Skip to content
BENIBIX Learn

Convolutional Neural Network (CNN)

AI📈 Machine LearningNeural Networks

Also known as CNN, ConvNet

A network made for photos. It slides small filters over the picture.

Think of it like

Think of it like a small window you slide over a photo. It checks one patch at a time.

Example

A farm app uses a CNN to look at photos of leaves. Its filters pick up spots and yellow edges, and it says which plants are sick.

Why it matters

CNNs made modern computer vision work. Face unlock, medical scans and self-driving cars all came to use them.

Definition

A convolutional neural network is a neural network built to work with images. It does not look at the whole picture at once. It slides small filters across the picture. Each filter looks for one simple thing, like an edge or a spot of colour. Later layers join these into bigger shapes, like eyes or wheels. So it can spot a thing anywhere in a photo.
The small window slides across the picture. Each stop writes one number into the feature map.

Common mistake

Thinking CNNs only work on photos. The same sliding idea works on sound and on other data laid out in a grid or a line.

Key takeaways

  1. 1
    Slides filters. Over small patches.
  2. 2
    Simple then complex. Edges become shapes.
  3. 3
    Made for grids. Images most of all.

Learn it properly