Cloud model-based method for range-constrained thresholding. (February 2015)

• Record Type: Journal Article
• Authors: Wu, Tao; Xiao, Jin; Qin, Kun; Chen, Yixiang
• Is Part Of: Computers & electrical engineering. Volume 42(2015)
• Page Start: 33
• Page End: 48
• Publication Date: 2015-02
• Subjects: Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2014.03.016
• Languages: English
• ISSNs: 0045-7906

Abstract

Graphical abstract: Highlights: A cloud model-based framework for range-constrained thresholding with uncertainty. Improving four traditional methods under the new framework. Representing the image using cloud model. Implementing image transformation to focus on mid-region of the image. Cloud model-based framework is efficient and effective. Abstract: Thresholding is a popular image segmentation method that converts a grayscale image into a binary image. In this paper, we propose a cloud model-based framework for range-constrained thresholding with uncertainty, and improve four traditional methods. The method involves four major steps, including representing the image using cloud model, estimating the automatic threshold for gray level ranges of object and background, implementing image transformation to focus on mid-region of the image, and determining the binary threshold within the constrained gray level range. Cloud model can effectively represent various visual properties of the image, such as intensity-based class uncertainty, intra-class homogeneity, and between-class contrast. The approach is validated both quantitatively and qualitatively. Compared with the traditional state-of-art algorithms on a variety of synthetic and real images, with or without noisy, as well as laser cladding images, the experimental results suggest that the presented method is efficient and effective.