Safety detection method, device and equipment for building construction, storage medium and product

Abstract

The invention discloses a safety detection method and device for building construction, equipment, a storage medium and a product, relates to the technical field of safety detection, and discloses a safety detection method for building construction, which comprises the following steps: acquiring a video image in a target detection area; performing feature extraction on the video image to obtain current behavior information and building state information of each constructor; and based on the current behavior information and the building state information, performing safety detection through a preset safety detection model to obtain a safety detection result. According to the method and the device, the safety detection result is obtained by performing safety analysis on the current behavior information and the building state information of each constructor in the current target detection area through the pre-trained safety detection model, and subjective judgment of safety risks by detection personnel is not needed, so that the accuracy of safety detection of building construction is improved.

Description

Technical Field

[0001] The present application relates to the field of safety detection technology, and in particular to safety detection methods, devices, equipment, storage media and products for construction. Background Art

[0002] During the construction process, safety inspection is a key link to ensure construction safety and project quality. The construction area has extremely high safety requirements for construction workers due to its special working environment and potential high risks. Therefore, how to conduct safety inspections in the construction area is a current concern.

[0003] In the related art, safety inspection of construction usually relies on manual inspection. However, manual inspection relies on the experience and judgment of the inspectors and is easily affected by subjective factors, resulting in low accuracy of safety inspection of construction.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a safety detection method for construction, aiming to solve the technical problem of low accuracy of safety detection of construction.

[0006] To achieve the above objectives, the present application proposes a construction safety detection method, the construction safety detection method comprising:

[0007] Acquire video images within the target detection area;

[0008] Extracting features from the video image to obtain current behavior information of each construction worker and building status information;

[0009] Based on the current behavior information and the building status information, a safety detection is performed through a preset safety detection model to obtain a safety detection result.

[0010] Optionally, the step of extracting features from the video image to obtain construction worker behavior information and building status information includes:

[0011] Obtaining historical behavior information of each construction worker in the video image;

[0012] Determine the historical behavior weight of each construction worker based on a preset risk weight mapping library and the historical behavior information;

[0013] Based on the current behavior information, the building status information and the historical behavior weight, a safety detection is performed through a preset safety detection model to obtain a safety detection result.

[0014] Optionally, the step of determining the historical behavior weight of each construction worker based on a preset risk weight mapping library and the historical behavior information includes:

[0015] Obtaining time interval information of the historical behavior information;

[0016] Determining a time decay weight based on the time interval information;

[0017] Determining the behavior superposition weight based on a preset risk weight mapping library and the historical behavior information;

[0018] Based on the time decay weight and the behavior superposition weight, the historical behavior weight of each construction worker is calculated.

[0019] Optionally, before the step of acquiring the video image within the target detection area, the method includes:

[0020] Obtaining a sample to be trained and a safety detection result label of the sample to be trained, wherein the sample to be trained includes historical personnel behavior information and historical building status information;

[0021] Based on the samples to be trained and the safety detection result labels, the preset model to be trained is iteratively trained to obtain a safety detection model.

[0022] Optionally, the step of iteratively training a preset model to be trained based on the sample to be trained and the safety detection result label to obtain a safety detection model includes:

[0023] Acquire historical event information of the target detection area;

[0024] Determining an event weight corresponding to the historical event information;

[0025] Based on the samples to be trained, the safety detection result labels and the event weights, the preset model to be trained is iteratively trained to obtain a safety detection model.

[0026] Optionally, the step of iteratively training a preset model to be trained based on the sample to be trained, the safety detection result label and the event weight to obtain a safety detection model includes:

[0027] Based on the samples to be trained and the event weights, security detection is performed using a preset model to be trained to obtain a predicted detection result;

[0028] Calculate the difference between the predicted detection result and the safety detection result label to obtain an error result;

[0029] Based on the error result, determining whether the error result meets an error standard indicated by a preset error threshold range;

[0030] If the error result does not meet the error standard indicated by the preset error threshold range, the model to be trained is updated, and the step of performing safety detection based on the sample to be trained and the event weight through the preset model to be trained to obtain a predicted detection result is returned, and the training is stopped until the error result meets the error standard indicated by the preset error threshold range to obtain a safety detection model.

