Receptive field

From Computer Science Wiki

Exclamation.png This is student work which has not yet been approved as correct by the instructor

Case study notes[1]

Introduction[edit]

Receptive fields used in an artificial neural networks are similar to the biological explanation. They are a sensory neuron where a prompt, or stimulation, changes the action of the neuron.[2] In other words, the fields gather information to be used by the neurons. They are used in relation to convolutional neural networks (CNNs). The neurons are arranged in a 3-D array. [3]

How does it work or a deeper look[edit]

The receptive field gathers information and sends the information to the neurons in the CNNs. Rather than a more traditional neural network where every input neuron is linked to every other neuron at the first level, Receptive fields allow for "pre-processing" of just a certain grid (or part) of the input image.

Examples[edit]

The Gaussian receptive fields are being used in test on cars to allow them to be detected. [4]

Pictures, diagrams[edit]

  1. An image showing the arrangement of neurons in the CNN.
  2. An image showing the receptive field and the CNN.

External links[edit]

References[edit]