This is the code repository for Hands-On Computer Vision with OpenCV 4, Keras and TensorFlow 2 [Video], published by Packt. It contains all the supporting project files necessary to work through the video course from start to finish.
Do you want to understand how computers see images and videos? Using artificial intelligence, we can enable computers and smart devices to interpret what is in an image (computer vision).
This can provide massive benefits when it comes to automating tasks for which images are vital, such as examining medical images or enabling self-driving cars to see. Already, these applications are creating a massive industry around computer vision—one that is set to grow rapidly, with some sources predicting that it will be worth over $43 billion by 2023.
This course provides you with a perfect foundation from which to understand computer vision and supports your professional development in this fast-growing arena. We first learn the basic concepts and explore these using OpenCV4, the most popular open-source computer vision library. Next, we explore using Machine Learning in computer vision, including the use of deep learning (using TensorFlow 2.0 and Keras) to implement advanced image classifiers.
This course is designed to help data scientists, and those who already have some familiarity with ML and DL (and experience with Python, Keras, and TensorFlow), to gain a solid understanding of OpenCV and train their own computer vision deep learning models.
- Image manipulations (dozens of techniques—such as transformations, blurring, thresholding, edge detection, and cropping)
- How to segment images using a variety of OpenCV algorithms, from contouring to blob and line detection
- Approximate contours and perform contour filtering, ordering, and approximations
- Perform object detection for faces, people, and cars
- Use Machine Learning in computer vision, including understanding Deep Learning models such as convolutional neural networks
- Create a varying range of image classifiers—for example, recognizing handwritten digits, gesture recognition, and other multi-class classifiers
- Perform facial recognition with deep learning
For successful completion of this course, students will require the computer systems with at least the following:
• Some prior knowledge of programming in Python, but is not required
• Familiarity with basic programming concepts
• Knowledge of basic machine learning would be helpful but is not required
For an optimal experience with hands-on labs and other practical activities, we recommend the following configuration:
• CPU: 2 x 64-bit, minimum RAM size of 4 GB
• Storage: Recommended minimum of 2 GB
• Internet access to download the files GitHub and view videos