DocumentCode :
231827
Title :
Fractal features for object recognition
Author :
Yan Zhou ; Jianning Liang
Author_Institution :
Dept. of Comput. Sci. & Eng., East China Univ. of Sci. & Technol., Shanghai, China
fYear :
2014
fDate :
19-23 Oct. 2014
Firstpage :
1225
Lastpage :
1229
Abstract :
The fractal dimension can measure the distribution of pixels´ intensities. Thus, it can be used as a tool to describe the image content. Three kinds of fractal dimensions i.e. box dimension, correlation dimension and information dimension are applied in the construction of fractal features. The object recognition is accomplished according to the similarities of the fractal features. The experiments on three data sets illustrate that fractal feature work well for objection recognition.
Keywords :
feature extraction; fractals; image recognition; object recognition; fractal dimension; fractal feature construction; image content; object recognition; pixel intensity distribution; Accuracy; Correlation; Feature extraction; Fractals; Object recognition; Shape; Training; Fractal features; Object recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing (ICSP), 2014 12th International Conference on
Conference_Location :
Hangzhou
ISSN :
2164-5221
Print_ISBN :
978-1-4799-2188-1
Type :
conf
DOI :
10.1109/ICOSP.2014.7015195
Filename :
7015195
Link To Document :
بازگشت