[0031] In addition, to achieve the above-mentioned purpose, the present application also proposes a safety detection device for construction, the safety detection device for construction comprising:

[0032] An acquisition module is used to acquire video images within the target detection area;

[0033] An extraction module is used to extract features from the video image to obtain current behavior information of each construction worker and building status information;

[0034] The detection module is used to perform safety detection based on the current behavior information and the building status information through a preset safety detection model to obtain a safety detection result.

[0035] In addition, to achieve the above-mentioned purpose, the present application also proposes a safety detection device for construction, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the safety detection method for construction as described above.

[0036] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the construction safety detection method as described above are implemented.

[0037] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the construction safety detection method as described above.

[0038] One or more technical solutions proposed in this application have at least the following technical effects:

[0039] Compared with the related art, the safety inspection of construction usually relies on manual inspection. However, manual inspection relies on the experience and judgment of the inspection personnel, which is easily affected by subjective factors, resulting in low accuracy of safety inspection of construction. In comparison, the present application obtains video images within the target detection area; performs feature extraction on the video images to obtain the current behavior information and building status information of each construction worker; based on the current behavior information and the building status information, performs safety inspection through a preset safety inspection model to obtain safety inspection results. It is understandable that the present application uses a pre-trained safety inspection model to perform safety analysis on the current behavior information and building status information of each construction worker in the current target detection area to obtain safety inspection results, without the need for inspection personnel to make subjective judgments on safety risks, thereby improving the accuracy of safety inspection of construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0042] Figure 1 A schematic diagram of a process flow provided for the first embodiment of the safety detection method for building construction of the present application;

[0043] Figure 2 A schematic diagram of a flow chart provided for the second embodiment of the safety detection method for building construction of the present application;

[0044] Figure 3 This is a schematic diagram of the module structure of the safety detection device for building construction according to an embodiment of the present application;

[0045] Figure 4 Schematic diagram of the equipment structure of the hardware operating environment involved in the construction safety detection method in the embodiment of the present application.

[0046] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0047] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0048] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0049] The main solution of the embodiment of the present application is: to obtain video images within the target detection area; to perform feature extraction on the video images to obtain current behavior information and building status information of each construction worker; based on the current behavior information and the building status information, to perform safety detection through a preset safety detection model to obtain safety detection results.

[0050] In this embodiment, the safety detection device for building construction is used as the execution subject. For the convenience of description, it will be described as "device" hereinafter.

[0051] In the related technologies, the safety inspection of construction usually relies on manual inspection. However, manual inspection relies on the experience and judgment of the inspectors and is easily affected by subjective factors, resulting in low accuracy of the safety inspection of construction.

[0052] The present application provides a solution to achieve safety inspection of building construction and improve the accuracy of safety inspection of building construction.

[0053] From the above embodiments, it can be seen that the present application performs a safety analysis on the current behavior information and building status information of each construction worker in the current target detection area through a pre-trained safety detection model to obtain a safety detection result. There is no need for detection personnel to make subjective judgments on safety risks, thereby improving the accuracy of safety detection of construction.

[0054] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a terminal system, etc. that can realize the above functions. The following takes the safety detection device for construction as an example to illustrate this embodiment and the following embodiments.

[0055] Based on this, the present application embodiment provides a safety detection method for building construction, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the construction safety detection method of the present application.

[0056] In this embodiment, the construction safety detection method includes steps S100 to S300:

[0057] Step S100, acquiring a video image within a target detection area;

[0058] It should be noted that the target detection area refers to a specific area in the construction site that is pre-set and needs to be monitored for safety. This area can be the entire construction site or a specific high-risk area, such as the scaffolding area, tower crane operation area, foundation pit perimeter, etc. Video images refer to real-time video data captured by cameras installed at the construction site, where these cameras can be fixed or mobile, and are used to continuously monitor the activities and status within the target detection area.

