|
1 | 1 | <html> |
2 | 2 | <article class="markdown-body entry-content container-lg" itemprop="text"><p align="center" dir="auto"> |
3 | | - </p><div class="markdown-heading" dir="auto"><h1 align="center" tabindex="-1" class="heading-element" dir="auto"><a target="_blank" rel="noopener noreferrer" href="/Sydney-Informatics-Hub/aigis/blob/main/docs/content/aigis.png"><img src="/Sydney-Informatics-Hub/aigis/raw/main/docs/content/aigis.png" alt="aigis logo" width="150px" style="max-width: 100%;"></a> <br> AIGIS</h1><a id="user-content---aigis" class="anchor" aria-label="Permalink: AIGIS" href="#--aigis"><svg class="octicon octicon-link" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true"><path d="m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z"></path></svg></a></div> |
| 3 | + </p><div class="markdown-heading" dir="auto"><h1 align="center" tabindex="-1" class="heading-element" dir="auto"><a target="_blank" rel="noopener noreferrer" href="/docs/content/aigis.png"><img src="/docs/content/aigis.png" alt="aigis logo" width="150px" style="max-width: 100%;"></a> <br> AIGIS</h1><a id="user-content---aigis" class="anchor" aria-label="Permalink: AIGIS" href="#--aigis"><svg class="octicon octicon-link" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true"><path d="m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z"></path></svg></a></div> |
4 | 4 | <p dir="auto"></p> |
5 | 5 | <p align="center" dir="auto"> |
6 | 6 | <em>AI annotation, segmentation, and conversion tools for GIS imagery</em> |
|
14 | 14 | </p><p align="center" dir="auto"> |
15 | 15 | </p> |
16 | 16 | <hr> |
17 | | -<p dir="auto"><a target="_blank" rel="noopener noreferrer" href="/Sydney-Informatics-Hub/aigis/blob/main/docs/content/Greater_Sydney_Dataset.jpg"><img src="/Sydney-Informatics-Hub/aigis/raw/main/docs/content/Greater_Sydney_Dataset.jpg" alt="Greater Sydney Dataset" style="max-width: 100%;"></a></p> |
| 17 | +<p dir="auto"><a target="_blank" rel="noopener noreferrer" href="/docs/content/Greater_Sydney_Dataset.jpg"><img src="/docs/content/Greater_Sydney_Dataset.jpg" alt="Greater Sydney Dataset" style="max-width: 100%;"></a></p> |
18 | 18 | <div class="markdown-heading" dir="auto"><h1 tabindex="-1" class="heading-element" dir="auto">aigis</h1><a id="user-content-aigis" class="anchor" aria-label="Permalink: aigis" href="#aigis"><svg class="octicon octicon-link" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true"><path d="m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z"></path></svg></a></div> |
19 | 19 | <p dir="auto"><code>aigis</code> is a comprehensive toolkit for aerial and satellite imagery acquisition, processing, annotation, and analysis using artificial intelligence. This repository contains three main components:</p> |
20 | 20 | <ol dir="auto"> |
|
66 | 66 | <p dir="auto">Inference and postprocessing both benefit from CUDA GPUs.</p> |
67 | 67 | <div class="markdown-heading" dir="auto"><h2 tabindex="-1" class="heading-element" dir="auto"><code>aigis</code> datasets</h2><a id="user-content-aigis-datasets" class="anchor" aria-label="Permalink: aigis datasets" href="#aigis-datasets"><svg class="octicon octicon-link" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true"><path d="m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z"></path></svg></a></div> |
68 | 68 | <p dir="auto">Explore our two featured example datasets using a <a href="https://felt.com/map/Sydney-Buildings-6drWs3o9ARViitVdt6qp60A?loc=-33.8282,150.8961,10.93z" rel="nofollow">Felt map</a>.</p> |
69 | | -<p dir="auto"><a target="_blank" rel="noopener noreferrer" href="/Sydney-Informatics-Hub/aigis/blob/main/docs/content/Felt_Map.jpeg"><img src="/Sydney-Informatics-Hub/aigis/raw/main/docs/content/Felt_Map.jpeg" alt="felt_map" style="max-width: 100%;"></a> |
