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{180181,
        author = {P Bharath and Soumya ranjan panda and Suhas P and Harshith B and Vinod Kumar S and Sowmya K N},
        title = {apple fruit detection model yolo v8 and CNN},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {995-1011},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=180181},
        abstract = {Precise identification of apple fruit contours and growth locations plays a vital role in enabling smart harvesting systems and accurate yield forecasting. This study introduces a hybrid edge detection framework named RED, which integrates convolutional neural networks with rough set theory. The approach begins by leveraging the Faster R-CNN model to extract individual apples from images containing multiple fruits, effectively minimizing interference from surrounding visual noise. Following this, K-means clustering is employed to further isolate the apple target within each cropped image, enhancing focus and reducing residual background effects.
To address challenges such as variable lighting, intricate backgrounds, and fruit occlusion, the rough set method is utilized to produce edge approximations—specifically, upper and lower bounds—that help to refine the detected fruit contours. The RED model's outcomes are then benchmarked against existing edge detection techniques commonly used in similar applications.
Results from extensive experiments reveal that the RED model delivers notably improved detection accuracy and reliability. Its performance remains stable even in visually complex and poorly lit environments, demonstrating its effectiveness over traditional edge detection operators. This robust performance positions the RED framework as a promising tool for advancing automation in fruit picking and yield estimation processes},
        keywords = {},
        month = {June},
        }