This paper proposes a novel approach for open-eye detection that can be used in driver drowsiness analysis based on computer vision techniques. The proposed method captures the driver video using a lowresolution camera. The proposed drowsiness detection system has three main stages. The first stage is face detection using elliptical approximation and template matching techniques. In the second stage, the open eye is detected using the proposed iris-sclera pattern analysis method. In the third stage, the drowsiness state of the driver is determined using PERcentage of eye CLOSure (PERCLOS) measure. The entire system is designed to be independent of any specific data sets for face or eye detection. The proposed method for open-eye detection uses basic image processing concepts of morphological and laplacian operations. The proposed system was evaluated with real-life images and videos. Open-eye detection accuracy of 93% was achieved.
Question classification is the process by which a system analyzes a question and labels the question based on the category to which it belongs. The automated categorization (or classification) of questions into predefined categories has witnessed a booming interest due to the increased popularity of web technologies. The recent advancement in the form of E-Learning calls for the need of question categorization. In network based learning the questions posted by students need to be categorized on the basis of the concerned concepts. This point to the relevance of question categorization in this area. Many approaches to question classification have been proposed and have achieved reasonable results. The dominant approaches are machine learning and context based classification. There are several Machine Learning methods for question categorization. Here we are extending the previous methods for text categorization to question categorization and making a comparative study of the performance of two approaches, Naïve Bayes and Support Vector Machine General TermsMachine Learning Algorithms
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.