Huanfeng Shen, Wuhan University ([email protected]), Jane Liu, University of Toronto ([email protected]), Wenping Yuan, Sun Yat-Sen University ([email protected]), Yongguang Zhang, Nanjing University ([email protected]), Holly Croft, University of Sheffield ([email protected]), Xiaobin Guan, Wuhan University ([email protected]). The dramatic increase in anthropogenic carbon emissions over the last five decades has already led to substantial damage to our environment, including increases in extreme weatherevents, loss of biodiversity, and a rise in sea level. Carbon neutrality, i.e., net-zero anthropogenic carbon emissions, is necessary to ensure the sustainable future of human beings, and hundreds of countries have pledged to achieve this goal by mid-century. Remote sensing techniques can acquire frequent observations of the Earth with various temporal and spatial resolutions, and provide substantial information for carbon emission monitoring and carbon cycle modeling. Remote sensing observations not only can be directly applied to retrieve the atmospheric concentrations of greenhouse gases (e.g., CO2, CO, CH4, CFCs, O3, et al.), but also can be employed to investigate the carbon budget of natural ecosystems.

Developed by Minderoo Foundation, the'Global Plastic Watch' tool uses advanced satellite data technology and machine learning to create a near-real-time, high resolution map of plastic pollution. The tool aims to help authorities better manage plastic leakage into the marine environment, and is said to provide the largest ever open source dataset of plastic waste across dozens of countries. Global Plastic Watch uses remote sensing satellite imagery from the European Space Agency and a novel machine learning model created in collaboration with digital product agency Earthrise Media. The tool can determine the size and scale of land-based plastic waste sites, which fuel the growing issue of plastic pollution in the world's rivers and oceans. By using the data, governments, industry and communities can evaluate and monitor the risk of land-based plastic waste sites as well as prioritise investment in solutions, Minderoo Foundation said.

This blog is accompanied by a Colab notebook which provides an in-depth look at how Raster Vision works and allows you to run each experiment discussed in this post yourself. Change detection is the computer-vision equivalent of the spot-the-difference game. Given two images, the model must detect all the points at which they differ. In the context of remote sensing, these images are usually satellite or aerial images of the same geographical location at two different points in time. Change detection has been an active research area for a long time and the literature is rich with algorithms that perform the task automatically, ranging from basic image processing techniques to present-day deep neural networks.

Since frequent severe droughts are lengthening the dry season in the Amazon Rainforest, it is important to detect wildfires promptly and forecast possible spread for effective suppression response. Current wildfire detection models are not versatile enough for the low-technology conditions of South American hot spots. This deep learning study first trains a Fully Convolutional Neural Network on Landsat 8 images of Ecuador and the Galapagos, using Green and Short-wave Infrared bands to predict pixel-level binary fire masks. This model achieves a 0.962 validation F2 score and a 0.932 F2 score on test data from Guyana and Suriname. Afterward, image segmentation is conducted on the Cirrus band using K-Means Clustering to simplify continuous pixel values into three discrete classes representing differing degrees of cirrus cloud contamination. Three additional Convolutional Neural Networks are trained to conduct a sensitivity analysis measuring the effect of simplified features on model accuracy and train time. The Experimental model trained on the segmented cirrus images provides a statistically significant decrease in train time compared to the Control model trained on raw cirrus images, without compromising binary accuracy. This proof of concept reveals that feature engineering can improve the performance of wildfire detection models by lowering computational expense.

The rapid urban growth and development have been putting increasing pressure on the environment including urban parks and green spaces. The green spaces are essential to improve the urban areas and to provide quality life to the urban population. Green spaces generally include lawns, public parks, gardens, street landscapes, forests, etc. In this regard, technologies such as satellite imagery and AI/ML can support the urban developers as well as the land managers to monitor and support decision-making for sustainable urban development in dense urban environments and prevent flooding conditions in urban areas by gathering high-resolution details of the urban area(s). Further, satellite images can provide detailed analysis for detecting major changes in the urban land cover and land use that allows frequent coverage and overlaying of different time sequences to classify environmentally safe and sustainable areas for any proposed development area(s).

