{"id":21318799,"url":"https://github.com/sydney-informatics-hub/aigis","last_synced_at":"2025-07-12T03:31:06.412Z","repository":{"id":220294164,"uuid":"751260428","full_name":"Sydney-Informatics-Hub/aigis","owner":"Sydney-Informatics-Hub","description":"Tools for aerial and satellite imagery acquisition, processing, annotation, and analysis using artificial intelligence","archived":false,"fork":false,"pushed_at":"2024-06-21T07:18:40.000Z","size":5699,"stargazers_count":4,"open_issues_count":5,"forks_count":1,"subscribers_count":7,"default_branch":"main","last_synced_at":"2024-06-22T00:38:27.224Z","etag":null,"topics":["aerial-imagery","computer-vision","geospatial","gis","python","pytorch","segmentation","transformers"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Sydney-Informatics-Hub.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null}},"created_at":"2024-02-01T08:54:22.000Z","updated_at":"2024-06-21T07:18:42.000Z","dependencies_parsed_at":"2024-04-02T01:43:30.899Z","dependency_job_id":null,"html_url":"https://github.com/Sydney-Informatics-Hub/aigis","commit_stats":null,"previous_names":["sydney-informatics-hub/aigis"],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sydney-Informatics-Hub%2Faigis","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sydney-Informatics-Hub%2Faigis/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sydney-Informatics-Hub%2Faigis/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sydney-Informatics-Hub%2Faigis/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Sydney-Informatics-Hub","download_url":"https://codeload.github.com/Sydney-Informatics-Hub/aigis/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":225789337,"owners_count":17524414,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["aerial-imagery","computer-vision","geospatial","gis","python","pytorch","segmentation","transformers"],"created_at":"2024-11-21T19:23:05.526Z","updated_at":"2025-07-12T03:31:06.399Z","avatar_url":"https://github.com/Sydney-Informatics-Hub.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n    \u003ch1 align=\"center\"\u003e\u003cimg src=\"docs/content/aigis.png\" alt=\"aigis logo\" width=150px\u003e \u003cbr\u003e AIGIS\u003c/h1\u003e\n\t\n\n\n\u003c/p\u003e\n\u003cp align=\"center\"\u003e\n    \u003cem\u003eAI annotation, segmentation, and conversion tools for GIS imagery\u003c/em\u003e\n\u003c/p\u003e\n\u003cp align=\"center\"\u003e\n\t\u003cimg src=\"https://img.shields.io/github/license/Sydney-Informatics-Hub/aigis?style=flat\u0026color=0080ff\" alt=\"license\"\u003e\n\t\u003cimg src=\"https://img.shields.io/github/last-commit/Sydney-Informatics-Hub/aigis?style=flat\u0026color=0080ff\" alt=\"last-commit\"\u003e\n\t\u003cimg src=\"https://img.shields.io/github/languages/top/Sydney-Informatics-Hub/aigis?style=flat\u0026color=0080ff\" alt=\"repo-top-language\"\u003e\n\t\u003cimg src=\"https://img.shields.io/github/languages/count/Sydney-Informatics-Hub/aigis?style=flat\u0026color=0080ff\" alt=\"repo-language-count\"\u003e\n\u003cp\u003e\n\u003cp align=\"center\"\u003e\n\t\u003c/p\u003e\n\u003chr\u003e\n\n![Greater Sydney Dataset](./docs/content/Greater_Sydney_Dataset.jpg)\n\n# aigis\n\n`aigis` is a comprehensive toolkit for aerial and satellite imagery acquisition, processing, annotation, and analysis using artificial intelligence. This repository contains three main components:\n\n1. **annotate:** Tools for annotating aerial imagery data.\n2. **convert:** Utilities for converting aerial imagery data to various formats, including COCO, GeoJSON, etc.\n3. **segment:** Scripts and notebooks for segmenting aerial imagery using deep learning models.\n\n## Why `aigis`?\n\nRather than develop models that depend on getting a time series of images, specific input resolutions, electromagentic radiation bands, or mission prameters, we've created an AI GIS toolkit that uses generalist visual light vision transformer and MaskRCNN models.\n\nTrain, test, deploy bespoke prediction models to classify, detect, and segment aerial \u0026 satellite imagery with `aigis`.\n\nBut wait, you don't have training data? No problem, use `aigis.annotate` module and generate annotations from scratch!\n\nTake those annotations and run them through `aigis.segment` pipeline and now you have lightweight, production ready models that can run at continental scale or be deployed on compute-limited edge applications for real-time inference in the field.\n\nThe glue linking this all together, `aigis.convert`, supports seemless transformation of geospatial data into the forms AI models need, and then bringing model predictions back into GIS analysis ready forms.