China
Africa
Explore chapters and articles related to this topic
Preprocessing in the Spatial Domain
Published in Sing-Tze Bow, Pattern Recognition and Image Preprocessing, 2002
Thinning is a necessary process in most pattern recognition problems because it offers a way to simplify the form for pattern analysis. In scanning an image, especially a text or drawing, high enough resolution is preferred to assure that no indispensable information is lost during digitization. In so doing, a width of more than two pixels will appear for each line. Thinning is the process to extract and apply additional constraints on the pixel elements that are to be preserved so that a linear structure of the input image will be recaptured without destroying its connectivity. See Figure 12.52 for the linear structure by medial axis transformation, which is covered in many books and is not discussed here.
Thinning and Skeletonizing
Published in Edward R. Dougherty, Digital Image Processing Methods, 2020
Thinning algorithms remain a useful tool in image processing and computer vision. The skeleton can give useful characterizations of an object, providing features for pattern recognition as well as reducing the amount of memory necessary for storing and transmitting data. Fast implementations on parallel machines make thinning algorithms an even more attractive and useful transform.
Mathematical Morphology
Published in Vipin Tyagi, Understanding Digital Image Processing, 2018
Thinning is a morphological operation that repeatedly erodes away foreground pixels from the boundary of binary images while preserving the end points of line segments. Thickening is the dual of thinning, i.e., thickening the foreground is equivalent to thinning the background.
A progressive method for the collapse of river representation considering geographical characteristics
Published in International Journal of Digital Earth, 2020
Yilang Shen, Tinghua Ai, Jingzhong Li, Lina Huang, Wende Li
The methods used for the skeletal extraction of geometric features based on raster data mainly include thinning algorithms (Lam, Lee, and Suen 1992; Jagna 2014; Deng, Iyengar, and Brener 2000; Rosenfeld and Kak 1976; Tarabek 2012; Zhang and Suen 1984; Ahmed and Ward 2002; Boudaoud, Sider, and Tari 2015; Jagna and Kamakshiprasad 2010; Kocharyan 2013; Palágyi 2002; Song et al. 2018; Tang and You 2003; Wu and Tsai 1992; Zhou, Quek, and Ng 1995; Chen et al. 2012; Lü and Wang 1986; Shen, Ai, and Yang 2019c; Boudaoud, Solaiman, and Tari 2018). The Zhang and Suen (ZS) thinning algorithm (Zhang and Suen 1984) is one of the most classical and popular algorithms. This iterative parallel thinning algorithm has been proven to be effective and simple. Subsequently, considering that the proposed ZS method has some disadvantages, such as generating redundant pixels at locations with acute angles, many improved image thinning algorithms for solving the problems (Chen et al. 2012; Lü and Wang 1986; Boudaoud, Solaiman, and Tari 2018) based on the ZS method have been proposed by various scholars. In addition, some scholars have proposed other thinning methods by considering speed (Deng, Iyengar, and Brener 2000), smoothing effect (Zhou, Quek, and Ng 1995), topology (Ahmed and Ward 2002; Wu and Tsai 1992), and map elements (Thomas 1998).
Robust user authentication model for securing electronic healthcare system using fingerprint biometrics
Published in International Journal of Computers and Applications, 2019
Sharmin Jahan, Mozammel Chowdhury, Rafiqul Islam
Pre-processing is required to enhance the quality of an image by filtering and removing unnecessary noises because the captured images may be of poor quality. This process removes the noises in the images and enhance them for better features extraction. For image filtering we employ a fuzzy filtering technique [30]. The filtered image was then binarized and thinned to make it more appropriate for feature extraction. Thinning is a morphological operation that is used to remove selected foreground pixels from binary images. It is used to eliminate the redundant pixels of ridges till the ridges are just one-pixel wide. Thinning is normally only applied to binary images, and produces another binary image as output. It is the final step prior to feature extraction.