DocumentCode
2268849
Title
Study on Acyclic Lateral Inhibition Network Model and Its Application for Image Processing
Author
Fu, Hongwei ; Li, Dongguang ; Guan, Renliang
Author_Institution
Sch. of Aerosp. Sci. & Eng., Beijing Inst. of Technol., Beijing
Volume
3
fYear
2008
fDate
20-22 Dec. 2008
Firstpage
301
Lastpage
305
Abstract
For overcoming the problems of algorithms in image edge detection, like distortion and shift of objectpsilas edge, easily losing the object detail information and higher demand for object detection in the modern war, this essay analyses the feasibility of lateral inhibition model in object detection, based on lateral inhibition theory, an acyclic lateral inhibition network model (ALINM) with biology vision information processing mechanism was introduced. This model has the merits of rapid calculating, easily real time operation, etc. Besides the correctness of ALINM is confirmed by two input cells, its transfer function is deduced. An algorithm of image edge detection based on this model is established finally, and simulative experiment with different parameters proves that this algorithm can preserve the farthest detail information of objects. It provides a useful method based on biology vision for object detection under difficult imaging conditions.
Keywords
biology computing; edge detection; acyclic lateral inhibition network model; biology vision information processing mechanism; image edge detection; lateral inhibition theory; object detection; transfer function; Algorithm design and analysis; Biological system modeling; Cells (biology); Computational biology; Image analysis; Image edge detection; Image processing; Information analysis; Information processing; Object detection; acyclic lateral inhibition network model; edge detection algorithm; lateral inhibition theory; simulative experiment;
fLanguage
English
Publisher
ieee
Conference_Titel
Intelligent Information Technology Application, 2008. IITA '08. Second International Symposium on
Conference_Location
Shanghai
Print_ISBN
978-0-7695-3497-8
Type
conf
DOI
10.1109/IITA.2008.554
Filename
4740006
Link To Document