A real-time high-precision monitoring method and system for safety equipment in complex construction environments

Abstract

The present application provides a method and system for real-time high-precision monitoring of safety equipment in a complex construction environment, which relates to the field of target detection technology. The method includes: obtaining construction site pictures and preprocessing them; constructing a construction site safety detection model based on amplitude parameters, an adaptive reduction deformation convolution layer, and an adaptive magnification deformation convolution layer; training the construction site safety detection model through the preprocessed construction site pictures; obtaining on-site work pictures of the construction site to be detected, and obtaining feature maps of different scales through the trained construction site safety detection model; sampling feature maps of different scales through the ARoI pooling method; inputting the feature maps of different scales after ARoI pooling into the decoupling head of the construction site safety detection model for identification, thereby realizing real-time monitoring of safety work equipment. Through the improved construction site safety detection model, high-precision real-time detection of safety equipment in complex construction environments can be achieved.

Description

Technical Field

[0001] The present application relates to the field of target detection technology, and in particular to a real-time high-precision monitoring method and system for safety equipment in complex construction environments. Background Art

[0002] With the steady growth of China's national economy and the continued advancement of development strategies in central and western regions, highway construction sites are experiencing rapid growth. However, behind this growth, traditional construction site management methods have exposed a series of problems, such as difficult supervision, complex personnel management, frequent safety accidents, and a lack of effective safety early warning mechanisms. These issues not only lead to poor coordination among supervisory parties, but also slow construction progress, weaken project traceability, and introduce potential safety risks.

[0003] Developed countries such as the United States, Canada, and Japan have long been developing smart construction sites, exploring and developing unique models tailored to their respective geographical and topographical characteristics. Although China started relatively late in developing smart construction sites, in recent years, driven by policies promoting artificial intelligence, digital and intelligent construction has rapidly become a trend in the industry. By integrating advanced technologies and management methods, smart construction sites have significantly improved construction safety management, effectively reduced the incidence of safety accidents, and are gradually becoming an indispensable component of the construction industry. Despite this, the development of smart construction sites still faces challenges. Construction sites are complex and dynamic environments, with overlapping operations and a widespread lack of safety awareness, making them a rife with safety hazards. Currently, construction site supervision still relies on traditional manual inspections, which suffer from inefficiencies, slow response times, and insufficient prevention. Furthermore, challenges remain, such as a shortage of supervisors, heavy workloads, an inability to effectively curb violations, delayed accident handling, and difficulty in obtaining evidence.

[0004] Chinese patent CN117975372A relates to a construction site safety detection system and method based on a combination of YOLOv8 and Transformer encoders. The system model is directly deployed on a cloud platform (computer). While ensuring recognition accuracy, it is unable to take into account computational efficiency. Summary of the Invention

[0005] The purpose of the present invention is to provide a real-time high-precision monitoring method and system for safety equipment in complex construction environments in order to solve the problem that existing construction site safety detection methods cannot take into account both detection accuracy and detection efficiency.

[0006] The above-mentioned purpose of this application is achieved through the following technical solutions:

[0007] S1: Obtain construction site pictures and perform preprocessing;

[0008] S2: Build a construction site safety detection model based on amplitude parameters, adaptive downscaling and upscaling convolutional layers; the construction site safety detection model includes a convolutional neural network, a feature fusion and detection network;

[0009] S3: Use pre-processed construction site images to train a construction site safety detection model;

[0010] S4: Obtain on-site operation pictures of the construction site to be inspected, and obtain feature maps of different scales through the trained construction site safety detection model;

[0011] S5: Sample feature maps of different scales through the ARoI pooling method;

[0012] S6: The feature maps of different scales after ARoI pooling are input into the decoupling head of the construction site safety detection model for recognition, realizing real-time monitoring of safety work equipment.

[0013] Optionally, step S1 includes: the preprocessing includes: mosaic enhancement, hybrid enhancement, random dithering, color dithering and size adjustment.

[0014] Optionally, step S3 includes:

[0015] The convolutional neural network includes: a first convolutional layer, a second convolutional layer, an adaptive shrinking and deforming convolutional layer, a fourth convolutional layer, and an adaptive enlarging and deforming convolutional layer;

[0016] The first convolution layer, the second convolution layer, the adaptive shrinking and deforming convolution layer, the fourth convolution layer, and the adaptive enlarging and deforming convolution layer are connected in sequence.

