Processing of Gray Scale Images
There are two major types of gray scale image processing: within class transformations, such as filtering and image enhancement, and class 1 image to class 2 image transforms, such as segmentation. Most methods for performing such processing use, directly or indirectly, statistics computed on images. We shall discuss two of them, the histogram of distribution of gray levels (in Section 3.2), and the cooccurrence matrix of pairs of gray levels at pixel pairs (in Section 3.3). Their applications in filtering are discussed in Sections 3.4 and 3.5, while their use for segmentation is presented in the next chapter.
KeywordsGray Level High Pass Filter Noisy Image Gray Scale Image Histogram Equalization
Unable to display preview. Download preview PDF.
- [3.AS]Abramatic, J. F. and Silverman, L. M. “Non-Stationary Linear Restoration of Noisy Images,” Proc. 18 th IEEE Decision and Control Conference, vol. 1, (December 1979), pp. 92–99.Google Scholar
- [3.HA]Haralick, R. M. “Statistical and Structural Approaches to Texture,” Proc. Fourth Intern. Joint Conf. on Pattern Recognition, (November 1978), pp. 45–69.Google Scholar
- [3.J1]Julesz, B. Foundations of Cyclopean Perception, Chicago: University of Chicago Press, 1971.Google Scholar
- [3.NM]Nagao, M. and Matsuyama, T. “Edge Preserving Smoothing,” Proc. Fourth Intern. Joint Conf. on Pattern Recognition, (November 1978), pp. 518–520.Google Scholar
- [3.ST]Steiglitz, K. An Introduction to Discrete Systems, New York: J. Wiley, 1974.Google Scholar