Feature detection in scale-space
The treatment in previous chapters gives a formal justification for using linear filtering as an initial step in early processing of image data. More important, it provides a catalogue of what filter kernels are natural to use, as well as an extensive theoretical explanation of how smoothing kernels of different order and at different scales can be related. In particular, the discretization problem is extensively treated. This forms the basis for a theoretically well-founded modelling of the smoothing operation.
KeywordsDirectional Derivative Local Extremum Gradient Direction Gradient Magnitude Differential Invariant
Unable to display preview. Download preview PDF.