Edge Computing-Enabled Crowd Density Estimation based on Lightweight Convolutional Neural Network

This paper focuses on crowd motion analysis and consists two parts. Firstly, we propose a new foreground detection approach called optical flow and background model (OFBM) based on Lucas-Kanade optical flow and Gaussian background model methods. This approach overcomes the shortages of optical flow and background subtract, such as sensitiveness of light changing and producing accumulate errors. Secondly, according to moment analysis, we propose a new feature based on the zeroth-order Tehebichef discrete orthogonal moment (TOM), which is employed for crowd density estimation. Some experimental results show that this approach is useful and efficient in crowd density estimation.

The rapid growth of metropolitan populations and large public gatherings has increased the need for intelligent surveillance systems capable of ensuring public safety. This study proposes a novel framework, Intelligent Anomaly Detection and Localization with Crowd Density Estimation (IADL-CDE), which jointly addresses crowd density classification and anomaly detection in real-time. Leveraging the ShanghaiTech dataset for crowd density estimation and the UCSD Pedestrian Database for anomaly detection, the framework integrates a Robust Maximum Correntropy Kalman Filter (RMCKF) for effective image pre-processing, followed by a Self-Modulating Convolutional Neural Network (SMCNN) that accurately classifies crowd density levels into sparse, medium, and dense categories. Subsequently, a Pixel-Associated Auto Encoder (PAAE) is employed to detect and localize anomalies such as cycling, skateboarding, and unauthorized vehicles in pedestrian environments. Experimental results demonstrate that the proposed IADL-CDE model achieves significant improvements in accuracy, precision, recall, F1-Score and Cohen’s Kappa Coefficient compared to the baseline methods. These findings highlight the model’s effectiveness and robustness in enhancing public safety through intelligent crowd monitoring and anomaly detection.

The estimation of crowd density is a crucial area of study in computer vision because of its extensive use in intelligence collection, urban planning, and surveillance. The crowd density prediction that comes from a thorough analysis considers several factors, including inter-blocking in dense crowds, background elements, and individual appearance similarity. To monitor populous areas and avoid congestion, we are interested in using machine learning for crowd control in this research. We suggest using a Single Convolutional Neural Network with Three Layers (S-CNN3) model to estimate the crowd size and count the number of individuals in a scene. Next, a comparison study for density counting determines how well the suggested model performs in comparison to switched convolutional neural networks (SCNN) and convolutional neural networks with four layers (single CNN4). This study makes use of the Shanghai Tech dataset, which is regarded as the biggest database for crowd counting. With an average test accuracy of 99.88% and an average validation loss of 0.02 for crowd density estimate, the suggested model demonstrates remarkable efficacy and efficiency. These outcomes outperform the state-of-the-art models already in use.

Fast crowd density estimation with convolutional neural networks

Due to the rapid increase in urbanization and population, crowd gatherings are frequently observed in the form of concerts, political, and religious meetings. HAJJ is one of the well-known crowding events that takes place every year in Makkah, Saudi Arabia. Crowd density estimation and crowd monitoring are significant research areas in Artificial Intelligence (AI) applications. The current research study develops a new Sparrow Search Optimization with Deep Transfer Learning based Crowd Density Detection and Classification (SSODTL-CD2C) model. The presented SSODTL-CD2C technique majorly focuses on the identification and classification of crowd densities. To attain this, SSODTL-CD2C technique exploits Oppositional Salp Swarm Optimization Algorithm (OSSA) with EfficientNet model to derive the feature vectors. At the same time, Stacked Sparse Auto Encoder (SSAE) model is utilized for the classification of crowd densities. Finally, SSO algorithm is employed for optimal fine-tuning of the parameters involved in SSAE mechanism. The performance of the proposed SSODTL-CD2C technique was validated using a dataset with four different kinds of crowd densities. The obtained results demonstrated that the proposed SSODTL-CD2C methodology accomplished an excellent crowd classification performance with a maximum accuracy of 93.25%. So, the proposed method will be highly helpful in managing HAJJ and other crowded events.

High-density crowd events like public concerts, sporting events, or religious festivals represent significant safety challenges due to high crowd density. Methods of monitoring crowds using manual observation or passive surveillance often don't provide real-time information. Therefore, we present a real-time crowd density estimation solution that utilizes YOLOv8 for people detection and CSRNet for density estimation. The crowd density estimation system utilizes live video feeds from direct surveillance cameras/CCTV or drone footage. The system will assess the crowd density in four levels across four distinct areas: Low, Medium, High, or Critical crowd density. The density estimator has a web-based dashboard that provides real-time analytics reports, heatmap density estimates, and historical records, which can assist in making quick, informed decisions and assessments following an event. The system is validated on benchmark datasets and real-world video streams with 95.3% detection accuracy, 7.4 MAE in crowd counting and 28 FPS processing with off-the-shelf GPU hardware. The results show high accuracy with low latency, making it feasible for real-world applications for large-scale events. The main contributions of the work include using YOLOv8 integrated with CSRNet to jointly detect and estimate crowds, developing a real-time dashboard to provide transparent crowd analytics, and system validation with quantitative metrics and real-world evidence.

Accurately estimate the crowd count from a still image with arbitrary perspective and arbitrary crowd density is one of the difficulties of crowd analysis in surveillance videos. Conventional methods are scene-specific and subject to occlusions. In this paper, we propose a Multi-task Multi-column Convolutional Neural Network (MMCNN) architecture for crowd counting and crowd density estimation in still images of surveillance scenes. The MMCNN architecture is an end-to-end system which is robust for images with different perspective and different crowd density. By promoting MCNN with \(3\times 3\) filter, the MMCNN could utilize local spatial features from each column. Furthermore, the ground truth density map is generated based on Perspective-Adaptive Gaussian kernels which can better represent the heads of pedestrians. Finally, we use an iterative switching process in our deep crowd model to alternatively optimize the crowd density map estimation task and crowd counting task. We conduct experiments on the WorldExpo’10 dataset and our method achieves better results.

