Please use this identifier to cite or link to this item:
http://localhost:8080/xmlui/handle/123456789/3210
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Khan, Mubashir | - |
dc.contributor.author | Sayed, Najneen Fatma Mustaq Ali [15CO07] | - |
dc.contributor.author | Kothari, Hozefa Abbas [15CO24] | - |
dc.contributor.author | Chougle, Zaid Noornabi [15CO15] | - |
dc.date.accessioned | 2019-08-01T11:54:54Z | - |
dc.date.available | 2019-08-01T11:54:54Z | - |
dc.date.issued | 2019-05 | - |
dc.identifier.uri | http://www.aiktcdspace.org:8080/jspui/handle/123456789/3210 | - |
dc.description.abstract | As action recognition problem is becoming a hot topic within computer vision, the detection of fights or in general aggressive behavior has been comparatively less studied. Such capability may be extremely useful in some video surveillance scenarios like in prisons, psychiatric centers or even embedded in camera phones Our System will be capable enough to detect aggressive behavior in a fight and various objects used in it. It will be using various computer vision technique and deep neural network to detect and recognize objects and actions in the violent outbreak. Our System is mainly dependent on how the model is trained and how good the quality of CCTV camera is, as we will be doing the processing on CCTV feed Keywords: Computer-vision, neural network, violent outbreak, action recognition, object detection, Keras, Numpy, Tensorflow, surveillance video, HMM(Hidden Markov Model) | en_US |
dc.language.iso | en | en_US |
dc.publisher | AIKTC | en_US |
dc.relation.ispartofseries | PE0588; | - |
dc.subject | Project Report - CO | en_US |
dc.title | Violence detection system | en_US |
dc.type | Project Report | en_US |
Appears in Collections: | Computer Engineering - Project Reports |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
PE0588.pdf | 60.66 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.