Affiliation:
1. Lanzhou University of Technology
Abstract
Aims at the problem that the threshold number and value are difficulty to determine automatically existing in multi-threshold color image segmentation method, a novel method of multi-threshold segmentation in HSV is proposed. First of all, the image is pre-processed in HSV, component H and V is projected to S and be quantified at the same time. Secondly, histogram and advanced Histon histogram (AHH) are constructed. According to concept of roughness in the theory of Rough Set, the histogram of roughness (RSH) is constructed. Finally, according to requirement of segmentation accuracy, set a threshold Hn on RSH to determine the number and scope of multi-threshold and the image is segmented with above thresholds. The experimental results show that this method can determine the threshold quantity automatically, segment image efficiently and robust against illumination variation.
Publisher
Trans Tech Publications, Ltd.
Subject
Mechanical Engineering,Mechanics of Materials,General Materials Science
Reference7 articles.
1. H.D. Cheng ,X.H. Jiang and Y. Sun. Color image segmentation: advances and prospects, Pattern Recognition. vol. 35, (2001), pp.2259-2281.
2. LIN Kai-yan , WU Jun- hui , XU Li-hong . A Survey on Color Image Segmentation Techniques, Journal of Image and Graphics. vol. 10, no. 1, ( 2005), pp.1-8.
3. C.C. Chang ,L.L. Wang . A fast multi level threshold method based on low pass and high pass filtering , Pattern Recognition Letters . vol. 18, (1977), p.1469~1478.
4. YE Qi-Xiang, GAO Wen, WANG Wei-Qiang, HUANGie-Jun , A Color Image Segmentation Algorithm by Using Color and Spatial Information, Journal of Software . vol. 15, no. 04(2004), pp.522-530.
5. Mohabey A , Ray A K. Fusion of Rough Set Theoretic Approximations and FCM for Color Image Segmentation. Systems, ManandCybernetics, 2000IEEE InternationalConference, (2000) , pp.1529-1534.