Behavior Recognition Based on Category Subspace in Crowded Videos

Deng, Chunhua; Kang, Xiaoge; Zhu, Ziqi; Wu, S. M.

doi:10.1109/access.2020.3043412

Cited by 4 publications

(3 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Event representation and anomaly measurement are the two major stages of anomaly detection. Spatial-temporal information can be used for the representation of abnormal events.A few examples are the Behavior Entropy Model (BE) [3], the Histogram of Motion Direction (HMD) [11], the spatial-temporal gradient [10], the chaotic invariant, the mixtures of dynamic textures, and the sparse representation. The majority of methods for measuring anomalies train normal samples using one-class learning techniques.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Real Time Crowd Analysis and Anomaly Detection

AMINA

Binu

2022

Preprint

View full text Add to dashboard Cite

Crowd analysis is gaining a lot of attention in both the technological and social aspects as the population expands. As certain crowdactivity cannot be assessed easily, These techniques differ dependingon the crowd. Strikes, protests, and other crowded events are commonin overpopulated cities. Parades and other public gatherings providea variety of security challenges. Security forces are frequently enlistedto keep an eye on these issues and ensuring the safety of those whoattend meetings. In these circumstances, ensuring the safety of thosewho attend meetings are challenging . Cutting-edge crowd control methods are necessary for an intelligent surveillance system. The goal of theproposed method is to find crowd analysis subfields, such as anomalydetection, that have not yet been well explored or that seem to beaddressed infrequently. Public data sets for indoor and outdoor useare used to evaluate the proposed system’s bespoke architecture. The Experimental findings demonstrate that the custom architecture performs better than the exixting Architectures.Our aim is to providerecent state-of-the-art algorithms to provide a novel strategy that canhelp with improved crowd surveillance and crowd activity analysis.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Crowd recognition has been the subject of numerous studies. This study [10] utilised a hybrid tracking model and a neural network with a genetic algorithm to identify crowd behaviour. The features extracted in this work [11] include conditional increment, decrement, irregular and speed, variance, and entropy.…”

Section: Related Workmentioning

confidence: 99%

Real Time Crowd Analysis and Anomaly Detection

AMINA

Binu

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Although there is a requirement for features in the dynamic sense for an efficient analysis of crowd. (4) A framework of deep learning that involves semi-supervision is proposed for the examination of events that are abnormal in crowd gatherings. This deep learning mechanism aids in distinguishing crowd behaviour irregularities.…”

Section: Introductionmentioning

confidence: 99%

Crowd Behavior Analysis and Prediction using the Feature Fusion Framework

Murthygowda¹,

Krishnegowda²,

Shashikala³

2022

Salud, Ciencia y Tecnología

View full text Add to dashboard Cite

The increasing number of people is a major cause of disasters that occur due to overcrowding. The gatherings of crowds in public places are a source of panic, which results in disaster. An analytical study was performed on crowd management. This is highly essential for the design of a well-planned public space, the possibility of surveillance in every area, and transportation systems. The disasters that occur due to uncontrollable crowd behaviour involve loss of property, fatalities, or casualties. To avoid this, the crowd’s behaviour was analysed. A MFF (multi-level feature fusion) framework was designed in this paper to predict behaviour. The first level of multi-level feature fusion employs motion and appearance, the second level employs spatial connections, and the third level employs temporal features. The combination of these characteristics aids in the exploitation of crowd behaviour. Furthermore, MFF was evaluated considering the web dataset, considering accuracy, precision, and recall as parameters. Comparative analysis was carried out with various existing methodologies with an accuracy of above 99 %.

show abstract

Real-time crowd behavior recognition in surveillance videos based on deep learning methods

Rezaei

Yazdi

2021

J Real-Time Image Proc

View full text Add to dashboard Cite

Behavior Recognition Based on Category Subspace in Crowded Videos

Cited by 4 publications

References 44 publications

Real Time Crowd Analysis and Anomaly Detection

Real Time Crowd Analysis and Anomaly Detection

Crowd Behavior Analysis and Prediction using the Feature Fusion Framework

Real-time crowd behavior recognition in surveillance videos based on deep learning methods

Contact Info

Product

Resources

About