ناشر تخصصی کنفرانس های ایران

لطفا کمی صبر نمایید

Publisher of Iranian Journals and Conference Proceedings

Please waite ..
Publisher of Iranian Journals and Conference Proceedings
Login |Register |Help |عضویت کتابخانه ها
Paper
Title

A new method for anomaly detection and localization in crowded scenes using deep laerning networks

Year: 1399
COI: MECECONF02_070
Language: EnglishView: 118
This Paper With 11 Page And PDF Format Ready To Download

Buy and Download

با استفاده از پرداخت اینترنتی بسیار سریع و ساده می توانید اصل این Paper را که دارای 11 صفحه است به صورت فایل PDF در اختیار داشته باشید.
آدرس ایمیل خود را در کادر زیر وارد نمایید:

Authors

Maedeh Bahrami - Department of electrical engineering, Yazd branch, Islamic Azad University,Iran.
Majid Pourahmadi - Department of electrical engineering, Yazd branch, Islamic Azad University,Iran.
Abbas Vafaei - School of Computer Engineering, University of Esfahan,Iran

Abstract:

Video anomaly detection is the learning of normal pattern in image sets. This is in contrast to most classifications which can guide the algorithm through the use of labels. Most deep learning based video anomaly detection techniques involve some form of video reconstruction. In other words, the network learns to reconstruct the video in training and lacks anomalies in training set. In the test phase, reconstruction happens again and it is expected that the network preserve normal pattern and degrade anomalies. Video anomaly detection is usually studied by considering the spatial and temporal contexts. This paper focuses on spatial context only and shows that while the overall detection rate falls without temporal context, localization gets better. There are two main contributions: employing a new deep network for reconstruction and introducing a new regularity score function. The new deep architecure is based on pyramid of input images and compared to UNet, this new architecture boosts AUC by ۱۵%. The new regularity scoring function is based on SSIM which in turn results in a noticable performance gain.

Keywords:

Paper COI Code

This Paper COI Code is MECECONF02_070. Also You can use the following address to link to this article. This link is permanent and is used as an article registration confirmation in the Civilica reference:

https://civilica.com/doc/1133792/

How to Cite to This Paper:

If you want to refer to this Paper in your research work, you can simply use the following phrase in the resources section:
Bahrami, Maedeh and Pourahmadi, Majid and Vafaei, Abbas,1399,A new method for anomaly detection and localization in crowded scenes using deep laerning networks,2nd Monaco Conference on Mechanical, Electrical, Computer and Engineering Sciences,https://civilica.com/doc/1133792

Research Info Management

Certificate | Report | من نویسنده این مقاله هستم
این Paper در بخشهای موضوعی زیر دسته بندی شده است:

اطلاعات استنادی این Paper را به نرم افزارهای مدیریت اطلاعات علمی و استنادی ارسال نمایید و در تحقیقات خود از آن استفاده نمایید.

Scientometrics

The specifications of the publisher center of this Paper are as follows:
Type of center: Azad University
Paper count: 7,455
In the scientometrics section of CIVILICA, you can see the scientific ranking of the Iranian academic and research centers based on the statistics of indexed articles.

New Papers

Share this page

More information about COI

COI stands for "CIVILICA Object Identifier". COI is the unique code assigned to articles of Iranian conferences and journals when indexing on the CIVILICA citation database.

The COI is the national code of documents indexed in CIVILICA and is a unique and permanent code. it can always be cited and tracked and assumed as registration confirmation ID.

Support