Module #6 of Metis' Data Science and Machine Learning bootcamp is all wrapped up! For this module we focused on Deep Learning, working with non-tabular data, and building models using Google's Tensorflow library. For our project, we were tasked with creating an image classification model to solve for a real-world problem. This module took place during the Russian invasion of Ukraine. The conflict has highlighted the use of satellite imagery by journalists, human rights organizations, and open-source intelligence analysts.

Crop maps are crucial for agricultural monitoring and food management and can additionally support domain-specific applications, such as setting cold supply chain infrastructure in developing countries. Machine learning (ML) models, combined with freely-available satellite imagery, can be used to produce cost-effective and high spatial-resolution crop maps. However, accessing ground truth data for supervised learning is especially challenging in developing countries due to factors such as smallholding and fragmented geography, which often results in a lack of crop type maps or even reliable cropland maps. Our area of interest for this study lies in Himachal Pradesh, India, where we aim at producing an open-access binary cropland map at 10-meter resolution for the Kullu, Shimla, and Mandi districts. To this end, we developed an ML pipeline that relies on Sentinel-2 satellite images time series. We investigated two pixel-based supervised classifiers, support vector machines (SVM) and random forest (RF), which are used to classify per-pixel time series for binary cropland mapping. The ground truth data used for training, validation and testing was manually annotated from a combination of field survey reference points and visual interpretation of very high resolution (VHR) imagery. We trained and validated the models via spatial cross-validation to account for local spatial autocorrelation and selected the RF model due to overall robustness and lower computational cost. We tested the generalization capability of the chosen model at the pixel level by computing the accuracy, recall, precision, and F1-score on hold-out test sets of each district, achieving an average accuracy for the RF (our best model) of 87%. We used this model to generate a cropland map for three districts of Himachal Pradesh, spanning 14,600 km2, which improves the resolution and quality of existing public maps.

Geospatial Data Analyses & Remote Sensing: 5 Classes in 1 Do you need to design a GIS map or satellite-imagery based map for your Remote Sensing or GIS project but you don't know how to do this? Have you heard about Remote Sensing object-based image analysis and machine learning or maybe QGIS or Google Earth Engine but did not know where to start with such analyses? Do you find Remote Sensing and GIS manuals too not practical and looking for a course that takes you by hand, teach you all the concepts, and get you started on a real-life GIS mapping project? I'm very excited that you found my Practical Geospatial Masterclass on Geospatial Data Analyses & Remote Sensing. This course provides and information that is usually delivered in 4 separate Geospatial Data Analyses & Remote Sensing courses, and thus you with learning all the necessary information to start and advance with Geospatial analysis and includes more than 9 hours of video content, plenty of practical analysis, and downloadable materials.

Become proficient in satellite remote sensing, spatial analysis principles, methods, applications, sensors, and GIS! Get this course for only 9.99. Do you find that other remote sensing courses are too short and vague, and do not prepare you for real world problems? Are you looking for a course that goes IN-DEPTH and teaches you all the fundamentals of remote sensing? My course provides a solid foundation to carry out practical, real life remote sensing spatial data analysis and gives you the techniques and knowledge to tackle a variety of geological and environmental problems. This course provides an introduction to remote sensing - the acquisition of information about the earth from a distance, typically via airborne and spaceborne sensors.

Learn to apply machine learning, remote sensing, big spatial data using the Google Earth Engine cloud computing. Do you want to learn how to access, process and analyze remote sensing data using open source cloud-based platforms? Do you want to master machine learning algorithms to predict Earth Observation big data? Do you want to start a spatial data scientist career in the geospatial industry? Enroll in my new course to master Machine Learning with Remote Sensing in Google Earth Engine.