Satellite Image Analysis for Customer Identification:
Objective:
Satellite-based customer identification is valuable for its ability to provide detailed and scalable insights into customer demographics and behavior across large regions. It enables businesses to optimize marketing strategies, allocate resources more effectively, and make informed decisions based on accurate spatial and environmental data. In this project, I conteibuted to the development of a customer identification tool based on analysis of satellite images.
Contributions:
- Trained a Unet algorithm on RGB-NIR Sentinel satellite images for land type segmentation.
- Developed an ARIMA based method for time series analysis of the segmented regions to forecast prospective customers for agricultural supplies.
- Designed a REST API application for segmentation and time series analysis to support the development of a cloud based application for Satellite Image based analytics.
Result:
- Reduced customer identification cost for the agriculture supplier client by 26%.