Cognitive Systems and Signal Processing in Image Processing

Chapter 3: A web application for crowd counting by building parallel and direct connection-based CNN architectures

Zhilin Hu    Department of Informatics, University of Leicester, Leicester, United Kingdom

Abstract

As deep learning techniques continue to iterate, the trend in the direction of convolutional neural network (CNN) applications is to be more relevant to life; in other words, the technology serves life. The research direction should be from the user’s point of view, and the designed and trained CNN model is combined with the web application to visually display how many people are in a crowd picture on a web page. In order to protect the user’s privacy needs, a registration module is developed, and only by entering the correct ...