Problem statement
Monitoring Earth’s ecosystems is essential, but traditional methods are slow and
expensive. Satellite imagery offers vast data, but manual analysis is impractical. The
challenge is to create an automated system for efficient and accurate satellite image
analysis to detect environmental changes like deforestation and land use shifts.
Abstract
This project aims to use advanced image analysis and machine learning on
satellite imagery to automate environmental monitoring. The system will identify
changes, like deforestation, through a scalable and robust platform. The goal is
to provide timely insights into environmental shifts by handling large datasets
efficiently.
Outcome
● Automated Change Detection
Identify environmental changes in satellite images such as deforestation or land cover
alterations.
● Data Visualization
Develop an intuitive interface for stakeholders with interactive maps and time-series
analysis.
● Scalable Architecture
Design a scalable system to process data from multiple satellite sources and cover
large geographical areas.
● Machine Learning Integration
Integrate machine learning algorithms to enhance the accuracy of environmental
change detection.
● Alerting System
Implement a real-time alerting mechanism for prompt response to significant
environmental changes.
Reference
Analysis of satellite images plays an increasingly vital role in environment and climate monitoring, especially in detecting and managing natural disaster. In this paper, we proposed an automatic disaster detection system by implementing one of the advance deep learning techniques, convolutional neural network (CNN), to analysis satellite images. The neural network consists of 3 convolutional layers, followed by max-pooling layers after each convolutional layer, and 2 fully connected layers. We created our own disaster detection training data patches, which is currently focusing on 2 main disasters in Japan and Thailand: landslide and flood. Each disaster’s training data set consists of 30000~40000 patches and all patches are trained automatically in CNN to extract region where disaster occurred instantaneously. The results reveal accuracy of 80%~90% for both disaster detection. The results presented here may facilitate improvements in detecting natural disaster efficiently by establishing automatic disaster detection system.
- Keigo Koizumi, Kenji Hirata, Kazuhiro Oda, Yukishige Fujita, Sadayuki Kamide and Tohru Watanabe, “Disaster Detection System Using Sensor Networks and Its Field Experiment Evaluations”, SICE Journal of Control Measurement and System Integration, vol. 5, no. 1, pp. 041-047, January 2012.
- K. Urabe and H. Saji, “Detection of Road Blockage Areas after Earthquake Disaster in Mountainous Districts Using Aerial Images”, Journal of Japan Associationfor Earthquake Engineering, vol. 9, no. 4, 2009.
- A. Krizhevsky, I. Sutskever and G. E. Hinton, “Imagenet classification with deep convolutional neural networks”, Advances in neural information processing systems, pp. 1097-1105, 2012.
- A. Toshev and C. Szegedy, “Deep pose: Human pose estimation via deep neural networks”, Computer Vision and Pattern Recognition (CVPR) 2014 IEEE Conference on, pp. 1653-1660, June 2014.
- C. Farabet, C. Couprie, L. Najman and Y. LeCun, “Learning hierarchical features for scene labeling”, Pattern Analysis and Machine Intelligence IEEE Transactions on, vol. 35, no. 8, pp. 1915-1929, 2013.
https://ieeexplore.ieee.org/document/7730352/references#references