Duration 3 days – 21 hrs
Overview
This course provides a hands-on introduction to computer vision using Python and deep learning frameworks. Participants will learn the fundamentals of image processing, explore Convolutional Neural Networks (CNNs), and build models for image classification and object detection. Designed for those with a basic understanding of machine learning, the course bridges theory with real-world applications in vision-based AI systems.
Objectives
- Understand key concepts of image representation and processing
- Build and train CNNs for image classification tasks
- Apply transfer learning with pre-trained vision models
- Implement basic object detection techniques using deep learning
- Use Python libraries like OpenCV, TensorFlow/Keras, or PyTorch for vision tasks
Audience
- Aspiring AI and computer vision developers
- Data scientists and machine learning engineers expanding into visual data
- Software developers and researchers working with image-related applications
- Professionals in industries like manufacturing, healthcare, and security interested in vision automation
Prerequisites
- Intermediate Python programming skills
- Knowledge of basic machine learning concepts (supervised learning, model evaluation)
- Familiarity with NumPy, Pandas, and either TensorFlow or PyTorch (recommended)
Course Content
Day 1: Image Processing & Foundations
- Introduction to computer vision and real-world use cases
- Image representation (pixels, color channels, grayscale)
- Image transformations (resizing, cropping, filtering, edge detection)
- Using OpenCV for basic image processing
- Hands-on: Load and manipulate images using OpenCV
Day 2: CNNs and Image Classification
- Introduction to Convolutional Neural Networks (CNNs)
- Layers: convolution, pooling, activation, and fully connected
- Building and training a CNN from scratch using Keras or PyTorch
- Image classification on datasets like CIFAR-10 or MNIST
- Hands-on: Build a CNN model to classify images
Day 3: Object Detection & Transfer Learning
- Difference between classification and object detection
- Introduction to object detection models (YOLO, SSD, Faster R-CNN – conceptual overview)
- Applying pre-trained CNNs with transfer learning (e.g., ResNet, MobileNet)
- Final project: Image classification or object detection on a custom dataset
- Review and evaluation
