Vision-Based System For Child Abuse Action Recognition Using Field Programmable Gate Array
Protecting children from violence is one of the top priorities in all societies. Despite the increasing number of reported abuse cases, no previous efforts have been made to detect and recognise physical abuses against children using vision-based system. Hence, it is substantial to have a promising...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 出版: |
2018
|
| 主題: | |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 總結: | Protecting children from violence is one of the top priorities in all societies. Despite the increasing number of reported abuse cases, no previous efforts have been made to detect and recognise physical abuses against children using vision-based system. Hence, it is substantial to have a promising solution to detect and recognise such actions in real-time with high accuracy so it would help in rescuing the victim child in shortest possible time. Depth sensors and Field Programmable Gate Array (FPGA) devices are highly promising tools in implementing the vision-based human action recognition systems. While depth sensors simplify pre-processing functions, FPGA devices accelerate features computation and classification functions. Thus, this thesis proposes to utilise a depth sensor (Microsoft Kinect) and FPGA-based device to form the basis of the vision-based child abuse recognition system. |
|---|