AKTU Image Analytics (KDS-061) Question Paper 2022-23
| University | Dr. A.P.J. Abdul Kalam Technical University (AKTU), Lucknow |
|---|---|
| Course | B.Tech, Semester VI |
| Subject | Image Analytics (KDS-061) |
| Exam | Theory Examination 2022-23 (even semester, held June 2023) |
| Time · Marks | 3 hours · 100 marks |
| Source | AKTU question paper, via Notes Gallery |
Note: Attempt all sections. If you require any missing data, then choose suitably.
Section A
Q1. Attempt all questions in brief · 2 × 10 = 20 marks
- What is the Digital Image Processing? [2]
- Name some applications of digital image processing. [2]
- Differentiate opening and closing. [2]
- Describe Hit or miss transform. [2]
- Define point detection process. [2]
- What do you mean by Edge detection algorithm? [2]
- Full form of SIFT and define it. [2]
- Define shape numbers. [2]
- What do you mean by Pattern Classes? [2]
- Define Optimum (Bayes) Statistical Classifiers. [2]
Section B
Q2. Attempt any three · 10 × 3 = 30 marks
- Define digital image. Explain fundamental steps in image processing system with the help of suitable diagram. [10]
- Explain morphological image processing step in detail. [10]
- Describe image segmentation step in detail. [10]
- Explain feature extraction process in detail. [10]
- Define patterns. Also discuss pattern classification by prototype matching. [10]
Section C
Q3 to Q7. Attempt any one part of each question · 10 × 5 = 50 marks
Q3
- Explain Image acquisition in detail with the help of diagram. [10]
- Differentiate between smoothing and sharpening spatial filters in detail. [10]
Q4
- Write short notes on: (i) Erosion (ii) Dilation [10]
- Explain Morphological reconstruction algorithm in detail. [10]
Q5
- Differentiate between Segmentation by Region Growing and by Region Splitting in detail. [10]
- Write short notes on: (i) Snakes sets (ii) Level Sets [10]
Q6
- Explain Topological and Texture Descriptors in detail. [10]
- Write short notes on: (i) Boundary Preprocessing (ii) Fourier Descriptors [10]
Q7
- Differentiate between Multilayer Feed forward neural networks and deep convolutional neural networks in detail. [10]
- Write short notes on: (i) Neural Networks (ii) Deep Learning [10]
Most repeated topics
Every one of these topics was asked in both the 2022-23 (KDS-061) and 2024-25 (BCDS061) papers. Prepare all of them.
- Fundamental steps of a digital image processing system (with block diagram)
- Morphology: erosion, dilation, opening, closing, hit-or-miss, reconstruction
- Smoothing and sharpening spatial filters
- Image segmentation: region growing, splitting and merging
- Feature extraction and SIFT
- Boundary preprocessing and representation
- Pattern classification and prototype matching
- Multilayer feed-forward networks and CNNs
Compared by LearnAIFast across the 2022-23 and 2024-25 even-semester theory papers.
Answer hints
Q1(c). Opening vs closing
Opening is erosion followed by dilation: it removes small bright objects and noise and smooths contours. Closing is dilation followed by erosion: it fills small holes and gaps and joins nearby objects. Both use the same structuring element.
Q1(g). SIFT
Scale-Invariant Feature Transform. It finds keypoints that stay the same when an image is scaled or rotated, and describes each with a 128-value vector. Steps: scale-space extrema detection, keypoint localisation, orientation assignment, and keypoint descriptor.
Q4(a). Erosion and dilation
Erosion shrinks foreground objects: a pixel stays 1 only if the structuring element fits entirely inside the object. It removes thin lines and noise. Dilation grows objects: a pixel becomes 1 if the structuring element touches the object at all. It fills gaps and thickens lines. Draw a small binary example with a 3×3 structuring element.
Q5(a). Region growing vs region splitting
Region growing starts from seed pixels and adds neighbours that meet a similarity rule (such as intensity within a threshold), so it builds regions bottom-up. Region splitting starts with the whole image and keeps dividing it into quadrants (a quadtree) until each region is uniform, then merges similar neighbours. Growing depends on good seeds; split-and-merge can give blocky boundaries.
Other papers
This paper is shared for study purposes only. Copyright belongs to AKTU. If you own the copyright and want it removed, email learnwithskillnest@gmail.com and we will take it down.