Embedded GPU-accelerated pedestrian detection for advanced driving assistance systems | |
| Author | Harischandra, Patikiri A.D. |
| Call Number | AIT Caps. Proj. no.EL-15-17 |
| Subject(s) | Driver assistance systems Pedestrians Protection |
| Note | A capstone project report submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Engineering Electronic Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Caps. Proj. ; no. EL-15-17 |
| Abstract | Advanced driving assistance systems (ADAS) aids the driver by providing alerts, warnings about possible collisions or lane departures real time so that the driver could take necessary maneuvers avoid any accidents. Pedestrian detection is an imperative area of ADAS which necessitates high speed computing. One way of getting higher speed is by increasing clock speed of a CPU (central processing unit). However the CPU speed is limited by gate delays and heat. In contrast, GPUs are prevalent for utilizing parallel processing in order to increase throughput. Typically GPUs are considered to be power hungry. However the emerge of Embedded GPUs made high speed computation with low power consumption is possible. In this research an Embedded-GPU is utilized for pedestrian detection by Histogram of Oriented Gradients descriptors. The results demonstrate an increase of detection speed by 3x compared to CPU implementation. |
| Year | 2015 |
| Corresponding Series Added Entry | Asian Institute of Technology. Caps. Proj. ; no. EL-15-17 |
| Type | Project |
| School | School of Engineering and Technology (SET) |
| Department | Bachelor Degree |
| Academic Program/FoS | Electronic Engineering (EL) |
| Chairperson(s) | Mongkol Ekpanyapong; |
| Examination Committee(s) | Dailey, Matthew;Manukid Parnichkun ; |
| Degree | Capstone Project (B.Sc.)-Asian Institute of Technology, 2015 |