[0017] Optionally, in the adaptive downscaling deformation convolution layer and the adaptive upscaling deformation convolution layer, the sampling points of the original convolution kernel are augmented using an offset, as follows:

[0018] During the training process of the construction site safety detection model, the coordinate points of the sampling pixels of the adaptive shrinking and enlarging convolution layers are transformed as follows:

[0019] X(P i )=∑w(p xi )*x(p x0 +p n +Δp xi ) (Formula 4)

[0020] Y(P i )=∑w(p yi )*y(p y0 +p n +Δp yi ) (Formula 5)

[0021] Where (X, Y) represents the position coordinates of the sampling point pixel after the change, P i =(p xi , p yi ) is the coordinate point of the sampling point pixel; w represents the weight that needs to be trained during the transformation process; p n is the offset of each sampling point in the convolution kernel relative to the center position, and the offset is {(-1, -1), (-1, 0), ..., (0, 1), (1, 1)}; Δp xi Indicates the adaptive offset in the x direction, Δp yi Indicates the adaptive offset in the y direction; p x0 represents the original x coordinate; p y0 Represents the original y coordinate.

[0022] Optionally, the construction site safety detection model further includes:

[0023] In each convolution process of the construction site safety detection model, an adaptive amplitude parameter is added;

[0024] The specific formula for adding adaptive amplitude parameters to the adaptive shrinking and enlarging convolution layers of the construction site safety detection model is as follows:

[0025] X(P i )=∑w(p xi )*x(p x0 +p n +Δp xi )*Δk (Equation 6)

[0026] Y(P i )=∑w(p yi )*y(p y0 +p n +Δp yi )*Δk (Equation 7)

[0027] Where Δk is the amplitude parameter that needs to be learned.

[0028] Step S3 includes:

[0029] Adaptively enlarge the feature map F output by the deformed convolutional layer in the convolutional neural network P5 , perform maximum pooling operations of different scales to obtain multi-scale feature maps;

[0030] Combine the multi-scale feature map with the feature map F output by the fourth convolutional layer in the convolutional neural network P4 Perform splicing to obtain feature map F1;

[0031] The feature map F1 is output by the adaptive shrinking and deformation convolution layer in the convolutional neural network after one upsampling and C2f layer. P3 Perform fusion to obtain feature map T1;

[0032] After the feature map T1 passes through a convolution layer and is fused with the feature map F1, it passes through a C2f layer to obtain the feature map T2;

[0033] After the feature map T2 is fused with the feature map T2 through a convolution layer, a C2f layer is used to obtain the feature map T3;

[0034] Feature maps T1, T2, and T3 are feature maps of different scales output by the construction site safety detection model;

[0035] When fusing feature maps through the fully connected layer, the amplitude parameter and scaling ratio r are used to adjust the fusion process. The calculation process is as follows:

[0036] s=∑r*w*f(z i ) (Formula 8)

[0037] Where r is the scaling ratio, w is the weight of each operation, f(z i ) is the current feature layer; s represents the output result.

[0038] A real-time, high-precision monitoring system for safety equipment in complex construction environments, comprising: a high-resolution camera, an embedded device, a touch display, and a wireless transmission module;

[0039] The high-resolution camera is connected to the embedded device through a wireless transmission module; the touch display is connected to the embedded development board;

[0040] High-resolution cameras are used to obtain on-site operation images of the construction site to be inspected;

[0041] The embedded device is used to monitor the on-site operation images of the construction site to be inspected in real time through the deployed construction site safety detection model and generate detection results;

[0042] The wireless transmission module is used to transmit image data in real time;

[0043] The touch display is used to visually display the test results.

[0044] A computer-readable storage medium stores instructions. When the instructions are executed, a real-time high-precision monitoring method for safety equipment in a complex construction environment is executed.

[0045] The beneficial effects of the technical solution provided by this application are:

[0046] 1. By using adaptive downscaling and upscaling convolutional layers, we build a construction site safety detection model, improving detection accuracy. This effectively enables real-time monitoring of workers and safety equipment such as helmets, reflective vests, and safety ropes at construction sites, thereby effectively reducing the risk of accidents.

