Edge Detection using Directional Filter Bank

Finding edges in digital images is an essential and important task in many imaging applications. This paper describes an edge detection using multi scale Directional Filter Bank (DFB). This method tries to solve two difficulties that edge finding algorithms must face: limited directions and combining the detected edges at different scales. The directional responses of DFB that represent the edge information can be used for edge detection. The steerable and scaled DFB has been presented to obtain directional along with scaled information. From the DFB based image decomposition, the scaled information is combined by scale multiplication. Finally, to evaluate edge detector performance the sensitivity, specificity, accuracy and Figure of merit ‘F’ parameters are used to compare with classical methods and the proposed approach provides better performance.