Face identification method based on half-face multi-feature fusion

Abstract

The invention discloses a face identification method based on half-face multi-feature fusion. Global features and local features are combined to be applied to a half of the face so that dimensionality of a processed image in face identification can be less. In feature extraction, the multi-feature fusion is used, therefore, the global features and the local features can be integrated when image features are extracted, the extracted features are comprehensive, and the identification rate of the image is greatly improved while identification time is shortened. The problems that in an original traditional method, a complete face is processed through a single feature, therefore, processing dimensionality is large, computation complexity is high, and feature extraction is incomplete are solved. The aim that a half-face image is extracted through multi-features is achieved. The extracted features are improved. The global features are extracted. Local, pivotal and subtle features can also be extracted. Meanwhile, processing time can be shortened, and rapid identification is achieved.

Description

technical field [0001] The invention relates to a face recognition method, which is a method for face feature extraction and recognition in the field of biological feature recognition, specifically a face recognition method based on half-face multi-feature fusion, which belongs to the field of pattern recognition technology. Background technique [0002] biometric identification [1] It is a technology that uses human-specific biological characteristics for identification, and it provides a highly reliable and stable identification method. Among all biometric identification methods, face recognition is currently the most concerned branch. It is a very active research direction in the field of computer vision and pattern recognition, and is widely used in public security, security, justice, government, Financial, commercial, security check, security and other identification systems. In foreign countries, institutions such as the Massachusetts Institute of Technology, the Uni...

AI Technical Summary

Problems solved by technology

It makes the dimensionality of image processing in face recognition less, and in feature extraction, using multi-feature fusion, so that when extracting image features, global features and local features can be integrated, and the extracted features are more comprehensive. At the same time, the recognition rate of the image is greatly improved, and the problem of using a single feature to process a complete face in the original traditional method, the processing dimension is large, the calculation complexity is high, and the feature extraction is incomplete, and the multi-feature extraction is realized. The face image makes the extracted features more perfect. Not only the global features are extracted, but also the local key subtle features can be extracted. At the same time, the processing time is relatively small, and rapid recognition can be realized.

Images