Detection in human crowd had been an important task requiring multiple visual tasks, such as tracking, counting, anomaly detection and action recognition. This problem is challenging due to its low resolution, small apparent size, non-uniform density and critical occlusions of the objects. Convolutional Neural Network (CNN) is a class of Deep Neural Network which has been steadily increased the performance of detection. The main aim of this paper is to detect the crowd from the street, evaluate the density level and count individual from the crowd. The Convolutional Neural Network (CNN) used in the system had demonstrated noticeable improvements as compared to Support Vector Machine (SVM), Viola Jones, Histogram of Oriented Gradients (HOG), and Local Binary Pattern (LBP).
CITATION STYLE
Chaudhari, M. D., & Ghotkar, A. S. (2019). Human Face Detection in Crowd and Density Analysis Using Neural Network Approach. In Lecture Notes on Data Engineering and Communications Technologies (Vol. 26, pp. 1208–1215). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-03146-6_141
Mendeley helps you to discover research relevant for your work.