Divyesh Tharakan, Dr. Nilima Kulkarni, Harshil Shah, Bilwa Vyas
DOI Number:
10.6084/m9.figshare.13650128
Keywords:
CNN ,Crowd Counting, Estimation, Prediction.
Abstract
Increase in population all over the world has opened a pathway to large gatherings for many causes. It might be grand openings of hotels and institutions, award ceremony, local football matches, riots and much more. Analysing the crowd formation before is quite a difficult task and therefore would not come under control. Due to this, many accidents have occurred in the past taking away innocent lives. The technique, which is Crowd Counting, is an approach to analyse and detect the crowd, that is, to estimate the number of people gathered in a certain area. In this paper, a CNN model is presented to estimate the count of a particular crowd. We have implemented the VGG-16 model for a deep layered image classification and henceforth a better understanding of each image. Shanghai Crowd Counting Dataset has been used which contains more than 400 images for training and testing.
