in an embodiment of the present application;
[0059] Figure 4 An example of a performance graph during training of an improved YOLO model provided in an embodiment of the present application;
[0060] Figure 5 An example of a performance graph during training of another improved YOLO model provided in an embodiment of the present application;
[0061] Figure 6 An example of a data analysis diagram after training of an improved YOLO model provided in an embodiment of the present application;
[0062] Figure 7 An example diagram of recognition results of an improved YOLO model provided in an embodiment of the present application;
[0063] Figure 8 An example diagram of recognition results of another improved YOLO model provided in an embodiment of the present application;
[0064] Fig. 9 An example diagram of recognition results of another improved YOLO model provided in an embodiment of the present application;
[0065] Fig.10 A structural block diagram of a deep foundation pit slope soil displacement monitoring device provided in an embodiment of the present application;
[0066] Fig.11 A structural block diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0067] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present application. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.
[0068] Figure 1 A flow chart of a method for monitoring soil displacement at a deep foundation pit slope according to an embodiment of the present application is shown.
[0069] like Figure 1 As shown, the method may include the following steps:
[0070] S110, acquiring a real-time high-definition image covering a target deep foundation pit slope and environmental sensing data of the target deep foundation pit slope.
[0071] In one embodiment, a high-resolution camera can be used to regularly shoot the target deep foundation pit slope to obtain the real-time high-definition image, wherein the shooting range of the high-resolution camera covers the target deep foundation pit slope. In this way, the micro cracks and subtle deformations of the target deep foundation pit slope can be captured by the high-resolution camera, which is convenient for the subsequent improvement of the accuracy of the deep learning model (such as the improved YOLO model) in identifying cracks and the ability to distinguish fine cracks. The real-time high-definition image can be a color image in RGB three-channel format.
[0072] As an example, the high-resolution camera may be a camera having at least 20 million pixels, so as to have sufficient resolution to capture tiny cracks and slight deformations of the target deep foundation pit slope.
[0073] In specific implementation, the image acquisition frequency of the high-resolution camera can be adjusted according to the activity level and historical data of the deep foundation pit project, such as shooting once a day or once an hour during periods of frequent activity, which is not limited in the embodiments of the present application.
[0074] In one embodiment, an environmental sensor may be used to collect relevant data of a target deep foundation pit slope to obtain the environmental sensing data, wherein the environmental sensing data may include displacement sensing data of the target deep foundation pit slope, that is, the environmental sensor may include a displacement sensor.
[0075] As an example, the environmental sensor data may also include temperature, humidity, rainfall and other sensor data, which are crucial for analyzing the environmental factors affecting soil cracks and soil displacement of the target deep foundation pit slope. That is, the environmental sensor may also include a temperature sensor, a humidity sensor and a rainfall sensor.
[0076] S120 , pre-processing the real-time high-definition image to obtain a processed real-time high-definition image.
[0077] In one implementation, the real-time high-definition image may be subjected to denoising, brightness and contrast adjustment, cropping and scaling processing to obtain a preliminarily processed real-time high-definition image.
[0078] In a specific implementation, an image denoising algorithm (such as a Gaussian blur algorithm) or a median filter may be used to remove noise points in the real-time high-definition image.
[0079] Exemplarily, when a median filter is used to remove noise in the real-time high-definition image, the denoising process can be as follows: call image editing software or a dedicated image processing tool to open the real-time high-definition image; select a denoising tool, i.e., a median filter, to denoise the real-time high-definition image, observe the denoising effect of the real-time high-definition image, and stop denoising when the key features in the real-time high-definition image (such as soil cracks and soil displacements) are clearly visible and the background noise is effectively suppressed.
[0080] In a specific implementation, the brightness and contrast of the real-time high-definition image can be automatically adjusted using image editing software to ensure the consistency of the real-time high-definition image under different lighting conditions.
[0081] For example, in the image editing software, an option to adjust brightness and contrast is selected. Then, according to the performance of the real-time high-definition image under different lighting conditions, the brightness and contrast sliders are gradually adjusted until the real-time high-definition image achieves an ideal visual effect, and then the brightness and contrast are stopped.
[0082] In specific implementation, the real-time high-definition image can be cropped to the area of interest (i.e., the area containing only the target deep foundation pit slope), which can reduce the consumption of computing resources. Afterwards, the cropped real-time high-definition image can be appropriately scaled to meet the input requirements of the trained improved YOLO model.
[0083] For example, a dedicated image processing tool can be used to select the area of interest of the real-time HD image for cropping, remove unnecessary background parts, and focus on key areas such as the foundation pit slope. Then adjust the size of the cropped real-time HD image according to the input requirements of the trained improved YOLO model. For example, if the trained improved YOLO model requires an input size of 416x416 pixels, the size of the cropped real-time HD image can be scaled to this size.
[0084] It should be noted that the embodiment of the present application does not limit the order of the above-mentioned preprocessing operations, that is, it does not limit the order of the three preprocessing processes of image denoising, brightness and contrast adjustment, and cropping and scaling.
[0085] In one implementation, the processed real-time high-definition image can be obtained by normalizing the preliminarily processed real-time high-definition image.