[0059] In a specific implementation, the device may obtain video images within the target detection area by capturing video images within the target detection area based on a preset camera, or by receiving video images uploaded by a capture device within the target detection area, which is not specifically limited here.

[0060] Step S200, extracting features from the video image to obtain current behavior information of each construction worker and building status information;

[0061] It should be noted that feature extraction is an image processing and analysis process, the purpose of which is to extract useful information from video images. This usually involves computer vision techniques, such as edge detection, shape recognition, motion analysis, etc., to identify and extract key features in the image; the current behavior information refers to the behavioral characteristics of the construction personnel in the video image. For example, whether they are wearing a safety helmet, whether they are wearing a safety belt, whether they are performing illegal operations (such as not using a safety rope when working at height), etc. By analyzing the video image, the device can automatically identify these behaviors and record them as current behavior information; building status information refers to the status characteristics of the building structure and equipment at the construction site. For example, whether the scaffolding is stable, whether there are cracks in the formwork, whether the tower crane is operating within a safe range, etc. By analyzing the video image, the device can automatically detect these states and record them as building status information.

[0062] In a specific implementation, the device extracts features from the video image to obtain the current behavior information of each construction worker by using a computer vision algorithm, such as a convolutional neural network (CNN) in deep learning, to identify the behavior of the construction worker in the video image. For example, a model is trained to identify whether a helmet is worn, whether a seat belt is fastened, and other behaviors.

[0063] Furthermore, the device extracts features from the video image to obtain the building status information by using image processing technology, such as edge detection and shape recognition, to detect the status of the building structure and equipment. For example, the building status can be evaluated by detecting the deformation of the scaffolding, cracks in the formwork, etc.

[0064] In a specific implementation, the device extracts features from the video image in a manner that includes, but is not limited to: data preprocessing, preprocessing the collected video image, including denoising, enhancement, cropping and other operations to improve the accuracy of feature extraction; feature extraction, extracting key features from the preprocessed video image, such as the behavioral characteristics of construction workers and the status characteristics of the building structure; data fusion, fusing the extracted feature information to generate comprehensive safety detection data.

[0065] For example, the device captures video images of construction workers through the camera and uses a deep learning model to identify that a construction worker is not wearing a safety helmet. This information is recorded as current behavior information. The device captures video images of scaffolding through the camera and uses image processing technology to detect that a part of the scaffolding is slightly deformed. This information is recorded as building status information.

[0066] In a specific implementation, the device extracts features from the video image to obtain construction worker behavior information and building status information, including:

[0067] Obtain historical behavior information of each construction worker in the video image; determine the historical behavior weight of each construction worker based on a preset risk weight mapping library and the historical behavior information; perform safety detection through a preset safety detection model based on the current behavior information, the building status information and the historical behavior weight to obtain a safety detection result.

[0068] It should be noted that the historical behavior information refers to the behavior records of construction workers in the past, including but not limited to whether they wear safety helmets, whether they comply with operating procedures, whether they have violated regulations, etc., where the historical behavior information can be obtained through long-term monitoring and data recording. The device will continuously record the behavior information of each construction worker and store it in the database. These data can be used to analyze the behavior patterns and safety habits of construction workers; the risk weight mapping library is a preset database in which the risk weights of different behaviors are defined. For example, the risk weight of not wearing a safety helmet is higher, while the risk weight of occasionally not fastening a seat belt is lower.

[0069] In the specific implementation, the device calculates the historical behavior weight of each construction worker based on the historical behavior information of the construction workers and the risk weight mapping library. This weight reflects the overall risk level of the construction worker's past behavior. The specific calculation method includes: assuming that a construction worker did not wear a helmet three times in the past month, and the risk weight of not wearing a helmet each time is 0.8, then the historical behavior weight of the construction worker can be calculated as 3×0.8=2.4.

