{"id":21239199,"url":"https://github.com/cutupdev/road-damage-detection","last_synced_at":"2025-03-15T03:40:29.611Z","repository":{"id":227037811,"uuid":"760304462","full_name":"cutupdev/Road-Damage-Detection","owner":"cutupdev","description":"This is repository for detection damage and judgement on road.","archived":false,"fork":false,"pushed_at":"2024-03-11T08:40:03.000Z","size":9731,"stargazers_count":3,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-21T19:22:39.714Z","etag":null,"topics":["image-processing","python","yolo"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/cutupdev.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":"2024-02-20T07:00:13.000Z","updated_at":"2024-04-18T17:56:00.000Z","dependencies_parsed_at":"2024-03-11T10:03:43.584Z","dependency_job_id":null,"html_url":"https://github.com/cutupdev/Road-Damage-Detection","commit_stats":null,"previous_names":["catlover75926/road-damage-detection","harmonitech/road-damage-detection","cutupdev/road-damage-detection"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cutupdev%2FRoad-Damage-Detection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cutupdev%2FRoad-Damage-Detection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cutupdev%2FRoad-Damage-Detection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cutupdev%2FRoad-Damage-Detection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/cutupdev","download_url":"https://codeload.github.com/cutupdev/Road-Damage-Detection/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243680974,"owners_count":20330154,"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":["image-processing","python","yolo"],"created_at":"2024-11-21T00:42:24.616Z","updated_at":"2025-03-15T03:40:29.595Z","avatar_url":"https://github.com/cutupdev.png","language":"Python","readme":"# Road Damage Detection\n-----\nThis repository contain codes for damage detection on the road.\nIn the city, it's important problem.\nThis project introduce in Oslo traffic center.\n\nThe solution is based on Deep CNN.\n\n![](doc/bigdata21.png)\n\n\n\u003cimg src=\"doc/road.JPG\" width=\"350\"\u003e\n\n\n\n\n\n### Usage\n\nConsider unisng a workspace for cleaner \n\n1. install required Libraries : \n1.1.\n- python\u003e= 3.6\n- pytorch 1.4 or 1.6\n- torchvision \u003e= 0.5\n- apex is also needed\n- timm \u003e= 1.28\n\n\n`pip install -r requirements.txt` \n\n1.2.\n\ninstall apex (2020-10-26): \n\n```\ngit clone https://github.com/NVIDIA/apex\npip install -r apex/requirements.txt\npip install -v --no-cache-dir --global-option=\"--cpp_ext\" --global-option=\"--cuda_ext\" ./apex\n```\n\n\n2. download Dataset :\nfor ease of use we have provided annotations and ... in coco format downloadable: \n- [train and validation sets](https://drive.google.com/file/d/1IHaqAxpMtFwPHia7msB_1QPAywPgg7fW/view?usp=sharing)\n- [test1 and test2 data](https://drive.google.com/uc?id=1apjJfNHUlKQS64IaHRg3qRp_T0NnZnnQ\u0026export=download)(no annotation). \n\n\notherwise one can download original data from sekilab github repo and convert using tools provided in utils folder.\n \n\n3. for training : \n\n- Train on single GPU : \n```\npython train.py ../data --model tf_efficientdet_d0 -b 40 --amp --lr .15 --sync-bn --opt fusedmomentum --warmup-epochs 3 --lr-noise 0.3 0.9 --model-ema --model-ema-decay 0.9998 -j 25 --epochs 300\n```\n\n- Distributed Training : (note you may need to make the file executable before training using `chmod +x distributed_train.sh`)\n\n```\n./distributed_train.sh 3 ../data --model tf_efficientdet_d0 -b 40 --amp --lr .15 --sync-bn --opt fusedmomentum --warmup-epochs 3 --lr-noise 0.3 0.9 --model-ema --model-ema-decay 0.9998 -j 25 --epochs 300 \n```\n\n4. for inference on testset and generating submission file :\n```\npython infer.py ./data --model tf_efficientdet_d0 --checkpoint ./path/to/model/checkpoint --use-ema --anno test1 -b 17 --threshold 0.300\n```\n5. Image Inference to generate detected images\n\n- first create image_info_annotations(e.g. if image folder is in `../data` path. One should first create image info in json format using `python utils/createimageinfo.py` then folder structure should be like )\n\n```\n..\n├── data\n│   └── annotations\n|       ├── image_info_test1.json\n├── test1\n│   ├── Japan_XXX.jpg\n│   └── Czech_xxx.jpg\n|   └── ....\n\n```\n\nfollowing command will create generated file with bounding boxes in ./predictions\n```\npython detector.py ../data --model tf_efficientdet_d0 --checkpoint path.to/modelfile.pth.tar --anno test1  -b 20 --use-ema  --tosave ./predictions \n```\n\nfor validation (AP scores) and benchmarking with `cuda.Event()` use the following command : \n```\npython validate.py ../data --model tf_efficientdet_d0 --checkpoint path/to/model/checkpoint.pth.tar --anno val  -b 20 --use-ema\n```\n\n\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcutupdev%2Froad-damage-detection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcutupdev%2Froad-damage-detection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcutupdev%2Froad-damage-detection/lists"}