{"id":20460261,"url":"https://github.com/madahetooo/hatdetector","last_synced_at":"2026-06-19T06:32:31.248Z","repository":{"id":115299719,"uuid":"294797831","full_name":"madahetooo/hatDetector","owner":"madahetooo","description":"This repository  is to build a model that detects people, heads, and hardhats in images. Each image can contain multiple people, multiple heads and multiple hardhats.  The solution will help build a safety app that will alert the user if there are workers in the field that do not comply with safety rules.","archived":false,"fork":false,"pushed_at":"2020-12-02T19:03:58.000Z","size":33,"stargazers_count":3,"open_issues_count":0,"forks_count":1,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-09-07T17:56:36.187Z","etag":null,"topics":["ai","detects-people","keras-retinanet","machine-learning","object-detection","opencv","retinanet","safety","safety-rules","wheet-heads","yolov3"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/madahetooo.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2020-09-11T19:41:34.000Z","updated_at":"2023-06-11T10:07:57.000Z","dependencies_parsed_at":"2023-05-15T10:00:23.657Z","dependency_job_id":null,"html_url":"https://github.com/madahetooo/hatDetector","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/madahetooo/hatDetector","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/madahetooo%2FhatDetector","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/madahetooo%2FhatDetector/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/madahetooo%2FhatDetector/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/madahetooo%2FhatDetector/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/madahetooo","download_url":"https://codeload.github.com/madahetooo/hatDetector/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/madahetooo%2FhatDetector/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34520431,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-19T02:00:06.005Z","response_time":61,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["ai","detects-people","keras-retinanet","machine-learning","object-detection","opencv","retinanet","safety","safety-rules","wheet-heads","yolov3"],"created_at":"2024-11-15T12:18:58.547Z","updated_at":"2026-06-19T06:32:31.226Z","avatar_url":"https://github.com/madahetooo.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# hatDetector\nThis repository  is to build a model that detects people, heads, and hardhats in images. Each image can contain multiple people, multiple heads and multiple hardhats.  The solution will help build a safety app that will alert the user if there are workers in the field that do not comply with safety rules.\nThis Notebook is for Training Purpose of Keras RetinaNet.\nRetinaNet is very slow as compared to F-RCNN so I've kept epochs and steps per epoch small for fast commiting purpose.\n\nStep-1 : Installing Keras-RetinaNet\nStep-2 : Let's look at the data\nStep-3 : EDA\nStep-4 : Visualizing images\n\n------------------\nWhat can we tell from visualizations:\n\nthere are plenty of overlappind bounding boxes\nall photos seem to be taken vertically\nall plants are can be rotated differently, there is no single orientation. this means that different flip and roration augmentations should probably help\ncolors of wheet heads are quite different and seem to depend a little bit on the source\nwheet heads themselves are seen from very different angles of view relevant to the observer\n----------------------\nStep-5 : Preprocessing Data for Input to RetinaNet\n\nStep-6 : Preparing Files to be given for training\nAnnotation file contains all the path of all images and their corresponding bounding boxes\nClass file contains the number of classes but in our case it is just 1 (Wheat)\n---------------------\nStep-7 : Downloading the pretrained model\n\nModel Parameters\n\nStep-8 : Training Model.\nStep-9 : Predictions.\n--------\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmadahetooo%2Fhatdetector","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmadahetooo%2Fhatdetector","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmadahetooo%2Fhatdetector/lists"}