{"id":16840587,"url":"https://github.com/mitmul/ssai","last_synced_at":"2025-04-11T05:41:48.481Z","repository":{"id":26909614,"uuid":"30371453","full_name":"mitmul/ssai","owner":"mitmul","description":"Semantic Segmentation for Aerial Imagery using Convolutional Neural Network","archived":false,"fork":false,"pushed_at":"2016-04-08T08:38:15.000Z","size":183,"stargazers_count":27,"open_issues_count":1,"forks_count":19,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-03-25T03:41:37.799Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"http://www.slideshare.net/mitmul/building-and-road-detection-from-large-aerial-imagery/1","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/mitmul.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2015-02-05T18:35:33.000Z","updated_at":"2024-01-05T01:52:57.000Z","dependencies_parsed_at":"2022-09-02T00:52:13.844Z","dependency_job_id":null,"html_url":"https://github.com/mitmul/ssai","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fssai","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fssai/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fssai/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mitmul%2Fssai/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/mitmul","download_url":"https://codeload.github.com/mitmul/ssai/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248351297,"owners_count":21089268,"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":[],"created_at":"2024-10-13T12:37:21.765Z","updated_at":"2025-04-11T05:41:48.449Z","avatar_url":"https://github.com/mitmul.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# This repo has been deprecated because whole things are re-implemented by using Chainer and I did refactoring for many codes. So please check this newer version: https://github.com/mitmul/ssai-cnn\n\n# Semantic Segmentation for Aerial Imagery\nExtract building and road from aerial imagery\n\n# Requirements\n- OpenCV 2.4.10\n- Boost 1.57.0\n- Boost.NumPy\n- Caffe (modified caffe: [https://github.com/mitmul/caffe](https://github.com/mitmul/caffe))\n  - NOTE: Build the `ssai` branch of the above repository\n\n# Data preparation\n\n```\n$ bash shells/donwload.sh\n$ python scripts/create_dataset.py --dataset multi\n$ python scripts/create_dataset.py --dataset single\n$ python scripts/create_dataset.py --dataset roads_mini\n$ python scripts/create_dataset.py --dataset roads\n$ python scripts/create_dataset.py --dataset buildings\n$ python scripts/create_dataset.py --dataset merged\n```\n\n## Massatusetts Building \u0026 Road dataset\n- mass_roads\n  - train: 8458173 patches\n    - epoch: 66079 mini-batches (mini-batch size: 128)\n\n  - valid: 126281 patches\n    - epoch: 987 mini-batches (mini-batch size: 128)\n\n  - test: 440932 patches\n    - epoch: 3445 mini-batches (mini-batch size: 128)\n\n- mass_roads_mini, mass_buildings, mass_merged\n  - train: 1119872 patches\n    - epoch: 8749 mini-batches (mini-batch size: 128)\n\n  - valid: 36100 patches\n    - epoch: 282 mini-batches (mini-batch size: 128)\n\n  - test: 89968 patches\n    - epoch: 703 mini-batches (mini-batch size: 128)\n\n# Create Models\n\n```\n$ python scripts/create_models.py --seed seeds/model_seeds.json --caffe_dir $HOME/lib/caffe/build/install\n```\n\n# Start training\n\n```\n$ bash shells/train.sh models/Mnih_CNN\n```\n\nwill create a directory named `results/Mnih_CNN_{started date}`.\n\n# Prediction\n\n```\n$ cd results/Mnih_CNN_{started date}\n$ python ../../scripts/test_prediction.py --model predict.prototxt --weight snapshots/Mnih_CNN_iter_1000000.caffemodel --img_dir ../../data/mass_merged/test/sat --channel 3\n```\n\n# Build Library for Evaluation\n\n```\n$ cd lib\n$ mkdir build\n$ cd build\n$ cmake ../\n$ make\n```\n\n# Evaluation\n\n```\n$ cd results/Mnih_CNN_{started date}\n$ python ../../scripts/test_evaluation.py --map_dir ../../data/mass_merged/test/map --result_dir prediction_1000000 --channel 3\n```\n\n# Model averaging\n\n```\n$ python ../scripts/batch_evaluation.py --offset True\n$ mkdir Mnih_CNN_Merged\n$ cd Mnih_CNN_Merged\n$ python ../../scripts/test_evaluation.py --map_dir ../../data/mass_merged/test/map --result_dir ./prediction_100000 --channel 3 --offset 0 --pad 31\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmitmul%2Fssai","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmitmul%2Fssai","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmitmul%2Fssai/lists"}