{"id":19002272,"url":"https://github.com/jboirazian/captchacv2","last_synced_at":"2026-04-20T10:30:18.800Z","repository":{"id":226101493,"uuid":"767756246","full_name":"jboirazian/CaptchaCV2","owner":"jboirazian","description":"A simple proof of concept for beating outdated Captcha methods using only opencv and numpy","archived":false,"fork":false,"pushed_at":"2024-03-05T21:47:11.000Z","size":50,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-01T18:42:35.913Z","etag":null,"topics":["captcha","computer-vision","opencv-python","termux"],"latest_commit_sha":null,"homepage":"","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/jboirazian.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}},"created_at":"2024-03-05T21:01:31.000Z","updated_at":"2024-09-03T04:17:55.000Z","dependencies_parsed_at":"2024-03-05T22:26:06.038Z","dependency_job_id":"fba73f80-ee19-4c75-87f0-644dec3acd83","html_url":"https://github.com/jboirazian/CaptchaCV2","commit_stats":null,"previous_names":["jboirazian/captchacv2"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jboirazian%2FCaptchaCV2","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jboirazian%2FCaptchaCV2/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jboirazian%2FCaptchaCV2/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jboirazian%2FCaptchaCV2/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jboirazian","download_url":"https://codeload.github.com/jboirazian/CaptchaCV2/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":240026444,"owners_count":19736202,"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":["captcha","computer-vision","opencv-python","termux"],"created_at":"2024-11-08T18:14:29.893Z","updated_at":"2026-04-20T10:30:18.545Z","avatar_url":"https://github.com/jboirazian.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# CaptchaCV2\nA simple proof of concept for beating outdated Captcha methods using only opencv and numpy\n\n## Summery\n\nMany outdated websites and mobile apps still use the following captcha method:\n\n![image](https://github.com/jboirazian/CaptchaCV2/assets/21143405/e0f80433-8e73-4cb9-a5b3-6f65eb995c65)\n\nWhile you might think that it can be solved with your favorite OCR framework , the reality is that it can be solved by using only [Template matching](https://docs.opencv.org/4.x/d4/dc6/tutorial_py_template_matching.html)\n\n\n### Method\n\nIn order to perform to perform Template matching... we are gonna need templates. In the folder **/selected_patterns** you will find the 3 pairs 0 to 9 diggits.\nThese images were reconstructed from 1000 digits of the same number and position in order to obtain the real number without that sky blue jitter :\n\n![image](https://github.com/jboirazian/CaptchaCV2/assets/21143405/75e70ec7-6f1a-4ce6-93b2-9ebf06ce5361) ![image](https://github.com/jboirazian/CaptchaCV2/assets/21143405/641b76f1-abc0-4802-b0ad-b7952faf7f60) ![image](https://github.com/jboirazian/CaptchaCV2/assets/21143405/47169727-ec31-4349-a487-bf2c0f9bc4b4)\n\n\n## How to run it\n\n+ Install numpy and cv2\n+ git clone https://github.com/jboirazian/CaptchaCV2.git\n+ python3 verify.py\n\n### Example\n\n```python3\nimport modules.digit_recogniton as dr\nimport cv2\n\nif __name__ == \"__main__\":\n    img=cv2.imread('example.png', cv2.IMREAD_GRAYSCALE)\n    print(dr.predict_number(image=img))\n```\n\n## Performance\n\nWithout the need of a GPU and with the provided image example , the prediction was achived in less than 5 ms on my PC.\n\nI was also able to run in on an really old Android table using [Termux](https://termux.dev/en/) and got around 50 ms \n\n\n\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjboirazian%2Fcaptchacv2","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjboirazian%2Fcaptchacv2","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjboirazian%2Fcaptchacv2/lists"}