\n\n## Installation\n\nCreate a python environment for the project, we recommend [`conda`](https://www.anaconda.com/download):\n\n```bash\nconda create -n aigis python==3.10\nconda activate aigis\n```\n\nClone the repository:\n\n```bash\ngit clone https://github.com/Sydney-Informatics-Hub/aigis.git\n\npip install -e aigis\n```\n\n## Usage\n\n### Recommended System Requirements\n\nAll tools were built for Linux or MacOS systems.\n\nHigh resolution aerial imagery quickly becomes very computationally intensive to work with, so we recommend using high powered workstations or cloud computing environments.\n\nModel fine tuning are best with CUDA GPUs. One or more RTX 4090 or A100 GPUs are recommended for best performance. \nModels can be fine tuned on Google Colab free T4 GPUs, but larger datasets and longer runs should run on other compute platforms.\n\nInference and postprocessing both benefit from CUDA GPUs.\n\n## `aigis` datasets\n\nExplore our two featured example datasets using a [Felt map](https://felt.com/map/Sydney-Buildings-6drWs3o9ARViitVdt6qp60A?loc=-33.8282,150.8961,10.93z).\n\n![felt_map](./docs/content/Felt_Map.jpeg)\nTo load the matching NSW SixMaps aerial photography you have to click 1. background layers, then select 2. NSW SixMaps, as in the image above.\n\n### Building Segmentation\n\nWe've used the full `aigis` 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. \n\nOur models's predictions are available as a shapefile in GeoParquet, with almost one million buildings.\n\n![Greater Sydney Buildings](./docs/content/gsu_buildings.png)\n\n### Tree Segmentation\n\nWe've used the full `aigis` 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. \n\nOur models's predictions are available as a shapefile in GeoParquet, with 2 million tree patches.\n\n![Greate Sydney Tree](./docs/content/gsu_trees.png)\n\n##  Contributing\n\nContributions are welcome! Here are several ways you can contribute:\n\n- **[Submit Pull Requests](https://github/Sydney-Informatics-Hub/aigis/blob/main/CONTRIBUTING.md)**: Review open PRs, and submit your own PRs.\n- **[Join the Discussions](https://github/Sydney-Informatics-Hub/aigis/discussions)**: Share your insights, provide feedback, or ask questions.\n- **[Report Issues](https://github/Sydney-Informatics-Hub/aigis/issues)**: Submit bugs found or log feature requests for Aigis.\n\n\u003cdetails closed\u003e\n    \u003csummary\u003eContributing Guidelines\u003c/summary\u003e\n\n1. **Fork the Repository**: Start by forking the project repository to your GitHub account.\n2. **Clone Locally**: Clone the forked repository to your local machine using a Git client.\n   ```sh\n   git clone https://github.com/Sydney-Informatics-Hub/aigis\n   ```\n3. **Create a New Branch**: Always work on a new branch, giving it a descriptive name.\n   ```sh\n   git checkout -b new-feature-x\n   ```\n4. **Make Your Changes**: Develop and test your changes locally.\n5. **Commit Your Changes**: Commit with a clear message describing your updates.\n   ```sh\n   git commit -m 'Implemented new feature x.'\n   ```\n6. **Push to GitHub**: Push the changes to your forked repository.\n   ```sh\n   git push origin new-feature-x\n   ```\n7. **Submit a Pull Request**: Create a PR against the original project repository. Clearly describe the changes and their motivations.\n\nOnce your PR is reviewed and approved, it will be merged into the main branch.\n\n\u003c/details\u003e\n\n---\n\n## About Us\n\nThis software was developed by the Sydney Informatics Hub, a core research facility of the University of Sydney.\n\nThe project team includes:\n\n- Henry Lydecker\n- Sahand Vahidnia\n- Xinwei Luo\n- Thomas Mauch\n\n##  License\n\nThis project is protected under the [MIT Licence](https://choosealicense.com/licenses) License. For more details, refer to the [LICENSE](https://choosealicense.com/licenses/) file.\n\n\n##  Acknowledgments\n\nAcknowledgements are an important way for us to demonstrate the value we bring to your research. Your research outcomes are vital for ongoing funding of the Sydney Informatics Hub. If you make use of this software for your research project, please include the following acknowledgement:\n\n\u003eThis research was supported by the Sydney Informatics Hub, a Core Research Facility of the University of Sydney.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsydney-informatics-hub%2Faigis","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsydney-informatics-hub%2Faigis","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsydney-informatics-hub%2Faigis/lists"}