Image Processing: Techniques and Applications

• CS389 covers Image Processing
• Prof. Tamer Moneir Mousa Nassef is a professor at the Faculty of Computer Science at October University for Modern Sciences and Arts (MSA University)
• Dr. Mohamed Nagy Saad is an Associate Professor at the Faculty of Computer Science at October University for Modern Sciences and Arts (MSA University)

Course Aim

• This module provides an understanding of image processing techniques and familiarizes you with computer vision applications.

Learning Outcomes

• Conceptualize image processing problems
• Demonstrate image processing concepts
• Illustrate standard image processing algorithms
• Compare/contrast the different applications of computer vision
• Apply concepts and methodologies for the formation, representation, enhancement, and analysis of digital images
• Use available tools to develop applications of image processing
• Use benchmark images for algorithm testing
• Improve the design and implementation of image processing algorithms to suit specific applications

Instructors

• Prof. Tamer Moneir Mousa Nassef is a Full Professor, email at [email protected], Office Hours: Office G
• Dr. Mohamed Nagy Saad is an Associate Professor, email at [email protected], Office Hours: Office H414

Grading Policy

• Attendance: 0%
• Lab work: 2%
• 3 Assignments: 6% (2+2+2)
• 3 Quizzes: 12% (4+4+4)
• Midterm Exam: 20%
• Project: 20%
• Final Exam: 40%
• Total: 100%
• Handouts: Lectures + Labs
• Textbook: "Digital Image Processing, Third edition”, by R. Gonzalez and R. Woods, 2008, Prentice Hall
• The grade scheme is subject to change.

Calendar and Syllabus

• Week 1: Image sensing and acquisition (Tutorial: Yes)
• Week 2: Image Sampling and Quantization (Tutorial: Yes)
• Week 3: Basic Relationship between Pixels (Tutorial: Yes, Assignment 1, Quiz 1)
• Week 4: Linear & Non-Linear Operations (Tutorial: Yes)
• Week 5: Image Enhancement in Spatial Domain (Tutorial: Yes, Assignment 2, Quiz 2)
• Week 6: Histogram Processing (Tutorial: Yes)
• Weeks 7 & 8: Midterm Weeks
• Week 9: Spatial Filtering (Smoothing, Sharpening) (Tutorial: Yes)
• Week 10: Image Restoration (Tutorial: Yes)
• Week 11: Edge detection (Tutorial: Yes, Assignment 3, Quiz 3)
• Week 12: Morphological operations (Tutorial: Yes)
• Week 13: Error-Free Compression and Loose Compression (Tutorial: Yes)
• Week 14: Image Segmentation (Tutorial: Yes, Project)
• Week 15: Final Examination

Project 1: Deep Learning-Based Medical Image Analysis

• For medical image analysis like lesion segmentation in histopathology images or tumor detection in MRI scans, design a machine learning model.
• For advanced authenticity and capability, investigate the upgraded systems.

Project 2: Artistic Style Transfer Application

• Implement artistic styles from prevalent paintings to photos using neural style transfer methods.
• Model an application that enables users with visual controls for personalization, executing real-time style transfer.

Project 3: Automated Document Analysis and Recognition

• Develop a system for automated document analysis and recognition from camera images or scanned documents.
• System must derive diagrams, text, and tables, incorporating computer vision techniques and OCR (Optical Character recognition) for document interpretation.

Project 4: Smart Image Cropping and Composition

• Recommend perfect crop areas and blueprints automatically based on content analysis and visual principles.
• Build a smart image cropping and composition tool that utilizes machine learning frameworks to interpret from user reviews and priorities.

Project 5: Wildlife Monitoring and Conservation

• Examine camera trap images to identify track activities, animal species, and population figures.
• Generate a wildlife monitoring system with image processing methods and apply efficient techniques for activity assessment and category recognition.

Project 6: Emotion Recognition from Facial Expressions

• Construct a model from facial expressions captured in videos or images for realizing up-to-the-minute emotions.
• Identify and categorize emotions like surprise, sadness, anger, and joy, using deep learning algorithms.

Project 7: Anomaly Detection in Surveillance Footage

• Create an anomaly detection system for surveillance footage to detect questionable scenarios or activities in crowded events.
• Identify differences and interpret basic behavior patterns through unsupervised learning methods.

Project 8: Food Quality Inspection for Agriculture

• Develop a food quality inspection system for agriculture using image processing to identify diseases, injuries, or drawbacks in images of fruits and vegetables.
• Employ machine learning algorithms for automated quality evaluation.

Project 9: Virtual Makeup Try-On Application

• Develop virtual makeup try-on applications using image processing and Augmented Reality (AR).
• Visualize various makeup products on users' faces by accessing their cameras, creating effective methods to precisely detect facial characteristics and makeup descriptions.

Project 10: Gesture Recognition for Human-Computer Interaction

• Build a gesture recognition system for communication between humans and computers.
• Use camera images or depth sensors to realize poses and hand signals, executing machine learning models for gesture categorization and management.

Evaluation of article reading and project

• Report
• Article reading: Submit a survey of the articles and a list of the articles
• Project:
• Submit an article including introduction, methods, experiments, results, and conclusions
• Submit the project code, the readme document, and some testing samples e.g. images, videos, et cetera for validation
• Presentation

Journals

• IEEE T IMAGE PROCESSING
• IEEE T MEDICAL IMAGING
• INTL J COMP. VISION
• IEEE T PATTERN ANALYSIS MACHINE INTELLIGENCE
• PATTERN RECOGNITION
• COMP. VISION AND IMAGE UNDERSTANDING
• IMAGE AND VISION COMPUTING

Conferences

• CVPR: Comp. Vision and Pattern Recognition
• ICCV: Intl Conf on Computer Vision
• ACM Multimedia
• ICIP
• SPIE
• ECCV: European Conf on Computer Vision
• CAIP: Intl Conf on Comp. Analysis of Images and Patterns

What is an Image

• Practically every scene involves images or image processing.
• An image is a spatial representation of a 2D or 3D scene
• Can be defined as an array or a matrix of pixels (picture elements) arranged in columns and rows.

Image Types

• Analog Image
• Continuous levels of brightness (shades of gray or colors) for human viewing
• Image seen is various levels of brightness or film density and colors
• Generally continuous, not broken into small individual pieces
• Digital Image
• A matrix of pixels for computer systems
• Each pixel is represented by a numerical value
• The pixel value related to the brightness or color that will be seen when converted into analog image for viewing

Impact Digital Image

• Required in modern medical imaging due to functions like:
• Image reconstruction (CT, MRI, SPECT, PET).
• Image reformatting (multi-plane, multi-view reconstructions).
• Wide (dynamic) range image data acquisition (digital radiography).
• Image processing (contrast adjustment).
• Fast image storage and retrieval.
• High-quality image distribution (teleradiology).
• Controlled viewing (windowing, zooming).
• Image analysis (measurements, computer-aided diagonisis).

Digital Image Processing Intro

• Digital Image
• A two-dimensional function f(x, y) where x and y are spatial coordinates.
• The amplitude of f is called intensity or gray level or color at the point (x, y).
• Digital Image Processing
• Digital images processed by means of computer
• Covers low-, mid-, and high-level processes:
• Low-level: Inputs and outputs are images, reduce noise, contrast enhancement, and image sharpening. - Mid-level: Outputs are attributes extracted from input images.
• Tasks are segmentation (partitioning an image into regions or objects), description of objects to reduce them to a form suitable for computer processing, classification (recognition) of individual objects. - High-level: Recognition of individual objects. Making sense of recognized objects, image analysis, cognitive functions