Implementing a foveal-pit inspired filter in a Spiking Convolutional Neural Network: a preliminary study
This is an incremental improvement for researchers in neuromorphic computing and computer vision, focusing on noisy image classification.
The authors tackled the problem of improving classification accuracy in Spiking Convolutional Neural Networks by incorporating retinal foveal-pit inspired filters and rank-order encoding, achieving up to 90% accuracy compared to 57% without filtering on digit and vehicle recognition tasks.
We have presented a Spiking Convolutional Neural Network (SCNN) that incorporates retinal foveal-pit inspired Difference of Gaussian filters and rank-order encoding. The model is trained using a variant of the backpropagation algorithm adapted to work with spiking neurons, as implemented in the Nengo library. We have evaluated the performance of our model on two publicly available datasets - one for digit recognition task, and the other for vehicle recognition task. The network has achieved up to 90% accuracy, where loss is calculated using the cross-entropy function. This is an improvement over around 57% accuracy obtained with the alternate approach of performing the classification without any kind of neural filtering. Overall, our proof-of-concept study indicates that introducing biologically plausible filtering in existing SCNN architecture will work well with noisy input images such as those in our vehicle recognition task. Based on our results, we plan to enhance our SCNN by integrating lateral inhibition-based redundancy reduction prior to rank-ordering, which will further improve the classification accuracy by the network.