[0070] In a specific implementation, the safety detection model not only considers the current behavior information and building status information, but also combines the historical behavior weights to conduct a comprehensive safety assessment. The safety detection model is a preset algorithm or model, which can be a model based on machine learning or deep learning, which is used to comprehensively analyze the current behavior information, building status information and historical behavior weights to generate safety detection results.

[0071] In the specific implementation, the device continuously collects video images of construction workers, records current behavior information, and obtains historical behavior information of construction workers from the database. Furthermore, the device uses a preset risk weight mapping library to calculate the historical behavior weight of each construction worker. For example, the historical behavior information contains multiple behaviors of not wearing a hard hat. According to the weights in the mapping library, the historical behavior weight is calculated. Finally, the device inputs the current behavior information, building status information, and historical behavior weights into the preset safety detection model. The model comprehensively analyzes this information and generates safety detection results. For example, the model detects that a construction worker is not currently wearing a hard hat. Combined with his high historical behavior weight, it is determined to be a high-risk behavior and an emergency alarm is issued.

[0072] For example, construction worker A did not wear a helmet three times and did not fasten his seat belt twice in the past month. The risk weight for not wearing a helmet is 0.8, and the risk weight for not fastening a seat belt is 0.5. The weight for not wearing a helmet is 3×0.8=2.4, and the weight for not fastening a seat belt is 2×0.5=1.0, so the total historical behavior weight is 2.4+1.0=3.4. Construction worker A is not wearing a helmet at present, and the building status information is that the scaffolding is in normal state. The device combines the current behavior information (not wearing a helmet), the building status information (the scaffolding is normal), and the historical behavior weight (3.4). The model determines that the behavior is medium-to-high risk, issues an emergency alarm, and recommends stopping construction immediately, requiring construction worker A to wear a helmet.

[0073] In a specific implementation, based on a preset risk weight mapping library and the historical behavior information, the step of determining the historical behavior weight of each construction worker includes:

[0074] Obtain the time interval information of the historical behavior information; determine the time decay weight based on the time interval information; determine the behavior superposition weight based on a preset risk weight mapping library and the historical behavior information; calculate the historical behavior weight of each construction worker based on the time decay weight and the behavior superposition weight.

[0075] It should be noted that the time interval information refers to the interval between the time point when each behavior in the historical behavior information occurs and the current time. For example, if a construction worker did not wear a safety helmet a month ago, then the time interval of this behavior is one month; the time decay weight is a weight calculated based on the time interval, which is used to reflect the degree of influence of historical behavior on the current safety risk. It is understandable that as time goes by, the impact of historical behavior on current safety risks will gradually decrease, so it is necessary to introduce time decay weight to adjust the contribution of historical behavior. Specific calculation method: The time decay weight can be calculated by a preset decay function, such as an exponential decay function.

[0076] In the specific implementation, the behavior superposition weight is a weight calculated based on the type and number of historical behaviors, which is used to reflect the overall risk level of historical behaviors. Furthermore, the historical behavior weight is the product of the time decay weight and the behavior superposition weight, which is used to comprehensively consider the time impact and risk level of historical behaviors.

[0077] In specific implementations, the device can use this method to more accurately assess the impact of construction workers' historical behaviors on current safety risks, thereby performing safety management more effectively.

[0078] Step S300, based on the current behavior information and the building status information, a safety detection is performed through a preset safety detection model to obtain a safety detection result.

[0079] In the specific implementation, the device collects current behavior information and building status information in real time through cameras and sensors. For example, the camera captures the behavior of construction workers, and the sensor monitors the stability of the scaffolding. Secondly, the device preprocesses the collected data, including denoising, format conversion and feature extraction, for example, edge detection and feature extraction of images, and filtering of sensor data. Then, the device inputs the preprocessed data into a preset safety detection model, for example, the extracted features are input into a trained CNN model. Next, the model analyzes the input data to identify potential safety hazards. For example, the model detects that a construction worker is not wearing a safety helmet and the scaffolding is slightly deformed. Finally, the model outputs the safety detection results, including the type, location, severity and recommended response measures of the safety hazard. For example, the output result is: "It is detected that the construction worker is not wearing a safety helmet. Location: Scaffolding area. Severity: High. It is recommended to stop construction immediately and require the construction worker to wear a safety helmet."

