Recognition and judgment method of unsafe action of blasting in coal mine based on gmm characterization

Abstract

A method for identifying and judging unsafe actions of blasting in coal mines based on GMM characterization. The specific process is: (1) Preprocessing the human skeleton sequence data, obtaining the human skeleton sequence data through kinect, obtaining hand joint data through data gloves, and analyzing the human skeleton sequence data. Data and hand joint data are preprocessed for denoising and normalization; (2) Human skeleton sequence data representation: ①GMM feature of action, ②Skeleton sequence time series representation, ③GMM feature extraction of action representation; (4) Based on shell Action Recognition with Yaesian Classifiers. The present invention understands an action as a sequential combination of several key gestures, and uses the mixed Gaussian model GMM model to represent the key gestures of the action, which simplifies the processing of the problem and has certain effectiveness. The identification and judgment accurately solve the action time of blasting in coal mines Problems with long sequences, complex actions, and difficult judgments.

Description

technical field [0001] The invention relates to a method for identifying and judging unsafe coal mine blasting actions in an immersive virtual reality environment, and belongs to the technical field of coal mine blasting safety operations. Background technique [0002] Unsafe actions in coal mines are the main cause of coal mine safety accidents, and among them, the gas explosion accidents caused by blasting by blasters are the ones with the highest frequency of occurrence and the largest number of deaths. [0003] Eliminate unsafe actions in the process of shooting, mainly through repeated training of standard shooting actions, discover and correct unsafe actions of trainees, so that they can consciously make safe actions when encountering the same situation in actual work . However, due to the particularity of the action training of blasting operation, it is impossible for the trainees to carry out repeated training or verification in the real gas explosion dangerous envi...

AI Technical Summary

Problems solved by technology

[0008] The present invention aims at many deficiencies in the existing unsafe action identification and judgment technology of blasting in coal mines, and considers the interdependence between frames and the interrelationships between nodes, proposes a process action that can effectively identify blasting, and A method for identifying and judging unsafe actions in coal mine blasting based on GMM (Gaussian Mixture Model) representation in an immersive virtual reality environment to determine the unsafe actions

Images