[0086] Exemplarily, by normalizing the preliminarily processed real-time high-definition image, the pixel value of the preliminarily processed real-time high-definition image can be scaled from [0, 255] to [0, 1].
[0087] In the embodiment of the present application, by executing step S120, the processed real-time high-definition image can be made more suitable for the input requirements of the trained improved YOLO model, thereby facilitating the improvement of the overall effect and accuracy of the trained improved YOLO model in monitoring the target deep foundation pit slope, and contributing to more efficient and reliable foundation pit slope monitoring.
[0088] S130, calling the trained improved YOLO model to perform recognition analysis on the processed real-time high-definition image to obtain visual information of the target deep foundation pit slope.
[0089] In one embodiment, the visual information includes displacement recognition data. In a specific implementation, the visual information may also include information such as a bounding box, a category, and a confidence level.
[0090] As an example, after obtaining the visual information of the target deep foundation pit slope, it can be stored in a database for further analysis and real-time monitoring.
[0091] In one embodiment, the trained improved YOLO model can be obtained through the following training process:
[0092] S131. Obtain an image sample set, where the image sample set includes a plurality of deep foundation pit slope images.
[0093] In a specific implementation, the multiple deep foundation pit slope images may correspond to the same foundation pit slope or to different foundation pit slopes, which is not limited in the embodiment of the present application.
[0094] As an example, the multiple deep foundation pit slope images may be images obtained by historically photographing the target deep foundation pit slope with a high-resolution camera.
[0095] S132, pre-processing each deep foundation pit slope image in the image sample set to obtain each processed deep foundation pit slope image.
[0096] In specific implementation, the implementation process of step S132 may be the same as or similar to the implementation process of the above-mentioned step S120, and will not be repeated here.
[0097] As an example, in step S132, LabelImg (an open source image annotation tool) may be used in combination with other tools to perform image denoising. The specific process may be as follows:
[0098] A. Use LabelImg for image annotation
[0099] a1. Open the image: Open LabelImg and load each deep foundation pit slope image that needs to be processed;
[0100] a2. Draw a bounding box: Use the mouse to draw a bounding box at the location of the key features of each deep foundation pit slope image. Add a label to each bounding box, such as "crack" or "displacement", and save the labeling results. Among them, LabelImg will save the coordinates and category information of the bounding box in an XML file in the same directory as each deep foundation pit slope image.
[0101] B. Image denoising using OpenCV (cross-platform computer vision library)
[0102] b1. Read the image: Use OpenCV to read each deep foundation pit slope image labeled with LabelImg;
[0103] b2. Apply median filtering: Use OpenCV's cv2.medianBlur() function to perform median filtering and denoising. Select an appropriate filter size k, usually 3 or 5;
[0104] b3. Save each processed deep foundation pit slope image: Use OpenCV's cv2.imwrite() function to save each processed deep foundation pit slope image;
[0105] b4. Observation effect: Check each processed deep foundation pit slope image to ensure that key features such as soil cracks and soil displacement are clearly visible, while background noise is effectively suppressed.
[0106] In the embodiment of the present application, by executing step S132, it is possible to ensure that each deep foundation pit slope image has a consistent brightness and contrast level after adjustment, so as to reduce the variability in the training of the improved YOLO model, and ensure that each deep foundation pit slope image after cropping and scaling meets the input size requirements of the improved YOLO model, and by normalizing each deep foundation pit slope image, it is helpful to improve the stability and convergence speed of the training of the improved YOLO model.
[0107] That is, by executing step S132, the quality of each deep foundation pit slope image can be optimized, irrelevant noise and variables can be reduced, and each deep foundation pit slope image can be ensured to be more suitable for the needs of the improved YOLO model, which helps to improve the improved YOLO model's ability to recognize important features, thereby improving the overall effect and accuracy of deep foundation pit monitoring, and helping to achieve more efficient and reliable foundation pit slope monitoring. For example, through preprocessing, the improved YOLO model can improve its recognition accuracy of cracks from the original 70% to more than 85%, greatly improving the improved YOLO model's recognition accuracy of cracks and its ability to distinguish fine cracks.
[0108] S133, marking each processed deep foundation pit slope image, annotating key features in each deep foundation pit slope image, and obtaining a marked image sample set.
[0109] In a specific implementation, the key characteristics include soil cracks and soil displacements.
[0110] In specific implementation, step S133 may include the following process:
[0111] (1) Labeling process
[0112] Open LabelImg: First open the LabelImg tool and load the image set that needs to be labeled, that is, the processed image sample set.
[0113] Select Image: Select one processed deep foundation pit slope image from the file list of the processed image sample set.
[0114] Draw a bounding box: Use the mouse to draw a bounding box (such as a rectangular box) at the location of the key features of each processed deep foundation pit slope image, so that the key features such as soil cracks and soil displacement in each processed deep foundation pit slope image can be selected; each time a bounding box is drawn, a label is added to the bounding box, such as "crack" or "displacement", and the coordinates and category information of the bounding box are saved in an XML file in the same directory as each processed deep foundation pit slope image. These XML files will be converted into the text format required by the improved YOLO model.
