License plate character segmentation using clahe enhancement and connected edge components

License plate images usually suffers from low illumination and poor contrast due to motion of vehicles and large depth of fields. Therefore, license plate image segmentation and number extraction is a challenging task. This paper proposed an fast and efficient license plate character segmentation method using CLAHE pre-enhancement and by finding the morphologically connected region. CLAHE method is used as pre-enhancement stage to improve the contrast and illumination of number plate images. This helps to improve the performance of character segmentation. In the second stage the wiener filter is used to remove the noise and blur present in the image. The proposed method uses the label connected neighbourhood with 8-conect mask to segment the desired characters form the edge image. The shift invariant edge detector mask is used for detecting the edges. Using the combination of pre enhancement and filtering improves the convergence rate of the standard segmentation method. The proposed character segmentation method is tested on the various kinds of licence plate images. It is found that entropy of segmented object is improved with the proposed method.