[0047] 2. Adding an amplitude parameter to the model's convolutional process allows the network to weight each input based on contextual information, suppressing the influence of task-irrelevant or noisy information, thereby better understanding the overall context and improving the network's robustness and generalization capabilities. This improvement effectively addresses issues such as blurred backgrounds and occlusions in construction scenes. Furthermore, integrating information across different spatial scales and channels helps the network better understand the structure and associations of objects, thereby improving the accuracy of object recognition and localization, particularly for multi-scale, multi-target detection in construction sites.

[0048] 3. Adopt the ARoI (Adaptive Region of Interest) pooling method. By eliminating quantization, similar offsets are used to determine sampling points, and then the actual candidate bounding boxes are averaged. This approach transforms the entire feature aggregation process into a continuous operation, thereby reducing errors and improving the detection accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The present application will be further described below with reference to the accompanying drawings and embodiments, in which:

[0050] Figure 1 is a step diagram in an embodiment of the present application;

[0051] Figure 2 is a structural diagram of a construction site safety detection model in an embodiment of the present application;

[0052] Figure 3 is a structural diagram of a common convolutional module of a construction site safety detection model in an embodiment of the present application;

[0053] Figure 4 : This is an effect diagram of the ASC convolution kernel and the ACC convolution kernel in the embodiment of the present application;

[0054] Figure 5 : This is an application effect diagram of the convolution kernel for adaptively detecting target size in an embodiment of the present application;

[0055] Figure 6 This is a diagram of the floating point conversion process in an embodiment of the present application;

[0056] Figure 7 This is a process diagram of AroI pooling in an embodiment of the present application;

[0057] Figure 8 It is a system module diagram in an embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to have a clearer understanding of the technical features, purposes and effects of this application, the specific implementation methods of this application are now described in detail with reference to the accompanying drawings.

[0059] The embodiments of the present application provide a method for real-time, high-precision monitoring of safety equipment in a complex construction environment.

[0060] Please refer to Figure 1 , Figure 1 This is a step diagram of a method for real-time, high-precision monitoring of safety equipment in a complex construction environment according to an embodiment of the present application, comprising:

[0061] S1: Obtain construction site pictures and perform preprocessing;

[0062] S2: Build a construction site safety detection model based on amplitude parameters, adaptive downscaling and upscaling convolutional layers; the construction site safety detection model includes a convolutional neural network, a feature fusion and detection network;

[0063] S3: Use pre-processed construction site images to train a construction site safety detection model;

[0064] S4: Obtain on-site operation pictures of the construction site to be inspected, and obtain feature maps of different scales through the trained construction site safety detection model;

[0065] S5: Sample feature maps of different scales through the ARoI pooling method;

[0066] Specifically, after fusion, the feature map will be regressed and predicted. However, when the model regresses to generate candidate boxes, the boundary coordinates are usually quantized to integer point values, which will cause a certain deviation between the generated candidate boxes and the initially regressed positions. This deviation may affect the accuracy of detection or segmentation. Based on the idea of ​​offset in the above improvement, this paper designs an ARoI (Adaptive Region of Interest) pooling method. The general pooling method converts feature maps of any size into feature vectors of fixed size. Two quantizations are used in the implementation process, which will cause errors in the detection of small-scale targets. At the construction site, the relative distances between different targets and the camera are different, and some targets will become very small in the picture. The general pooling method will increase the probability of false detection or missed detection in such cases.

[0067] The ARoI pooling method first divides the target candidate box into equal parts, generally into 2x2 regions. After the division, the corner points of each region are not necessarily on the pixel points. Different from the traditional method of shaping, this method chooses to retain such corner points and float them to achieve the purpose of preserving real information. The floating point process is as follows Figure 6 On this basis, if it is a normal pooling operation, the pixels corresponding to the four points in each box will be averaged. However, ARoI first performs bilinear interpolation on the sampling points in each box, just like the adaptive deformable convolution mentioned above, to obtain pixels closer to the target. It then averages the updated sampling points to obtain a candidate box that is more suitable for the target size. Figure 7 This shows the ARoI processing process. This adaptive pooling method uses bilinear interpolation to ensure that feature points within each region of interest are accurately aligned in the output feature map. This more accurately preserves information within the region of interest, reduces information loss introduced during the pooling process, and helps improve the system's detection accuracy for small targets or targets with fine structures in construction scenarios.