[0115] (2) Annotation output format
[0116] The annotation format required by the improved YOLO model usually includes the category index and four parameters of the bounding box (such as center coordinates and width and height). These data need to be converted from XML format to a simple text format, with each line representing an object in the following format:
[0117] <class_id><x_center><y_center> <width> <height>
[0118] Among them, the coordinates and sizes are normalized to the interval [0,1], relative to the width and height of each deep foundation pit slope image after processing.
[0119] In the specific implementation, by using LabelImg to label each processed deep foundation pit slope image, it is easy to generate a labeling format compatible with the improved YOLO model.
[0120] It should be noted that, in the process of preprocessing each deep foundation pit slope image, if LabelImg is used in combination with other tools for image denoising, each deep foundation pit slope image needs to be labeled before denoising, as described above for details.
[0121] S134, training a preset YOLO model based on the labeled image sample set, and during the training process, using a GPU to accelerate the training process, continuously adjusting the network structure and loss function of the preset YOLO model, and obtaining a trained improved YOLO model.
[0122] In a specific implementation, the adjustment of the network structure of the preset YOLO model includes: adjusting the anchor box aspect ratio and anchor box size of the preset YOLO model, the number of convolutional layers of the preset YOLO model, the filter size of the preset YOLO model, and the image input size of the preset YOLO model.
[0123] As an example, the process of adjusting the anchor box aspect ratio and anchor box size of the preset YOLO model can be as follows:
[0124] The anchor box aspect ratio and anchor box size of the preset YOLO model are adjusted and optimized by the following formulas (1) and (2).
[0125] AR=w / h (1)
[0126] S=w×h (2)
[0127] Among them, AR represents the aspect ratio of the anchor box, S represents the size of the anchor box, w represents the width of the anchor box, and h represents the height of the anchor box.
[0128] In the specific implementation, adjusting the anchor frame strategy in the preset YOLO model is a key step to improve the accuracy of soil crack and soil displacement detection. In the present application, by adjusting and optimizing the anchor frame aspect ratio and anchor frame size of the improved YOLO model, the sensitivity of the enhanced loss function to small objects can be optimized, and the shape and size of the anchor frame can be better matched to the actual monitored object, thereby improving the monitoring capability of small-scale deformation or cracks. Such adjustments help to more accurately locate and identify smaller or specially shaped cracks when performing bounding box prediction through the trained improved YOLO model in the future. In other words, the trained improved YOLO model can have accurate target positioning capabilities.
[0129] As an example, the process of adjusting the number of convolutional layers of the preset YOLO model may be as follows: adding an additional convolutional layer to the network structure of the preset YOLO model.
[0130] As an example, the process of adjusting the filter size of the preset YOLO model can be as follows: using smaller filters in the additional convolutional layers.
[0131] In a specific implementation, by adding convolutional layers and adjusting the filter size in the network structure of the preset YOLO model, the improved YOLO model can recognize more subtle textures and changes in the primary processing stage, thereby improving the overall detection performance of the trained improved YOLO model.
[0132] As an example, the process of adjusting the image input size of the preset YOLO model can be as follows:
[0133] The multi-scale training method is used to dynamically adjust the input image size of the preset YOLO model at different scales (k) (e.g., Dimension at scale k = 146 × 1.1 k ), which can improve the real-time processing speed of the improved YOLO model, and at the same time enable the improved YOLO model to adapt to monitoring objects of different sizes, which not only improves the adaptability of the improved YOLO model to changing sizes, but also enhances the application flexibility of the improved YOLO model in actual monitoring scenarios.
[0134] In specific implementation, the loss function of the preset YOLO model consists of several parts, involving position, size, confidence and classification probability. Based on this, in order to further optimize the performance of the model, especially in terms of accuracy and robustness, the adjustment of the loss function of the preset YOLO model includes: the position loss (error) of the loss function, the size loss (error) of the loss function, the confidence loss (error) of the loss function and the category probability loss (error) of the loss function. The following will introduce in detail how to adjust each part to meet the specific needs of deep foundation pit monitoring.
[0135] As an example, the process of adjusting the position loss of the loss function can be as follows:
[0136] The position loss of the loss function is expressed as follows:
[0137]
[0138] Among them, L coord Expressed as the position loss of the loss function, is an indicator function that indicates that the jth bounding box in the i-th image is responsible for predicting the object, λ coord Expressed as the weight of the location parameter, (x i ,y i )and are the predicted coordinates and the actual coordinates respectively, and B represents the number of bounding boxes predicted for the i-th image.
[0139] In specific implementation, it is very important to improve the accuracy of position prediction for small-sized targets such as soil cracks and soil displacements. The position loss of this loss function mainly focuses on the accurate prediction of the center position of the bounding box, and the weight of the position loss is usually large because accurate positioning is the key to object detection. Based on this, in the embodiment of the present application, by adjusting λ coord To adjust the weight of the center coordinates, the improved YOLO model can learn the crack location more accurately, so that the improved YOLO model can pay more attention to these details.
