An Adaptive Kernels Layer for Deep Neural Networks Based on Spectral Analysis for Image Applications
Open Access
Abstract
As the pixel resolution of imaging equipment has grown larger, the images’ sizes and the number of pixels used to represent objects in images have increased accordingly, exposing an issue when dealing with larger images using the traditional deep learning models and methods, as they typically employ mechanisms such as increasing the models’ depth, which, while suitable for applications that have to be spatially invariant, such as image classification, causes issues for applications that relies on the location of the different features within the images such as object localization and change detection. This paper proposes an adaptive convolutional kernels layer (AKL) as an architecture that adjusts dynamically to images’ sizes in order to extract comparable spectral information from images of different sizes, improving the features’ spatial resolution without sacrificing the local receptive field (LRF) for various image applications, specifically those that are sensitive to objects and features locations, using the definition of Fourier transform and the relation between spectral analysis and convolution kernels. The proposed method is then tested using a Monte Carlo simulation to evaluate its performance in spectral information coverage across images of various sizes, validating its ability to maintain coverage of a ratio of the spectral domain with a variation of around 20% of the desired coverage ratio. Finally, the AKL is validated for various image applications compared to other architectures such as Inception and VGG, demonstrating its capability to match Inception v4 in image classification applications, and outperforms it as images grow larger, up to a 30% increase in accuracy in object localization for the same number of parameters.Keywords
Funding Information
- Natural Science and Engineering Research Council
- Defence Research and Development Canada
This publication has 28 references indexed in Scilit:
- Gabor Convolutional NetworksIEEE Transactions on Image Processing, 2018
- ImageNet classification with deep convolutional neural networksCommunications of the ACM, 2017
- Food/Non-food Image Classification and Food Categorization using Pre-Trained GoogLeNet Model2016
- Deep Residual Learning for Image Recognition2016
- Rethinking the Inception Architecture for Computer Vision2016
- Going deeper with convolutions2015
- Building Change Detection in Multitemporal Very High Resolution SAR ImagesIEEE Transactions on Geoscience and Remote Sensing, 2014
- Cats and dogs2012
- Robust Object Recognition with Cortex-Like MechanismsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2007
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998