A real-time CAM-based Hough transform algorithm and its performance evaluation

This paper proposes a real-time content addressable memory (CAM)-based Hough transform algorithm for straight line detection and evaluates its performance. Both voting and peak extraction, which compose the Hough transform, are directly executed by CAM. The CAM acts as a processing element (PE) array that performs highly parallel processing for the Hough transform and also as a memory for two-dimensional Hough space. To achieve high-speed processing, voting is executed in every scanning line, not every pixel. The Hough space is mapped into the CAM in folded form to reduce the size of the CAM hardware. Moreover, CAM-based weighted voting achieves more accurate line detection in spite of the quantization error and noise in the image space. Simulations of CAM hardware size, processing time and the accuracy of line detection show that a real-time and high-resolution Hough transform for a 256/spl times/256 picture can be achieved using a single CAM chip with current VLSI technology. This CAM-based Hough transform algorithm promises to be an important step towards the realization of a real-time and compact image understanding system.