[0068] S6: The feature maps of different scales after ARoI pooling are input into the decoupling head of the construction site safety detection model for recognition, realizing real-time monitoring of safety work equipment.

[0069] Specifically, the structure of the decoupling head is as follows Figure 8 As shown in the figure, it includes four 3×3 convolutional layers and two 1×1 convolutional layers. In order to improve the detection accuracy, the WIOU (Weighted Intersection over Union) loss function is added to the regression branch of the detection head. The main task of the regression head is to calculate the position offset between the predicted box and the true box.

[0070] Specifically, the regression head first calculates the four coordinate offsets between the predicted box and the true box, and then inputs these offsets into the regression head for loss calculation. Finally, the regression head outputs a four-dimensional vector, representing the upper left corner coordinates (x1, y1) and the lower right corner coordinates (x2, y2) of the target box. Next, after a series of convolution operations, the classification head generates a classifier output tensor. The value at each position of the tensor represents the probability distribution of the candidate box belonging to different categories. In order to ensure the accuracy of the final detection result, the maximum value suppression (NMS) method is applied in the last step. Maximum value suppression suppresses the candidate boxes with higher overlap and retains the candidate boxes with the highest confidence, thereby screening out the final detection results. The deep, middle, and shallow prediction results (i.e., feature maps T1, T2, and T3) are processed and used to detect target objects and draw target boxes.

[0071] Step S1 includes: the preprocessing includes: mosaic enhancement, hybrid enhancement, random dithering, color dithering and size adjustment.

[0072] Specifically, the data is expanded online in real time through data augmentation methods, and then the processed image is adjusted to the fixed size required by the model while maintaining the original aspect ratio.

[0073] Step S3 includes:

[0074] The convolutional neural network includes: a first convolutional layer, a second convolutional layer, an adaptive shrinking and deforming convolutional layer, a fourth convolutional layer, and an adaptive enlarging and deforming convolutional layer;

[0075] The first convolution layer, the second convolution layer, the adaptive shrinking and deforming convolution layer, the fourth convolution layer, and the adaptive enlarging and deforming convolution layer are connected in sequence.

[0076] Specifically, such as Figure 3 The standard convolutional module of the construction site safety detection model shown in the figure performs multiple feature extractions on construction site images. Each convolutional layer in the convolutional module uses a convolution kernel with a stride of 2 for downsampling to reduce the size of the feature map and increase the number of channels. Feature extraction produces feature maps at five scales. The convolution operations for the first, second, and fourth convolutional layers are as follows:

[0077] F P1 =Conv 3×3 (Input) (Formula 1)

[0078] F P2 =Conv 3×3 (F P1 ) (Formula 2)

[0079] F P4 =Conv 3×3 (F P3 ) (Formula 3)

[0080] Where Conv represents the convolution operation, and F represents the feature map obtained after the convolution operation;

[0081] In the third convolution, the original convolution kernel is replaced by an adaptive shrinkage convolution kernel (ASC) for convolution, and in the fifth convolution, the ordinary convolution operation is replaced by an adaptive amplification convolution kernel (ACC) for convolution.

[0082] In the adaptive shrinking deformation convolution layer and the adaptive enlarging deformation convolution layer, the sampling points of the original convolution kernel are augmented by using the offset, as follows:

[0083] During the training process of the construction site safety detection model, the coordinate points of the sampling pixels of the adaptive shrinking and enlarging convolution layers are transformed as follows:

[0084] X(P i )=∑w(p xi )*x(p x0 +p n +Δp xi ) (Formula 4)

[0085] Y(P i )=∑w(p yi )*y(p y0 +p n +Δp yi ) (Formula 5)

[0086] Where (X, Y) represents the position coordinates of the sampling point pixel after the change, P i =(p xi , p yi ) is the coordinate point of the sampling point pixel; w represents the weight that needs to be trained during the transformation process; p n is the offset of each sampling point in the convolution kernel relative to the center position, and the offset is {(-1, -1), (-1, 0), ..., (0, 1), (1, 1)}; Δp xi Indicates the adaptive offset in the x direction, Δp yi Indicates the adaptive offset in the y direction; p x0 represents the original x coordinate; p y0 Represents the original y coordinate.