[0140] As an example, the process of adjusting the size loss of the loss function can be as follows:
[0141] The size loss of the loss function is expressed as follows:
[0142]
[0143] Among them, L size Expressed as the size loss of the loss function, λ size用 To adjust the weights of width and height prediction, (w i ,h i ) represents the predicted width and height of the anchor box, Represented as the actual width and height of the anchor box.
[0144] In practice, cracks are usually long and thin, and reducing the size error is particularly important for cracks. The ability to accurately predict the width and height of the bounding box is very important for accurately assessing the actual size of the object, especially when the object size has an important impact on subsequent processing (such as tracking or behavior analysis). Based on this, in this application, by adjusting λ size The weights of the width and height prediction of the anchor box are adjusted to emphasize the sensitivity to small-sized cracks, so that the improved YOLO model can predict the actual size of the crack more accurately.
[0145] As an example, the process of adjusting the confidence loss of the loss function can be as follows:
[0146] The confidence loss needs to consider both the target and non-target situations. Based on this, the confidence loss of the loss function can be expressed as follows:
[0147]
[0148] Among them, L conf Expressed as the confidence loss of the loss function, λ noobj Expressed as the weight of the confidence parameter, C i is expressed as the confidence of the prediction, is the actual confidence level, is an indicator function used to indicate that the i-th bounding box in the i-th image does not contain an object.
[0149] In specific implementation, the confidence level indicates whether the predicted bounding box contains the target and the prediction accuracy of the bounding box. Based on this, in this application, by adjusting λ noobj To measure the confidence of whether the bounding box contains the target and the prediction quality of the bounding box, the confidence loss can help the improved YOLO model distinguish the bounding box containing the target and the background, so that the improved YOLO model can improve the prediction accuracy of the bounding box.
[0150] As an example, the process of adjusting the class probability loss of the loss function can be as follows:
[0151] The following formula (6) is used to represent the category probability loss of the loss function.
[0152]
[0153] Among them, L class Expressed as the category probability loss of the loss function, p ic Expressed as the predicted class probability, Expressed as the actual class probability, λ class Expressed as the weight of the class probability.
[0154] In specific implementation, it may not be the main concern for specific applications, because there may be only one category in crack monitoring, but it is still necessary to ensure the accuracy of category prediction. When the improved YOLO model needs to classify objects within the bounding box, the category probability error is particularly important. Based on this, in this application, by adjusting λ class To improve the category probability, the confidence loss can ensure that the improved YOLO model can correctly identify and classify the detected objects, so that the improved YOLO model can improve the detection accuracy of the category of the bounding box.
[0155] In summary, in the specific implementation, the loss function of the improved YOLO model is the result of combining the position loss, size loss, confidence loss and category probability loss. This merging process involves weighted summation of various types of errors to form a comprehensive indicator that reflects the performance of the improved YOLO model in various aspects. This is done to simultaneously optimize the performance of the improved YOLO model in positioning accuracy, size prediction, confidence accuracy and classification effect. It can also be understood that the improved YOLO model divides the input image into an S×S grid, and each grid cell predicts B bounding boxes, each of which includes the position (center coordinates (x, y), width w, height h), confidence, and category probability distribution.
[0156] Among them, the total loss function of the improved YOLO model can usually be expressed as follows:
[0157] L=L coord +L size +L conf +L class (7)
[0158] In practical applications, the weights of these losses may be adjusted according to the needs of specific tasks. For example, in some applications with high-precision positioning requirements, λ may be increased. coord Similarly, if classification is not a primary concern, one might decrease λ class In this way, the total loss function can comprehensively reflect the performance of the improved YOLO model in multiple important aspects, thereby guiding the comprehensive optimization of the performance of the improved YOLO model during the training process. That is, the optimization of the loss function directly drives the improved YOLO model towards the expected performance goal.
[0159] In this application, the preset YOLO model is improved by continuously adjusting the network structure and loss function of the preset YOLO model. For example, after the comprehensive improvement in the above step S134, the improved YOLO model is not only more suitable for deep foundation pit slope monitoring tasks in theory, but also can provide higher accuracy and reliability in practical applications. The adjustment of these network structures and the optimization of the loss function work together to significantly improve the adaptability and detection efficiency of the improved YOLO model to complex monitoring scenarios.
[0160] In a specific implementation, step S134 can be performed online at leisure time, that is, the preset YOLO model can be trained online.
[0161] For example, the preset YOLO model can be trained online through the following steps:
[0162] (1) Data preparation:
[0163] (1.1) Prepare a set of labeled image samples;
[0164] (1.2) Data loader implementation: Use Python with a deep learning framework (such as PyTorch or TensorFlow) to write a data loading script that can automatically read image files and annotation data from a directory and convert them into a format suitable for the preset YOLO model input;
[0165] (1.3) Data augmentation: Implement data augmentation techniques in the data loader, such as random rotation, scaling, color jitter, etc., to enhance the adaptability and generalization ability of the preset YOLO model to various environmental changes.
