An Infrared Image Segmentation Algorithm Based on Spatial Correlative Information
Image segmentation is an important technique for image processing and computer vision. The principles of 1-D Otsu’s algorithm and thresholding through index of fuzziness are described. Since the infrared images of tank have low object-background contrasts and blurred boundaries in the complex background condition, an adaptive algorithm for image thresholding through index of fuzziness, which is combined with the spatial correlative information, is proposed. The new method makes full use of the spatial correlation of pixels, so that it can extract the detail of the image from the complex background effectively, and improve the accuracy of the segmentation. The results of experiments prove that the presented algorithm has better performance and better robustness against noise.
K. Y. Wang et al., "An Infrared Image Segmentation Algorithm Based on Spatial Correlative Information", Applied Mechanics and Materials, Vols. 44-47, pp. 3274-3278, 2011