[0087] Specifically, when extracting features, convolutional neural networks generally use regular square convolution kernels such as 1x1, 3x3, and 5x5. This will result in the activation units of the same layer of CNN having the same receptive field size during the convolution process. Traditional methods will limit the location of the receptive field. These two convolution methods add a separate direction vector to each convolution kernel, allowing the convolution kernel to adapt to any shape, achieving the following: Figure 4 The designed convolution kernels for adaptive detection of target sizes (ASC convolution kernel and ACC convolution kernel) can more efficiently extract and abstract the semantic information and visual features of detection objects of different sizes in the image, achieving the following results: Figure 5 These features contain various structural and contextual information of workers and safety work equipment, providing key clues for subsequent target recognition and positioning.

[0088] The construction site safety detection model also includes:

[0089] In each convolution process of the construction site safety detection model, an adaptive amplitude parameter is added;

[0090] The specific formula for adding adaptive amplitude parameters to the adaptive shrinking and enlarging convolution layers of the construction site safety detection model is as follows:

[0091] X(P i )=∑w(p xi )*x(p x0 +p n +Δp xi )*Δk (Equation 6)

[0092] Y(P i )=∑w(p yi )*y(p y0 +p n +Δp yi )*Δk (Equation 7)

[0093] Where Δk is the amplitude parameter that needs to be learned;

[0094] Specifically, construction scenes may be subject to background blur and interference from external objects. To address this issue, in addition to sampling point offsets, this paper adds a weight coefficient as an amplitude parameter to distinguish whether the introduced area is the area of ​​interest. Specifically, an adaptive amplitude parameter is added to each convolution component of feature maps T1, T2, and T3. This enhances the expressiveness and adaptability of features when fusion involves cross-level or cross-scale features, reduces the interference of background errors, and produces more accurate, specific, and diverse feature representations.

[0095] During the convolution process, the addition of the amplitude parameter adds an additional channel to each convolution layer on the same input feature map. This primarily involves the aforementioned spatial resolution offset and the newly added weight channel. Unlike the offset weight parameter w, k is a parameter used to determine whether a sample point is in a region of interest. If the sample point is not of interest, k is 0; otherwise, k is initialized to 0.5 for subsequent training.

[0096] Step S3 includes:

[0097] Adaptively enlarge the feature map F output by the deformed convolutional layer in the convolutional neural network P5 , perform maximum pooling operations of different scales to obtain multi-scale feature maps;

[0098] Combine the multi-scale feature map with the feature map F output by the fourth convolutional layer in the convolutional neural network P4 Perform splicing to obtain feature map F1;

[0099] The feature map F1 is output by the adaptive shrinking and deformation convolution layer in the convolutional neural network after one upsampling and C2f layer. P3 Perform fusion to obtain feature map T1;

[0100] After the feature map T1 passes through a convolution layer and is fused with the feature map F1, it passes through a C2f layer to obtain the feature map T2;

[0101] After the feature map T2 is fused with the feature map T2 through a convolution layer, a C2f layer is used to obtain the feature map T3;

[0102] Feature maps T1, T2, and T3 are feature maps of different scales output by the construction site safety detection model;

[0103] When fusing feature maps through the fully connected layer, the amplitude parameter and scaling ratio r are used to adjust the fusion process. The calculation process is as follows:

[0104] s=∑r*w*f(z i ) (Formula 8)

[0105] Where r is the scaling ratio, w is the weight of each operation, f(z i ) is the current feature layer; s represents the output result.

[0106] Specifically, the C2f layer is a special convolutional layer used to fuse feature maps of different scales to improve the accuracy of object detection. Feature fusion and detection network are key steps in the object detection process. At this stage, features at different levels or scales are integrated and fused to produce a more comprehensive, rich, and diverse feature representation.

