Selective Deep Convolutional Neural Network for Low Cost Distorted Image Classification
This addresses the issue of high computational and energy costs for image classification in distorted scenarios, but it is incremental as it builds on existing methods.
The paper tackles the problem of degraded classification accuracy in distorted images by proposing a selective deep convolutional neural network topology, achieving similar accuracy to state-of-the-art networks but with significantly lower cost through fewer weight parameters.
Deep convolutional neural networks have proven to be well suited for image classification applications. However, if there is distortion in the image, the classification accuracy can be significantly degraded, even with state-of-the-art neural networks. The accuracy cannot be significantly improved by simply training with distorted images. Instead, this paper proposes a multiple neural network topology referred to as a selective deep convolutional neural network. By modifying existing state-of-the-art neural networks in the proposed manner, it is shown that a similar level of classification accuracy can be achieved, but at a significantly lower cost. The cost reduction is obtained primarily through the use of fewer weight parameters. Using fewer weights reduces the number of multiply-accumulate operations and also reduces the energy required for data accesses. Finally, it is shown that the effectiveness of the proposed selective deep convolutional neural network can be further improved by combining it with previously proposed network cost reduction methods.