Course Code Course Title L T P C

MACHINE VISION 3 0 0 3

a) Course Category
Program Elective

b) Preamble
This course provides the basic knowledge about implementation of various algorithms
related to machine vision systems. In addition, students will also get an opportunity to
know the fundamentals of calibrating machine vision systems.

c) Prerequisite
Nil
d) Related Courses
Digital Image Processing, Machine learning

e) Course Outcome
Upon the successful completion of the course, student will be able to:

Knowledge Level
CO Nos. Course Outcomes (Based on Revised
Bloom’s Taxonomy)
Apply the geometric properties, algorithms and
CO1 distance measures of binary images to facilitate pre- K3
processing of images in machine vision
Implement the concept of region splitting and
CO2 merging for image segmentation in machine vision K3
images
Implement the edge detection operators to extract and
CO3 track relevant features from images and evaluate its K3
performance
Outline the concepts involved in geometrical optics
CO4 K2
and techniques for computing the depth from images
Discuss dynamic vision and new developments in
CO5 K2
object recognition systems
 f) Correlation of COs with POs

PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 PO9 PO10 PO11 PO12 PSO1 PSO2
CO1 M L M - M - - L H L - L - -
CO2 M L - L M - - L H - - M M M
CO3 M L - L M - - L H - - H M M
CO4 L L M L - - L - M L - L L L
CO5 L L M L - - L - M L - M L L

Unit I
Introduction to Machine vision & Binary Image Processing Techniques
Introduction – Machine vision –Relationship to other fields –Image definitions - Levels of computation- Binary
image processing – Geometric properties, size, position & orientation - Component labeling – Binary Algorithms:
Run length encoding - Size filter - Euler number - Region boundary - Area perimeter - compact Distance measures
- Distance transforms - Medial axis - Morphological operators - Thinning- Expanding and Shrinking- Matlab
Exercises on binary algorithms & morphological operators. [R1, R2]
9
Unit II
Segmentation Techniques
Segmentation – Automatic thresholding - Limitations of Histogram methods – Region representation – array
representation - Hierarchical representation - Region based segmentation: Split and merge – region merging –
Removing weak edges –Region splitting - split and merge – Region growing- Matlab exercises on various methods
for automatic thresholding [R1,R2]

9
Unit III
Edge detection
Gradient - Steps in edge deduction - Robert’s operator - Sobel operator - Prewitt operator - Comparison Second
derivative operator - Laplacian operator - Gaussian edge Detection - Canny edge detector - Image Approximation -
Subpixel location estimation –Edge detector performance- methods of Evaluating performance – Figure of merit –
Matlab exercises on edge detectors and its performance evaluation [R1, R3]

9
Unit IV
Geometrical Optics & Depth Computing Techniques
Optics - Lens equation - Image resolution - Depth of Field - View volume - Exposure - Shading - Image Inductance
- Illumination - Reflector - Surface orientation - Shape from shading depth - Stereo imaging - Cameras in arbitrary
position and orientation - Stereo matching - Shape from X - Range imaging - Structural lighting - Imaging Radar-
Active vision. [R1, R2]
9
Unit V
Dynamic vision & object recognition
Change detection –Difference pictures – Static segmentation and matching –object recognition – system components
– complexity of object recognition – object representation -observer centered & object centered representations –
feature detection –recognition strategies – classification – Matching Feature indexing - verification – Template
matching –morphological approach – symbolic – analogical methods. [R1, R4]

9
Total: 45 Hours

References
1. Ramesh Jain, Rangachar Kasturi and Brian G. Schunck, Machine Vision, McGraw Hill International
Edition, 2006.
2. Wataru Ohyama, Soon Ki Jung, Frontiers of Computer Vision: 26th International Workshop, IW-FCV
2020, Ibusuki, Kagoshima, Japan, 2020
3. González,RafaelC.andWoods,RichardEugene,DigitalImageProcessing,3rdEdition, Prentice Hall,2008
4. David Forsyth and Jean Ponce, Computer Vision: A modern Approach, Prentic Hall India 2004

Online Resources
1. https://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-801-machine-vision-fall-2004/
lecture-notes/
2. http://faculty.salina.k-state.edu/tim/mVision/Introduction.html
3. https://faculty.ucmerced.edu/mcarreira-perpinan/teaching/ee589/lecture-notes.pdf