[0166] (2) Configure training parameters to optimize the training process and model performance:
[0167] (2.1) Batch Size Configuration: Choose 16 to 32, taking into account GPU memory limitations and model complexity to ensure efficient use of computing resources;
[0168] (2.2) Learning Rate configuration: Initially set to 0.001, and adopt a learning rate decay strategy, such as reducing the learning rate by 10% every 20 training rounds, to help the model adjust weights more finely in the later stages of training;
[0169] (2.3) Training Epochs configuration: set to 50 to 100 epochs, adjusted according to the performance of the model on the validation set to avoid overfitting.
[0170] (3) GPU acceleration
[0171] (3.1) Configure the GPU environment: Make sure all necessary GPU drivers and support libraries are installed and correctly configured;
[0172] (3.2) Specify GPU resources: Explicitly specify the GPU device to be used in the training script to optimize resource allocation;
[0173] (3.3) Parallel processing: If conditions permit, use multiple GPUs for data parallel processing to further improve training efficiency.
[0174] In this way, the powerful computing power of the GPU can be used to accelerate the training process of the preset YOLO model.
[0175] (4) Training execution
[0176] (4.1) Forward propagation: The preset YOLO model accepts a batch of data, performs a series of calculations, including convolution, activation, pooling, etc., and finally outputs the prediction result;
[0177] (4.2) Loss calculation: According to the preset YOLO model output and the true label, the total loss is calculated, including position loss, size loss, confidence loss and category probability loss.
[0178] (4.3) Back propagation: Using the calculated total loss, the gradient of each parameter in the network is automatically calculated through the chain rule.
[0179] (4.4) Parameter update: Use the Adam optimizer to update the network parameters according to the gradient to minimize the total loss.
[0180] In this way, the model parameters of the preset YOLO model can be optimized through continuous iteration during the actual training process.
[0181] (5) Model verification:
[0182] (5.1) Setting of cross-validation method
[0183] Cross-validation is a statistical analysis method that evaluates the ability of a model to generalize to independent data sets and can effectively avoid model overfitting.
[0184] (5.1.1) Dividing the Dataset
[0185] Dataset partitioning: Randomly divide the labeled image sample set into K mutually exclusive subsets (usually 5 or 10). Each subset should maintain the consistency of data distribution as much as possible, that is, the data type and proportion in each subset should be similar to the entire dataset.
[0186] (5.1.2) Perform K training and validation
[0187] K iterations: In each iteration, one subset is selected as the validation set, and the remaining K-1 subsets are combined as the training set. In this way, each subset will be used as a validation set exactly once and as a training set K-1 times;
[0188] Training and Validation: In each iteration, the model is trained using K-1 subsets of the data, and then the performance of the model is tested on the held-out validation set.
[0189] (5.2) Evaluate the generalization ability and accuracy of the model
[0190] (5.2.1) Determine the model performance indicators:
[0191] Accuracy: measures the proportion of correct labels predicted by the improved YOLO model;
[0192] Precision: measures the proportion of samples predicted as positive by the improved YOLO model that are actually positive;
[0193] Recall: measures the proportion of positive samples correctly identified by the improved YOLO model to all positive samples;
[0194] F1 score: The harmonic mean of precision and recall, an important indicator for evaluating the accuracy of the improved YOLO model.
[0195] (5.2.2) Performance analysis
[0196] Record and analyze: In each validation, record the above performance indicators and analyze the performance differences of the improved YOLO model on different validation sets to evaluate the generalization ability of the improved YOLO model;
[0197] Average performance: Calculate the average of the improved YOLO model performance metrics across all iterations to provide a comprehensive performance evaluation.
[0198] (5.3) Adjust model parameters
[0199] Parameter adjustment: According to the results of cross-validation, especially the mean and standard deviation of those performance indicators, adjust the key parameters of the improved YOLO model (such as learning rate, batch size, etc.), or modify the network architecture (such as increasing or decreasing the number of layers, adjusting the filter size);
[0200] Iterative optimization: Based on the adjusted parameters, a new round of cross-validation is performed to verify the effect of the parameter adjustment;
[0201] In this application, by performing the above-mentioned model training and verification steps, it not only helps to optimize the performance of the improved YOLO model, but also lays a good foundation for the actual deployment and use of the improved YOLO model.
[0202] To further illustrate, the following experimental data verifies that the improved YOLO model can have good performance.
[0203] (I) Using the on-site images of the deep foundation pit slope of a project (e.g. Figure 2 and Figure 3 The on-site image of the deep foundation pit slope shown in FIG. 1 is used as the input image of the improved YOLO model. The improved YOLO model is then trained using the above online training method. During the training period, the following images can be obtained: Figure 4 and Figure 5 Performance during training is shown in Figure 2. Figure 4 and Figure 5 The performance plots during training shown here allow you to check the effect of data preprocessing, label accuracy, and model performance.
[0204] (ii) After completing the training, we can get Figure 6 The data analysis diagram shown below. Figure 6 The training and validation accuracy metrics in are explained.
[0205] Precision refers to the proportion of samples predicted as positive by the improved YOLO model to actually be positive. The calculation formula is: Precision = TP / (TP+FP), where TP is a true positive and FP is a false positive. Figure 6 The performance in the figure is: there is a slight fluctuation in the early stage of training, but it remains at a high level overall, indicating that the improved YOLO model has a strong prediction ability for the positive class.
