GEOSPECTRA - Advanced Geospatial Analysis Platform

GEOSPECTRA is a cutting-edge web-based platform for advanced geospatial analysis, specializing in AI-powered roof segmentation, building classification, and 3D solar potential analysis. Built for the ASEAN Geospatial Challenge, this platform combines satellite imagery processing with deep learning to deliver precise spatial intelligence.

Features

Building Segmentation Tool

• AI-Powered Detection: Advanced U-Net architecture for precise building segmentation
• Interactive ROI Selection: Draw rectangles, polygons, or circles on satellite imagery
• Multi-Source Imagery: Support for Google Satellite, Landsat 8, and custom uploads
• Real-time Processing: Live progress tracking with step-by-step analysis

Roof Classification

• Material Analysis: Utilizing U-Net for Roof material detection and classification
• Statistical Insights: Comprehensive building analytics and metrics
• Export Capabilities: Download results in multiple formats

3D Solar Analysis

• Solar Potential Mapping: Advanced 3D analysis for solar panel placement
• Shadow Analysis: Comprehensive shadow modeling throughout the year
• Energy Estimation: Accurate solar energy potential calculations
• Interactive 3D Visualization: Immersive 3D building models

🛠️ Technology Stack

• Frontend: HTML5, CSS3, JavaScript (ES6+)
• Styling: TailwindCSS for responsive design
• Mapping: MapLibre GL JS for interactive maps
• Icons: Feather Icons for consistent UI elements
• Fonts: Inter font family for modern typography
• AI Models: U-Net architecture for segmentation and classification
• Data Sources: Google Satellite, OpenStreetMap, Landsat 8

Project Structure

• GEOSPECTRA/
  • components/ - Reusable UI components
  • images/ - Images and assets
  • Scripts/ - Source Code used and Deep Learning Models
  • .gitattributes - Git LFS configuration
  • 3d-viewer.html - 3D viewer interface
  • 3d-viewer2.html - 3D viewer interface (v2)
  • 3d-viewer3.html - 3D viewer interface (v3)
  • LICENSE - Project license file
  • README.md - Project documentation
  • about.html - About page
  • bandung_buildings_3d_fixed.json - 3D building data (Bandung)
  • index.html - Main landing page
  • malang_buildings_3d_fixed.json - 3D building data (Malang)
  • segmentation.html - Legacy segmentation page
  • semarang_reproject_3d_fixed.json - 3D building data (Semarang)
  • style.css - Custom styles
  • team.html - Team information page
  • tool1-segmentation.html - Main segmentation tool
  • tool2-classification.html - Building classification tool
  • tools.html - Tools overview page
  • api.html - Api and references used
  • report-issue.html - Form report for issues in web

🚀 Quick Start

Prerequisites

• Modern web browser (Chrome, Firefox, Safari, Edge)
• Internet connection for map tiles and external resources

Installation

1. Clone the repository

   git clone https://github.com/yourusername/geospectra.git
   cd geospectra

2. Open in browser

   # Option 1: Direct file opening
   open index.html
   
   # Option 2: Local server (recommended)
   python -m http.server 8000
   # Then visit http://localhost:8000

3. Start analyzing!

   • Navigate to the Tools section
   • Select your analysis type
   • Draw your region of interest
   • Run the analysis

Usage Guide

Roof Segmentation Workflow

1. Select Analysis Area

   • Choose your preferred basemap (Satellite/OSM/Terrain)
   • Use drawing tools to define your Region of Interest (ROI)
   • Configure analysis parameters

2. Configure Parameters

   • Set date range for imagery
   • Adjust cloud coverage threshold
   • Select AI segmentation model

3. Run Analysis

   • Click "Run GEOSPECTRA Segmentation"
   • Monitor real-time processing status
   • View results in interactive gallery

4. Export Results

   • Download segmented imagery
   • Export statistical data
   • Continue to classification workflow

Building Classification

1. Import segmentation results or upload new imagery
2. Configure classification parameters
3. Run AI-powered building type detection
4. Review and export classification results

3D Solar Analysis

1. Load 3D building models
2. Configure solar analysis parameters
3. Run comprehensive solar potential analysis
4. Explore interactive 3D visualizations

Key Capabilities

• 🎯 Precision: Sub-meter accuracy in building detection
• ⚡ Speed: Real-time processing with optimized algorithms
• 🌐 Scalability: Handle large-scale urban analysis
• 📊 Analytics: Comprehensive statistical insights
• 🔄 Integration: Seamless workflow between analysis tools
• 📱 Responsive: Works on desktop, tablet, and mobile devices

Team

GEOSPECTRA Team

• Mohammad Zulfi Rahadi Putra
• Raffi Satya Nugraha
• Najieda Azka
• Salzabilla Enzal Putri
• Department of Geodetic Engineering
• Universitas Gadjah Mada
• Yogyakarta, Indonesia

Performance Metrics

• Processing Speed: < 10 seconds for typical ROI
• Resolution Support: 0.5m/pixel trained using Bing satellite maps