SETHU INSTITUTE OF TECHNOLOGY, KARIAPATTI

(An Autonomous Institution Affiliated to Anna University, Chennai)

Regulation 2019(Question Bank)
Department: Biomedical Subject name : Image Processing Techniques
Engineering
Subject code : 19UBM703 Question Pattern : Part A: (2 x 10 = 20 marks)
Part B: (5 x 16 = 80 marks)
Course Coordinator : Mrs. Time Duration : 3 hours
Mary James
Approved by: HoD/BME

Font size : Times new roman :12 Line spacing: 1.15

PART – B ( 2 Marks)
UNIT - I (Minimum 8 Questions)
1. State the primary application of the DCT (Discrete Cosine
CO2 A -
Transform) in image compression?
2. Explain the significance of the HSI color model in color image CO1-U -
processing.
3. Define image sampling and explain its role in digital image CO1-U -
processing.
4. Compute the distance between the two pixels q(1,1) p (2,2) using CO2 A -
the city block distance measurement.
5. Describe the distribution of cones and rods in the human retina. Are CO2 A -
they evenly distributed, or is there a specific pattern?
6. Use a step-by-step diagram to depict the progression of the Mach CO3-An -
band effect as it manifests in an image with varying levels of
intensity.
7. How is image acquisition done in a digital image processing? CO1-U -
8. Write the transform pair for 3D-DFT. CO1-U -
9. Explain the significance of image acquisition in biomedical image CO1-U -
processing in two sentences.
UNIT - II (Minimum 8 Questions)
1. Demonstrate the formula for the transfer function of a Butterworth CO1-U -
high-pass filter.

2. Specify the objective of an image enhancement technique in two CO1-U -

sentences.

3. Demonstrate the two categories of image enhancement and briefly CO1-U -

explain each.
4. Define contrast stretching in the context of image processing. CO1-U -
5. Spatial filters are also called averaging filters. Provide a reason for CO3-An -
this nomenclature in two sentences.
6. Define intensity level slicing and explain its purpose in image CO1-U -
processing.
7. Sketch a block diagram for homomorphic filtering, highlighting its CO3-An -
key components.
8. Discuss the concept of histogram equalization and its role in CO2-A -
enhancing image details and improving visualization.

UNIT - III (Minimum 8 Questions)

1. Sketch the basic components of the model of the image CO3-An -
degradation/restoration process in digital image processing.
2. Describe the different types of noise commonly found in images. CO3-An 3
3. Why is it important to estimate the degradation function in image CO2-A 2
restoration?
4. Write down the mathematical formula for Salt and pepper noise. CO1-U -
5. Provide a definition of inverse filtering in the context of image CO1-U -
restoration.
6. Discuss the characteristics of uniform noise and its impact on image CO1-U -
quality.
7. State the significance of the signal-to-noise ratio (SNR) in the CO1-U -
Wiener filter?
8. State the probability density function of exponential noise. CO1-U -
9. Define Rayleigh noise. CO1-U -
UNIT - IV (Minimum 8 Questions)
1. Explain thresholding in image segmentation. CO1-U -
2. Classify the types of discontinuity in digital image? CO2-A 2
3. Demonstrate the steps involved in splitting and merging? CO1-U -
4. Define medical image segmentation and explain why it is essential CO1-U -
in the field of healthcare.
5. Explain the concept of image representation schemes and their CO1-U -
importance in image analysis.
6. Compare an edge with line in an image. CO3-An 3
7. Define region growing? CO1-U -
8. Illustrate the steps in polygonal approximation method. CO2-A 2
9. How are the derivatives obtained in edge detection during CO3-An 3
formulation?
UNIT - V (Minimum 8 Questions)
1. Define image compression? CO1 U 2
2. Illustrate the need for Compression? CO3 An 3
3. Classify the different Compression Methods? CO2 A -
4. Define Huffman coding CO1 U 2
5. Define bit plane coding. CO1 U -
6. Define run length coding. CO1 U 2
7. Classify the types of decoder? CO2 A 3
8. Sketch the two equivalent representations of the perception model CO3 An 3
for two pattern classes.

2
 PART – C
UNIT - I (Minimum 5 Questions either or choice )
1. (a Classify the different transforms used in digital image CO3- An PO 1 (16)
)
processing. Explain in detail.

Or
(b Illustrate the conversion of RGB color model to HSI model CO3- An PO 1 (16)
) and vice versa in detail.

2. (a Define digital image processing. Explain various functional CO1- U PO 1 (16)

)
block of digital image processing with diagram.

Or
(b Explain the concept of spatial domain sampling in digital CO1- U PO 1 (16)
) images and how it relates to pixel resolution and image
size.Describe the process of quantization in image processing,
highlighting its role in reducing continuous intensity values to
discrete levels.

3. (a Analyze the concept of pixel connectivity in digital images, CO3 - An PO 3 (16)

)
and discuss how it influences the representation and
processing of objects in an image. Explain the basic
relationships between pixels in detail.

Or
(b Analyze the elements of visual perception in the context of CO3 - An PO 3 (16)
) human vision and describe the process of image formation in
the eye.

4. (a Explain the color model which is suitable for hardware CO1- U PO 1 (16)
) implementation with neat diagram.
Or
(b i) Explain Discrete Fourier Transform (DFT) of an image in CO1- U PO 1 (16)
) detail.
ii) Explain Discrete Cosine Transform of an image in detail.

5. (a Analyze the fundamental differences between the RGB and CO3 - An PO 3 (16)
) HSI color models, explaining how they represent color
information and their respective advantages in different
applications.
Or
(b Sketch and illustrate the elements of visual perception and CO3 - An PO 3 (16)
) also describe image formation in the eye.

