geoai-py 0.8.2__tar.gz → 0.9.0__tar.gz

{geoai_py-0.8.2 → geoai_py-0.9.0}/.github/workflows/docs-build.yml

@@ -16,6 +16,9 @@ jobs:
               with:
                   fetch-depth: 0
 
+            # - name: Install GDAL system dependencies
+            #   run: sudo apt-get update && sudo apt-get install -y gdal-bin libgdal-dev
+
             - name: Install uv
               uses: astral-sh/setup-uv@v6
               with:
@@ -33,15 +36,15 @@ jobs:
 
             - name: Install optional dependencies
               run: |
-                  uv pip install --find-links https://girder.github.io/large_image_wheels GDAL pyproj
+                  # uv pip install --find-links https://girder.github.io/large_image_wheels GDAL pyproj
                   uv pip install pytest
                   uv pip install "geoai-py[extra]"
 
-            - name: Test import
-              run: |
-                  uv run python -c "import geoai; print('geoai import successful')"
-                  uv run python -c "from osgeo import gdal; print('gdal import successful')"
-                  uv run gdalinfo --version
+            # - name: Test import
+            #   run: |
+            #       uv run python -c "import geoai; print('geoai import successful')"
+            #       uv run python -c "from osgeo import gdal; print('gdal import successful')"
+            #       uv run gdalinfo --version
 
             - name: Running pytest
               run: |

{geoai_py-0.8.2 → geoai_py-0.9.0}/.github/workflows/docs.yml

@@ -15,6 +15,9 @@ jobs:
               with:
                   fetch-depth: 0
 
+            # - name: Install GDAL system dependencies
+            #   run: sudo apt-get update && sudo apt-get install -y gdal-bin libgdal-dev
+
             - name: Install uv
               uses: astral-sh/setup-uv@v6
               with:
@@ -32,15 +35,15 @@ jobs:
 
             - name: Install optional dependencies
               run: |
-                  uv pip install --find-links https://girder.github.io/large_image_wheels GDAL pyproj
+                  # uv pip install --find-links https://girder.github.io/large_image_wheels GDAL pyproj
                   uv pip install pytest
                   uv pip install "geoai-py[extra]"
 
-            - name: Test import
-              run: |
-                  uv run python -c "import geoai; print('geoai import successful')"
-                  uv run python -c "from osgeo import gdal; print('gdal import successful')"
-                  uv run gdalinfo --version
+            # - name: Test import
+            #   run: |
+            #       uv run python -c "import geoai; print('geoai import successful')"
+            #       uv run python -c "from osgeo import gdal; print('gdal import successful')"
+            #       uv run gdalinfo --version
 
             - name: Running pytest
               run: |

{geoai_py-0.8.2 → geoai_py-0.9.0}/.github/workflows/ubuntu.yml

@@ -19,6 +19,9 @@ jobs:
         steps:
             - uses: actions/checkout@v4
 
+            # - name: Install GDAL system dependencies
+            #   run: sudo apt-get update && sudo apt-get install -y gdal-bin libgdal-dev
+
             - name: Install uv
               uses: astral-sh/setup-uv@v6
               with:
@@ -35,15 +38,15 @@ jobs:
 
             - name: Install optional dependencies
               run: |
-                  uv pip install --find-links https://girder.github.io/large_image_wheels gdal pyproj
+                #   uv pip install --find-links https://girder.github.io/large_image_wheels gdal pyproj
                   uv pip install pytest
                   uv pip install "geoai-py[extra]"
 
-            - name: Test import
-              run: |
-                  uv run python -c "import geoai; print('geoai import successful')"
-                  uv run python -c "from osgeo import gdal; print('gdal import successful')"
-                  uv run gdalinfo --version
+            # - name: Test import
+            #   run: |
+            #       uv run python -c "import geoai; print('geoai import successful')"
+            #       uv run python -c "from osgeo import gdal; print('gdal import successful')"
+            #       uv run gdalinfo --version
 
             - name: Running pytest
               run: |

{geoai_py-0.8.2 → geoai_py-0.9.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: geoai-py
-Version: 0.8.2
+Version: 0.9.0
 Summary: A Python package for using Artificial Intelligence (AI) with geospatial data
 Author-email: Qiusheng Wu <giswqs@gmail.com>
 License: MIT License
@@ -19,6 +19,7 @@ License-File: LICENSE
 Requires-Dist: albumentations
 Requires-Dist: buildingregulariser
 Requires-Dist: contextily
+Requires-Dist: ever-beta
 Requires-Dist: geopandas
 Requires-Dist: huggingface_hub
 Requires-Dist: jupyter-server-proxy
@@ -35,6 +36,7 @@ Requires-Dist: rioxarray
 Requires-Dist: scikit-image
 Requires-Dist: scikit-learn
 Requires-Dist: torch
+Requires-Dist: torchange
 Requires-Dist: torchgeo
 Requires-Dist: torchinfo
 Requires-Dist: tqdm

