BackgroundExtracting features from the colonoscopic images is essential for getting the features, which characterizes the properties of the colon. The features are employed in the computer-assisted diagnosis of colonoscopic images to assist the physician in detecting the colon status.MethodsEndoscopic images contain rich texture and color information. Novel schemes are developed to extract new texture features from the texture spectra in the chromatic and achromatic domains, and color features for a selected region of interest from each color component histogram of the colonoscopic images. These features are reduced in size using Principal Component Analysis (PCA) and are evaluated using Backpropagation Neural Network (BPNN).Results Features extracted from endoscopic images were tested to classify the colon status as either normal or abnormal. The classification results obtained show the features' capability for classifying the colon's status. The average classification accuracy, which is using hybrid of the texture and color features with PCA (τ = 1%), is 97.72%. It is higher than the average classification accuracy using only texture (96.96%, τ = 1%) or color (90.52%, τ = 1%) features.ConclusionIn conclusion, novel methods for extracting new texture- and color-based features from the colonoscopic images to classify the colon status have been proposed. A new approach using PCA in conjunction with BPNN for evaluating the features has also been proposed. The preliminary test results support the feasibility of the proposed method.
Abstract-A computer-assisted endoscopic analysis is intended and facilitates the diagnosis process. Segmentation of the image is an important step and a novel approach is proposed to segment clinical endoscopic images based on homogeneity and color feature hue. In the first stage, the regions are segmented using a peak-finding algorithm on a 2-D histogram of homogeneity and intensity values. In the second stage, histogram analysis of the color feature hue is performed to subdivide the segmented regions obtained from the first stage. The subdivisions of different segmented regions having similar CIE (L* a* b*) color measure are merged. The proposed scheme was evaluated on a database of clinical endoscopic images. Keywords-clinical endoscopic image, color image segmentation, homogeneity, color space, region merging. I. INTRODUCTIONMinimally invasive endoscopic procedures are increasingly employed for diagnostic and surgical purposes in cases such as the gastrointestinal and respiratory ailments. An endoscopist, who analyzes the acquired images, performs these procedures. A computer-aided scheme will help considerably in the image analysis and quantitative characterization of abnormalities, thereby improving the overall efficiency of the diagnosis. A crucial step in such an automated diagnostic scheme is proper segmentation of the endoscopic images.Color of an image carries much more information than the gray levels. In many pattern recognition and computer vision applications, the additional information provided by color can help in image analysis and provide better results than approaches using merely gray scale information.The scales employed to differentiate pixels in a color image include uniform chromaticity scale (UCS), OHTA, and Hue, Saturation, and Intensity (HSI).The UCS approximates human perceptions when a Euclidean distance measure in the b*) * a * (L , and *) * u * (L v color spaces is used to differentiate between pixels compare to B) G, (R, and Z) Y, (X, color spaces. The UCS matches the sensitivity of human eye, and can mimic human perception.OHTA color space was derived as a result of a search of completely statistically independent components on a representative sample of images. The OHTA components are good approximations of the results of the KarhunenLoeve transformation, which is very good with respect to decorrelation of RGB components [3].HSI uses psychological attributes instead of human perceptions to differentiate pixels. The perceptual color space is used to capture the psychological attributes, and it usually described by Hue, Saturation, and Intensity (HSI). The hue is the attribute of color perception denoted by red, yellow, green, blue, and so on. Saturation is used to describe how pure a color is or how much white is added to a pure color. The intensity is a measure of total reflectance in the visible region of spectrum, and it is an achromatic component of color. Using HSI color space for color image segmentation has two advantages: (1) specifying and controlling color is more suit...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.