DocumentCode :
2245021
Title :
Optimal selection of fractal features for man-made object detection from infrared images
Author :
Liu, Jun ; Wei, Hong
Author_Institution :
Sch. of Autom., Hangzhou Dianzi Univ., Hangzhou, China
Volume :
2
fYear :
2010
fDate :
6-7 March 2010
Firstpage :
177
Lastpage :
180
Abstract :
In this paper, a review of man-made object detection algorithms is presented based on various fractal features which are derived from the blanket covering method. These fractal features include fractal dimension (D), fractal model fitting error (FE), D-dimension area (K), multi-scale fractal feature related with D (MFFD), and multi-scale fractal feature related with K (MFFK). To choose the optimal fractal feature for man-made object detection from infrared images, a performance evaluation method for these algorithms is proposed in criterion of overlapped regions between ground truth and segmented image. The analysis and comparison of these algorithms are performed in terms of detection accuracy and computation cost. The results have revealed that different fractal features have different capability in discriminating between natural and man-made objects, and MFFK has the highest detection accuracy among all evaluated fractal features.
Keywords :
feature extraction; fractals; infrared imaging; object detection; D-dimension area; MFFD; MFFK; blanket covering method; computation cost; detection accuracy; fractal dimension; fractal model fitting error; ground truth image; infrared images; man-made object detection; multiscale fractal feature related; optimal fractal feature selection; performance evaluation; segmented image; Computational efficiency; Computer vision; Equations; Fractals; Infrared detectors; Infrared imaging; Iron; Object detection; Robotics and automation; Signal to noise ratio; feature selection; fractal feature; man-made object detection;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Informatics in Control, Automation and Robotics (CAR), 2010 2nd International Asia Conference on
Conference_Location :
Wuhan
ISSN :
1948-3414
Print_ISBN :
978-1-4244-5192-0
Electronic_ISBN :
1948-3414
Type :
conf
DOI :
10.1109/CAR.2010.5456575
Filename :
5456575
Link To Document :
بازگشت