Projects

Caribou Lichen Time-Series Analysis in Northern Ontario

Here is a a Google Earth Engine app that I created for leveraging Landsat imagery and AI to perform time-series analysis of lichen-cover in Northern Ontario over three caribou ranges. To analyze these changes, a combination of Convolutional Neural Networks (CNNs) and trend analysis were employed. The CNNs were instrumental in estimating the fraction of lichen cover within each pixel footprint of Landsat imagery. The trend analysis, on the other hand, helped in quantifying the changes in lichen cover over time and identifying the factors driving these changes. The results revealed significant variations in lichen abundance across the three ranges, with some areas showing a decline in lichen cover due to environmental pressures such as climate change and human activities. The implications of these findings are significant for caribou conservation efforts. Caribou rely heavily on lichen as a primary food source, especially in winter. Thus, understanding the dynamics of lichen cover is essential for developing effective management and conservation strategies.

Client-Side Segment Anything Model (SAM) for Remote Sensing Imagery

This is a client-side deep learning image classifier that I developed for classifying remote sensing images. It is still under development and will eventually work with different deep learning architectures and can be used for classifying land cover in remote sensing images without the need for server-side processing. This makes it ideal for applications where remote sensing data are to be processed locally. The model should be hosted locally and imported to the app along with the image. The model should be converted to an ONNX format to be used in this app. Such a web tool can be used to classify different types of land cover, such as forests, water bodies, urban areas, and agricultural lands.

4D Urban Heat Island Analysis

This is a platform to monitor and analyze the urban heat island effect in 4D using multi-sensor data and AI. The goal of this platform is to help build climate-smart cities.