An Identification Algorithm for Underwater Vehicle Infrared Wake Based on GLCM Minimum Difference of Entropy

摘要

In order to distinguish the underwater vehicle from the surface vehicle by their infrared thermal wakes, an identification algorithm for underwater vehicle infrared wake based on GLCM (Gray Level Co-occurrence Matrix) minimum difference of entropy is proposed. First, the infrared image is processed and the wake area is segmented, to construct the GLCM which only contains the information of the wake in the next step. Then, four GLCMs are generated by four different positional relationships of the pixel pairs. Finally, the entropy values of the GLCMs corresponding to the vertical pixel pairs (0° and 90°, 45° and 135°) are subtracted and the absolute values are normalized. The minimum value of the two values is selected for judgment. If it is less than 0.02, the wake is identified as the wake of the underwater vehicle, otherwise it is determined as the wake of the surface vehicle. Experimental results show that the algorithm proposed in this paper can effectively determine the category of infrared wakes.