The Research On Ear Detection Technology Based On Greyscale Images

With the recognition technology on biology characteristics developing in recent years, the research on ear recognition has gained more attention after the recognition technology on fingerprint, face, palm, iris, etc. Although,the ear recognition technology hasn't grown to be a complete theory system, and, it's practical application is still in exporatory stage, which is not like such biology recognition technology as fingerprint recognition, face recognition and so on. Because of its advantages such as collecting easilly, the simple portrait and fast handle speed, etc, the ear recognition technology is becomming the research with a vast development foreground.At present, ear recognition technology is realized on some specifical ear picture databases, which are built up with a amount of ear pictures collectted with cameras or digital cameras. Dynamic ear pictures collecting and detecting is very difficult. Generally, a set of ear automatic recognition technology system includes the steps as follows: collectting the ear picture, pretreating the picture, edge detection and partition, withdrawing characteristic, identifying persons'ear. Ear detection technology as the emphasis of this treatise is a part of ear edge detection and partition.This treatise focuses on ear detection technology on greyscale images, which is on the background of identification technology on biology characteristics and ears. It experimentalizes ear detection technology on greyscale images by template matching method and brings forward a better detection method which is based on the normal detection method to reducing time consuming by increasing it's space comsuming.According to the treatise, the experiment has detected ear images in three groups of images with diffdrent scale. The experiment demonstrates that the performance is better. The correct rate of ear detecting is high. The improvement of reducing the time consuming is obvious.