geoai_py-0.9.0/docs/change_detection.md

@@ -0,0 +1,3 @@
+# change_detection module
+
+::: geoai.change_detection

geoai_py-0.9.0/docs/examples/change_detection.ipynb

@@ -0,0 +1,422 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Change Detection with Instance Segmentation\n",
+    "\n",
+    "[![image](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/opengeos/geoai/blob/main/docs/examples/change_detection.ipynb)\n",
+    "\n",
+    "This notebook demonstrates the change detection functionality in GeoAI, which provides instance segmentation and confidence scoring for individual change objects. \n",
+    "\n",
+    "The change detection functionality builds upon the [torchange](https://github.com/Z-Zheng/pytorch-change-models) package developed by Dr. Zhuo Zheng. We have made it much easier to analyze remote sensing imagery and visualize the results.\n",
+    "\n",
+    "## Overview\n",
+    "\n",
+    "The change detection system provides:\n",
+    "\n",
+    "- **Instance Segmentation**: Each change object gets a unique ID\n",
+    "- **Confidence Scores**: Individual confidence values for each detected instance\n",
+    "- **Proper GeoTIFF Output**: Maintains spatial reference information\n",
+    "\n",
+    "## Key Features\n",
+    "\n",
+    "- Instance-level change detection with unique IDs\n",
+    "- Confidence scoring for quality assessment\n",
+    "- Support for large GeoTIFF files\n",
+    "- Comprehensive analysis capabilities"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Install packages"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# %pip install geoai-py"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Import libraries"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import geoai\n",
+    "import os\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "from pathlib import Path"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Setup"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Check if CUDA is available\n",
+    "device = geoai.get_device()\n",
+    "print(f\"Using device: {device}\")\n",
+    "\n",
+    "# Set up paths\n",
+    "out_folder = \"change_detection_results\"\n",
+    "Path(out_folder).mkdir(exist_ok=True)\n",
+    "\n",
+    "print(f\"Working directory: {out_folder}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Download sample data\n",
+    "\n",
+    "We'll use NAIP imagery for Las Vegas to demonstrate change detection."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Download NAIP imagery\n",
+    "naip_2019_url = \"https://huggingface.co/datasets/giswqs/geospatial/resolve/main/las_vegas_naip_2019_a.tif\"\n",
+    "naip_2022_url = \"https://huggingface.co/datasets/giswqs/geospatial/resolve/main/las_vegas_naip_2022_a.tif\"\n",
+    "\n",
+    "naip_2019_path = geoai.download_file(naip_2019_url)\n",
+    "naip_2022_path = geoai.download_file(naip_2022_url)\n",
+    "\n",
+    "print(f\"Downloaded 2019 NAIP: {naip_2019_path}\")\n",
+    "print(f\"Downloaded 2022 NAIP: {naip_2022_path}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Visualize sample data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Check raster information\n",
+    "geoai.get_raster_info(naip_2019_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# View the images\n",
+    "geoai.view_raster(naip_2019_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "geoai.view_raster(naip_2022_path)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Initialize Change Detection\n",
+    "\n",
+    "Create the change detection system with optimal parameters."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Initialize change detection\n",
+    "detector = geoai.ChangeDetection(sam_model_type=\"vit_h\")\n",
+    "\n",
+    "# Configure parameters (following the torchange example)\n",
+    "detector.set_hyperparameters(\n",
+    "    change_confidence_threshold=145,\n",
+    "    use_normalized_feature=True,\n",
+    "    bitemporal_match=True,\n",
+    ")\n",
+    "\n",
+    "detector.set_mask_generator_params(\n",
+    "    points_per_side=32,\n",
+    "    stability_score_thresh=0.95,\n",
+    ")\n",
+    "\n",
+    "print(\"Change detection system initialized!\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Run Change Detection\n",
+    "\n",
+    "Execute change detection with instance segmentation and confidence scoring."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Run change detection\n",
+    "results = detector.detect_changes(\n",
+    "    naip_2019_path,\n",
+    "    naip_2022_path,\n",
+    "    output_path=f\"{out_folder}/binary_mask.tif\",\n",
+    "    export_probability=True,\n",
+    "    probability_output_path=f\"{out_folder}/probability_mask.tif\",\n",
+    "    export_instance_masks=True,\n",
+    "    instance_masks_output_path=f\"{out_folder}/instance_masks.tif\",\n",
+    "    return_detailed_results=True,\n",
+    "    return_results=False,\n",
+    ")\n",
+    "\n",
+    "print(f\"Change detection completed!\")\n",
+    "print(f\"Total instances detected: {results['summary']['total_masks']}\")\n",
