Two-Phase Image Segmentation with Nonlocal Mean Filter and Kernel Evolutionary Clustering in Local Learning

2014 ◽  
Vol 536-537 ◽  
pp. 172-175
Author(s):  
Bing Chen ◽  
Dong Dong Yang ◽  
Gang Lu ◽  
Hong Xiao Feng
Keyword(s):  
Image Segmentation ◽  
Learning Strategy ◽  
Phase Image ◽  
Local Learning ◽  
Two Phase ◽  
Evolutionary Clustering ◽  
Kernel Clustering ◽  
Mean Filter ◽  
Image Segmentation Algorithm

In this study, a novel two-phase image segmentation algorithm (TPIS) by using nonlocal mean filter and kernel evolutionary clustering in local learning is proposed. Currently, the difficulties for image segmentation lie in its vast pixels with overlapping characteristic and the noise in the different process of imaging. Here, we want to use nonlocal mean filter to remove different types of noise in the image, and then, two kernel clustering indices are designed in evolutionary optimization. Besides, the local learning strategy is designed using local coefficient of variation of each local pixels or image patch is employed to update the quality of the local segments. The new algorithm is used to solve different image segmentation tasks. The experimental results show that TPIS is competent for segmenting majority of the test images with high quality.

scholarly journals Fast Mumford-Shah Two-Phase Image Segmentation Using Proximal Splitting Scheme

2021 ◽  
Vol 2021 ◽  
pp. 1-13
Author(s):  
Nilima Shah ◽  
Dhanesh Patel ◽  
Pasi Fränti
Keyword(s):  
Image Segmentation ◽  
Level Set ◽  
Energy Functional ◽  
Phase Image ◽  
Splitting Scheme ◽  
Two Phase ◽  
Crucial Step ◽  
Error Metrics ◽  
Initial Segmentation

The Mumford-Shah model is extensively used in image segmentation. Its energy functional causes the content of the segments to remain homogeneous and the segment boundaries to become short. However, the problem is that optimization of the functional can be very slow. To attack this problem, we propose a reduced two-phase Mumford-Shah model to segment images having one prominent object. First, initial segmentation is obtained by the k-means clustering technique, further minimizing the Mumford-Shah functional by the Douglas-Rachford algorithm. Evaluation of segmentations with various error metrics shows that 70 percent of the segmentations keep the error values below 50%. Compared to the level set method to solve the Chan-Vese model, our algorithm is significantly faster. At the same time, it gives almost the same or better segmentation results. When compared to the recent k-means variant, it also gives much better segmentation with convex boundaries. The proposed algorithm balances well between time and quality of the segmentation. A crucial step in the design of machine vision systems is the extraction of discriminant features from the images, which is based on low-level segmentation which can be obtained by our approach.

Spatial Information Based on the Multi-Objective Programming of Fuzzy Kernel Clustering Image Segmentation Algorithm

2013 ◽  
Vol 791-793 ◽  
pp. 1337-1340
Author(s):  
Xue Zhang Zhao ◽  
Ming Qi ◽  
Yong Yi Feng
Keyword(s):  
Image Segmentation ◽  
Clustering Algorithm ◽  
Spatial Information ◽  
Segmentation Algorithm ◽  
Multi Objective ◽  
Multi Objective Programming ◽  
Kernel Clustering ◽  
Noise Image ◽  
The Relationship ◽  
Image Segmentation Algorithm

Fuzzy kernel clustering algorithm is a combination of unsupervised clustering and fuzzy set of the concept of image segmentation techniques, But the algorithm is sensitive to initial value, to a large extent dependent on the initial clustering center of choice, and easy to converge to local minimum values, when used in image segmentation, membership of the calculation only consider the current pixel values in the image, and did not consider the relationship between neighborhood pixels, and so on segmentation contains noise image is not ideal. This paper puts forward an improved fuzzy kernel clustering image segmentation algorithm, the multi-objective problem, change the single objective problem to increase the secondary goals concerning membership functions, Then add the constraint information space; Finally, using spatial neighborhood pixels corrected membership degree of the current pixel. The experimental results show that the algorithm effectively avoids the algorithm converges to local extremism and the stagnation of the iterative process will appear problem, significantly lower iterative times, and has good robustness and adaptability.

