摘要: Proposed in 1962, the Hough transform (HT) has been widely applied and investigated for detecting curves, shapes, motions fields of image processing computer vision. However, HT several shortcomings, including high computational cost, low detection accuracy, vulnerability to noise, possibility missing objects. Many efforts target at solving some problems decades, while key idea remains more or less same. 1989 further developed thereafter, Random-ized Transform (RHT) manages considerably overcome these shortcomings via innovations on fundamental mechanisms, with random sampling place pixel scanning, converging mapping diverging mapping, dynamic storage accumulation array. This article will provides an overview advances applications RHT past one half decades.