Digital Medical Image Fundamentals
Medical imaging involves radiological, nuclear, photon and positron emission, nuclear magnetic resonance, ultrasound and light photon images. However, concepts and image quality measurement techniques of these various types of image are derived from conventional radiographic imaging and fundamental digital image processing principles. For an extensive review of the these concepts and quality measuring techniques, see: Bankman (2008), Barrett and Swindell (1981), Benedetto, Huang, and Ragan (1990), Bertram (1970), Beutel et al. (2000), Bracewell (1965), Brigham (1979), Cochran et al. (1967), Curry, Dowdey, and Murry (1987), Dainty and Shaw (1984), Dhawan (2003, 2008), Gonzalez and Cointz (1982), Hendee and Wells (1997), Huang (1987, 1996, 1999, 2004), Leondes (1997), Prince and Links (2006), Robb (1995), Rosenfeld and Kak (1976), and Rossman (1969).
2.1.1 Digital Image
A digital image is a two-dimensional array of nonnegative integer function, f (x, y), where 1 ≤ x ≤ M and 1 ≤ y ≤ N, and M and N are positive integers representing the number of columns and rows, respectively. For any given x and y, the small square in the image represented by the coordinates (x, y) is called a picture element, or a pixel, and f (x, y) is its corresponding pixel value. If M = N, then f becomes a square image; most sectional images in a three-dimensional (3-D) image volume used in medicine are square images. If the digital image f (x,