[0080] For example, the current behavior information includes: construction worker A did not wear a safety helmet, and construction worker B did not use a safety rope when working at height. The building status information includes: the scaffolding is slightly deformed, and the tower crane is operating normally. The device uses a convolutional neural network (CNN) model based on deep learning to identify the wearing of safety helmets and the stability of the scaffolding in the image. The device inputs the current behavior information and building status information into the CNN model. The model analyzes the image and recognizes that construction worker A is not wearing a safety helmet and the scaffolding is slightly deformed. Finally, the model outputs safety detection results including: construction worker A is not wearing a safety helmet, location: scaffolding area, severity: high, it is recommended to stop construction immediately and require construction workers to wear safety helmets; the scaffolding is slightly deformed, location: scaffolding area, severity: medium, it is recommended to inspect and repair.

[0081] Compared with the related art, the safety inspection of construction usually relies on manual inspection. However, manual inspection relies on the experience and judgment of the inspection personnel, which is easily affected by subjective factors, resulting in low accuracy of safety inspection of construction. In comparison, the present application obtains video images within the target detection area; performs feature extraction on the video images to obtain the current behavior information and building status information of each construction worker; based on the current behavior information and the building status information, performs safety inspection through a preset safety inspection model to obtain safety inspection results. It is understandable that the present application uses a pre-trained safety inspection model to perform safety analysis on the current behavior information and building status information of each construction worker in the current target detection area to obtain safety inspection results, without the need for inspection personnel to make subjective judgments on safety risks, thereby improving the accuracy of safety inspection of construction.

[0082] Based on the above first embodiment, the present application also proposes another embodiment, referring to Figure 2 , the safety detection method for building construction includes:

[0083] In a specific implementation, before the step of the device acquiring a video image within the target detection area, the method includes:

[0084] Step A100, obtaining a sample to be trained and a safety detection result label of the sample to be trained, wherein the sample to be trained includes historical personnel behavior information and historical building status information;

[0085] It should be noted that the samples to be trained refer to the data sets used to train the model, which contain historical personnel behavior information and historical building status information. The historical personnel behavior information includes the behavior records of construction personnel in the past period of time, such as whether they wear safety helmets, whether they fasten their safety belts, whether they have violated regulations, etc. The historical building status information includes the status records of building structures and equipment in the past period of time, such as the stability of scaffolding, the integrity of formwork, the operating status of tower cranes, the settlement of foundation pits, etc.

[0086] In the specific implementation, the safety test result label refers to the safety test result corresponding to each training sample, which is usually annotated by experts or obtained through other reliable methods. The label can be binary classification (safe / unsafe) or multi-classification (low risk, medium risk, high risk).

[0087] Step A200, based on the samples to be trained and the safety detection result labels, iteratively train the preset model to be trained to obtain a safety detection model.

[0088] It should be noted that the preset model to be trained is an initial machine learning or deep learning model used to learn the patterns and features in the samples to be trained. This model can be a decision tree, support vector machine (SVM), neural network, etc. Furthermore, iterative training refers to the process of gradually improving the prediction accuracy of the model on the training data by adjusting the parameters of the model multiple times. Each iteration updates the parameters of the model based on the difference between the model's prediction results and the true labels (loss function). Specifically, the iterative training process includes (1) forward propagation: input the samples to be trained into the model to obtain the prediction results. (2) Calculate the loss: calculate the difference between the prediction results and the true labels (loss function). (3) Back propagation: update the parameters of the model through the back propagation algorithm according to the loss function. (4) Repeat: repeat the above steps until the performance of the model no longer improves or reaches the preset number of iterations.

[0089] In the specific implementation, after iterative training, the model can accurately predict the safety detection results based on the input personnel behavior information and building status information. This trained model is called a safety detection model. Specifically, the safety detection model can be used to monitor the safety status of the construction site in real time and promptly discover and deal with potential safety hazards.

