China 
Africa 
Explore chapters and articles related to this topic
Medical and Mathematical Background
Published in Arwa Ahmed Gasm Elseid, Alnazier Osman Mohammed Hamza, Computer-Aided Glaucoma Diagnosis System, 2020
Arwa Ahmed Gasm Elseid, Alnazier Osman Mohammed Hamza
Image segmentation, one of the most important aspects of image processing, is still a research area in computer vision. Image segmentation is used to extract the Region of Interest for image analysis, and the division of an image into meaningful structures. The image segmentation is a basic step in image analysis, object representation, visualization, and many other image processing tasks; thus, segmenting an image into several parts makes further processing simpler, and reduces the amount of information. Mainly, segmentation depends on several different features, such as the color or texture contained in an image. Before de-noising an image, it is segmented to recover the original image. There are several image segmentation techniques that partition the image into several parts based on image features like pixel intensity value, color, texture, etc. Many segmentation methods have been proposed and used (Manjula, 2015).
Multimodal Imaging Radiomics and Machine Learning
Published in Ayman El-Baz, Jasjit S. Suri, Big Data in Multimodal Medical Imaging, 2019
Gengbo Liu, Youngho Seo, Debasis Mitra, Benjamin L. Franc
Four general steps of an ideal machine learning model over radiomic features have been described in the literature [3,6] and are summarized in Figure 1.1: (a) high-quality standardized imaging data acquisition; (b) region of interest image segmentation; (c) high-throughput feature extraction from images; (d) clinical predictive model establishment. Machine learning algorithms were mainly used in the fourth step to predict the patient outcomes. In the following sections, the first three steps will be described briefly and the fourth step will be described in detail.
Skin Cancer Detection using Imagery Techniques
Published in P. C. Thomas, Vishal John Mathai, Geevarghese Titus, Emerging Technologies for Sustainability, 2020
Jose J. Edathala, Vishnupriya Mohanan
Image segmentation is the process of partitioning the image into set of objects and background. Image segmentation is used to extract region of interest. K-means clustering is an unsupervised machine learning algorithm that is iterative. It is used to divide a group of data points into clusters wherein points inside one cluster are similar to each other.
A sonography image processing system for tumour segmentation
Published in Enterprise Information Systems, 2020
Chung-Ming Chen, Shu-Wei Zhang, Chih-Yu Hsu
Active contour model is used to describe how a contour evolutes to find the boundary of Region of Interest (ROI) for image segmentation. Because the deformation of level set function at zero level represents the deformation of active contour (Osher and Fedkiw 2002), the active contour can be represented by a level set function with the zero level . The active contour separates the inside region and the outside region .
A model for screening eye diseases using optical coherence tomography images
Published in International Journal of Computers and Applications, 2022
Sanchay Gupta, Siddharth Chandra, N. Maheswari, M. Sivagami
Retinal thickness is one of the most important features when dealing with OCT images [3]. It defines most of the concerns regarding various diseases and help individuals to correctly identify whether a person is having a disease or not. Using computer vision techniques one can extract the required region of interest like choroidal region or whole retina if required. Here this process involves the use of methods like contour extraction and region identification using segmentation to extract choroid and then find its thickness which can be depicted using the area under the curve of surrounding contours, shown in the figure.
Cotton flow velocity measurement based on image cross-correlation and Kalman filtering algorithm for foreign fibre elimination
Published in The Journal of The Textile Institute, 2022
Yang Dai, Tao Xu, Zhongqiang Feng, Xing Gao
The purpose of image segmentation is to obtain the ROI (region of interest). The current method for the NCC image registration is to segment the image into several blocks on average. However, in this method the horizontal velocity is omitted which has a great impact on the cotton flow trajectory. Furthermore, every single cotton has a different velocity. Here, we segmented the cotton image into discrete single cotton. Image segmentation goes through the following steps, shown in Figure 6.