Explore chapters and articles related to this topic
An Enhanced and Dynamic Crowd Estimation System
Published in Durgesh Kumar Mishra, Nilanjan Dey, Bharat Singh Deora, Amit Joshi, ICT for Competitive Strategies, 2020
Niranjan C. Kundur*, Praveen M. Dhulavvagol, P. B. Mallikarjuna, M. Sreenatha
The aim of Crowd counting is to count the exact number of people in the mass images. Counting crowd is one of the biggest challenge that every researchers are facing today. Huge varieties in scale and presence of the objects which as serious twisting of the scene makes it troublesome recognize and count the crowd. Head count is of most importance to politicians and event organizers. Estimating large numbers accurately is very difficult even if the intention is right. If we start counting the jellybeans in jar three or four times to get the correct count, we would probably end up with three or four different count. This is because we cannot count large numbers without any error. Now, counting huge mass of crowd that has distortion is extremely difficult. It’s important to provide public safety by giving a precise and robust crowd count evaluation [1].
A point and density map hybrid network for crowd counting and localization based on unmanned aerial vehicles
Published in Connection Science, 2022
Lei Zhao, Zhengwei Bao, Zhijun Xie, Guangyan Huang, Zeeshan Ur Rehman
The most common supervised method in crowd counting is to employ the density map first proposed by Lempitsky and Zisserman (2010), which sums up the prediction density map as the counting result. In recent years, the accuracy of crowd counting methods has been enhanced through network construction and density map innovations. Zhang et al. (2016) employs a multi-branch network structure with different convolutional kernel sizes to address the scale variations of the crowd. Sam et al. (2017) uses a multi-branch network like (Zhang et al., 2016), but the difference is that (Sam et al., 2017) includes a switch layer to determine which branch to generate the prediction density map. Li et al. (2018) and Wang et al. (2022) use dilated convolutions with different dilation rates to obtain deeper features and larger perceptual fields to adapt scale variations. Cao et al. (2018) employs a multi-branch convolutional module at each stage to extract features at various scales and then uses deconvolutions to generate the predicted density map. Jiang and Jin (2019) and Fan et al. (2020) proposed improvement schemes for generating density maps in the feature extraction and loss supervised stages, respectively.
Deep Learning Technique Based Surveillance Video Analysis for the Store
Published in Applied Artificial Intelligence, 2020
Qingyang Xu, Wanqiang Zheng, Xiaoxiao Liu, Punan Jing
Nowadays, video surveillance system is an essential part of the store. Surveillance video contains abundant information, such as customer preferences, etc. The intelligent surveillance video analysis technique can be used for crowd estimation and gathering intelligence for further analysis and inference such as gauge people’s interest in a product of a store and this information can be used for appropriate product placement, etc. Therefore, the commercial prediction problem can be realized based on the analysis of earlier surveillance video of the store. Estimating the customer preference and drawing density maps by the surveillance video are a crowd counting and density estimation problem. Crowd counting aims to count the number of people in a crowd image and the density estimation aims to map the corresponding density map of crowd image which indicates the density of people at certain position (Sindagi and Patel 2018).
MHAMD-MST-CNN: multiscale head attention guided multiscale density maps fusion for video crowd counting via multi-attention spatial-temporal CNN
Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2023
Santosh Kumar Tripathy, Subodh Srivastava, Rajeev Srivastava
Recent years have witnessed the exponential growth of the worldwide population, which has caused a tremendous increase of crowd formations in places like rallies, stadiums, public speeches, pilgrim places, and malls. Such places are prone to crowd disasters in very dense crowds or panic situations. In such a case, video-based crowd counting and density estimation (CCDE) has become a valuable surveillance tool to provide essential crowd density information, which can be used to manage crowds and also helps to minimise casualties. However, occlusion, cluttered crowd background, varying crowd scenarios, illumination or daylight changes, varying crowd head shapes due to perspective distortion, and varying viewing angle make the video-based CCDE a tedious and still a challenging task.