[0090] In a specific implementation, the device iteratively trains a preset model to be trained based on the sample to be trained and the safety detection result label to obtain a safety detection model, including:

[0091] Acquire historical event information of the target detection area; determine event weights corresponding to the historical event information; iteratively train a preset model to be trained based on the samples to be trained, the safety detection result labels, and the event weights to obtain a safety detection model.

[0092] It should be noted that historical event information refers to the records of various safety-related events that have occurred in the target detection area in the past. These events may include but are not limited to: 1. Safety accidents (such as falls, collisions, fires, etc.); 2. Safety hazards (such as scaffold deformation, formwork cracks, equipment failures, etc.); 3. Safety violations (such as not wearing a safety helmet, not wearing a safety belt, illegal operations, etc.).

[0093] In the specific implementation, the event weight is a weight calculated based on the severity and impact of historical events, which is used to reflect the impact of each historical event on the current security risk. Event weight can help the model more accurately assess the importance of historical events. Specific calculation method: Event weight can be determined by preset rules or expert experience. For example, a weight table can be defined to assign weights based on the type and severity of the event.

[0094] In the specific implementation, this application introduces historical event information and its corresponding event weights, so that the safety detection model can more comprehensively consider the impact of historical data, improve the accuracy and reliability of prediction, and thus more effectively manage the safety of the construction site.

[0095] In a specific implementation, the device iteratively trains the preset model to be trained based on the sample to be trained, the safety detection result label and the event weight to obtain the step of the safety detection model, including:

[0096] Based on the samples to be trained and the event weights, a security check is performed using a preset model to be trained to obtain a predicted test result; a difference is calculated between the predicted test result and the security test result label to obtain an error result; based on the error result, a judgment is made as to whether the error result meets the error standard indicated by a preset error threshold range; if the error result does not meet the error standard indicated by the preset error threshold range, the model to be trained is updated, and the step of performing a security check based on the samples to be trained and the event weights and obtaining a predicted test result using a preset model to be trained is returned to, and training is stopped until the error result meets the error standard indicated by the preset error threshold range to obtain a security test model.

[0097] In a specific implementation, the device inputs the samples to be trained and the event weights into the preset model to be trained, and the model performs security detection based on these inputs and outputs the predicted detection results, which refer to the model's prediction of the security status of each sample.

[0098] In the specific implementation, the device calculates the difference between the predicted detection result and the true label, and usually uses a loss function (such as mean square error, cross entropy loss, etc.) to quantify this difference. The error result refers to the result of the difference calculation, which indicates the gap between the accuracy of the model prediction and the true label. That is, it verifies whether the results obtained by the model in training are consistent with the known results, and calculates the difference between the results to obtain the error result.

[0099] It should be noted that the device determines whether the error result meets the error standard indicated by the preset error threshold range based on the error result. Specifically, since there is an error between the result after model training and the actual result, the error result is allowed to be within the preset error threshold range, so as to further determine whether the error result meets the error standard indicated by the preset error threshold range.

[0100] It is understandable that if the error result does not meet the error standard indicated by the preset error threshold range, it means that the model has too large an error in this training, and the device needs to update the parameters of the model, which is usually achieved through a back-propagation algorithm and an optimizer (such as gradient descent, etc.). After updating the model parameters, the safety test is performed again, and the difference between the predicted test result and the true label is calculated again until the error result meets the preset error threshold range. That is, iterative training is performed until the error result meets the error standard indicated by the preset error threshold range, and the training is stopped to obtain a safety detection model. Through this iterative training and optimization process, the present application can ensure that the safety detection model gradually improves the prediction accuracy on the training data, and finally reaches the predetermined performance standard, so as to more effectively manage the safety of the construction site.

[0101] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the safety inspection method for building construction of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0102] The present application also provides a safety detection device for construction, referring to Figure 3 , the safety detection device for building construction comprises:

[0103] An acquisition module 10 is used to acquire a video image within a target detection area;

[0104] An extraction module 20 is used to extract features from the video image to obtain current behavior information of each construction worker and building status information;

[0105] The detection module 30 is used to perform safety detection based on the current behavior information and the building status information through a preset safety detection model to obtain a safety detection result.

