Change Detection from SAR Images Based on Convolutional Neural Networks Guided by Saliency Enhancement

Change detection is an important task in identifying land cover change in different periods. In synthetic aperture radar (SAR) images, the inherent speckle noise leads to false changed points, and this affects the performance of change detection. To improve the accuracy of change detection, a novel automatic SAR image change detection algorithm based on saliency detection and convolutional-wavelet neural networks is proposed. The log-ratio operator is adopted to generate the difference image, and the speckle reducing anisotropic diffusion is used to enhance the original multitemporal SAR images and the difference image. To reduce the influence of speckle noise, the salient area that probably belongs to the changed object is obtained from the difference image. The saliency analysis step can remove small noise regions by thresholding the saliency map, and interest regions can be preserved. Then an enhanced difference image is generated by combing the binarized saliency map and two input images. A hierarchical fuzzy c-means model is applied to the enhanced difference image to classify pixels into the changed, unchanged, and intermediate regions. The convolutional-wavelet neural networks are used to generate the final change map. Experimental results on five SAR data sets indicated the proposed approach provided good performance in change detection compared to state-of-the-art relative techniques, and the values of the metrics computed by the proposed method caused significant improvement.

Forecasting Analysis of Covid-19 Cases with Wavelet Neural Network and Time Series Approach

COVID-19 is a respiratory disease caused by a novel coronavirus first detected in December 2019. As the number of new cases increases rapidly, pandemic fatigue and public disinterest in different response strategies are creating new challenges for government officials in tackling the pandemic. Therefore, government officials need to fully understand the future dynamics of COVID-19 to develop strategic preparedness and flexible response planning. In the light of the above-mentioned conditions, in this study, autoregressive integrated moving average (ARIMA) time series model and Wavelet Neural Networks (WNN) methods are used to predict the number of new cases and new deaths to draw possible future epidemic scenarios. These two methods were applied to publicly available data of the COVID-19 pandemic for Turkey, Italy, and the United Kingdom. In our analysis, excluding Turkey data, the WNN algorithm outperformed the ARIMA model in terms of forecasting consistency. Our work highlighted the promising validation of using wavelet neural networks when making predictions with very few features and a smaller amount of historical data.