3
 UNIT - II (Minimum 5 Questions either or choice )

1. (a Evaluate the impact of different histogram equalization CO3 - An PO 3 (16)

) algorithms on images with varying statistical distributions and
assess their ability to improve contrast.
Or

(b Utilize a point processing technique of your choice to CO3 - An PO 3 (16)

)
enhance the visibility of faint structures in astronomical
images. Discuss the various point processing techniques in
detail.

2. (a Illustrate the design of image sharpening and smoothing filters CO3 - An PO 3 (16)
) in frequency domain.
Or
(b Analyze the basic intensity transformation functions along CO3 An (16) (16)
) with diagram.

3. (a Write short notes on the following: CO1 - U PO 1 (16)

) i) Contrast stretching
ii) Intensity level slicing
iii) Bit level slicing
iv) Image negatives
Or
(b Discuss the image smoothing and sharpening filters with its CO1- U (16) (16)
) model in the spatial domain.

4. (a Illustrate the mathematical principles behind homomorphic CO3 - An PO 3 (16)

) filtering and how it enables the separation of illumination and
reflectance components in an image. How is the Fourier
transform utilized in this process?
Or
(b Illustrate the design of image sharpening and smoothing filters CO3 - An P (16)
) in spatial domain. O
3

5. (a CO1 - U P (16)
Explain histogram and Illustrate the steps in histogram
) O
equalization of the image.
1
Or

(b With the help of a block diagram, illustrate the principle of CO1 - U P (16)
) homomorphic filtering. O
1
UNIT - III (Minimum 5 Questions either or choice )

4
1. (a Evaluate the impact of inverse filtering advancements on fields CO3 - An PO (16)
) such as medical diagnostics, forensics, and remote sensing, 3
highlighting both the benefits and potential risks.
Or
(b Analyze how frequency domain filtering techniques can be CO3 - An PO (16)
) used to reduce periodic noise in images? Provide a brief 1
overview of the approach.

2. (a Evaluate the advantages and disadvantages of using order CO3 - An PO (16)

) statistics 1
filters in image processing applications, comparing them with
other filtering methods such as median filters and Gaussian
filters.
Or
(b Illustrate the minimum mean square error filtering in detail. CO3 - An PO (16)
) 3

3. (a Analyze the different types of noise models along with their CO3 - An PO (16)
) probability density functions. 3
Or
(b Design weiner filtering approach for segmentation in detail. CO3 - An PO (16)
) 3

4. (a Analyze the statistical properties of mean filters, including their CO3 - An PO (16)
) ability to reduce different types of noise (e.g., Gaussian, salt- 3
and-
pepper) and their impact on image features and texture.
Compare
and contrast these properties with other noise reduction
techniques.
Or
(b Classify the different types of order statistics filtering in detail. CO3 - An PO (16)
) 3

5. (a Demonstrate the different types of mean filtering in detail. CO3 - An PO (16)

) 3
Or
(b Illustrate the minimum mean square error filtering in detail. CO3 - An PO (16)
) 3

UNIT - IV (Minimum 5 Questions either or choice )

1. (a Illustrate how hough transforms can be used to link edges. CO3 - An PO (16)
) 3
Or
(b Illustrate the techniques used for region representation in detail. CO3 - An PO (16)
) 3
5
2. (a Examine region based segmentation with relevant examples. CO3 - An PO (16)
) 3
Or
(b Illustrate with suitable examples how detection of edges done CO3 - An PO (16)
) in medical images are done. 3

3. (a Demonstrate how edge detection is performed? Explain hough CO3 - An PO (16)

) transform and discuss how the edge points are linked. 3
Or
(b Classify and explain the various image representation CO3 - An PO (16)
) approaches in detail. 3

4. (a Examine image segmentation based on various thresholding CO1 - U PO (16)

) techniques. 1
Or
(b Assess how an image is segmented using region splitting and CO1 - U PO (16)
) merging algorithm in detail and how the segmented object is 1
represented by chain codes.

5. (a Illustrate how hough transforms can be used to link edges. CO1 - U PO (16)
) 1
Or
(b Examine region based segmentation and region growing with CO1 - U PO (16)
) an 1
example.

UNIT - V (Minimum 5 Questions either or choice )

1. (a Explain the fundamental concepts of image recognition, CO1 - U PO (16)

) emphasizing the role of patterns and pattern classes. Provide 1
examples to illustrate your explanation.
Or
(b Explain the fundamental principles of image compression. CO1 - U PO (16)
) Differentiate between lossless and lossy compression 1
techniques, focusing on their applications and trade-offs.

2. (a Critically assess the optimality of variable-length coding CO3 - An PO (16)

) methods such as Huffman and Arithmetic coding in terms of 3
compression efficiency. Discuss scenarios where one might be
preferred over the other.
Or
(b Propose a scenario where an optimum statistical classifier CO3 - An PO (16)
) would be the most suitable image recognition method. 3

3. (a Compare and contrast decision theoretic methods, minimum CO3 - An PO (16)

6
 ) distance classifiers, and matching by correlation in the context 3
of image recognition. Assess the strengths and weaknesses of
each approach.
Or
(b Develop a step-by-step explanation of how transform coding CO3 - An PO (16)
) works in image compression. Use an example to illustrate the 3
transformation process and its impact on the image.

4. (a Write short notes on the following in detail, CO1 - U PO (16)

) i . Patterns and Pattern classes 3
ii . minimum distance classifier
Or
(b Explain the need of image compression. How is run length CO1 - U PO (16)
) encoding approach used for compression?it is lossy? Justify. 3

5. (a Explain the need of image compression. How are bit plane CO1 - U PO (16)
) coding and constant area coding approaches used for 3
compression?
Or
(b Write short notes on the following in detail, CO1 - U PO (16)
) i . Matching by correlation, 3
ii . Optimum statistical classifier.