DCT-Domain Image Retrieval Via Block-Edge-Patterns

A new algorithm for compressed image retrieval is proposed in this paper based on DCT block edge patterns. This algorithm directly extract three edge patterns from compressed image data to construct an edge pattern histogram as an indexing key to retrieve images based on their content features. Three feature-based indexing keys are described, which include: (i) the first two features are represented by 3-D and 4-D histograms respectively; and (ii) the third feature is constructed by following the spirit of run-length coding, which is performed on consecutive horizontal and vertical edges. To test and evaluate the proposed algorithms, we carried out two-stage experiments. The results show that our proposed methods are robust to color changes and varied noise. In comparison with existing representative techniques, the proposed algorithms achieves superior performances in terms of retrieval precision and processing speed.

[1]  C. Won,et al.  Efficient Use of MPEG‐7 Edge Histogram Descriptor , 2002 .

[2]  Guizhong Liu,et al.  An effective approach to edge classification from DCT domain , 2002, Proceedings. International Conference on Image Processing.

[3]  Kyeongok Kang,et al.  A compressed domain scheme for classifying block edge patterns , 2005, IEEE Transactions on Image Processing.

[4]  Lei Guo,et al.  A shape-based image retrieval method using salient edges , 2003, Signal Process. Image Commun..

[5]  Haiming Gu,et al.  Image retrieval in various domains , 2003, Comput. Graph..

[6]  Naceur Kerkeni,et al.  Use of two-dimensional discrete cosine transform for an adaptive approach to image segmentation , 1996, Electronic Imaging.

[7]  Yeong-Ho Ha,et al.  Spatial color descriptor for image retrieval and video segmentation , 2003, IEEE Trans. Multim..

[8]  Malay Kumar Kundu,et al.  Edge based features for content based image retrieval , 2003, Pattern Recognit..

[9]  Sethuraman Panchanathan,et al.  A critical evaluation of image and video indexing techniques in the compressed domain , 1999, Image Vis. Comput..

[10]  Y. S. Hsu,et al.  Pattern Recognition Experiments in the Mandala/Cosine Domain , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  John J. Metzner,et al.  Performance improvement of a frequency hopping-CDMA system utilizing memorized prior data , 1991, IEEE Trans. Commun..

[12]  Dong-Gyu Sim,et al.  Fast texture description and retrieval of DCT-based compressed images , 2001 .

[13]  Sang Uk Lee,et al.  Image vector quantizer based on a classification in the DCT domain , 1991, IEEE Trans. Commun..

[14]  Jianmin Jiang,et al.  JPEG compressed image retrieval via statistical features , 2003, Pattern Recognit..

[15]  Michael Shneier,et al.  Exploiting the JPEG Compression Scheme for Image Retrieval , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  Chong-Wah Ngo,et al.  Exploiting image indexing techniques in DCT domain , 2001, Pattern Recognit..

[17]  Ling Guan,et al.  Image retrieval based on energy histograms of the low frequency DCT coefficients , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[18]  Shih-Fu Chang,et al.  Compressed-domain techniques for image/video indexing and manipulation , 1995, Proceedings., International Conference on Image Processing.

[19]  Young-Min Kim,et al.  Fast Scene Change Detection using Direct Feature Extraction from MPEG Compressed Videos , 2000, IEEE Trans. Multim..

[20]  Irek Defée,et al.  DCT histogram optimization for image database retrieval , 2005, Pattern Recognit. Lett..

[21]  Bo Shen,et al.  Direct feature extraction from compressed images , 1996, Electronic Imaging.