The trace transform for object identification

• Main Author: Leong, Wei Siong.
• Other Authors: School of Computer Engineering
• Format: Final Year Project
• Language: English
• Published: 2012
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
• Online Access: http://hdl.handle.net/10356/48510
• Institution: Nanyang Technological University

We know that Trace transform is is a novel algorithm with properties like scale and rotation invariance and occlusion robustness [1] that has been shown to be effective in recognition tasks using generalisation of Radon transform. However its computation complexity has been a problem. A Trace transform of the image is a 2D function of the parameters of each tracing line. In this project, exploring the uses of Hough transform which extract primitives, such as lines, edges and curves, is often a key step in an image analysis procedure. We designed a function that take in Polar Coordinates (r,θ) inputs and converting to Cartesian Coordinates (x,y). As Hough transform is the simplest example for detection of lines. Experiment result from this function will allow us to get all the pixel and its coordinate from a image base on the given inputs.