IBM Introduction to Computer Vision and Image Processing

Welcome to the repository for my course project IBM Introduction to Computer Vision and Image Processing, completed through Coursera. I've learned a lot about computer vision and image processing through this course, and I earned a certificate upon completion. This repository contains various Jupyter notebooks and files demonstrating the topics covered throughout the course.

Course Highlights

The course covered a wide range of topics related to computer vision and image processing, including:

Basic Image Processing

• Working with Python Libraries (PIL & OpenCV): Loading, manipulating, and saving images using PIL and OpenCV.
• Geometric Transformations: Resizing, cropping, and rotating images.
• Histograms and Intensity Transformations: Analyzing image histograms and implementing contrast stretching.

Advanced Image Processing Techniques

• Spatial Filtering: Applying smoothing and sharpening filters to images.
• Support Vector Machines (SVM): Training and using SVM models for image classification.

Machine Learning & Deep Learning in Computer Vision

• K-Nearest Neighbors (KNN): Training and using KNN models for classification.
• Logistic Regression & Softmax Classification: Implementing logistic regression with mini-batch gradient descent.
• Neural Networks & Convolutional Neural Networks (CNNs):
  • Building simple neural networks for binary classification.
  • Comparing ReLU and Sigmoid activation functions.
  • Using CNNs with PyTorch for image classification.
• Data Augmentation: Techniques to expand image datasets.

Object Detection

• Object Detection with Haar Cascades: Detecting objects using Haar-like features.
• Faster R-CNN: Building a Faster R-CNN model for object detection.

Repository Structure

The repository is organized as follows:

• basic_image_processing/ - Notebooks covering the basics like PIL and OpenCV usage, geometric transformations, and histograms.
• machine_learning/ - Notebooks on machine learning algorithms like KNN, SVM, and logistic regression.
• deep_learning/ - Neural network and CNN implementations with data augmentation techniques.