Face recognition method under complicated illumination conditions of underground coal mine

Abstract

The invention discloses a face recognition method under complicated illumination conditions of an underground coal mine. The method mainly comprises an initialization stage, a training stage and a recognition stage, wherein image collection, image storage, image denoising, image enhancement and feature description are included at the initialization stage, feature vector dimension reduction and classifier model establishment are included at the training stage, and classified recognition is performed on to-be-recognized faces according to an established classifier model at the recognition stage;and image denoising and image enhancement are realized by the adoption of a fuzzy enhancement algorithm based on wavelet decomposition, an ALBP operator is adopted to perform feature description on aface image after wavelet processing, and a classifier is adopted to construct a face model and establish a face sample database. Through the method, the problem that the recognition rate is sharply lowered due to image shadows, bright and dark areas, dark light and high light under complicated underground illumination conditions can be overcome, and the face attendance recognition accuracy in theunderground coal mine is improved.

Description

technical field [0001] The invention relates to a face recognition method under complex lighting conditions in coal mines, in particular to an adaptive face recognition method using image enhancement and online training through a classifier, belonging to the technical field of image pattern recognition. Background technique [0002] The current general process of face recognition is as follows: the recognition system inputs a face image containing an unidentified identity as a sample to be recognized, and several face images with known identities in the face database as training samples, and outputs the image to be recognized through an algorithm. Similarity of samples to indicate the identity of persons in unidentified face images. The face recognition method mainly includes two parts: feature extraction and similarity calculation. [0003] Face recognition technology is of great significance in applications such as video surveillance, work attendance, and personnel positi...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art, provide a face recognition method in coal mines, and solve the problem that the existing face recognition technology cannot meet the image shadows, bright and dark areas, and dark light caused by complex lighting conditions in the mine. , The sharp reduction in the recognition rate caused by high light

Images