Title :
A new prediction method for edge detection based on human visual feature
Author :
Lin, Mingxiu ; Chen, Shuai
Author_Institution :
Coll. of Inf. Sci. & Eng., Northeastern Univ., Shenyang, China
Abstract :
The edge detection rate of the traditional edge detection operators is relatively high, so that some edge points with special characteristics can´t be separated from the normal state edge points. With the specific machine vision applications, a new prediction method for edge detection based on human visual feature is proposed in this paper, and the normal state edge points can be removed by changing the method´s parameter. After the experimental verification, the laceration edge points of the sausage can be extracted precisely.
Keywords :
computer vision; edge detection; feature extraction; prediction theory; edge detection operators; edge detection rate; human visual feature; machine vision; normal state edge point removal; prediction method; sausage laceration edge points; Convolution; Feature extraction; Humans; Image edge detection; Machine vision; Prediction methods; Visualization; edge detection; machine vision; prediction;
Conference_Titel :
Control and Decision Conference (CCDC), 2012 24th Chinese
Conference_Location :
Taiyuan
Print_ISBN :
978-1-4577-2073-4
DOI :
10.1109/CCDC.2012.6244235
