Automatic detection of visual defects in image intensifiers

Visible defects in Night Vision (NVD) Device images can act as visual distractions and may be large enough to mask critical information of normal night vision operations. In this paper we present a new method for detection of visual defects which will in turn help in the evaluation of Micro Channel Plate used in image intensifiers. The proposed method adopts a hybrid scheme using Circular Hough Transform and Shape classifier with Connected Component Analysis. The statistical and geometrical properties over a connected region of boundaries are explored for the purpose of defect detection. The performance is evaluated based on the noise withstanding capability of the algorithm.