{"id":17716595,"url":"https://github.com/rageworx/libsrcnn","last_synced_at":"2025-04-28T17:19:47.911Z","repository":{"id":78866051,"uuid":"143988444","full_name":"rageworx/libsrcnn","owner":"rageworx","description":"Super-Resolution imaging with Convolutional Neural Network library for G++, Non-OpenCV model.","archived":false,"fork":false,"pushed_at":"2024-06-20T04:52:38.000Z","size":1348,"stargazers_count":16,"open_issues_count":6,"forks_count":1,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-04-28T17:19:40.034Z","etag":null,"topics":["c","cpp","gcc","gpp","imaging","mingw-w64","no-opencv","resolution","super","super-resolution"],"latest_commit_sha":null,"homepage":"http://rageworx.pe.kr/search/SRCNN","language":"C","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"lgpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/rageworx.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,"publiccode":null,"codemeta":null}},"created_at":"2018-08-08T09:03:35.000Z","updated_at":"2024-06-20T04:52:41.000Z","dependencies_parsed_at":"2024-10-25T17:04:43.253Z","dependency_job_id":"47374c3d-1361-44d5-94a9-97a807d0101b","html_url":"https://github.com/rageworx/libsrcnn","commit_stats":null,"previous_names":[],"tags_count":9,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rageworx%2Flibsrcnn","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rageworx%2Flibsrcnn/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rageworx%2Flibsrcnn/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/rageworx%2Flibsrcnn/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/rageworx","download_url":"https://codeload.github.com/rageworx/libsrcnn/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":251352638,"owners_count":21575865,"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":["c","cpp","gcc","gpp","imaging","mingw-w64","no-opencv","resolution","super","super-resolution"],"created_at":"2024-10-25T13:42:19.082Z","updated_at":"2025-04-28T17:19:47.894Z","avatar_url":"https://github.com/rageworx.png","language":"C","readme":"# libsrcnn\r\n### Super-Resolution imaging with Convolutional Neural Network\r\nA stand-alone library for Super-Resolution, Non-OpenCV model related in these projects:\r\n* https://github.com/rageworx/SRCNN_OpenCV_GCC\r\n* https://github.com/shuwang127/SRCNN_Cpp.\r\n\r\n## Introduction\r\nThis is an open source project from original of this:\r\n**SRCNN_Cpp** is a C++ Implementation of Image Super-Resolution using SRCNN which is proposed by Chao Dong in 2014.\r\n - If you want to find the details of SRCNN algorithm, please read the paper:  \r\n\r\n   Chao Dong, Chen Change Loy, Kaiming He, Xiaoou Tang. Learning a Deep Convolutional Network for Image Super-Resolution, in Proceedings of European Conference on Computer Vision (ECCV), 2014\r\n - If you want to download the training code(caffe) or test code(Matlab) for SRCNN, please open your browse and visit http://mmlab.ie.cuhk.edu.hk/projects/SRCNN.html for more details.\r\n - And thank you very much for Chao's work in SRCNN.\r\n\r\n## WIKI\r\n* See [Wiki](https://github.com/rageworx/libsrcnn/wiki) page.\r\n* Includes OpenMP support for macOS.\r\n\r\n## Additional references\r\n* Fast resizing methods for BiCubic filtering \r\n    * [Free Image Project](http://freeimage.sourceforge.net/)\r\n\t* [librawprocessor](https://github.com/rageworx/librawprocessor)\r\n    * [fl_imgtk](https://github.com/rageworx/fl_imgtk)\r\n\r\n \r\n## Features\r\n* Faster about 400% or more than OpenCV GCC version of SRCNN, with OpenMP.\r\n    - references :\r\n    - commit/b340b885a58605f793aa000eebc7f96c19c8e9fe#commitcomment-103507343\r\n\t- commit/b340b885a58605f793aa000eebc7f96c19c8e9fe#commitcomment-103507802\r\n* None-OpenCV, no OpenCV required !\r\n* Compilation availed for almost of POSIX g++.\r\n* Simply optimized for basic OpenMP.\r\n* Works well even without OpenMP like macOS.\r\n* Not support M$VC, sorry MS guyz.\r\n\r\n## Sample images\r\n* Original 100%\r\n\r\n    ![IMG_0](Pictures/butterfly.png)\r\n\r\n* Bicubic 150%\r\n\r\n    ![IMG_1](Pictures/butterfly_bicubic.png)\r\n\r\n* SRCNN 150%\r\n\r\n    ![IMG_2](Pictures/butterfly_srcnn.png)\r\n\r\n* SRCNN (Convolution Y channel) 150%\r\n\r\n    ![IMG_3](Pictures/butterfly_srcnn_convolution.png)\r\n\r\n## Supporting platforms\r\n* Windows 32, 64 with MSYS2 + MinGW-W64\r\n* Almost any Linux, x86_32, x86_64, arm, armhf, aarch64\r\n* macOS ( clang, llvm )\r\n\r\n## Latest Changes\r\n\r\n### Verison 0.1.10.40\r\n* Better speed, less memory usage by convolution I+II\r\n* Regards to zvezdochiot@github\r\n\r\n## Previous Changes\r\n\r\n### Verison 0.1.9.35\r\n* Fixed memory bug in float images from RGB case.\r\n* header version flag fixed.\r\n\r\n### Verison 0.1.9.34\r\n* Fixed don't use color space scaling with bicubic filter.\r\n* Now supporting alpha channel.\r\n\r\n### Verison 0.1.8.30\r\n* Precision step scaling bug fixed.\r\n### Verison 0.1.8.28\r\n* Precision step scaling option availed.\r\n* included option by reason of libsrcnn trained for maximum double multiply.\r\n### Verison 0.1.6.23\r\n* Fixed a small bug of wrong internal copying size.\r\n### Verison 0.1.6.22\r\n* Fixed bug of original source (ShuWang's SRCNN).\r\n   - Use last layer (3) to Y channel at last construction.\r\n* Changed ProcessSRCNN() method to get optional convolutional result.\r\n### Verison 0.1.6.20\r\n* Fixed memory leak after convolution55.\r\n* Changed ProcessSRCNN() method to get convolutional gray.\r\n### Version 0.1.4.17\r\n* Bug fixed for color space conversion.\r\n### Verison 0.1.5.18\r\n* Supports variable filters for interpolation.\r\n    1. Nearest\r\n    1. Bilinear\r\n    1. Bicubic\r\n    1. Lanczos-3\r\n    1. B-Spline\r\n\r\n## License\r\n* Follows original source GPLv2, but this project is LGPLv3.\r\n\r\n## Requirements\r\n* Your G++.\r\n\r\n## How to build ?\r\n* Make a symlink from `Makefile.{your platform}` in `makefiles` directory.\r\n    - eg.) `ln -s makefiles/Makefile.macos Makefile`\r\n* Then build with `make`.\r\n* Testing applications may one of these,\r\n\t- `make -f makefiles/Makefile.test`\r\n\t- or\r\n\t- `make -f makefiles/Makefiles.testmac`\r\n\r\n## Dependency\r\n* Testing application by `make -f makefiles/Makefile.test` may requires FLTK and fl_imgtk libraries.\r\n* FLTK should be installed by anyway, but recommend to my below FLTK-custom with fl_imgtk.\r\n* [FLTK-custom](https://github.com/rageworx/fltk-custom) and [fl_imgtk](https://github.com/rageworx/fl_imgtk) for build test program for read and write image files.\r\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frageworx%2Flibsrcnn","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Frageworx%2Flibsrcnn","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Frageworx%2Flibsrcnn/lists"}