Detection of foreground structured objects in the images is an essential task in many image processing applications. This paper presents a region merging approach for automatic detection of the foreground objects in the image. The foreground objects are the structured objects with an independent and detectable boundary. The proposed approach identifies objects in the given image based on general properties of the objects without depending on the prior knowledge about specific objects. The regions of the structured objects in the image are separated from the background based on region contrast information. The perceptual organization laws of human visual system are used in the region merging process to identify the boundaries of various objects. The system is adaptive to the image content. The results of the experiments show that the proposed scheme can efficiently extract object boundary from the background.

Contrast, Segmentation, Histogram, Thresholding, Region Merging

[1] E. Borenstein and E. Sharon, �Combining top-down and bottom-up
segmentation,� in Proc. IEEE Workshop Perceptual Org. Comput. Vis.,
CVPR, 2004.
[2] X. Ren, �Learning a classification model for segmentation,� in Proc.
IEEE ICCV, 2003.
[3] P. Felzenszwalb and D. Huttenlocher, �Efficient graph-based image
segmentation,� Int. Journal of Computer Vision, vol. 59, no. 2, Sep.2004.
[4] J. B. Shi and J. Malik, �Normalized cuts and image segmentation,�
IEEE Transaction Pattern Anal. Mach. Intell., vol. 22, no. 8, Aug. 2000.
[5] S. Gould, J. Rodgers, D. Cohen, G. Elidan, and D. Koller, �Multi-class segmentation with relative location prior,� Int. J. Computer Vision, vol. 80, no. 3,Dec. 2008.
[6] J. D. McCafferty, Human and Machine Vision: Computing Perceptual Organization. Chichester, U.K.: Ellis Horwood, 1990.
[7] R. Mohan and R. Nevatia, �Perceptual organization for scene segmentation and description,� IEEE Trans. Pattern Analysis & Machine Intelligence, vol.14, no. 6,Jun. 1992.
[8] T. Brinkhoff, H. P. Kriegel, and R. Schneider, �Measuring the complexity of polygonal objects,� in Proc. 3rd ACM Int. Workshop, 1995.
[9] I. H. Jermyn and H. Ishikawa, �Globally optimal regions and boundaries as minimum ratio weight cycles,� IEEE Trans. Pattern Analysis & Machine Intelligence, vol. 23,Oct. 2001.
[10] B. C. Russell, �Using multiple segmentations to discover objects and their extent in image collections,� in Proc. IEEE CVPR, 2006, vol. 2,pp. 1605�1614.
[11] D. R. Martin, C. C. Fowlkes, and J. Malik, �Learning to detect natural image boundaries using local brightness, color, and texture cues,� IEEE Trans. Pattern Analysis & Machine Intelligence., vol. 26, no. 5, pp. 530�549, May 2004.
[12] Z. L. Liu, D. W. Jacobs, and R. Basri, �The role of convexity in perceptual completion: Beyond good continuation,� Vis. Res., vol. 39, no. 25, pp. 4244�4257, Dec. 1999
[13] V. Bruce and P. Green, Visual Perception: Physiology, Psychology and Ecology. Hillsdale, NJ: Lawrence Erlbaum Associates Ltd., 1990.
[14] C. Cheng, A. Koschan, D. L. Page, and M. A. Abidi, �Outdoor Scene image segmentation based on background recognition and perceptual organization,� IEEE Transaction on Image Processing VOL 21, No 3.March 2012.
[15] C. Cheng, A. Koschan, D. L. Page, and M. A. Abidi, �Scene image
segmentation based on perceptual organization,� in Proc. IEEE ICIP,
2009, pp. 1801�1804.
[16] L. Yang, P. Meer, and D. J. Foran, �Multiple class segmentation using a unified framework over man-shift patches,� in Proc. IEEE CVPR, 2007, pp. 1�8.
[17] S Jayaraman,�Image Segmentation� in Digital Image Processing: McGraw-Hill,2010