A Randomized Curve Detection Platform Based on Design Patterns

• Main Authors: Wei Han Chiou, 邱暐瀚
• Other Authors: J S Wang
• Format: Others
• Language: zh-TW
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/90915729065523999769

碩士 === 國立臺北科技大學 === 電腦通訊與控制研究所 === 88 === Curve detection is an important research area in computer vision. In developing efficient and effective curve detection methods, it is necessary to conduct large numbers of experiments on randomly generated images. Such a process can be very tedious and time consuming. In this thesis, a platform for developing curve detection methods based on randomized Hough transform (RHT) is designed and implemented. The objective is to expedite the process of developing RHT-based curve detection methods. In so doing, we have made use of design patterns to create a framework that provides support for random image generation and result visualization. With this platform, developers can easily plug in new RHT-based methods and conduct experiments. To demonstrate this, two examples using RHT and coaxal transform are given.