INTRODUCTION: Breast cancer is the most hazardous disease among women worldwide. A simple, cost-effective, and efficient screening called mammographic imaging is used to find the breast abnormalities to detect breast cancer in the early stages so that the patient's health can be improved. OBJECTIVES: The main challenge is to extract the features by using a novel technique called Advanced Gray-Level Cooccurrence Matrix (AGLCM) from pre-processed images and to classify the images using machine learning algorithms. METHODS: To achieve this, we proposed a four-step process: image acquisition, pre-processing, feature extraction, and classification. Initially, a pre-processing technique called Contrast Limited Advanced Histogram Equalization (CLAHE) is used to increase the contrast of images and the features are retrieved using AGLCM which extracts texture, intensity and shape-based features as these are important to identify the abnormality. RESULTS: In our framework, a classifier called eXtreme Gradient Boosting (XGBoost) is applied on mammograms and the results are compared with other classifiers such as Random Forest (RF), K-Nearest Neighbor (KNN), Artificial Neural Networks (ANN), and Support Vector Machine (SVM). The experiments are done on the Mammographic Image Analysis Society (MIAS) dataset. CONCLUSION: The outcome achieved with CLAHE+ AGLCM+ XGBoost classifier is better than the existing methods.In future, we experiment on large datasets and also concentrate on optimal features selection to increase the classification.
Abstract-Probability distributions formulate the basic framework for developing several segmentation algorithms.Among the various segmentation algorithms, skin colour segmentation is one of the most important algorithms for human computer interaction. Due to various random factors influencing the colour space, there does not exist a unique algorithm which serve the purpose of all images. In this paper a novel and new skin colour segmentation algorithms is proposed based on bivariate Pearson type II a mixture model since the hue and saturation values always lies between 0 and 1. The bivariate feature vector of the human image is to be modeled with a Pearson type II a mixture (bivariate Beta mixture) model. Using the EM Algorithm the model parameters are estimated. The segmentation algorithm is developed under Bayesian frame. Through experimentation the proposed skin colour segmentation algorithm performs better with respect to segmentation quality metrics such as PRI, VOI and GCE. The ROC curves plotted for the system also revealed that the proposed algorithm can segment the skin colour more effectively than the algorithm with Gaussian mixture model for some images.Index Terms-Bivariate Pearson type II a mixture model, skin colour segmentation, HSI colour space, segmentation quality metrics
Skin colour segmentation plays an important role in computer vision, face detection and human related systems. Much work has been reported in literature regarding skin colour detection using Gaussian mixture model. The Gaussian mixture model has certain limitations regarding the assumptions like pixels in each component are mesokurtic, having negative range and it doesn't adequately represent the variance of the skin distribution under illumination conditions. In this paper we develop and analyze a new skin colour segmentation based on HSI colour space using bivariate Pearsonian type-IIb mixture model. The model parameters are estimated by deriving the updated equation of EM-Algorithm. The initialization of the model parameters is done through K-means algorithm and method of moments. The segmentation algorithm is obtained using component maximum likelihood under Bayes frame. The experimental results using hue and saturation as feature vector revealed that the developed method perform better with respect to segmentation performance metrics than that of Gaussian mixture model. This method is useful in face detection and medical diagnostics.
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.