Extension of Convolutional Neural Network with General Image Processing Kernels
This work addresses efficiency and performance issues in image classification for researchers and practitioners, though it is incremental as it builds on existing CNN architectures.
The authors tackled the problem of reducing training time and improving accuracy in convolutional neural networks by using pre-defined general-purpose image processing kernels in the first layer, achieving a 30% reduction in training time and 99.56% accuracy on MNIST.
We applied pre-defined kernels also known as filters or masks developed for image processing to convolution neural network. Instead of letting neural networks find its own kernels, we used 41 different general-purpose kernels of blurring, edge detecting, sharpening, discrete cosine transformation, etc. for the first layer of the convolution neural networks. This architecture, thus named as general filter convolutional neural network (GFNN), can reduce training time by 30% with a better accuracy compared to the regular convolutional neural network (CNN). GFNN also can be trained to achieve 90% accuracy with only 500 samples. Furthermore, even though these kernels are not specialized for the MNIST dataset, we achieved 99.56% accuracy without ensemble nor any other special algorithms.