Human beings perceive images through their properties, like colour, shape, size, and texture. Texture is a fertile source of information about the physical environment. Images of low density crowds tend to present coarse textures, while images of dense crowds tend to present fine textures. This paper describes a new technique for automatic estimation of crowd density, which is a part of the problem of automatic crowd monitoring, using texture information based on grey-level transition probabilities on digitised images. Crowd density feature vectors are extracted from such images and used by a self organising neural network which is responsible for the crowd density estimation. Results obtained respectively to the estimation of the number of people in a specific area of Liverpool Street Railway Station in London (UK) are presented.

Vanishing region loss for crowd density estimation

Aggregating crowd density in grids from big mobile datasets is a basic but critical work in urban computing and mobile computing. The error of position estimation in raw mobile data, including spatial deviation and temporal deviation, is inevitable and directly impacts the accuracy of aggregated crowd density results. In this case, a key modifiable areal unit problem is raised to understand the relationship among the crowd density accuracy, raw mobile data error, grid shape, and size, but few studies focused on it. This paper analyzes this modifiable areal unit problem of the error in crowd density estimation from big mobility data. By regarding the error as the result of a convolution operation, an optimization model based restoration method was proposed to fix the error of the estimated result, and we analyzed the restoration effect under different circumstances by several simulation experiments. A real application for grided population distribution map construction and restoration from Call Detail Record was conducted to prove the reliability of the whole analysis, which demonstrates the restoration method can reduce the error by nearly 40% under certain conditions.

Estimation crowd density from surveillance video is a significant research filed in the area of computer vision. Crowd density is one of the critical crowd monitoring parameters which can represent the space that is possessed by the crowd in the scene under surveillance. The space available for crowd gathering in a particular area may be limited, and the authorities need to restrict the crowd in that area within that predefined limit. If the density of the crowd exceeds that predefined limit, some congestion may occur in that scene. To avoid this congestion, we can incorporate the density parameter to the existing surveillance system to make the surveillance system more intelligent. In this paper, we propose a new crowd density estimation strategy using deep learning frameworks like convolution neural network and long short-term memory network. The experiment is tested with CUHK Crowd dataset, and the result is compared with existing crowd density estimation methods available in the literature.

Estimating crowd density and counting people are essential for crowd control, urban planning, and public safety. This research study utilizes a Multi-Column Convolutional Neural Network (MC-CNN) as a crowd counting technique trained on crowd datasets. The MC-CNN predicts crowd density maps from input images, providing precise crowd density estimates. Additionally, post-processing methods like clustering analyze the spatial crowd distribution. The proposed system incorporates object detection algorithms to generate individual coordinates, aiding in crowd analysis through clustering and then classifying them into three categories i.e. sparse, medium, and dense. Experimental findings showcase the method’s efficacy in accurately estimating crowd densities and discerning crowd patterns. The proposed system provides valuable insights for crowd monitoring, resource allocation, and decision-making in densely populated areas.

Aiming at the problems of crowd distribution, scale feature, and crowd feature extraction difficulties in exhibition centers, this paper proposes a crowd density estimation method using deep learning for passenger flow detection systems in exhibition centers. Firstly, based on the pixel difference symbol feature, the difference amplitude feature and gray feature of the central pixel are extracted to form the CLBP feature to obtain more crowd group description information. Secondly, use the LR activation function to add nonlinear factors to the convolution neural network (CNN) and use dense blocks derived from crowd density estimation to train the LR-CNN crowd density estimation model. Finally, experimental results show that the mean absolute error (MAE) and mean square error (MSE) of the proposed method in the UCF_CC_50 dataset are 325.6 and 369.4, respectively. Besides, MAE and MSE in part_A of the Shanghai Tech dataset are 213.5 and 247.1, respectively, and they in part_B are 85.3 and 99.7, respectively. The proposed method effectively improves the accuracy of crowd density estimation in exhibition centers.

Crowd density estimation in wide areas is a challenging problem for visual surveillance. Because of the high risk of degeneration, the safety of public events involving large crowds has always been a major concern. In this paper, we propose a video-based crowd density analysis and prediction system for wide-area surveillance applications. In monocular image sequences, the Accumulated Mosaic Image Difference (AMID) method is applied to extract crowd areas having irregular motion. The specific number of persons and velocity of a crowd can be adequately estimated by our system from the density of crowded areas. Using a multi-camera network, we can obtain predictions of a crowd's density several minutes in advance. The system has been used in real applications, and numerous experiments conducted in real scenes (station, park, plaza) demonstrate the effectiveness and robustness of the proposed method.

The task of crowd counting and density map estimation is riddled with many challenges, such as occlusions, non-uniform density, intra-scene and inter-scene variations in scale and perspective. Due to the development of deep learning and large crowd datasets in recent years, most crowd counting methods have achieved notable success. This paper aims to solve crowd density estimation problem for both sparse and dense conditions. To this end, we make two contributions: (1) a network named Patch Scale Discriminant Regression Network (PSDR). Given an input crowd image, it divides the image into patches and sends image patches of different density levels into different regression networks to get the corresponding density maps. It combines all patch density maps to predict the entire density map as the output. (2) A person classification activation map (CAM) method. CAM provides person location information and guides the generation of the entire density map in the final stage. Experiment confirms that CAM allows PSDR to gain another round of performance boost. For instance, on the SmartCity dataset, we achieve (8.6–1.1) MAE and (11.6–1.4) MSE. Our method combining above two methods performs better than state-of-the-art methods.