USING FOCUS OF ATTENTION WITH THE HOUGH TRANSFORM FOR ACCURATE LINE PARAMETER-ESTIMATION

In this paper we describe a Hough transform algorithm for finding line s in an image that can have an order of magnitude denser sampling in b oth parameters of the parameter space without significant increase in required memory or overhead in CPU time. By increasing the sampling de nsity we determine the line parameters for each line segment much more accurately. The method circumvents the problems of requiring large am ounts of memory to store the accumulator array, and is not significant ly slower than other Hough algorithms by using a focus of attention ap proach. We illustrate the benefits that can be obtained on a wide vari ety of images, and quantify the improvements in line parameter estimat ion by performing statistical tests. We also discuss the benefits that can be obtained by algorithms attempting to locate higher level featu res in the image, and determining the 3D model of the image scene.