DocumentCode :
3501811
Title :
A comparative study between different new techniques for object detection on a structured background
Author :
Serhane, Khaled ; Devars, Jean
Author_Institution :
Lab. PARC, Univ. Pierre et Marie Curie, Paris, France
Volume :
2
fYear :
1996
fDate :
14-18 Oct 1996
Firstpage :
934
Abstract :
A comparative study between different techniques for object detection on a structured background is considered. A summary of existing works is presented and the concept of a structured background is introduced with an original concept of object detection. We describe the proposed methods and then a comparison was done to choose the method which performed best according to some commonly used criterion in object detection such as processing time, false alarms, missed detection, and the average distance between the theoretic zone and detected zone. The proposed methods have been tested on a variety of synthetic images and yielded promising results
Keywords :
image recognition; object detection; FFT method; algorithms; average distance; cooccurrence matrices; detected zone; false alarms; hit or miss method; missed detection; morphological method; object detection; processing time; structured background; superposition method; synthetic images; theoretic zone; zeros submatrices; Image processing; Layout; Object detection; Robustness; Shape; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing, 1996., 3rd International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-2912-0
Type :
conf
DOI :
10.1109/ICSIGP.1996.566243
Filename :
566243
Link To Document :
بازگشت