An Infrared Image Preprocessing Method Using Improved Adaptive Neighborhoods

摘要

By analyzing features of infrared images, a new image preprocessing method based on adaptive neighborhoods is presented. The improved filter establishes a variable neighborhoods whose size and shape depend on statistical properties around the given pixel. The method appends an inspection to impulsive noise and filters Gaussian noise used median filter. And to accelerate the algorithm, a size limitation is added to the adaptive neighborhoods. At last an antinoise filter based on adaptive neighborhoods and a local contrast enhancement function are presented. Results show that this infrared image preprocessing method enhances contrast and details in the image, suppresses noise rised synchronously and makes the visual effect of infrared image better.