Automated Retail Analytics:-Object Detection for Product Visibility and Market Share Analysis

Object detection and data analytics for market presence assessment of business products

Keywords: Satelite Data Analysis, GIS, Google Earth Engine(GEE), Sentinel satelite Data

Brief Description

The project leverages computer vision and data analytics to provide insights into product placement and market presence in retail environments. By analyzing images of store shelves captured by field agents, we detect and quantify the presence of our products and those of competitors. This data is then used to generate actionable insights on market share, product availability, and potential restocking needs across various regions and time periods.

Key Technologies Used:

  • Object Detection: Custom-trained model to identify specific product brands in shelf images
  • Cloud Computing: AWS S3 and Terraform for image storage
  • Data Analytics: Automated analysis of detection results, combining with geographical data
  • Backend Infrastructure: Server deployment for running the object detection model and data processing pipelines

Simple illustration of the project

Object detection output Object detection output

N.B: The code for this project can’t be made public for propritory reasons