+    "print(f\"Image size: {results['summary']['original_shape']}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Analyze Results\n",
+    "\n",
+    "Display key statistics and quality metrics."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Display statistics\n",
+    "if \"statistics\" in results and results[\"statistics\"]:\n",
+    "    print(\"Quality Statistics:\")\n",
+    "    for metric, stats in results[\"statistics\"].items():\n",
+    "        print(f\"  {metric}: mean={stats['mean']:.3f}, std={stats['std']:.3f}\")\n",
+    "\n",
+    "# Show top instances\n",
+    "if \"masks\" in results and len(results[\"masks\"]) > 0:\n",
+    "    print(\"\\nTop 5 detected instances:\")\n",
+    "    for i, mask in enumerate(results[\"masks\"][:5]):\n",
+    "        print(\n",
+    "            f\"  {i+1}. Instance {mask['mask_id']}: \"\n",
+    "            f\"IoU={mask['iou_pred']:.3f}, \"\n",
+    "            f\"Stability={mask['stability_score']:.3f}, \"\n",
+    "            f\"Area={mask['area']} pixels\"\n",
+    "        )"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Visualizations\n",
+    "\n",
+    "Use the integrated visualization methods for comprehensive analysis."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# probability visualization\n",
+    "detector.visualize_results(\n",
+    "    naip_2019_path,\n",
+    "    naip_2022_path,\n",
+    "    f\"{out_folder}/binary_mask.tif\",\n",
+    "    f\"{out_folder}/probability_mask.tif\",\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "![](https://github.com/user-attachments/assets/e7c00b50-c456-4653-b8ce-0c9ec8f05b7f)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Create split comparison visualization\n",
+    "detector.create_split_comparison(\n",
+    "    naip_2019_path,\n",
+    "    naip_2022_path,\n",
+    "    f\"{out_folder}/binary_mask.tif\",\n",
+    "    f\"{out_folder}/probability_mask.tif\",\n",
+    "    f\"{out_folder}/split_comparison.png\",\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Analyze individual instances\n",
+    "instance_stats = detector.analyze_instances(\n",
+    "    f\"{out_folder}/instance_masks.tif\",\n",
+    "    f\"{out_folder}/instance_masks_scores.tif\",\n",
+    "    f\"{out_folder}/instance_analysis.png\",\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "![](https://github.com/user-attachments/assets/ea1f8a51-ea14-415a-9733-78b243061dd3)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Comprehensive Analysis Report\n",
+    "\n",
+    "Generate a detailed analysis report combining all metrics."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Create comprehensive analysis report\n",
+    "detector.create_comprehensive_report(results, f\"{out_folder}/comprehensive_report.png\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "![](https://github.com/user-attachments/assets/629caf85-0713-4e04-8023-f4273edbbb4c)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## One-Click Complete Analysis\n",
+    "\n",
+    "For ultimate simplicity, use the complete analysis method."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Alternative: Run complete analysis in one step\n",
+    "# This method does everything - detection, file outputs, and all visualizations\n",
+    "complete_results = detector.run_complete_analysis(\n",
+    "    naip_2019_path, naip_2022_path, \"complete_analysis_results\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Summary\n",
+    "\n",
+    "This notebook demonstrated the change detection functionality in GeoAI with integrated visualization methods:\n",
+    "\n",
+    "### Key Features Used:\n",
+    "1. **Change Detection**: Instance segmentation with confidence scoring\n",
+    "2. **Integrated Visualizations**: Built-in methods for comprehensive analysis\n",
+    "3. **Simplified API**: Clean, streamlined interface following geoai patterns\n",
+    "4. **Complete Analysis**: One-click method for full analysis workflow\n",
+    "\n",
+    "### Output Files Generated:\n",
+    "- `binary_mask.tif`: Traditional binary change detection\n",
+    "- `probability_mask.tif`: Probability-weighted change detection  \n",
+    "- `instance_masks.tif`: Instance segmentation with unique IDs\n",
+    "- `instance_masks_scores.tif`: Confidence scores for each instance\n",
+    "- `enhanced_probability_results.png`: Comprehensive visualization\n",
+    "- `split_comparison.png`: Before/after split comparison\n",
+    "- `instance_analysis.png`: Individual instance analysis\n",
+    "- `comprehensive_report.png`: Complete analysis report\n",
+    "\n",
+    "### Advantages Over Traditional Methods:\n",
+    "1. **Instance-Level Analysis**: Each change object has unique ID and metrics\n",
+    "2. **Quality Assessment**: Confidence scores for filtering and ranking\n",
+    "3. **Rich Visualizations**: Multiple analysis perspectives in one package\n",
+    "4. **Simplified Workflow**: Integrated methods reduce code complexity\n",
+    "5. **Flexible Usage**: From simple detection to comprehensive analysis"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "geo",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}