Infrared Electric Image Thresholding Using Two-Dimensional Fuzzy Renyi Entropy

Infrared thermograph is of great significance in electric equipment monitoring, but due to both the limitation in thermograph technology and surrounding interferences, infrared electric images are always fuzzy by nature, and thus the segmentation of infrared electric image is a challenging task. To handle this ambiguity, we first calculate the two-dimensional histogram of image to make full use of spatial information, and then transform the two-dimensional histogram into fuzzy domain employing fuzzy membership. A new kind of two-dimensional fuzzy entropy, namely two-dimensional fuzzy Renyi entropy, is defined and employed to compute the fuzzy entropy of object and background respectively, and thus the image is segmented following maximum entropy principle. Compared with other typical methods by experiments, the presented method is verified to be more effective and robust.