Vision-Based Activity Tracking and Event Detection to Locate Forgotten Objects at Home
碩士 === 國立中正大學 === 電機工程研究所 === 105 === Recently, the advancement of electronic information processing and inter-networking technologies have fostered the concept of intelligent home. Among many initiatives that have been brought up for intelligent home, an important one is to exploit the cameras that...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2017
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/geqr83 |
| Summary: | 碩士 === 國立中正大學 === 電機工程研究所 === 105 === Recently, the advancement of electronic information processing and inter-networking technologies have fostered the concept of intelligent home. Among many initiatives that have been brought up for intelligent home, an important one is to exploit the cameras that were already deployed for home security. Along this line, we propose to utilize these cameras to help user locate missing objects at home. Algorithms are proposed to establish the geographical relation between the user and the object, and to record the potential locations based on calibrated coordinates. For the cases where the object is blocked from video recording, we also propose to predict the potential locations using the on-line learning technique. Experimental results on a prototype recording demonstrate that the proposed system satisfactorily fulfills the purpose of providing potential locations of missing objects.
|
|---|