Circular spot detection and segmentation with rotating line local binary analysis

Author

Dongbo Zhang ; Xia Liu

Author_Institution

Coll. of Inf. Eng., Xiangtan Univ., Xiangtan, China

fYear

2014

fDate

27-30 Oct. 2014

Firstpage

1683

Lastpage

1687

Abstract

The circular spots appear as round local peaks/valleys in image space, and they reflect omnidirectional symmetry and Gaussian distribution characteristics. According to these observations, a novel rotating line scan operator is proposed to achieve feature analysis of the spot object. Experimental tests show that the proposed algorithm is robust to spots´ appearance variation (size, local contrast and intensity). The method presents promising performance in various real images.

Keywords

Gaussian distribution; image segmentation; object detection; Gaussian distribution characteristics; circular spot detection; circular spot segmentation; feature analysis; image space; omnidirectional symmetry; real images; rotating line local binary analysis; rotating line scan operator; round local peaks; round local valleys; spot appearance variation; spot intensity; spot local contrast; spot object; spot size; Biomedical imaging; Gaussian distribution; Image color analysis; Lesions; Object detection; Retina; Robustness; Circular spot detection; local binary analysis; rotating line;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing (ICIP), 2014 IEEE International Conference on

Conference_Location

Paris

Type

conf

DOI

10.1109/ICIP.2014.7025337

Filename

7025337