Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{186781,
        author = {Suryansh Tripathi and Meenu Garg},
        title = {Hybrid HSV and Deep Learning Method for Blood Scene Identification in Videos},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {2796-2803},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=186781},
        abstract = {This research introduces a hybrid system that combines HSV color analysis with a MobileNetV2-based convolutional neural network (CNN) integrated into an intuitive Tkinter graphical user interface (GUI) for the real-time detection and skipping of blood sequences in films. The system, which targets applications in forensic investigation, healthcare teaching, and content moderation, scores 98% precision, recall, and F1-score on a 500-image test set. This is a huge improvement above the 70% precision and 75% recall of a preliminary HSV prototype. The CNN was trained using a balanced dataset of 5,750 images that were tagged, enhanced, and integrated with HSV detection to maximize accuracy and efficiency. The Tkinter GUI makes it simple to choose, process, and produce films containing only non-blood content.
Videos are processed by the system at about 20 frames per second (FPS), making it acceptable for real-world use. A web application and browser extension utilizing React.js and TensorFlow.js are among the planned extensions that would improve accessibility for websites like YouTube. By balancing computing efficiency, accuracy, and usability with ethical deployment issues, this work advances content-aware video processing.},
        keywords = {blood detection, HSV color space, convolutional neural networks, hybrid approach, video processing, graphical user interface, content moderation, deep learning},
        month = {November},
        }