WebNova Container Challenge
WebNova II Space App Hackathon
A computer vision solution developed for the WebNova Container Challenge hackathon. The system counts differences in shipping containers between two satellite images of the same geographic location using advanced image processing techniques including SIFT feature detection and BruteForce matching algorithms.
The Objective
To count the differences in shipping containers between two images of the same Latitude/Longitude coordinates. This solution helps in:
- Inventory Tracking - Monitor container movements at ports and shipping yards
- Change Detection - Identify additions and removals of containers over time
- Logistics Optimization - Provide data for supply chain management decisions
- Automated Monitoring - Enable satellite-based surveillance of shipping infrastructure
Technical Solution
1. Feature Detection
The two images are processed using SIFT (Scale-Invariant Feature Transform) to detect and extract key features from the container imagery.
2. Feature Matching
Features are compared using BruteForce matching algorithm (BFCompare) to identify corresponding points between the two images.
3. Difference Calculation
The distance difference against a configurable threshold is used to calculate variations and identify container changes.
Output
The system generates differential imagery highlighting the changes and reports the count of container differences. Results are output to a designated directory for further analysis, enabling:
- Visual representation of container additions/removals
- Quantitative count of differences detected
- Confidence scoring based on feature matching quality
- Exportable results for integration with other systems
Technology Stack
Team
Team Members: Pankhuri Gupta, Sathya Krishansamy
Achievement Certificate
Impact
- Scalable Solution - Can be applied to any satellite imagery for container detection
- Cost-Effective - Automated analysis reduces manual inspection costs
- Real-World Application - Directly applicable to logistics and supply chain monitoring