[0206] Recall refers to the proportion of samples that are actually positive that are correctly predicted as positive by the improved YOLO model. The calculation formula is: Recall = TP / (TP+FN), where FN is a false negative example. Figure 6 The performance in is: Similar to the accuracy, there are some fluctuations in the early stage of training, but it remains at a high level overall, indicating that the improved YOLO model has a higher recognition rate for positive samples.
[0207] mAP50 refers to the average precision of all categories when the IOU threshold is 0.5. The higher the mAP value, the better the overall performance of the improved YOLO model. Figure 6 The performance in the test is: it has remained close to full score since the beginning of training, indicating that the overall detection effect of the improved YOLO model is very good.
[0208] mAP50-95 refers to the average accuracy of all categories under different IOU thresholds (such as from 0.5 to 0.95, with a step size of 0.05). This indicator is more rigorous and comprehensively measures the detection ability of the improved YOLO model. Figure 6 The performance in the figure is also maintained at a high level during the training process, indicating that the improved YOLO model has good detection effect at different IOU thresholds.
[0209] from Figure 6 From the data performance, the improved YOLO model maintains a high level of accuracy, recall, mAP50 and mAP50-95 as the number of training rounds increases, indicating that the improved YOLO model performs well on both training and validation data. For example, in terms of fluctuations, although the accuracy and recall fluctuate somewhat in the early stages of training, this is normal, usually because the improved YOLO model is still learning and adjusting parameters in the initial stage. In terms of stability, in general, the various indicators of the improved YOLO model tend to be stable during the training process, indicating that the improved YOLO model is well trained and has good performance.
[0210] Right now Figure 6 The changes in the accuracy indicators of the improved YOLO model in different training rounds are intuitively demonstrated, providing strong support for the performance evaluation and optimization of the improved YOLO model.
[0211] (III) Finally, the crack image recognition test is completed by combining the on-site image of the deep foundation pit slope, and the following results can be obtained: Figure 7-Figure 9 The following is an example of the recognition results shown in Figure 1. Figure 7-Figure 9 The middle refers to the crack, and the number behind it represents the corresponding recognition probability.
[0212] From the above description, it can be seen that the above experimental data well verifies that the improved YOLO model can have good performance.
[0213] S140, fusing the visual information and the environmental sensor data to obtain the final displacement data of the target deep foundation pit slope.
[0214] In one implementation, the displacement recognition data in the visual information and the displacement sensing data in the environmental sensing data may be substituted into a specified formula for fusion calculation to obtain the final displacement data.
[0215] As an example, the specified formula can be shown as the following formula (8).
[0216] d=αdv+(1-α)ds (8)
[0217] Among them, dv is the displacement corresponding to the displacement recognition data, ds is the displacement corresponding to the displacement sensing data, and α is a weight coefficient between 0 and 1, which is used to adjust the influence of the displacement recognition data and the displacement sensing data in the final output. In specific applications, the optimal value of α can be determined based on experiments or historical data analysis to ensure the reasonable integration of the two data sources of displacement recognition data and displacement sensing data.
[0218] In an embodiment of the present application, by executing step S140, the visual information recognized by the trained improved YOLO model can be fused with the environmental sensor data, thereby enhancing the accuracy and robustness of detecting the displacement of soil on the deep foundation pit slope, providing more comprehensive monitoring information, and improving the accuracy and reliability of the monitoring data of the deep foundation pit slope.
[0219] In summary, in the deep foundation pit slope soil displacement monitoring method of the present application, by executing steps S110 to S140, the following beneficial effects can be achieved:
[0220] 1. Improve the level of monitoring automation: By using a well-trained deep learning model (such as the trained improved YOLO model) to automatically analyze the real-time high-definition images of the real-time target deep foundation pit slope, reduce manual intervention and improve the level of automation of the monitoring process.
[0221] 2. Expand monitoring scope and accuracy: Obtain real-time high-definition images covering the target deep foundation pit slope. For example, high-resolution cameras can be used to cover a wider area. Combined with the high-precision recognition function of the trained improved YOLO model, fine-grained monitoring of the entire deep foundation pit area can be achieved.
[0222] 3. Realize real-time data processing and early warning: The trained and improved YOLO model can process image data in real time, promptly identify soil cracks, soil displacement and other safety risks on the slopes of deep foundation pits, and quickly issue early warnings, greatly improving response speed and timeliness.
[0223] 4. Improve safety: Real-time and comprehensive monitoring can promptly detect potential risks, so that corresponding safety measures can be taken to reduce the probability of accidents and protect the safety of workers and equipment.
[0224] 5. Reduce costs: Automated monitoring reduces dependence on manpower. At the same time, rapid identification and response to potential problems reduces the cost of long-term maintenance and emergency treatment.
[0225] 6. Enhance the accuracy and reliability of monitoring data: By combining multi-source data through data fusion technology, the comprehensiveness and accuracy of monitoring data are improved, making engineering management decisions more scientific and effective.
