Abstract：After analyzing the texture feature extraction on MPEG-7 standard, a new Markov Edge Descriptor (MED) was proposed based on Edge Histogram Descriptor (EHD), which was in virtue of the Markov chain. It adopted initial distribution of Markov chain model to represent the spatial information between the same kinds of edges. And it adopted stationary distribution of Markov chain model to represent the spatial information between different edges. The experimental results show that MED has the merits of EHD, and MED can describe the spatial information between the image edges. It has corking retrieval performance for the same or different kinds of edges. Its retrieval performance is better than EHD.
MANJUNATH B S, OHM J R, VASUDEVAN V V, et al. Color and texture descriptors ［J］. IEEE Transactions on CAS for
Video Technology, 2001, 11(6): 703-715.
[3]
CHEE S W, DONG K P, SOO-JUN P. Efficient use of MPEG-7 edge histogram descriptor ［J］. ETRI Journal, 2002, 24(1): 35-42.
[4]
SHIM O, CHOI T S. Edge color histogram for image retrieval ［C］// Proceedings of International Conference on Image Processing. Piscataway: IEEE, 2002, 3: 957-960.
LANTAGNE M, PARIZEAU M, BERGEVIN R. VIP: Vision tool for comparing images of people ［C］// Proceedings of IEEE Conference on Vision Interface. Piscataway: IEEE, 2003: 35-42.
[7]
Hichem F, GADER P. Detection and discrimination of land mines in ground-penetrating radar based on edge histogram descriptors and possibilistic K-nearest neighbor classifier ［J］. IEEE Transactions on Fuzzy Systems, 2009, 17(1): 185-199.
WEI PIAN-PIAN, WANG BEI-ZHAN, QU CHENG. Improved generalized cooccurrence matrix for image retrieval using MPEG-7 texture edge ［C］// ISISE 08： International Symposium on Information Science and Engineering. Piscataway: IEEE, 2008: 205-208.
[10]
MANJUNATH B S, OHM J-R, VASUDEVAN V V. Color and texture descriptors ［J］. IEEE Transactions on Circuits and Systems for Video Technology, 2001, 11(6): 703-715.
[11]
TAKAHSHI N, IWASAKRI M, KUNIEDAL T. Image retrieval using spatial intensity features ［J］. Signal Processing: Image Communication, 2000, 16(1): 45-57.
2012, Vol. 32 
