Machine learning approaches are increasingly used to extract patterns and insights from the ever-increasing stream of geospatial data, but current approaches may not be optimal when system behaviour is dominated by spatial or temporal context. Here, rather than amending classical machine learning, we argue that these contextual cues should be used as part of deep learning (an approach that is able to extract spatio-temporal features automatically) to gain further process understanding of Earth system science problems, improving the predictive ability of seasonal forecasting and modelling of long-range spatial connections across multiple timescales, for example. The next step will be a hybrid modelling approach, coupling physical process models with the versatility of data-driven machine learning.

/pdf/deep-learning-and-process-understanding-for-data-driven-3so9cx99my.pdf

Deep learning and process understanding for data-driven Earth system science

Predicting electricity consumption for commercial and residential buildings using deep recurrent neural networks

https://escholarship.org/content/qt48x0t0b1/qt48x0t0b1.pdf?t=q8he9y

Modeling and simulating of reservoir operation using the artificial neural network, support vector regression, deep learning algorithm.

Use Long Short-Term Memory to Enhance Internet of Things for Combined Sewer Overflow Monitoring

Time-series forecasting is applied to many areas of smart factories, including machine health monitoring, predictive maintenance, and production scheduling. In smart factories, machine speed prediction can be used to dynamically adjust production processes based on different system conditions, optimize production throughput, and minimize energy consumption. However, making accurate data-driven machine speed forecasts is challenging. Given the complex nature of industrial manufacturing process data, predictive models that are robust to noise and can capture the temporal and spatial distributions of input time-series signals are prerequisites for accurate forecasting. Motivated by recent deep learning studies in smart manufacturing, in this article, we propose an end-to-end model for multistep machine speed prediction. The model comprises a deep convolutional LSTM encoder–decoder architecture. Extensive empirical analyses using real-world data obtained from a metal packaging plant in the United Kingdom demonstrate the value of the proposed method when compared with the state-of-the-art predictive models.

/pdf/a-deep-learning-model-for-smart-manufacturing-using-3zqpl67wdb.pdf

A Deep Learning Model for Smart Manufacturing Using Convolutional LSTM Neural Network Autoencoders

The aim of this paper is to present a deep neural network architecture and use it in time series weather prediction. It uses multi stacked LSTMs to map sequences of weather values of the same length. The final goal is to produce two types of models per city (for 9 cities in Morocco) to forecast 24 and 72 hours worth of weather data (for Temperature, Humidity and Wind Speed). Approximately 15 years (2000-2015) of hourly meteorological data was used to train the model. The results show that LSTM based neural networks are competitive with the traditional methods and can be considered a better alternative to forecast general weather conditions.

/pdf/sequence-to-sequence-weather-forecasting-with-long-short-4jp0ep5ntt.pdf

Sequence to Sequence Weather Forecasting with Long Short-Term Memory Recurrent Neural Networks

This work addresses the problem of recovering lost or damaged satellite image pixels (gaps) caused by sensor processing errors or by natural phenomena like cloud presence. Such errors decrease our ability to monitor regions of interest and significantly increase the average revisit time for all satellites. This paper presents a novel neural system based on conditional deep generative adversarial networks (cGAN) optimized to fill satellite imagery gaps using surrounding pixel values and static high-resolution visual priors. Experimental results show that the proposed system outperforms traditional and neural network baselines. It achieves a normalized least absolute deviations error of (  &  decrease in error compared with the two baselines) and a mean squared error loss of  (  &  decrease in error) over the test set. The model can be deployed within a remote sensing data pipeline to reconstruct missing pixel measurements for near-real-time monitoring and inference purposes, thus empowering policymakers and users to make environmentally informed decisions.

https://ijai.iaescore.com/index.php/IJAI/article/download/20740/13081

Satellite image inpainting with deep generative adversarial neural networks

This article aims to present an end-to-end software solution capable of providing up to date weather and pollution values and health recommendations based on User profiles and personal health data, while making use of environmental satellite data processed in the back-end. this system demonstrates the possible range of applications of satellite-backed environmental systems that can assist and potentially replace the current expensive sensor-based systems, especially in developing countries in Africa.

/pdf/air2day-an-air-quality-monitoring-adviser-in-morocco-xul28nwrk0.pdf

Air2Day: An Air Quality Monitoring Adviser in Morocco

. We propose a deep learning method for Atmospheric Ozone Interpolation. Our method directly learns an end-to-end mapping between classically interpolated satellite ozone images and the real ozone measurements. The model's architecture represents a deep stack of convolutions (CNN) that takes the already interpolated images (Using the classical state-of-the-art interpolation method) as Input and outputs a more precise Interpolation of the Region of Interest. Our deep CNN has a lightweight structure, yet demonstrates state-of-the-art interpolation quality, and achieves optimal data processing latency (∆T) for production-ready near-real-time Atmospheric Image Interpolation, which has a big advantage over the state of the art classical interpolation algorithms. We explore different network structures and parameter settings to achieve trade-offs between performance and speed. This method showcases the potential applications of deep learning in Remote Sensing and Climate Science.

/pdf/ozonet-atmospheric-ozone-interpolation-with-deep-2g1oi9nnl8.pdf

OzoNet: Atmospheric Ozone Interpolation with Deep ConvolutionalNeural Networks

This paper presents a data processing system based on an architecture comprised of multiple stacked layers of computational processes that transforms Raw Binary Pollution Data coming directly from Two EUMETSAT MetOp satellites to our servers, into ready to interpret and visualise continuous data stream in near real time using techniques varying from task automation, data preprocessing and data analysis to machine learning using feedforward artificial neural networks. The proposed system handles the acquisition, cleaning, processing, normalizing, and predicting of Pollution Data in our area of interest of Morocco.

/pdf/metop-satellites-data-processing-for-air-pollution-4br44n8zj6.pdf

Mohamed Akram Zaytar

Papers

Sequence to Sequence Weather Forecasting with Long Short-Term Memory Recurrent Neural Networks

Satellite image inpainting with deep generative adversarial neural networks

Air2Day: An Air Quality Monitoring Adviser in Morocco

OzoNet: Atmospheric Ozone Interpolation with Deep ConvolutionalNeural Networks

MetOp Satellites Data Processing for Air Pollution Monitoring in Morocco