From the course: Computer Vision on the Raspberry Pi 4
Join today to access over 24,800 courses taught by industry experts.
Introducing convolutional neural networks (CNNs)
From the course: Computer Vision on the Raspberry Pi 4
Introducing convolutional neural networks (CNNs)
- [Instructor] In this video, I'll introduce a new type of neural network called a convolutional neural network, or CNN. CNNs are used extensively in machine learning, and we're going to use them to recognize objects. In earlier videos, I presented a neural network that returned a set of probabilities. CNNs accomplish essentially the same result. If a CNN is trained to recognize n objects, it will return n probabilities. If a probability is close to one, it means the CNN recognizes the corresponding object. The main difference between CNNs and regular neural networks is that CNNs accept images as input and process them using convolution. The values in the convolution kernels serve as the network's weights. Another difference between CNNs and regular neural networks involves training. When training a CNN, the goal is to determine which kernels are best suited for recognizing the objects of interest. This can be hard…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.