Copyright © 2025 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{184240,
        author = {Prof.Santhosh SG and Priyanka GV},
        title = {Automated Detection and Classification of Text and Non-Text Elements in Images using Python and OpenCV},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {4},
        pages = {722-727},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=184240},
        abstract = {This study introduced a lightweight and modular system for the automated detection and classification of textual and non-textual elements contained within digital images which synthesized deep learning-based OCR and classical computer vision processes. The outlined process combined EasyOCR for the detection and recognition of printed and handwritten text with OpenCV- based contour analysis to detect non-text elements including logos, diagrams, or other shapes. Implemented as a web application using Python and Flask, the system accepts user uploaded images, detects all textual and non-textual elements, outputs annotated digital images that denotes textual regions in green and non-text regions in blue. The device further measures the ratio of the text area relative to the image area, a useful metric for understanding how an image's content is displayed in regard to layout. The ratio of the text area is especially important for applications in document analysis, media summarization, and digital content moderation processes. The open-source tooling of the system promotes accessibility as well as modularity within the context of larger interpretations of the system. Overall, the project contributes a useful, interpretable hybrid classification of the text and non-text nature of content in an image which also raises opportunities for improving many of the items outlined in the future work section such as: real-time processing; multilingual amounts of content output; and structured data output regarding the detected textual and non-textual elements that can be integrated into pipeline systems.},
        keywords = {EasyOCR, Image Classification, Non-Text Segmentation, OpenCV, Python, Text Detection, Web Application.},
        month = {September},
        }