State recognition scheme using feature vector and geometric area ratio techniques

Abstract

The state recognition based on the image processing can identify whether or not the target is in normal state. In this paper, there are three creative works in our scheme. Firstly, the improved threshold segmentation (ITS) method can obtain the optimal parameters of the foreground and the background, and it will be favorable for the feature extraction. Secondly, we construct the geometric area ratio (GAR) feature vector to intensify the patterns to simplify the successive state recognition. Thirdly, a novel state recognition algorithm (NSRA) can correctly classify the states of the unknown patterns. Experiments demonstrate the ITS has a best edge effect than the Wavelet method. The proposed GAR feature vector is effective to reflect the similarity of the samples in same Log operator method. The presented NSRA is suitable for the state recognition of the target in an image. In the other words, the proposed algorithm can recognize effectively and correctly the unknown patterns.

All Rights Reserved © 2015 Universidad Nacional Autónoma de México, Centro de Ciencias Aplicadas y Desarrollo Tecnológico. This is an open access item distributed under the Creative Commons CC License BY-NC-ND 4.0.