DocumentCode :
1715050
Title :
On the small ship target detection based on panoramic vision
Author :
Li Su ; Na Zhou ; Congying Xu ; Zhilin Liu
Author_Institution :
Coll. of Autom., Harbin Eng. Univ., Harbin, China
fYear :
2013
Firstpage :
3575
Lastpage :
3579
Abstract :
The small target detection is an important and difficult problem in the image processing field. According the problem of small ship target detection in complex background which is gathered by panoramic vision system, this paper presents a practical method of small ship target detection based on the panoramic visible image. On the basis of the panoramic image preprocessing, this paper designs a detection method based on ellipses fitting, which can detect and extract sea-sky-line. Then, using the brightness characteristic of the target on extracted sea-sky-line, this paper designs a single window threshold method based on accurate sea-sky-line area constraints, which can lock the small ship target. The results show that this method can detect the small ship target in the background of complex sea effectively.
Keywords :
computer vision; edge detection; feature extraction; object detection; ships; brightness characteristic; complex sea background; ellipse fitting; panoramic image preprocessing; panoramic visible image; panoramic vision system; sea-sky-line area constraints; sea-sky-line detection; sea-sky-line extraction; single window threshold method; small ship target detection method; Abstracts; Automation; Educational institutions; Electronic mail; Fitting; Marine vehicles; Object detection; Area Constraints; Ellipses Fitting; Image Preprocessing; Sea-Sky-Line Extraction; Small Target Detection;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control Conference (CCC), 2013 32nd Chinese
Conference_Location :
Xi´an
Type :
conf
Filename :
6640041
Link To Document :
بازگشت