Due to the significance of edge detection in image analysis, object recognition and many applications, many edge detection techniques are used to detect edges of objects in the image. Our aim is to estimate the CPU and memory requirements of edge detection on the TMS320C6416 platform to verify the feasibility of implementing DSP-based edge recognition systems. The experimental results demonstrate that a generic C implementation with a modest C level optimization effort consequence in a edge detection software prototype that has low CPU and memory necessities on TMS320C64x, and runs with high speed to facilitate realtime applications. Therefore, well optimized edge detection implementations on TMS320C64x are an effective design option for embedded edge detection systems.
Volume 11 | 02-Special Issue
Pages: 1326-1333