Finding Edge Features Using the Fast Level Set Transform and the Helmholtz Principle

There is a need to develop automated methods to analyze images. A method that relies on the Helmholtz principle to discover edge features that is weakly dependent on the parameters is presented. An application of the Helmholtz principle requires a method of finding groups of independent and identically distributed random variables, and the fast level set transform was used to find such groups. The ability of the method to locate contrasted sections of curves is reviewed, and an original technique to find sections of curves with uniform curvature is demonstrated.