[0107] like Figure 8 As shown, a real-time high-precision monitoring system for safety equipment in complex construction environments includes: a high-resolution camera, an embedded device, a touch display, and a wireless transmission module;

[0108] The high-resolution camera is connected to the embedded device through a wireless transmission module; the touch display is connected to the embedded development board;

[0109] High-resolution cameras are used to obtain on-site operation images of the construction site to be inspected;

[0110] The embedded device is used to monitor the on-site operation images of the construction site to be inspected in real time through the deployed construction site safety detection model and generate detection results;

[0111] The wireless transmission module is used to transmit image data in real time;

[0112] The touch display is used to visually display the test results.

[0113] Specifically, complete the construction site safety detection model (such as Figure 2 Once the design (as shown) is complete, it can be deployed on embedded devices for on-site inspections. This device utilizes a high-performance NVIDIA chip as its core computing platform for data analysis and processing. First, the embedded development board will be equipped with a suitable image acquisition device, combined with a high-resolution camera and wireless transmission module to achieve efficient data acquisition and processing. Furthermore, during real-time monitoring, the development board's core chip must possess powerful computing capabilities and low power consumption, making it suitable for use in resource-constrained environments. The high-resolution camera provides high-definition images and a wide-angle field of view, ensuring wide monitoring coverage and clear data. The wireless transmission module enables real-time data upload and processing. Processing large amounts of data generates heat, so to ensure computational efficiency, this project incorporates a dedicated cooling fan for the chip to dissipate heat and maintain a suitable temperature. Finally, the development board incorporates a multi-functional interface, facilitating the installation of a display and control panel. This allows users to view monitoring data, test results, and alarm messages in real time, adjust monitoring parameters, and view historical data via the touchscreen or keyboard, achieving user-friendly human-computer interaction.

[0114] Specifically, after building the object detection model (construction site safety detection model), a suitable dataset needed to be selected for training. Work data from multiple construction scenes on the construction site was collected, and the collected image data underwent preprocessing, including data cleaning, denoising, and format conversion. These steps ensured data quality and consistency, resulting in a dataset covering multiple construction scenes. This dataset was not only rich and diverse, but also highly representative, reflecting various situations encountered in actual construction. After the dataset was constructed, multiple training runs were conducted on the dataset using the previously designed and constructed deep learning model. To achieve optimal results, the training process employed various optimization techniques, such as data augmentation, transfer learning, and hyperparameter tuning, to improve the model's generalization and accuracy. After each training run, the model's performance was evaluated, and model parameters were adjusted and optimized based on the evaluation results. After multiple iterations of training and optimization, an optimal model was obtained. This optimal model can efficiently and accurately identify and analyze various work situations on the construction site, and accurately detect in real time whether construction workers are wearing safety equipment such as hardhats and reflective vests.

[0115] The present application also discloses a computer-readable storage medium storing a plurality of instructions suitable for loading by a processor to execute the above-mentioned method for real-time and high-precision monitoring of safety equipment in a complex construction environment.

[0116] The above are merely exemplary embodiments of the present disclosure and are not intended to limit the scope of the present disclosure. In other words, any equivalent variations and modifications made in accordance with the teachings of the present disclosure are still within the scope of the present disclosure. Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the disclosure and the practical implications thereof.

[0117] This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not described herein. The description and examples are to be considered as exemplary only, and the scope and spirit of the present disclosure are to be defined by the claims.

