Discussion on Image Contour Extracting Method of Homogeneous Objects

Abstract:

Article Preview

Contour extracting plays an important role in image processing and computer vision, and it’s one of the most important components in low level processing. Based on many tests, this paper applies hue and texture as feature combined with mathematical morphology to have a rough segmentation of homogeneous objects, then restores the obtained contour based on h component. Tests prove that acceptable contour can be obtained. Compared with traditional method of contour extracting, the algorithm mentioned here includes self-differing, contour restoring based on hue and so on, so it has better robustness to light variation.

Info:

Periodical:

Key Engineering Materials (Volumes 392-394)

Edited by:

Guanglin Wang, Huifeng Wang and Jun Liu

Pages:

935-940

Citation:

Q.G. Zeng et al., "Discussion on Image Contour Extracting Method of Homogeneous Objects", Key Engineering Materials, Vols. 392-394, pp. 935-940, 2009

Online since:

October 2008

Export:

Price:

$41.00

[1] S.J. Song, L.D. Shi and C.F. Qiao: Chinese Journal of Construction Machinery, Vol. 4 (2007) No. 2, pp.229-234 (In Chinese).

[2] S.J. Song, L.D. Shi and C.F. Qiao: Chinese Journal of Construction Machinery, Vol. 5 (2007) No. 1, pp.16-21 (In Chinese).

[3] F.M. Zhang: Study on Algorithms of Field Road Detection and Stereovision-based Guidance Algorithm for a Field (Ph.D., Zhejiang University, China 2006), pp.41-53 (In Chinese).

[4] Y.L. Zhang and A.X. Liu: Journal of PLA University of Science and Technology, Vol. 6 (2005) No. 1, pp.44-46 (In Chinese).

[5] H.C. Yang, J.X. Hong and H.H. Chen: Journal of Xiamen University, Vol. 45 (2006) No. 1, pp.60-63 (In Chinese).

[6] Y.D. Jia: Machine Vision (Science Press, China 2000) (In Chinese).