{"id":18470016,"url":"https://github.com/greyhatguy007/hazeclear","last_synced_at":"2025-05-12T00:24:13.356Z","repository":{"id":234032211,"uuid":"694266267","full_name":"greyhatguy007/hazeclear","owner":"greyhatguy007","description":null,"archived":false,"fork":false,"pushed_at":"2023-09-20T17:49:05.000Z","size":13,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-05-09T00:09:14.082Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"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/greyhatguy007.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":"2023-09-20T16:46:30.000Z","updated_at":"2023-09-21T05:56:39.000Z","dependencies_parsed_at":null,"dependency_job_id":"5d2a2ae7-41a3-4055-8ff3-f2b3b38eb62b","html_url":"https://github.com/greyhatguy007/hazeclear","commit_stats":null,"previous_names":["greyhatguy007/hazeclear"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/greyhatguy007%2Fhazeclear","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/greyhatguy007%2Fhazeclear/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/greyhatguy007%2Fhazeclear/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/greyhatguy007%2Fhazeclear/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/greyhatguy007","download_url":"https://codeload.github.com/greyhatguy007/hazeclear/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253166521,"owners_count":21864482,"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":[],"created_at":"2024-11-06T10:12:44.779Z","updated_at":"2025-05-09T00:09:16.761Z","avatar_url":"https://github.com/greyhatguy007.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# AI-ML based Intelligent de-hazing algorithm\n\n## Introduction\n\nThis project aims to develop a deep learning model to dehaze images. Image dehazing is pivotal for enhancing visual clarity in various applications in computer vision tasks. The model is constructed using an encoder-decoder architecture in PyTorch.\n\n## Dataset\n\nWe are utilizing the [Dense-Haze dataset](https://www.kaggle.com/datasets/rajat95gupta/hazing-images-dataset-cvpr-2019) available on Kaggle. This dataset contains 55 pairs of hazy and non-hazy images, making it a suitable benchmark for image dehazing tasks.\n\n## Project Steps\n\n1. **Data Preprocessing**:\n    - Resize images and normalize pixel values.\n\n2. **Model Development**:\n    - Design an encoder-decoder neural network architecture.\n    - Initialize model layers, including convolutional layers, pooling layers, and upsampling layers.\n    - Define the forward propagation mechanism.\n\n3. **Training the Model**:\n    - Define loss function (Mean Squared Error) and optimizer (Adam).\n    - Train the model using the training dataset.\n\n4. **Model Evaluation**:\n    - Compute evaluation metrics, including Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM).\n\n5. **Results Visualization**:\n    - Showcase original, hazy, and dehazed images side by side.\n    - Plot training loss and validation loss across epochs.\n\n6. **Optimization (if necessary)**:\n    - Tune hyperparameters, such as learning rate, batch size, and number of epochs.\n    - Modify the model architecture by adding more layers or changing layer configurations.\n    - Implement data augmentation to artificially increase the training dataset size and robustness.\n\n7. **Deployment**:\n    - Convert the trained model into a format suitable for deployment.\n    - Integrate the model into a web or mobile application for real-time image dehazing.\n\n## Technical Details\n\n- **Framework**: PyTorch\n- **Model Architecture**: [Encoder-Decoder CNN](./docs/encoder-decoder-architecture/readme.md)\n- **Loss Function**: Mean Squared Error (MSE)\n- **Optimizer**: Adam\n- **Evaluation Metrics**: [PSNR, SSIM](./docs/evaluation-metric/readme.md)\n\n## Future Work\n\n1. Experiment with advanced architectures like U-Net or ResNet.\n2. Implement a custom loss function for improved performance.\n3. Enhance the deployment solution to process video streams for real-time dehazing.\n\n## Contributions\n\nContributions to this project are welcome. Please ensure that you adhere to the project's coding and documentation standards.\n\n## Acknowledgments\n\nSpecial thanks to the creators of the Dense-Haze dataset for making it publicly available on Kaggle.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgreyhatguy007%2Fhazeclear","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgreyhatguy007%2Fhazeclear","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgreyhatguy007%2Fhazeclear/lists"}