Module 4

1. What is a digital image? Explain different types of images.

Answer:

A digital image is a numerical representation of a visual scene. It is composed of a finite number of picture elements called pixels, each of which has a specific location (coordinates) and a value (intensity). These pixel values are stored in a computer as a matrix of numbers, allowing them to be processed, stored, and transmitted digitally.

Different types of images are:

1. Binary Images:
   • These are the simplest type of images, where each pixel can have only one of two possible values, typically 0 (Black) or 1 (White).
   • They are also known as 1-bit images because only one bit is required to represent each pixel.
   • Applications: Digital signatures, barcodes, simple logos, and masks in image processing.
2. Grayscale Images:
   • In these images, each pixel carries only intensity information. The pixel value typically ranges from 0 (pure black) to 255 (pure white), with intermediate values representing different shades of gray.
   • They are often called 8-bit images because 2⁸ = 256 intensity levels.
   • Applications: Black and white photography, medical imaging (like X-rays), and many scientific imaging techniques.
3. Color Images:
   • These images contain color information for each pixel. The most common model is the RGB (Red, Green, Blue) model, where each pixel is a combination of its Red, Green, and Blue intensity values.
   • A true-color image is typically 24-bit, with 8 bits for each of the three color channels (resulting in over 16 million possible colors).
   • Applications: Digital photos, web graphics, and all forms of digital color media.
4. Multispectral and Hyperspectral Images:
   • These images capture information from across the electromagnetic spectrum, far beyond just the visible light (RGB). Each pixel contains data for many wavelength bands (spectral bands).
   • Applications: Satellite remote sensing, astronomical imaging, and environmental monitoring to identify materials based on their spectral signatures.

2. What is the principle of image compression?

Answer:

The fundamental principle of image compression is to reduce the amount of data required to represent a digital image by eliminating redundant and/or irrelevant information. The goal is to store or transmit an image in an efficient form without significantly degrading its perceptual quality.

This principle is based on three main types of data present in an image:

1. Coding Redundancy: This refers to the use of sub-optimal code words to represent pixel values. For example, using 8 bits for every pixel in an image with only 16 colors is inefficient. Compression techniques like Huffman coding or Arithmetic coding assign shorter codes to more frequent pixel values and longer codes to less frequent ones, reducing the average code length.
2. Spatial and Temporal Redundancy: Pixels in an image are often highly correlated with their neighbors. In a picture of a blue sky, many adjacent pixels have very similar values. This is called spatial redundancy. In video, consecutive frames are very similar (temporal redundancy). Compression algorithms like Run-Length Encoding (RLE) or transform-based methods (e.g., DCT) exploit this by describing large, uniform areas more efficiently.
3. Irrelevant Information: The human visual system (HVS) is less sensitive to certain types of information. For instance, we are less sensitive to very high-frequency changes and small color details. Lossy compression techniques deliberately discard this perceptually irrelevant information to achieve much higher compression ratios.

3. Explain different approaches to image compression.