Intelligent video surveillance with abandoned object detection and multiple pedestrian counting

2010 ◽  
Author(s):  
Taekyung Kim ◽  
Joonki Paik
Keyword(s):  
Object Detection ◽  
Video Surveillance ◽  
Intelligent Video Surveillance ◽  
Abandoned Object ◽  
Pedestrian Counting

scholarly journals Abandoned Object Detection in Video-Surveillance: Survey and Comparison

Sensors ◽  
2018 ◽  
Vol 18 (12) ◽  
pp. 4290 ◽  
Author(s):  
Elena Luna ◽  
Juan San Miguel ◽  
Diego Ortego ◽  
José Martínez
Keyword(s):  
Object Detection ◽  
Video Surveillance ◽  
Moving Objects ◽  
Automatic Monitoring ◽  
Experimental Comparison ◽  
Public And Private ◽  
Extensive Evaluation ◽  
Abandoned Object ◽  
Configuration System ◽  
The Impact

During the last few years, abandoned object detection has emerged as a hot topic in the video-surveillance community. As a consequence, a myriad of systems has been proposed for automatic monitoring of public and private places, while addressing several challenges affecting detection performance. Due to the complexity of these systems, researchers often address independently the different analysis stages such as foreground segmentation, stationary object detection, and abandonment validation. Despite the improvements achieved for each stage, the advances are rarely applied to the full pipeline, and therefore, the impact of each stage of improvement on the overall system performance has not been studied. In this paper, we formalize the framework employed by systems for abandoned object detection and provide an extensive review of state-of-the-art approaches for each stage. We also build a multi-configuration system allowing one to select a range of alternatives for each stage with the objective of determining the combination achieving the best performance. This multi-configuration is made available online to the research community. We perform an extensive evaluation by gathering a heterogeneous dataset from existing data. Such a dataset allows considering multiple and different scenarios, whereas presenting various challenges such as illumination changes, shadows, and a high density of moving objects, unlike existing literature focusing on a few sequences. The experimental results identify the most effective configurations and highlight design choices favoring robustness to errors. Moreover, we validated such an optimal configuration on additional datasets not previously considered. We conclude the paper by discussing open research challenges arising from the experimental comparison.

scholarly journals A Fuzzy rule- based Abandoned Object Detection using Image Fusion for Intelligent Video Surveillance Systems

2021 ◽  
Vol 12 (3) ◽  
pp. 3694-3702
Author(s):  
Preetha K G Et.al
Keyword(s):  
Object Detection ◽  
Video Surveillance ◽  
Fuzzy Rule ◽  
Super Resolution ◽  
Threat Assessment ◽  
Video Enhancement ◽  
Shopping Malls ◽  
Railway Stations ◽  
Rule Based ◽  
Abandoned Object

Abandoned object/luggage is a major threat in all public scenes like hospitals, railway stations, airports and shopping malls.  Abandoned luggage may contain explosive, biological warfare or smuggled goods. Abandoned object detection is the process to identify the unattended strange object within a specific time.  It is also crucial to identify the person who has abandoned the luggage in the scene. Video surveillance is one of the essential techniques for automatic video analysis to extract crucial information or relevant scenes. The main objectives of this work is the automatic detection of abandoned objects and related persons in public areas like airports, railway stations, shopping malls etc.  Video enhancement techniques like residual dense networks are adopted to improve the quality of the image before applying it to detect the abandoned objects and related humans. The scenario of abandoned objects and related humans are identified through distance differencing methods.  Once the scene is identified, the method is capable of producing alert messages or alarms in real-time through automated means.  A fuzzy rule based threat assessment module is also incorporated in this work which reduces the false alarm rate. The related person is identified through reconstruction of the face through super-resolution techniques. Experiments are found to be appreciable in terms of the metrics in video enhancement, detection, fuzzification and face super-resolution.

Sign in / Sign up

Export Citation Format

Share Document