Digital Image Processing and Computer Vision Lecture Notes

Comprehensive practice questions covering the fundamental steps of Digital Image Processing, system components, human visual perception, and image model mathematics.

What is the primary difference between Image Enhancement and Image Restoration?

Image Restoration uses mathematical or probabilistic models to improve an image, while Image Enhancement is based on human subjective preferences regarding what constitutes a "good" result.

What are the three principal Image Processing Fields and how are they defined?

1. Computer Graphics: The creation of images. 2. Image Processing: Enhancement or manipulation of images. 3. Computer Vision: Analysis of image content.

Describe the characteristics of Mid-Level Processes in DIP.

The inputs are generally images, but the outputs are attributes extracted from those images (such as edges or object identities). Tasks include segmentation, description, and classification (recognition).

What is the function of an ALU in specialized image processing hardware?

The Arithmetic Logic Unit (ALU) performs arithmetic and logical operations in parallel on entire images.

Identify the three principal categories of digital storage for DIP applications.

1. Short-term storage (used during processing). 2. On-line storage (for relatively fast recall). 3. Archival storage (infrequent access).

What is the function of the Iris and Pupil in the human eye?

The Iris is the colored area that surrounds the Pupil (a black circle/window); the Pupil expands and contracts to control the amount of light entering the eye.

Differentiate between Photopic and Scotopic vision.

Photopic vision (bright-light vision) is served by cones and is highly sensitive to color. Scotopic vision (dim-light vision) is served by rods and provides a general view of the scene.

Define Brightness Adaptation.

The process by which the human eye adjusts to different light intensities to maintain visibility, as the eye cannot operate over the entire range of light intensities ($10^{-6}$ to $10^4 \text{ cd/m}^2$) simultaneously.

What is the Weber ratio and what does it indicate?

The Weber ratio is \frac{\text{\Delta I}}{I}, where $I$ is the background illumination and \text{\Delta I} is the increment of illumination. A small ratio indicates good brightness discrimination, while a large ratio indicates poor discrimination.

In the image model $f(x,y) = i(x,y) \times r(x,y)$, what do the terms represent?

$i(x,y)$ represents illumination (energy source) and $r(x,y)$ represents reflectance (property of the object).

What are the specific constraints on the illumination component $i(x,y)$ and the reflectance component $r(x,y)$?

$0 < i(x,y) < \infty$ and $0 < r(x,y) < 1$.

Define Sampling and Quantization in the context of digital image formation.

Sampling is the process of selecting discrete points from a continuous signal. Quantization is the process of assigning discrete intensity values (gray levels) to those sampled points.

A square image is represented by $131072$ bits. If each pixel is represented by $8$ bits, what is the spatial size ($N$)?

$N^2 = \frac{131072}{8} = 16384$, so $N = \sqrt{16384} = 128$. The spatial size is $128 \times 128$ pixels.

What formula is used to calculate the time required to transmit an image over a modem using a specific Baud rate?

$\text{Time} = \frac{\text{Total bits}}{\text{Baud rate}}$, where total bits often include overhead (e.g., packets with a start bit, 8 bits of data, and a stop bit).

What are the two major technologies used to read the output voltage of sensors in a digital image matrix?

CMOS (Complementary Metal Oxide Semi-conductor) and CCD (Charge Coupled Device).

What is the advantage of using a Sensor Array over a Single Sensing Element for image acquisition?

A Sensor Array is two-dimensional and can obtain a complete image at once by focusing energy onto the array surface, whereas a Single Sensing Element requires relative mechanical displacement in both $x$ and $y$ directions, making it slow.

In the human eye, what is the distance commonly referred to as the focal length when focusing on distant objects?

The focal length varies from approximately $17 \, \text{mm}$ to about $14 \, \text{mm}$ as refractive power increases; it is at its minimum when focusing on objects farther than $3 \, \text{m}$ away.

Which DIP step is responsible for techniques that reduce the required storage or transmission bandwidth?

Step 6: Image Compression.

What does Morphological Processing involve?

It consists of tools for extracting image components that are useful for the representation and description of shape.