DocumentCode :
231944
Title :
A novel quantum-inspired algorithm for edge detection of sonar image
Author :
Wang Xingmei ; Liu Guangyu ; Li Lin ; Liu Zhipeng
Author_Institution :
Coll. of Comput. Sci. & Technol., Harbin Eng. Univ., Harbin, China
fYear :
2014
fDate :
28-30 July 2014
Firstpage :
4836
Lastpage :
4841
Abstract :
In order to extract the underwater object contours of sonar image accurately, a novel quantum-inspired edge detection algorithm is proposed. This algorithm use parameters of anisotropic second-order distribution characteristics MRF (Markov Random Field, MRF) model to describe the texture feature of original sonar image to smooth noise. Based on the conditions mentioned above, sonar image is represented by quantum bit on quantum theory, structure edge detection operator of sonar image by establishing a quantum superposition relationship between pixels. Evaluation the results of quantum-inspired edge detection by PSNR (Peak Signal to Noise Ratio, PSNR), and then complete the quantum-inspired edge detection of sonar image. The comparison different experiments demonstrate that the proposed algorithm get good smoothing result of original sonar image and underwater object contours can be extracted accurately. And it has better adaptability.
Keywords :
Markov processes; edge detection; feature extraction; image representation; image texture; oceanographic techniques; smoothing methods; sonar imaging; Markov random field; PSNR; anisotropic second-order distribution characteristics MRF model; noise smoothing; peak signal-to-noise ratio; quantum bit; quantum superposition relationship; quantum theory; quantum-inspired edge detection algorithm; sonar image edge detection; sonar image representation; structure edge detection operator; texture feature; underwater object contour extraction; Histograms; Image edge detection; PSNR; Quantum mechanics; Sonar detection; Edge Detection; Peak Signal to Noise Ratio; Quantum-inspired; Sonar image;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control Conference (CCC), 2014 33rd Chinese
Conference_Location :
Nanjing
Type :
conf
DOI :
10.1109/ChiCC.2014.6895759
Filename :
6895759
Link To Document :
بازگشت