Improved Line Detection based-on Pixel Coordinates: A New Approach to Line Detection

Determining the location and orientation of straight lines in images is a problem of great importance in the field of image processing. The Hough transform, has been widely used to solve this problem for binary images, and is still by far dominating the scene when it comes to line detection and its concept was further developed to detect other shapes like circles. In this paper, a new method for straight line detection is proposed. The formulation of this new method is based on the use of the coordinates of pixels in the digital image, which relates the parameters determining the location and orientation of the lines in the image. The advantage of this method is that it provides an approach to the problem of line detection within a framework that enhances the detection of lines by incorporation of prior information of the images nature. This approach proves useful and efficient in computational schemes where it is needed to detect lines in 0, 45, 90, and 135 degree directions, and results in optimized performance. Experimental results of this new approach shows how the image coordinates method can be viewed as an alternate optimal method for noisy images and dotted lines in terms of detection accuracy and accumulator size reduction. Keywords—line detection; Hough Transform; digital image; processing; line location; line direction.