Scribd
Upload
34 views5 pages

Final Project Requriment

The Digital Image Processing (DIP) project aims to provide students with practical experience in classical DIP techniques without using deep learning. Students must work in pairs, implement at least three different DIP methods, and create a GUI for their project, while submitting a report and code. The project includes various ideas for implementation, such as edge detection, image filtering, and object recognition, with specific grading criteria outlined for assessment.

Uploaded by

amnashoaib0319
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
34 views5 pages

Final Project Requriment

The Digital Image Processing (DIP) project aims to provide students with practical experience in classical DIP techniques without using deep learning. Students must work in pairs, implement at least three different DIP methods, and create a GUI for their project, while submitting a report and code. The project includes various ideas for implementation, such as edge detection, image filtering, and object recognition, with specific grading criteria outlined for assessment.

Uploaded by

amnashoaib0319
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Digital Image Processing (DIP) Project

Requirement Document

Objective:
The goal of this project is to provide students with hands-on experience in implementing
classical Digital Image Processing (DIP) techniques. Students will apply classical
computer vision algorithms without the use of deep learning.

Project Guidelines:
1. Scope & Constraints:
• You must not use deep learning methods (e.g., CNNs, neural networks).
• Only classical image processing techniques should be used (e.g., edge detection,
thresholding, morphological operations, filtering, feature extraction, segmentation,
etc.).
• You may use OpenCV, NumPy, SciPy, PIL, or any other classical computer vision
libraries.
2. Group Members:
• Students should work in groups of 2 for better collaboration.

3. Dataset & Inputs:


• You can either use standard datasets (e.g., MNIST for handwritten digits, medical
images, etc.) or collect your own dataset (e.g., real-world images).
• If using your own images, at least 20 images must be included in your dataset.
4. Implementation Requirements:
• Implement at least three DIP techniques from different categories (e.g., filtering,
segmentation, feature extraction, etc.).
• Your project must allow for runtime modifications (e.g., adjusting threshold values,
changing filter sizes, etc.).
• Provide a graphical interface (GUI) using Tkinter, Streamlet, Gradio, OpenCV GUI,
or Matplotlib
5. Report & Code Submission:
• Submit a project report explaining:
o Introduction & problem statement
o Techniques used and why
o Results & analysis
o Challenges faced & conclusions
• Submit well-commented code and dataset.

Deliverables:
• Project Report (PDF format)
• Python Code (.py / Jupyter Notebook)
• Dataset link

Grading Rubric (Total: 30 Marks)


Component Marks Description
Project Submission 5 Proper submission of code,
report, and dataset
Implementation Quality 7.5 Correct use of classical DIP
techniques and quality of
results
Runtime 7.5 Ability to adjust parameters
Modifications/Viva at runtime (e.g., filter size,
thresholding values)
Analysis & Report 5 Proper explanation of
results, challenges, and
conclusions
Code Structure & 5 Well-commented,
Readability structured, and properly
formatted code

Project Ideas for Digital Image Processing


Below are some examples ideas so you can choose from it or can choose on your own but
the topic should be Unique of each group

1. Edge Detection for Object Boundaries

• Use Sobel, Prewitt, or Canny edge detection to detect the boundaries of objects
in an image.
• Possible Applications: Handwriting detection, signature verification, object shape
extraction.

2. Image Filtering & Noise Removal

• Implement Gaussian, Median, or Bilateral filtering to remove noise from images.

• Compare the performance of different filtering techniques.

• Possible Applications: Medical image enhancement, document scanning.

3. Automatic Number Plate Recognition (ANPR)

• Use thresholding, morphological operations, and character segmentation to


extract and process vehicle number plates.

• Can be extended by integrating OCR (Optical Character Recognition) for number


extraction.

4. Face Detection using Haar Cascades

• Use OpenCV’s Haar Cascade Classifier to detect faces in images.

• Extend it by detecting eyes, nose, and mouth separately.

5. Color-Based Object Detection

• Use color thresholding (HSV-based segmentation) to detect objects of a specific


color in an image.

• Example: Detecting a red ball from an image.

6. Image Segmentation using Thresholding

• Use Otsu’s Thresholding, Adaptive Thresholding, or Watershed Algorithm for


segmenting objects in an image.

• Possible Applications: Leaf disease detection, tumor segmentation in medical


images.

7. Optical Character Recognition (OCR) with Classical Methods

• Use connected component analysis, contour detection, and template matching


to recognize handwritten or printed text.

• Example: Detecting numbers from speedometers, meters, or billboards.

8. Coin or Object Counting in an Image


• Use morphological operations and contour detection to detect and count circular
objects (like coins) in an image.

• Extend it by calculating the total value based on coin size.

9. Image Stitching for Panorama Generation

• Use feature detection (SIFT, ORB) and homography transformation to stitch


multiple images together.

• Example: Creating a 360-degree panorama from multiple images.

10. Hand Gesture Recognition using Contours

• Detect hand gestures using contour detection and convex hull approach.

• Example: Counting fingers, recognizing thumbs up/down.

11. Image Forgery Detection

• Use keypoint matching (ORB, SIFT) and histogram analysis to detect tampered
images.

• Example: Identifying if an image has been modified or duplicated.

12. Barcode & QR Code Scanner

• Use Zebra Crossing (ZXing) or OpenCV techniques to detect and read QR codes
from images.

13. Lane Detection for Self-Driving Cars

• Use edge detection (Canny), Hough Transform, and region of interest masking to
detect road lanes in real-world images.

14. Image Enhancement (Histogram Equalization)

• Implement histogram equalization, contrast stretching, and CLAHE to improve


low-quality images.

• Example: Enhancing old faded photographs.

15. Medical Image Processing (X-Ray, MRI)

• Use segmentation, filtering, and feature extraction to analyze medical images.

• Example: Detecting fractures in X-rays, brain tumor segmentation.

16. Signature Verification using Contour Analysis


• Extract features from a signature using edge detection and contour properties.

• Compare signatures to verify authenticity.

17. Image Watermarking

• Implement invisible or visible watermarking techniques using Fourier Transform


or spatial domain processing.

18. Image Cartoonization

• Convert an image into a cartoon-style sketch using edge detection and bilateral
filtering.

19. Traffic Light Detection

• Use color thresholding and shape detection to recognize and classify traffic lights.

20. Skin Disease Detection

• Use color segmentation and texture analysis to identify skin disease patterns.

You might also like