Claims

1. A real-time high-precision monitoring method for safety equipment in a complex construction environment, characterized in that: The method comprises the following steps: S1: Obtain construction site pictures and perform preprocessing; S2: Build a construction site safety detection model based on amplitude parameters, adaptive downscaling and upscaling convolutional layers; the construction site safety detection model includes a convolutional neural network, a feature fusion and detection network; S3: Use pre-processed construction site images to train a construction site safety detection model; Step S3 includes: The convolutional neural network includes: a first convolutional layer, a second convolutional layer, an adaptive shrinking and deforming convolutional layer, a fourth convolutional layer, and an adaptive enlarging and deforming convolutional layer; The first convolution layer, the second convolution layer, the adaptive shrinking and deforming convolution layer, the fourth convolution layer, and the adaptive enlarging and deforming convolution layer are connected in sequence; In the adaptive shrinking deformation convolution layer and the adaptive enlarging deformation convolution layer, the sampling points of the original convolution kernel are augmented by using the offset, as follows: During the training process of the construction site safety detection model, the coordinate points of the sampling pixels of the adaptive shrinking and enlarging convolution layers are transformed as follows: X(P i ) = Σw(p xi ) * x(p x0 + p n + Δp xi ) (Equation 4) Y(P i ) = Σw(p yi ) * y(p y0 + p n + Δp yi ) (Equation 5) Where (X, Y) represents the position coordinates of the sampling point pixel after the change, P i =(p xi ,p yi ) is the coordinate point of the sampling point pixel; w represents the weight that needs to be trained during the transformation process; p n is the offset of each sampling point in the convolution kernel relative to the center position, and the offset is {(-1,-1), (-1,0),…, (0,1), (1,1)}; Δp xi Indicates the adaptive offset in the x direction, Δp yi Indicates the adaptive offset in the y direction; p x0 represents the original x coordinate; p y0 Represents the original y coordinate; The construction site safety detection model also includes: In each convolution process of the construction site safety detection model, an adaptive amplitude parameter is added; The specific formula for adding adaptive amplitude parameters to the adaptive shrinking and enlarging convolution layers of the construction site safety detection model is as follows: X(P i ) = Σw(p xi ) * x(p x0 + p n + Δp xi ) * Δk (Equation 6) Y(P i ) = Σw(p yi ) * y(p y0 + p n + Δp yi ) * Δk (Equation 7) Where Δk is the amplitude parameter that needs to be learned; Step S3 further includes: Adaptively enlarge the feature map F output by the deformed convolutional layer in the convolutional neural network P5 , perform maximum pooling operations of different scales to obtain multi-scale feature maps; Combine the multi-scale feature map with the feature map F output by the fourth convolutional layer in the convolutional neural network P4 Perform splicing to obtain feature map F1; The feature map F1 is output by the adaptive shrinking and deformation convolution layer in the convolutional neural network after one upsampling and C2f layer. P3 Perform fusion to obtain feature map T1; After the feature map T1 passes through a convolution layer and is fused with the feature map F1, it passes through a C2f layer to obtain the feature map T2; After the feature map T2 is fused with the feature map T2 through a convolution layer, a C2f layer is used to obtain the feature map T3; Feature maps T1, T2, and T3 are feature maps of different scales output by the construction site safety detection model; When fusing feature maps through the fully connected layer, the amplitude parameter and scaling ratio r are used to adjust the fusion process. The calculation process is as follows: s = Σr * w * f(z i ) (Equation 8) Where r is the scaling ratio, w is the weight of each operation, f(z i ) is the current feature layer; s represents the output result; S4: Obtain on-site operation pictures of the construction site to be inspected, and obtain feature maps of different scales through the trained construction site safety detection model; S5: Sample feature maps of different scales through the ARoI pooling method; S6: The feature maps of different scales after ARoI pooling are input into the decoupling head of the construction site safety detection model for recognition, realizing real-time monitoring of safety work equipment.

2. The method for real-time high-precision monitoring of safety equipment in a complex construction environment according to claim 1, characterized in that: Step S1 includes: the preprocessing includes: mosaic enhancement, hybrid enhancement, random dithering, color dithering and size adjustment.

3. A real-time high-precision monitoring system for safety equipment in a complex construction environment, used to implement the real-time high-precision monitoring method for safety equipment in a complex construction environment as described in claims 1-2, characterized in that: The system includes: high-resolution camera, embedded device, touch display, wireless transmission module; The high-resolution camera is connected to the embedded device through a wireless transmission module; the touch display is connected to the embedded development board; High-resolution cameras are used to obtain on-site operation images of the construction site to be inspected; The embedded device is used to monitor the on-site operation images of the construction site to be inspected in real time through the deployed construction site safety detection model and generate detection results; The wireless transmission module is used to transmit image data in real time; The touch display is used to visually display the test results.

4. A computer-readable storage medium, characterized in that The computer-readable storage medium stores instructions, and when the instructions are executed by a computer, the method according to any one of claims 1 to 2 is executed.

Images