Psycho-Visual Redundancy in Image Compression

Podcast Beta

Play an AI-generated podcast conversation about this lesson

Questions and Answers

What is the compression ratio if the first data set has 10 information carrying units for every 1 unit in the second data set?

50:1

10:1 (correct)

5:1

20:1

If n1 = n2, then CR = 0 and RD = 1.

False

What is the probability of a pixel to have a certain intensity value rk?

p(rk) = h(rk) / n = nk / n

In Image compression, there are 3 basic redundancies: Coding Redundancy, _______________, and Psycho-visual Redundancy.

Inter-pixel Redundancy Signup and view all the answers

Match the following types of redundancy in Image Compression with their descriptions:

Coding Redundancy = Redundancy due to the coding scheme used Inter-pixel Redundancy = Redundancy due to the correlation between pixels Psycho-visual Redundancy = Redundancy due to the human visual system's limitations Signup and view all the answers

If n1 >> n2, then CR = 1 and RD = 0.

False Signup and view all the answers

What is the human visual system more sensitive to?

edges and texture Signup and view all the answers

The human visual system is more sensitive to huge variations in gray levels or colors.

False Signup and view all the answers

What is the formula to calculate the total error between two images?

∑∑[f(x, y) - f'(x, y)] Signup and view all the answers

The compression ratio (CR) of the middle picture is ___________.

2 Signup and view all the answers

What is the fidelity criterion that measures the mean square signal to noise ratio of the compressed-decompressed image?

Mean Square Signal to Noise Ratio (SNRms) Signup and view all the answers

The root-mean-square error averaged over the whole image is a subjective fidelity criterion.

False Signup and view all the answers

Match the following image compression techniques with their corresponding descriptions:

Uniform Quantization = Divides the range of intensity values into equal intervals Improved Gray Level Quantization (IGS) = Smoothing and quantization correction Signup and view all the answers

What is the purpose of fidelity criteria in image compression?

To measure the quality of the compressed image Signup and view all the answers

What is the main purpose of data compression?

To reduce the amount of data required to represent a given quantity of information Signup and view all the answers

Data and information are the same thing.

False Signup and view all the answers

What is data redundancy?

Elements in data that provide no relevant information Signup and view all the answers

The compression ratio CR can be defined as _______________________.

n1 / n2 Signup and view all the answers

If the compression ratio CR is 2.5, what is the data redundancy RD?

60% Signup and view all the answers

Match the following terms with their definitions:

Data Compression = Reducing the amount of data required to represent a given quantity of information Data Redundancy = Elements in data that provide no relevant information Compression Ratio = The ratio of the total number of storage units for an original digital image to the total number of storage units of the same digital image after compression Signup and view all the answers

Data redundancy is an abstract concept.

False Signup and view all the answers

What is the purpose of calculating the compression ratio and data redundancy?

To quantify the amount of data reduction achieved through compression Signup and view all the answers

Study Notes

Fundamentals of Data Compression

Data compression is the process of reducing the amount of data required to represent a given quantity of information.
Data and information are not the same; various amounts of data can be used to represent the same information.
Data redundancy refers to the presence of elements in data that provide no relevant information.

Data Redundancy

Data redundancy is a central issue in image compression and can be quantified mathematically.
The compression ratio (CR) is defined as the ratio of the original data size to the compressed data size: CR = n1 / n2.
The relative redundancy (RD) is defined as RD = 1 - 1/CR.

Example of Data Redundancy

If an image is compressed from 1MB to 0.4MB, the compression ratio is 2.5, and the relative redundancy is 60%.

Psycho-visual Redundancy

The human visual system is less sensitive to huge variations in gray levels or colors, but more sensitive to edges and texture.
This redundancy can be exploited in image compression.

Fidelity Criteria

The error between two functions is given by e(x, y) = f(x, y) - f'(x, y).
The total error between two images is the sum of the squared errors over all pixels.
The root-mean-square error (erms) is a measure of the fidelity of the compressed image.
The mean square signal to noise ratio (SNRms) is another fidelity criterion.

Data Redundancy and Compression Ratio

If n1 = n2, CR = 1, and RD = 0, indicating no redundancy.
If n1 >> n2, CR → ∞, and RD → 1, indicating high redundancy.
A compression ratio of 10 means that the original data set has 10 information-carrying units for every 1 unit in the compressed data set.

Types of Redundancy

Coding redundancy refers to the use of inefficient coding schemes.
Inter-pixel redundancy refers to the correlation between adjacent pixels.
Psycho-visual redundancy refers to the redundancy in the human visual system's perception of images.

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Description

This quiz covers the concept of psycho-visual redundancy in image compression, including the human visual system's sensitivity to edges and texture, and various methods of gray level quantization. It includes examples of uniform quantization and improved gray level quantization with smoothing and quantization correction.