The Hough Transform and Its Nature
It has already been seen that the Hough transform can be used to locate straight-line and circle features in digital images. It would be useful to know whether the method can be generalized to cover all shapes and whether it is always as robust as it is for the original two examples. This chapter discusses these questions, showing that the method can be generalized and is broadly able to retain its robustness properties.
• the generalized Hough transform technique.
• its relation to spatial matched filtering.
• how sensitivity is optimized by gradient rather than uniform weighting.
• the effect of applying the generalized Hough transform to line detection.
• how speed has been improved by a variety of methods.
• the value of the Gerig and Klein back-projection technique in cutting down the effects of extraneous clutter.
This chapter not only describes the generalized Hough transform but also generalizes our view of the transform as a generic machine vision technique. The generalized method will be employed in later chapters for the detection of further types of macroscopic image feature.