Weighted variational two-phase image segmentation based on Fuzzy Region Competition

2011 ◽  
Author(s):  
Vinicius R. P. Borges ◽  
Celia A. Zorzo Barcelos ◽  
Denise Guliato ◽  
Marcos Aurelio Batista
Keyword(s):  
Image Segmentation ◽  
Phase Image ◽  
Two Phase ◽  
Fuzzy Region

Fuzzy Clustering Color Image Segmentation Algorithm Based on CPSO

2012 ◽  
Vol 461 ◽  
pp. 526-531
Author(s):  
Xiao Hong Zhang ◽  
Hong Mei Ning
Keyword(s):  
Image Segmentation ◽  
Color Image ◽  
Segmentation Algorithm ◽  
Color Image Segmentation ◽  
Local Optimum ◽  
Fcm Clustering ◽  
Clustering Center ◽  
Membership Matrix ◽  
Image Segmentation Algorithm

Fuzzy C-mean algorithm (FCM) has been well used in the field of color image segmentation. But it is sensitive to initial clustering center and membership matrix, and likely converges into the local minimum, which causes the quality of image segmentation lower. By use of the properties-ergodicity, randomicity of chaos, a new image segmentation algorithm is proposed, which combines the chaos particle swarm optimization (CPSO) and FCM clustering. Some experimental results are shown that this method not only has the ability to prevent the particles to convergence to local optimum, but also has faster convergence and higher accuracy for segmentation. Using the feature distance instead of Euclidian distance, robustness of this method is enhanced.

Research on the Automatic Multilevel Thresholding for Athletic Image Segmentation Based on Block Sampling and Genetic Algorithm

2013 ◽  
Vol 791-793 ◽  
pp. 2007-2012
Author(s):  
Xiao Guang Li
Keyword(s):  
Image Processing ◽  
Image Segmentation ◽  
Meaningful Work ◽  
Image Processing Algorithm ◽  
Computer Image Processing ◽  
Ga Algorithm ◽  
Block Sampling ◽  
The Cost ◽  
Image Segmentation Algorithm

The research on new image segmentation algorithm is a very meaningful work in the processing of image. In the process, it will produce large amount of data redundancy. The efficient algorithm not only can greatly improve the quality of image treatment but also can greatly reduce the time and cost of the treatment. In this context, the paper analyzes several image processing algorithms commonly used in recent years and presents a new computer image processing algorithm--AMT-GA algorithm. In order to verify the effectiveness of AMT-GA algorithm, this paper takes the process of athletics image for example and compares the consistency and time of image segmentation with other literature results and ultimately finds that the consistency of AMT-GA algorithm reaches 0.99. The time in the algorithm execution is only 0.81 which not only achieves effective segmentation of the image but also saves the cost of computing. It also provides a theoretical reference for the research of computer graphics technology.

Image Segmentation Research Based on Genetic Algorithm

2011 ◽  
Vol 403-408 ◽  
pp. 1622-1625
Author(s):  
Xiao Wei Guan ◽  
Xia Zhu
Keyword(s):  
Genetic Algorithm ◽  
Image Processing ◽  
Image Segmentation ◽  
Feature Representation ◽  
The Other ◽  
Ga Algorithm ◽  
Representation Method ◽  
Optimal Feature ◽  
Image Segmentation Algorithm

As one of the difficulties and hot of computer vision and image processing, Image segmentation is highly valued by the research workers. Yet there is no image segmentation algorithm which is generic, and it is difficult to obtain an optimal feature representation method. In this paper, genetic algorithm (GA) has proposed to segment the image. GA algorithm can improve the efficiency and quality of the picture some extent through the experimental results. The algorithm has some versatility, as long as the corresponding parameters are adjusted, it can also handle the other images. The results show that GA algorithm is very stable, and the fusion result is more satisfactory. Thus, GA can be applied in image segmentation and this algorithm will have good prospects in image processing.

VARIATIONAL TWO-PHASE IMAGE SEGMENTATION BASED ON REGION COMPETITION BY LEVEL SET AND FUZZY APPROACHES

2011 ◽  
Author(s):  
Marcus Aurelio Vinicus R. Pereira Borges ◽  
Marcus Aurelio Batista ◽  
Celia A Zorzo Barcelos
Keyword(s):  
Image Segmentation ◽  
Level Set ◽  
Phase Image ◽  
Two Phase
Sign in / Sign up

Export Citation Format

Share Document