Contrast Context Histogram – A Discriminating Local Feature for Object Recognition

Author(s):

Chun-Rong Huang, Chu-Song Chen, and Pau-Choo Chung

Introduction

This paper presents a new invariant local descriptor, contrast context histogram (CCH), for object recognition. It represents the contrast distributions of a local region, and serves as a local distinctive descriptor of this region. Object recognition can be considered as matching salient corners with similar contrast context histogram descriptors on two or more images. Our experimental results show that CCH is insensitive to image scales, rotations, viewing directions, and illumination variations. Moreover, the computation of CCH is efficient and potentially suitable for real-time applications.

Software Download

Download_CCH Software (ver 1.2)

NOTE

The input image needs to be a 24-bit bitmap, jpeg or png image. Both the width and height of the input image need to be the multiple of 4, for example, 640x480, 320x240 or 372x388, etc.

If the program lacks some DLL files, please download Inter Performance Primitives Run-Time Installer (22MB) and install it (in our program, the IPP library is only used to detect Harris-Laplacian corners and compute the two-norms of descriptors).

Copyrights

This software for the detection of contrast context histogram features is available for personal research usage only. Any commercial use or any redistribution of this software requires a license from the Academia Sinica. This license file must be retained with all copies of the software, including any modified or derivative versions. For further details on obtaining a commercial license, please contact Chun-Rong Huang ( nckuos@iis.sinica.edu.tw ) or the Multimedia Lab of the Institute of Information Science , Academia Sinica. Copyright (2007), ACADEMIA SINICA. THE ACADEMIA SINICA AND THE AUTHORS MAKE NO REPRESENTATIONS OR WARRANTIES OF ANY KIND CONCERNING THIS SOFTWARE.

This software also uses the FreeImage open source image library. See http://freeimage.sourceforge.net for details. FreeImage is used under the (GNU GPL or FIPL).

Release Notes

V1.2 Add a matching function to match two CCH feature files without reading images.

V1.1 First release of CCH console mode with Matlab demos and fix the bug of the image height problem.

V1.0 First release of CCH GUI mode.

Feedback

Chun-Rong Huang nckuos at iis.sinica.edu.tw

Publications

[1] Chun-Rong Huang , Chu-Song Chen, and Pau-Choo Chung, "Contrast Context Histogram - An Efficient Discriminating Local Descriptor for Object Recognition and Image Matching," Pattern Recognition , vol.41, no.10, pp. 3071-3077, 2008. (SCI)

[2] Chun-Rong Huang , Chu-Song Chen, and Pau-Choo Chung, "Contrast Context Histogram – A Discriminating Local Descriptor for Image Matching," In Proceedings of 18 th IAPR International Conference on Pattern Recognition , ICPR'06, vol.4, pp. 53-56, Hong Kong, Aug., 2006.