Summary of "DIP#8 Sampling and Quantisation of Digital image || EC Academy"

Summary of "DIP#8 Sampling and Quantisation of Digital Image || EC Academy"

This lecture focuses on the fundamental concepts of sampling and Quantisation in the context of converting continuous images into digital form. It builds upon previous discussions related to image digitization and explains how continuous signals (images) are transformed into discrete digital data.

Main Ideas and Concepts:

Image Sampling and Quantisation:
- An image, originally continuous in both spatial dimensions (x and y axes) and amplitude (intensity), needs to be converted into a digital signal.
- Sampling involves selecting discrete points (samples) from the continuous image along spatial coordinates.
- Quantisation refers to converting the continuous amplitude values at these sample points into discrete digital values (levels).
Continuous to Discrete Conversion:
- The continuous image function is sampled at specific intervals along both axes.
- Each sample point's amplitude is then quantised into a finite set of digital levels.
- This process results in a digital representation of the image, which can be stored, processed, or transmitted digitally.
Sampling Function:
- Sampling locations are represented by vertical tick marks on the image.
- The sampled data points form a sequence of discrete values.
- The density of sampling points (number of samples) directly affects the quality of the Digital Image.
Quantisation Levels:
- The amplitude at each sample is mapped to a Quantisation level.
- More Quantisation levels mean finer amplitude resolution and better image quality.
- Fewer levels lead to loss of detail and potential distortion.
Quality of Digital Image:
- The quality depends on the sampling rate (number of samples per unit area).
- Increasing the number of samples improves spatial resolution.
- Increasing the number of Quantisation levels improves amplitude resolution.
- There is a trade-off between image quality and data size.
Illustrations and Examples:
- The video uses figures showing continuous images and their sampled and quantised versions.
- One-dimensional and two-dimensional signals are discussed to explain the concept.
- Practical references to images and sampling points are used to illustrate the process.
Additional Notes:
- The lecture briefly mentions the importance of sampling and Quantisation in Digital Image processing.
- The process is essential for converting analog images into digital format for further processing.

Methodology / Steps for Sampling and Quantisation:

Start with a continuous image defined over spatial coordinates (x, y) with continuous amplitude values.
Sampling:
- Select discrete points along the spatial dimensions.
- Represent these sample points by vertical tick marks on the image.
- Extract amplitude values at these sample points.
Quantisation:
- Map each sampled amplitude value to the nearest Quantisation level.
- Use a finite number of levels to represent the amplitude.
Form a Digital Image:
- Combine the sampled and quantised values into a digital representation.
- This Digital Image is a matrix of discrete amplitude values at sampled locations.
Adjust sampling density and Quantisation levels to balance image quality and data size.