[0226] Fig.10 FIG. 2 shows a structural block diagram of a deep foundation pit slope soil displacement monitoring device according to an embodiment of the present application. Fig.10 As shown, the device may include:
[0227] The processing unit 210 is used to obtain a real-time high-definition image covering a target deep foundation pit slope and environmental sensing data of the target deep foundation pit slope; pre-process the real-time high-definition image to obtain a processed real-time high-definition image;
[0228] The monitoring unit 220 is used to call the trained improved YOLO model to perform recognition and analysis on the processed real-time high-definition image to obtain the visual information of the target deep foundation pit slope, and the visual information includes displacement recognition data; the visual information and the environmental sensor data are fused and calculated to obtain the final displacement data of the target deep foundation pit slope.
[0229] In one embodiment, when the processing unit 210 is used to obtain a real-time high-definition image covering a target deep foundation pit slope and environmental sensor data of the target deep foundation pit slope, it is specifically used to:
[0230] Use a high-resolution camera to regularly shoot the target deep foundation pit slope to obtain real-time high-definition images. The shooting range of the high-resolution camera covers the target deep foundation pit slope;
[0231] Environmental sensors are used to collect relevant data of the target deep foundation pit slope to obtain environmental sensing data.
[0232] In one implementation, when the processing unit 210 is used to pre-process the real-time high-definition image to obtain the processed real-time high-definition image, it is specifically used to:
[0233] De-noising, brightness and contrast adjustment, cropping and scaling are performed on the real-time HD image to obtain a real-time HD image after preliminary processing;
[0234] The real-time high-definition image after preliminary processing is normalized to obtain a processed real-time high-definition image.
[0235] In one embodiment, the trained improved YOLO model is obtained by the monitoring unit 220 through the following training process:
[0236] Acquire an image sample set, the image sample set including a plurality of deep foundation pit slope images;
[0237] Preprocessing each deep foundation pit slope image in the image sample set to obtain each processed deep foundation pit slope image;
[0238] Each processed deep foundation pit slope image is marked, and the key features in each deep foundation pit slope image are annotated to obtain a marked image sample set, wherein the key features include soil cracks and soil displacement;
[0239] The preset YOLO model is trained based on the labeled image sample set. During the training process, the GPU is used to accelerate the training process, and the network structure and loss function of the preset YOLO model are continuously adjusted to obtain a trained and improved YOLO model.
[0240] In one embodiment, adjusting the network structure of the preset YOLO model includes: adjusting the anchor box aspect ratio and anchor box size of the preset YOLO model, the number of convolutional layers of the preset YOLO model, the filter size of the preset YOLO model, and the image input size of the preset YOLO model.
[0241] In one embodiment, adjusting the loss function of the preset YOLO model includes: adjusting the position loss of the loss function, the size loss of the loss function, the confidence loss of the loss function, and the category probability loss of the loss function.
[0242] In one embodiment, the monitoring unit 220 is used to fuse the visual information and the environmental sensor data to obtain the final displacement data of the target deep foundation pit slope, specifically for:
[0243] The displacement recognition data in the visual information and the displacement sensing data in the environmental sensing data are substituted into the specified formula for fusion calculation to obtain the final displacement data.
[0244] The functions of each unit in the deep foundation pit slope soil displacement monitoring device of the embodiment of the present application can be found in the corresponding description in the above method, and will not be repeated here.
[0245] Fig.11 FIG. 2 shows a block diagram of a computer device according to an embodiment of the present application. Fig.11 As shown, the computer device includes: a memory 310 and a processor 320. The memory 310 stores instructions, and the instructions are loaded and executed by the processor 320 to implement the deep foundation pit slope soil displacement monitoring method in the above embodiment. The number of the memory 310 and the processor 320 can be one or more.
[0246] The computer device also includes:
[0247] The communication interface 330 is used to communicate with external devices and perform data exchange transmission.
[0248] If the memory 310, the processor 320 and the communication interface 330 are implemented independently, the memory 310, the processor 320 and the communication interface 330 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.11 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0249] Optionally, in a specific implementation, if the memory 310, the processor 320 and the communication interface 330 are integrated on a chip, the memory 310, the processor 320 and the communication interface 330 can communicate with each other through an internal interface.
[0250] An embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program runs on a computer, the method provided in the embodiment of the present application is implemented.
[0251] An embodiment of the present application also provides a chip, which includes a processor for calling and executing instructions stored in a memory, so that a communication device equipped with the chip executes the method provided by the embodiment of the present application.
[0252] An embodiment of the present application also provides a chip, including: an input interface, an output interface, a processor and a memory, wherein the input interface, the output interface, the processor and the memory are connected via an internal connection path, and the processor is used to execute the code in the memory. When the code is executed, the processor is used to execute the method provided in the embodiment of the application.
[0253] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. It is worth noting that the processor may be a processor supporting the Advanced RISC Machines (ARM) architecture.
[0254] Further, optionally, the above-mentioned memory may include a read-only memory and a random access memory, and may also include a non-volatile random access memory. The memory may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may include a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may include a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available. For example, static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct rambus RAM (DR RAM).
[0255] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
[0256] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.
[0257] In the description of this application, "multiple sheets" means two or more sheets, unless otherwise clearly and specifically defined.