| 69 | +<p dir="auto"><a target="_blank" rel="noopener noreferrer" href="/docs/content/Felt_Map.jpeg"><img src="/docs/content/Felt_Map.jpeg" alt="felt_map" style="max-width: 100%;"></a> |
70 | 70 | To load the matching NSW SixMaps aerial photography you have to click 1. background layers, then select 2. NSW SixMaps, as in the image above.</p> |
71 | 71 | <div class="markdown-heading" dir="auto"><h3 tabindex="-1" class="heading-element" dir="auto">Building Segmentation</h3><a id="user-content-building-segmentation" class="anchor" aria-label="Permalink: Building Segmentation" href="#building-segmentation"><svg class="octicon octicon-link" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true"><path d="m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z"></path></svg></a></div> |
72 | 72 | <p dir="auto">We've used the full <code>aigis</code> toolkit to fine tune segmentation models to detect and segment building outlines in aerial imagery. Our fine tuned model was then run on 37,000 high reslution aerial images across the entire Greater Sydney Urban area in New South Wales, Australia.</p> |
73 | 73 | <p dir="auto">Our models's predictions are available as a shapefile in GeoParquet, with almost one million buildings.</p> |
74 | | -<p dir="auto"><a target="_blank" rel="noopener noreferrer" href="/Sydney-Informatics-Hub/aigis/blob/main/docs/content/gsu_buildings.png"><img src="/Sydney-Informatics-Hub/aigis/raw/main/docs/content/gsu_buildings.png" alt="Greater Sydney Buildings" style="max-width: 100%;"></a></p> |
| 74 | +<p dir="auto"><a target="_blank" rel="noopener noreferrer" href="/docs/content/gsu_buildings.png"><img src="/docs/content/gsu_buildings.png" alt="Greater Sydney Buildings" style="max-width: 100%;"></a></p> |
75 | 75 | <div class="markdown-heading" dir="auto"><h3 tabindex="-1" class="heading-element" dir="auto">Tree Segmentation</h3><a id="user-content-tree-segmentation" class="anchor" aria-label="Permalink: Tree Segmentation" href="#tree-segmentation"><svg class="octicon octicon-link" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true"><path d="m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z"></path></svg></a></div> |
76 | 76 | <p dir="auto">We've used the full <code>aigis</code> toolkit to fine tune segmentation models to detect and segment tree patch outlines in aerial imagery. Our fine tuned model was then run on 37,000 high reslution aerial images across the entire Greater Sydney Urban area in New South Wales, Australia.</p> |
77 | 77 | <p dir="auto">Our models's predictions are available as a shapefile in GeoParquet, with 2 million tree patches.</p> |
78 | | -<p dir="auto"><a target="_blank" rel="noopener noreferrer" href="/Sydney-Informatics-Hub/aigis/blob/main/docs/content/gsu_trees.png"><img src="/Sydney-Informatics-Hub/aigis/raw/main/docs/content/gsu_trees.png" alt="Greate Sydney Tree" style="max-width: 100%;"></a></p> |
| 78 | +<p dir="auto"><a target="_blank" rel="noopener noreferrer" href="/docs/content/gsu_trees.png"><img src="/docs/content/gsu_trees.png" alt="Greate Sydney Tree" style="max-width: 100%;"></a></p> |
79 | 79 | <div class="markdown-heading" dir="auto"><h2 tabindex="-1" class="heading-element" dir="auto">Contributing</h2><a id="user-content-contributing" class="anchor" aria-label="Permalink: Contributing" href="#contributing"><svg class="octicon octicon-link" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true"><path d="m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z"></path></svg></a></div> |
80 | 80 | <p dir="auto">Contributions are welcome! Here are several ways you can contribute:</p> |
81 | 81 | <ul dir="auto"> |
|
0 commit comments