DSpace Repository

Stolen object detection from fixed video surveillance cameras

Show simple item record

dc.contributor.advisor Dailey, Matthew
dc.contributor.author Soe, Myo Thu
dc.contributor.other Guha, Sumanta
dc.contributor.other Ekpanyapong, Mongkol
dc.date.accessioned 2015-12-14T04:51:38Z
dc.date.available 2015-12-14T04:51:38Z
dc.date.issued 2015-12
dc.identifier.other AIT
dc.identifier.uri http://www.cs.ait.ac.th/xmlui/handle/123456789/803
dc.description 39 p en_US
dc.description.abstract This research study develops a stolen object detection system that is based on motion detection with improved Camshift tracking algorithm. The algorithm takes the first frame of the video as a background model and calculates difference with the current frame. All the motion regions in the frame are tracked as they are subtracted from the background, but only objects in first frame are classified as stolen objects or not. The system was tested on object stealing scenarios captured in indoor and outdoor environments and found to be robust against occlusion but some harsh lighting changes and shadow conditions may impair the detection of the system. en_US
dc.description.sponsorship AIT Fellowship en_US
dc.publisher AIT en_US
dc.subject Stolen object en_US
dc.subject surveillance camera en_US
dc.subject object tracking en_US
dc.subject object detection en_US
dc.subject removed object en_US
dc.subject motion detection en_US
dc.subject image processing en_US
dc.title Stolen object detection from fixed video surveillance cameras en_US
dc.type Research report en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Advanced Search

Browse

My Account