ABSTRACT
This chapter focuses on some of the fundamental image processing methods for grayscale images. It draws on fundamental methods in image processing would be incomplete without a discussion on histogram-based methods. The chapter explains the image’s histogram, including the definition of the histogram, histogram statistics, and modification of the histogram to achieve segmentation and contrast enhancement (CE). It is based on the types of degradations present in images and ways to reduce these effects via denoising or enhancement. The chapter also focuses on CE and edge-preserving smoothing where the performance of these algorithms are objectively quantified by two image quality assessment techniques. It deals with the definition of histogram statistics, histogram thresholding for image segmentation, CE via histogram equalization and a robust edge strength measure that is obtained with the histogram of the gradient values. Many imaging modalities produce images of superior quality, noise and/or low-contrast objects can cause challenges in automated recognition tasks.
