[1]LI Hai-feng,DU Jun-ping.Image classification technology based on color features[J].CAAI Transactions on Intelligent Systems,2008,3(2):155-158.
Copy

Image classification technology based on color features

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 3 Number of periods: 2008 2 Page number: 155-158 Column: 学术论文—人工智能基础 Public date: 2008-04-25
Title:
Image classification technology based on color features
Author(s):
LI Hai-feng1;  DU Jun-ping2
1. School of Computer Science, Beijing Business and Technology University, Beij ing 100037, China;
2. School of Computer and Technology, Beijing University of Posts and Teleco mmunications, Beijing 100876,China
Keywords:
image classification;  color features;  boosting algorithm
CLC:
TP391
DOI:
-
Abstract:
This paper studies the effects of an image’s color features on image classification. We developed a new classification method based on positional distribution of colors and compare it with other methods based on RGB and HSV histograms. The random forest, the Boosting algorithm, and the MLP neural network were applied respectively to classify images and a natural image classification system built up. The advantages and disadvantages of these three classification algorithms are discussed according to experimental results, showing that the boosting algorithm gives the best performance and is more suitable to image classification. 
References:
［1］章毓晋.图像分割［M］. 北京:科学出版社, 2001.
?［2］贾富仓,李 华. 基于随机森林的多谱磁共振图像分割［J］. 计算机工程, 2005 ,31(10):159161.
?JIA Fucang,LI Hua. Multispectral magnetic resonance Image segmentation using r andom forests［J］. Computer Engineering，2005,31(10):159161.
［3］吕 坤,高 珊.一种基于改进的BP神经网络算法的布匹瑕疵分类器［J］.微电子学与计算机，2006,23(3):8890.
?LU Kun,GAO Shan. A fabric flaws classifier based on improved BP neural network algorithm［J］. Microelectronics & Computer, 2006,23(3):8890.
［4］BREIMAN L. Random forests［J］. Machine Learning, 2001,45(1):532.
［5］SCHAPIRE R E. The strength of weak learnability［J］. Machine Learnin g,1990, 5(2): 197227.
［6］SCHAPIRE R E. A brief introduction to boosting［C］// Proceedings of the Si xteenth International Joint Conference on Artificial Intelligence. Stockho lm, 1999：14011406.
［7］FABRIZIO S, ALESSAMDRO S, NICOLA V.An improved boosting algorithm and its ap plication［C］// Proceedings of the 2000ACM CIKM International Conference on I nformation and Knowledge Management.Mclean,VA,USA,2000：7885.
［8］HAN Jiawei, KAMBER M. Data mining: concepts and techniques ［M］ . 北京:机械工业出版社, 2001.
［9］杨再华,李玉和,李庆祥.基于边缘特征提取的图像清晰度评价函数［J］. 计算机工程与应用, 2005 (10) : 3536.
?YANG Zaihua,LI Yuhe,LI Qingxiang. Image definition criterion based on edge grad ient feature［J］. Computer Engineering and Applications, 2005 (10) : 3536. 
［10］MANJUNATH B S, OHM J R, VASUDENVAN V V, et al. Color and text ure descriptors［J］.IEEE Trans on CSVT, 2001,11(6): 703715.
Similar References:

Memo

-

Last Update: 2009-05-11

Copyright © CAAI Transactions on Intelligent Systems