[0258] Any process or method description in the flow chart or otherwise described herein can be understood to represent a module, fragment or portion of a code including one or more executable instructions for implementing the steps of a specific logical function or process. And the scope of the preferred embodiment of the present application includes other implementations, in which the functions may not be performed in the order shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the functions involved.
[0259] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium for use by an instruction execution system, apparatus or device (such as a computer-based system, a system including a processor or other system that can fetch instructions from an instruction execution system, apparatus or device and execute instructions), or used in combination with these instruction execution systems, apparatuses or devices.
[0260] It should be understood that the various parts of the present application can be implemented with hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented with software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above embodiment method can be completed by instructing the relevant hardware through a program, which can be stored in a computer-readable storage medium, and when the program is executed, it includes one of the steps of the method embodiment or a combination thereof.
[0261] In addition, each functional unit in each embodiment of the present application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a disk or an optical disk, etc.
[0262] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.< / height> < / width>
Claims
1. A method for monitoring soil displacement of a deep foundation pit slope, characterized in that: include: Acquire a real-time high-definition image covering a target deep foundation pit slope and environmental sensing data of the target deep foundation pit slope; Preprocessing the real-time high-definition image to obtain a processed real-time high-definition image; Calling the trained improved YOLO model to perform recognition analysis on the processed real-time high-definition image to obtain visual information of the target deep foundation pit slope, wherein the visual information includes displacement recognition data; The visual information and the environmental sensor data are fused and calculated to obtain the final displacement data of the target deep foundation pit slope.
2. The method according to claim 1, characterized in that: Acquiring real-time high-definition images covering a target deep foundation pit slope and environmental sensing data of the target deep foundation pit slope includes: Using a high-resolution camera to regularly photograph the target deep foundation pit slope to obtain the real-time high-definition image, the shooting range of the high-resolution camera covering the target deep foundation pit slope; The environmental sensor is used to collect the relevant data of the target deep foundation pit slope to obtain the environmental sensing data.
3. The method according to claim 1, characterized in that Preprocessing the real-time high-definition image to obtain the processed real-time high-definition image includes: De-noising, adjusting brightness and contrast, cropping and scaling the real-time high-definition image to obtain a real-time high-definition image after preliminary processing; The real-time high-definition image after the preliminary processing is normalized to obtain the processed real-time high-definition image.
4. The method according to claim 1, characterized in that: The trained improved YOLO model is obtained through the following training process: Acquire an image sample set, wherein the image sample set includes a plurality of deep foundation pit slope images; Preprocessing each deep foundation pit slope image in the image sample set to obtain each processed deep foundation pit slope image; Marking each processed deep foundation pit slope image, marking key features in each deep foundation pit slope image, and obtaining a marked image sample set, wherein the key features include soil cracks and soil displacement; A preset YOLO model is trained according to the labeled image sample set, and during the training process, a GPU is used to accelerate the training process, and the network structure and loss function of the preset YOLO model are continuously adjusted to obtain the trained improved YOLO model.
5. The method according to claim 4, characterized in that The adjustment of the network structure of the preset YOLO model includes: adjusting the anchor box aspect ratio and anchor box size of the preset YOLO model, the number of convolutional layers of the preset YOLO model, the filter size of the preset YOLO model, and the image input size of the preset YOLO model.
6. The method according to claim 4, characterized in that The adjustment of the loss function of the preset YOLO model includes: adjusting the position loss of the loss function, the size loss of the loss function, the confidence loss of the loss function, and the category probability loss of the loss function.
7. The method according to any one of claims 1 to 6, characterized in that The visual information and the environmental sensor data are fused and calculated to obtain the final displacement data of the target deep foundation pit slope, including: The displacement recognition data in the visual information and the displacement sensing data in the environmental sensing data are substituted into a specified formula for fusion calculation to obtain the final displacement data.
8. A deep foundation pit slope soil displacement monitoring device, characterized in that: include: A processing unit, used to obtain a real-time high-definition image covering a target deep foundation pit slope and environmental sensing data of the target deep foundation pit slope; Preprocessing the real-time high-definition image to obtain a processed real-time high-definition image; A monitoring unit is used to call the trained improved YOLO model to perform recognition analysis on the processed real-time high-definition image to obtain visual information of the target deep foundation pit slope, wherein the visual information includes displacement recognition data; The visual information and the environmental sensor data are fused and calculated to obtain the final displacement data of the target deep foundation pit slope.
9. A computer device, characterized in that: include: A memory and a processor, wherein the memory stores instructions, and the instructions are loaded and executed by the processor to implement the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed on a computer, the method according to any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Method for identifying concrete cracks based on yolov3 deep learning model
AU2020101011A4
Beidou positioning-based real-time foundation pit slope deformation monitor and method
CN110332890A
Landslide risk inspection method and system based on deep learning
CN113065455A
High slope monitoring and early warning system based on unmanned aerial vehicle data checking
CN113936228A
Foundation pit real-time monitoring method and system based on image collector
CN114708416A
Cited By
Slope high-risk stone identification method based on oblique photography technology
CN120543943A
