{"id":23802717,"url":"https://github.com/aritrakar/trafficsigndetection","last_synced_at":"2026-04-15T22:34:08.687Z","repository":{"id":173357391,"uuid":"548119197","full_name":"aritrakar/TrafficSignDetection","owner":"aritrakar","description":"🚦Traffic sign detection using CNNs.","archived":false,"fork":false,"pushed_at":"2022-10-09T14:58:35.000Z","size":2048,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-11-10T22:31:09.998Z","etag":null,"topics":["cnn","python","tensorflow"],"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/aritrakar.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":"2022-10-09T00:35:32.000Z","updated_at":"2024-06-04T01:42:16.000Z","dependencies_parsed_at":null,"dependency_job_id":"a45a0218-d1f8-4ab1-bc81-497bb2f265b0","html_url":"https://github.com/aritrakar/TrafficSignDetection","commit_stats":null,"previous_names":["aritrakar/trafficsigndetection"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/aritrakar/TrafficSignDetection","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aritrakar%2FTrafficSignDetection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aritrakar%2FTrafficSignDetection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aritrakar%2FTrafficSignDetection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aritrakar%2FTrafficSignDetection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/aritrakar","download_url":"https://codeload.github.com/aritrakar/TrafficSignDetection/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/aritrakar%2FTrafficSignDetection/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31863493,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-15T15:24:51.572Z","status":"ssl_error","status_checked_at":"2026-04-15T15:24:39.138Z","response_time":63,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":["cnn","python","tensorflow"],"created_at":"2025-01-01T22:27:30.979Z","updated_at":"2026-04-15T22:34:08.669Z","avatar_url":"https://github.com/aritrakar.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Traffic Sign Detection\n\nThis repository contains the models trained on the German Traffic Sign Recognition Benchmark ([GTSRB](https://www.kaggle.com/datasets/meowmeowmeowmeowmeow/gtsrb-german-traffic-sign)) dataset to detect traffic signs. There are 43 types of signs in the dataset. One of the things I wanted to explore in this dataset was different image preprocessing techniques and how they lead to significantly better performances than training without image preprocessing.\n\nThe final code can be found in `final.ipynb` and the scratch/experimental code can be found in `scratch.ipynb`.\n\nI experimented with different Convolutional-Neural-Network (CNN) models with different hyperparameters and settled on the following model with a learning rate of 0.01:\n\n```\nModel: \"sequential\"\n_________________________________________________________________\n Layer (type)                Output Shape              Param #   \n=================================================================\n conv2d (Conv2D)             (None, 32, 32, 8)         608       \n                                                                 \n batch_normalization (BatchN  (None, 32, 32, 8)        32        \n ormalization)                                                   \n                                                                 \n max_pooling2d (MaxPooling2D  (None, 16, 16, 8)        0         \n )                                                               \n                                                                 \n conv2d_1 (Conv2D)           (None, 16, 16, 16)        1168      \n                                                                 \n batch_normalization_1 (Batc  (None, 16, 16, 16)       64        \n hNormalization)                                                 \n                                                                 \n conv2d_2 (Conv2D)           (None, 16, 16, 16)        2320      \n                                                                 \n batch_normalization_2 (Batc  (None, 16, 16, 16)       64        \n hNormalization)                                                 \n                                                                 \n max_pooling2d_1 (MaxPooling  (None, 8, 8, 16)         0         \n...\nTotal params: 107,147\nTrainable params: 106,427\nNon-trainable params: 720\n_________________________________________________________________\n```\n\nTHe highest validation accuracy achieved by the best model during training is 94.78% so far as shown below:\n\n```\nEpoch 30/30\n613/613 [==============================] - 58s 94ms/step - loss: 0.3170 - accuracy: 0.9505 - val_loss: 0.1937 - val_accuracy: 0.9478\n```\n\nFinal prediction results:\n\n![Predictions](./predictions.png)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faritrakar%2Ftrafficsigndetection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Faritrakar%2Ftrafficsigndetection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Faritrakar%2Ftrafficsigndetection/lists"}