Color image reconstruction and identification method and system

Abstract

The invention discloses a color image reconstruction and identification method and system. The method includes: performing color divergence matrix construction by an unlabeled color image training sample, performing two-dimensional principal component learning on a color image, acquiring a projection matrix through a feature decomposition problem to complete feature dimension reduction, inputting data after dimension reduction to a nearest neighbor classifier, embedding a test sample into a projection space for feature extraction, inputting extracted features to the nearest neighbor classifier for similarity measurement, and outputting the category of the test sample to obtain a most accurate color image identification result. Direct dimension reduction of the color image features is realized by optimizing one unsupervised feature problem, time complexity is lowered, the color image feature extraction based identification process is fast, and meanwhile important color information and a topological structure included in image pixels can be kept effectively, besides, the projection direction can be effectively used for color image reconstruction, and a certain number of projection vectors are selected to obtain the clear color image reconstruction result.

Description

technical field [0001] The present invention relates to the technical fields of computer vision and image recognition, and more specifically, relates to a color image reconstruction and recognition method and system. Background technique [0002] With the advent of the information age and the Internet, the amount of data and information to be processed is increasing. In the process of daily communication and surfing the Internet, color images are very common, but there are many types and it is difficult to accurately calibrate the categories, making color image recognition technology It has developed into one of the difficult problems in the field of computer vision and pattern recognition. Color image recognition technology realizes image digitization through computers, and then performs data analysis to obtain color features and automatically classify color images. It is of great significance in the fields of machine vision systems and other fields. social and economic be...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the present invention provides a color image reconstruction and recognition method and system to overcome the problem of increased computational complexity due to the high sample complexity in the prior art, and realize color image feature extraction in the true sense Program

Images