Vision-based operator dangerous behavior detection method and system
A technology for operators and detection methods, applied in neural learning methods, image data processing, biological neural network models, etc., can solve problems such as inability to effectively detect dangerous behaviors of operators, and inability to warn of dangerous behaviors, so as to prevent malignant accidents, improve Efficiency, the effect of ensuring personal safety
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0053] Such as figure 1 As shown, this embodiment proposes a method for detecting dangerous behaviors of workers based on vision, which specifically includes the following steps:
[0054] Step S1, collecting real-time video surveillance data of the production area to obtain video streams.
[0055] In this embodiment, binocular cameras installed in every corner of the production plant are used to collect video monitoring data of the production area in real time to obtain video streams.
[0056] Step S2, intercepting monitoring pictures from the video stream at a preset frequency, and performing preprocessing on the monitoring pictures. Specifically include:
[0057] Step S2.1. Cutting the monitoring picture to meet the size requirements of the model for the input picture, and obtaining the cut monitoring picture.
[0058] In this embodiment, the YOLOV5 deep network model is used to detect the human body targets of the workers in the monitoring picture. Therefore, when the mo...
Embodiment 2
[0091] Such as Figure 4 As shown, this embodiment proposes a vision-based dangerous behavior detection system for workers. The functions of each module of the system are the same as the steps of the method in Embodiment 1 and correspond one-to-one. The system specifically includes:
[0092] The video stream acquisition module M1 is used to collect real-time video surveillance data in the production area to obtain video streams.
[0093] The surveillance picture interception and preprocessing module M2 is configured to intercept surveillance pictures from the video stream at a preset frequency, and perform preprocessing on the surveillance pictures.
[0094] The human object detection module M3 is used to detect the human object of the operator in the monitoring picture, and obtain the location information of the operator.
[0095] The human body posture key point positioning module M4 is used to perform human body posture key point positioning on the monitoring pictures afte...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


