| Hi-index | 0.01 |
Presented is a hierarchical neural network that using some highly simplified ideas from the neurophysiology of the visual cortex can simultaneously recognise multiple objects within the same field of view without recourse to feedback or a competitive learning type algorithm. It employs ensembles of neurons that within each layer are isolated from one another and can adapt itself either with or without supervision, requiring only one forward pass for learning or recognition. The adaption algorithm for each ensemble or neuron is purely local and is independent from the system's output response. Experiments verified the network's ability for simultaneous multiple-object recognition and also gave an indication of its performance with noisy images. A method of automatic neuron threshold dropoff was found to give increased performance in the recognition of noisy images.