Safety helmet identification incremental learning and role determination method based on deep learning

Abstract

The invention discloses a safety helmet identification incremental learning and role judgment method based on deep learning, and the method comprises the steps: carrying out the stream extraction operation of a camera through an RTSP, and carrying out the frame skipping processing of a collected video; recording batch information of the training data added into the data set, and performing fine adjustment on the network after new training data is added to realize incremental learning; determining learning rules according to batches, and avoiding catastrophic forgetting; and detecting and identifying safety helmets with different colors in the video and the situation of not wearing the safety helmets, and determining personnel roles of personnel wearing the safety helmets through the colors of the safety helmets while judging the unsafe behavior of not wearing the safety helmets. According to the invention, a target detection technology and a safety mechanism of a factory are combined, a risk detection early warning system is completed, a lot of manpower and material resources are saved, a safety protection mechanism of the factory is improved, a factory management system is perfected, and life health and safety of personnel are maintained.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and relates to a monitoring video incremental learning method based on deep learning, identification of the wearing condition of a helmet, distinction of the color of the helmet, judgment of characters and design of an alarm system. Background technique [0002] Wearing hard hats has become mandatory and essential in many factories. The safety helmet can effectively prevent sudden flying objects from hitting the head, prevent the head from being shocked by electric shocks, and prevent hair from being drawn into the machine or exposed to dust. But its importance is easily overlooked by some staff, resulting in safety accidents. [0003] Relying on staff such as security guards to manually check the wearing of helmets and identify unsafe behaviors has great instability and subjectivity, and it is difficult to monitor in real time without dead ends. At the same time, different colors of he...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a safety helmet recognition incremental learning and role determination method based on deep learning, which solves the technical problem of how to improve the factory safety protection mechanism. By providing a method that can obtain camera video in time, and process the video, identify It is a method to find out the wearing status of the safety helmet of the staff, distinguish the role of the staff through the different colors of the safety helmet, detect in time the unsafe behavior of not wearing the safety helmet, and whether it complies with the safety production regulations. In addition, it can also support incremental learning, avoid catastrophic forgetting

Images