[0106] Optionally, the detection module 30 includes:

[0107] A historical behavior information acquisition module, used to acquire historical behavior information of each construction worker in the video image;

[0108] A determination module, used to determine the historical behavior weight of each construction worker based on a preset risk weight mapping library and the historical behavior information;

[0109] The safety detection module is used to perform safety detection based on the current behavior information, the building status information and the historical behavior weight through a preset safety detection model to obtain a safety detection result.

[0110] Optionally, the determining module includes:

[0111] A time interval information acquisition module, used to acquire the time interval information of the historical behavior information;

[0112] A time decay weight determination module, used to determine the time decay weight based on the time interval information;

[0113] A behavior superposition weight determination module, used to determine the behavior superposition weight based on a preset risk weight mapping library and the historical behavior information;

[0114] The calculation module is used to calculate the historical behavior weight of each construction worker based on the time decay weight and the behavior superposition weight.

[0115] Optionally, the construction safety detection device further comprises:

[0116] A sample acquisition module, used to acquire samples to be trained and safety detection result labels of the samples to be trained, wherein the samples to be trained include historical personnel behavior information and historical building status information;

[0117] The training module is used to iteratively train the preset model to be trained based on the sample to be trained and the safety detection result label to obtain a safety detection model.

[0118] Optionally, the training module includes:

[0119] A historical event information acquisition module, used to acquire historical event information of the target detection area;

[0120] An event weight determination module, used to determine the event weight corresponding to the historical event information;

[0121] The iterative training module is used to iteratively train the preset model to be trained based on the sample to be trained, the safety detection result label and the event weight to obtain a safety detection model.

[0122] Optionally, the iterative training module includes:

[0123] A prediction module, used to perform security detection based on the samples to be trained and the event weights through a preset model to be trained to obtain a prediction detection result;

[0124] A difference calculation module, used to calculate the difference between the predicted detection result and the safety detection result label to obtain an error result;

[0125] A judgment module, configured to judge, based on the error result, whether the error result satisfies an error standard indicated by a preset error threshold range;

[0126] The safety detection model training module is used to update the model to be trained if the error result does not meet the error standard indicated by the preset error threshold range, and return to the step of performing safety detection based on the sample to be trained and the event weight through the preset model to be trained to obtain a predicted detection result, and stop training until the error result meets the error standard indicated by the preset error threshold range to obtain a safety detection model.

[0127] The safety detection device for construction provided by the present application adopts the safety detection method for construction in the above-mentioned embodiment, and can solve the technical problem of safety detection of construction. Compared with the prior art, the beneficial effects of the safety detection device for construction provided by the present application are the same as the beneficial effects of the safety detection method for construction provided by the above-mentioned embodiment, and the other technical features of the safety detection device for construction are the same as the features disclosed in the above-mentioned embodiment method, which will not be described in detail here.

[0128] The present application provides a safety detection device for construction, which includes: at least one processor; and a memory that is communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the safety detection method for construction in the above-mentioned embodiment one.

[0129] Reference below Figure 4, which shows a schematic diagram of the structure of the safety detection equipment for building construction suitable for implementing the embodiment of the present application. The safety detection equipment for building construction in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The safety detection equipment for construction shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0130] like Figure 4 As shown, the safety detection equipment for building construction may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the safety detection equipment for building construction are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the safety detection device of the construction to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows the safety detection device of the construction with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have instead.

[0131] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0132] The safety detection equipment for construction provided by the present application adopts the safety detection method for construction in the above-mentioned embodiment, and can solve the technical problem of safety detection of construction. Compared with the prior art, the beneficial effects of the safety detection equipment for construction provided by the present application are the same as the beneficial effects of the safety detection method for construction provided by the above-mentioned embodiment, and the other technical features of the safety detection equipment for construction are the same as the features disclosed in the method of the previous embodiment, which will not be described in detail here.

