Email RIS file
Bookmark
Report an Issue

Interactive video surveillance as an edge service / by Seyed Yahya Nikouei and Yu Chen

Author
Nikouei, Seyed Yahya
Published
Bellingham, Washington (1000 20th St. Bellingham WA 98225-6705 USA) : SPIE, 2024.
Physical Description
1 online resource (59 pages) : illustrations
Additional Creators
Chen, Yu and Society of Photo-Optical Instrumentation Engineers
Access Online

Availability

I Want It
I Want It

Finding items...

Series
Restrictions on Access
Restricted to subscribers or individual electronic text purchasers.
Contents
Preface -- 1 Introduction -- 1.1 Migration from cloud to edge -- 1.2 Key components -- 2 Video Surveillance for Public Safety: An Overview -- 2.1 Human detection at the edge -- 2.2 Object tracking at the edge -- 2.3 Behavior recognition -- 3 Interactive Surveillance -- 3.1 Rationales and design principles -- 3.2 Enabling technologies -- 3.3 Object tracking -- 3.4 Fuzzy mathematics -- 4 Loitering Detection: A Case Study -- 4.1 System overview -- 4.2 Key design tradeoffs -- 4.3 Implementation -- 4.4 Performance and evaluation -- 4.5 Limitations and open issues -- 5 Conclusions and Future Opportunities -- References
Summary
"The safety of the community is a vital issue as the crime rate is reaching 50 per 100,000 people in the United States. The surveillance communities are looking into new ways to protect members and possibly help with the reduction in the numbers. Today, surveillance video analysis mostly relies on teams of specially trained officers manually watching thousands of hours of video captured in real-time surveilled locations, while looking into historical records in case of an incident in locations with lower priority. Starting with the background and fundamentals of state-of-the-art video surveillance for public safety in the era of Smart Cities, this Spotlight Series book highlights the rationale, principle, and case study of interactive video surveillance as an edge service based on unsupervised feature queries. We demonstrate the use of readily accessible IoT hardware and showcase how compact computing units can efficiently execute deep learning models for near real-time to real-time performance in object detection and behavior classification tasks."--
Subject(s)
Genre(s)
ISBN
9781510668492 PDF
Related Titles
SPIE digital library
Bibliography Note
Includes bibliographical references.
Other Forms
Also published in print version.
Technical Details
Mode of access: World Wide Web.
System requirements: Adobe Acrobat Reader.
View MARC record | catkey: 43976664