Papers by Keyword: Over-Segmentation

Paper TitlePage

Authors: Jin Xi Wang, Lin Xiang Liu, Xiu Zheng Li
Abstract: The watershed algorithm has been widely used in image segmentation for its characteristics of accurately positioning edge, simple operation and etc. But it also has drawbacks of easy to over-segmentation and loss important outline for the character of sensitive to noise. Aiming at the problem of over-segmentation of watershed algorithm, the paper brought out an improved image segmentation algorithm based on watershed, which can limit the number of existing regions that are allowed with combination pre-processing steps, so that the over-segmentation problem can be better solved. The result of experiment also verifies the correctness and feasibility of the proposed algorithm in the paper.
Authors: Qiang Qiang Li, Wei Li
Abstract: In order to solve the problem of over-segmentation of traditional watershed algorithm, an improved watershed segmentation algorithm of the bridge image was proposed in this paper. First, the input image was filtered by top-hat transformation and bottom-hat transformation, and then, a multiscale algorithm for computing morphological gradient images is proposed, and the threshold for marker-extraction is automatically calculated according to the statistics of local extreme points in the gradient map. The watershed algorithm is applied on the modified gradient map to segment the image. Then, the over-segmentation regions of the initial watershed segmentation is settled by region merging based on fisher distance.Region merging is ended according to divergence principle. Many contrast experimental results verified the feasibility and validity of the method.
Authors: Zhu Lin Wang, Bin Fang, Xi Wei Guo
Abstract: Abstract. Image segmentation is a key technology in image engineering, it occupy an important position. This paper introduces the watershed transform to Image of monolithic circuit processing method, and then introduced the watershed transform to Image of monolithic circuit segmentation and sample. The results show that, by using the watershed algorithm and morphological processing function, which is connected with a plurality of object images are segmented into a plurality of single object, to achieve the image segmentation, and as far as possible to reduce or eliminate the phenomenon of over-segmentation. Finally it points out the further direction of research.
Authors: Wei Dong Lei, Kai Li, Xue Ping Wang
Abstract: Reasonable distribution of the rock fragmentation can reduce the overall cost in a mining industry. Digital image processing technique can be used to determine the rock fragmentation distribution function. This paper proposes an improved watershed segmentation method in digital image processing technique. The seven key steps in the new method are presented. Meanwhile, the detailed rock fragmentation analysis is performed on an example digital photo of fragmented rock pile taken from a practical underground mine, to further explain the mechanism of the new method. It is found that the new method can overcome the phenomenon of over-segmentation and over-lapping in digital photos, and exactly detect boundaries for every individual target in the photo.
Authors: Ping Wang, Rong Chen, Wei Luo, Rui Hua Xia
Abstract: In the process of macrophage image manufacture, cell adhesion often appears because of equipment or man-made reason and then influences subsequent automatic detection and analysis. The traditional watershed algorithm is easy to cause over-segmentation for the volatility of gray extremum. To overcome these phenomena, an improved watershed algorithm which was based on macrophages image segmentation was introduced in this paper. The simulation results show that the improved algorithm can effectively segment adhesive macrophages and restrain the over-segmentation phenomena with an acceptable computation speed.
Showing 1 to 5 of 5 Paper Titles