[0133] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0134] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of 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.

[0135] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the construction safety detection method in the above-mentioned embodiment.

[0136] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0137] The computer-readable storage medium may be included in the safety detection equipment for building construction; or it may exist independently without being assembled into the safety detection equipment for building construction.

[0138] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the safety detection equipment for building construction, the safety detection equipment for building construction can perform safety detection of building construction.

[0139] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0140] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0141] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0142] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned construction safety detection method, and can solve the technical problems of construction safety detection. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the construction safety detection method provided by the above-mentioned embodiment, and will not be elaborated here.

[0143] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned construction safety detection method when executed by a processor.

[0144] The computer program product provided in this application can solve the technical problem of safety detection of construction. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the safety detection method of construction provided in the above embodiment, which will not be repeated here.

[0145] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A safety detection method for building construction, characterized in that: The method for safety detection of building construction comprises: Acquire video images within the target detection area; Extracting features from the video image to obtain current behavior information of each construction worker and building status information; Based on the current behavior information and the building status information, a safety detection is performed through a preset safety detection model to obtain a safety detection result.

2. The safety detection method for building construction according to claim 1, characterized in that: The step of extracting features from the video image to obtain construction worker behavior information and building status information includes: Obtaining historical behavior information of each construction worker in the video image; Determine the historical behavior weight of each construction worker based on a preset risk weight mapping library and the historical behavior information; Based on the current behavior information, the building status information and the historical behavior weight, a safety detection is performed through a preset safety detection model to obtain a safety detection result.

3. The safety detection method for building construction as claimed in claim 2, characterized in that: The step of determining the historical behavior weight of each construction worker based on the preset risk weight mapping library and the historical behavior information includes: Obtaining time interval information of the historical behavior information; Determining a time decay weight based on the time interval information; Determining the behavior superposition weight based on a preset risk weight mapping library and the historical behavior information; Based on the time decay weight and the behavior superposition weight, the historical behavior weight of each construction worker is calculated.

4. The safety detection method for building construction according to claim 1, characterized in that: Before the step of acquiring the video image within the target detection area, the method includes: Obtaining a sample to be trained and a safety detection result label of the sample to be trained, wherein the sample to be trained includes historical personnel behavior information and historical building status information; Based on the samples to be trained and the safety detection result labels, the preset model to be trained is iteratively trained to obtain a safety detection model.

5. The safety detection method for building construction as claimed in claim 4, characterized in that: The step of iteratively training the preset model to be trained based on the sample to be trained and the safety detection result label to obtain the safety detection model includes: Acquire historical event information of the target detection area; Determining an event weight corresponding to the historical event information; Based on the samples to be trained, the safety detection result labels and the event weights, the preset model to be trained is iteratively trained to obtain a safety detection model.

6. The safety detection method for building construction as claimed in claim 5, characterized in that: The step of iteratively training the preset model to be trained based on the sample to be trained, the safety detection result label and the event weight to obtain the safety detection model includes: Based on the samples to be trained and the event weights, security detection is performed using a preset model to be trained to obtain a predicted detection result; Calculate the difference between the predicted detection result and the safety detection result label to obtain an error result; Based on the error result, determining whether the error result meets an error standard indicated by a preset error threshold range; If the error result does not meet the error standard indicated by the preset error threshold range, the model to be trained is updated, and the step of performing safety detection based on the sample to be trained and the event weight through the preset model to be trained to obtain a predicted detection result is returned, and the training is stopped until the error result meets the error standard indicated by the preset error threshold range to obtain a safety detection model.

7. A safety detection device for building construction, characterized in that: The device comprises: An acquisition module is used to acquire video images within the target detection area; An extraction module is used to extract features from the video image to obtain current behavior information of each construction worker and building status information; The detection module is used to perform safety detection based on the current behavior information and the building status information through a preset safety detection model to obtain a safety detection result.

8. A safety detection device for building construction, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the construction safety detection method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the construction safety detection method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the construction safety detection method according to any one of claims 1 to 6 are implemented.

Images