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{169452,
        author = {Supriya Parashuramsingh Rajput and Suraj Kadli},
        title = {Machine Learning and Neural Network Approach to Fruit Quality Assessment Using Polarization Features of the Image},
        journal = {International Journal of Innovative Research in Technology},
        year = {2024},
        volume = {11},
        number = {6},
        pages = {1728-1733},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=169452},
        abstract = {The project titled “Machine Learning and Neural Network Approach to Fruit Quality Assessment Using Polarization Features of the Image” aims to generate a non-destructive and non-invasive technique for determining the freshness and quality of apples by leveraging image polarization characteristics. Images of apples are captured using a smartphone camera, resized to 256X256 pixels, and cropped to highlight the fruit's edges. These images are then polarized at four specific angles 0°, 45°, 90°, and 135° from which Stokes parameters are computed to extract features like Degree of Linear Polarization (DoLP) and Angle of Polarization (AoP). Following the application of demosaicking to refine color precision, these features are analyzed by a “Radial Basis Function Neural Network (RBFNN)”, which is trained to accurately estimate the apple’s age with perfect precision based on real-world data. This method facilitates the early identification of apples that might not be fit for consumption before external signs of decay are visible, offering a means to evaluate fruit quality without causing any harm to the fruit and establishing a standardized approach for assessing apple edibility. The technology offers possibilities for use in retail and consumer environments to select fresher apples and in production and quality control settings, with future opportunities for scaling, adapting the approach for other types of fruit, and developing user-friendly applications.},
        keywords = {“Radial Basis Function Neural Network (RBFNN)”, Polarization Features, Degree of Linear Polarization(DoLP), Angle of Polarization(AoP), Mean Square Error (MSE).},
        month = {November},
        }