Student information
MSc thesis topic: Commodity crops recognition in satellite images using Fourier transform: A new look at signal to frequency domain for detecting commodity crops.
Commodity crops play a significant role in global agricultural production and economic development. Accurate detection and monitoring of commodity crops from satellite imagery are essential for various applications, including land use planning, agricultural management, and environmental monitoring. Traditional methods often rely on spectral analysis, which may overlook important spatial and structural characteristics of crop fields. This research proposes a novel approach that leverages Fourier Transform to analyze satellite images in the frequency domain for improved detection of commodity crops, specifically commodity crops in agroforestry systems.
The recent advance and availability of earth observation data and deep learning technologies provide an opportunity to monitor the earth in high detail and all weather. In this thesis you will use high resolution Planet-NICFI images, Sentinel-1 radar data to detect and assess the performance of fourier transform based deep learning method for mapping commodity crops.
Software: Tensorflow, Pytorch (python) and Google Earth Engine
Objectives
- To explore the feasibility of using Fourier Transform for recognizing commodity crops in satellite images.
- To develop a methodology for extracting frequency domain features from satellite imagery to enhance crop detection accuracy.
- To compare the performance of Fourier Transform-deep learning based crop recognition with traditional spectral and spatial analysis methods.
- To assess the robustness and scalability of the proposed approach across different crop types and geographical regions.
Literature
- Masolele, R.N., et al. (2024). Mapping the diversity of land uses following deforestation across Africa. Sci Rep 14, 1681.
- Du-Ming, et al. (2017). Coffee plantation area recognition in satellite images using Fourier transform. (2020). Computers and Electronics in Agriculture. 35, 115-127.
Requirements
- Mathematics
- Advanced Earth Observation course
- Machine learning course
- Geo-scripting course (Good knowledge in scripting is an asset; e.g. python, Google Earth Engine, and java script)
Theme(s): Sensing & measuring, Modelling & visualisation