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{209251,
        author = {Appurva Rajendra Kapil and Dr. Manika Verma},
        title = {Transfer Learning-based Coal Classification through Enhanced CNN Model Architecture & Workflow},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {5},
        pages = {1102-1108},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=209251},
        abstract = {In this work, a coal image classification technique using a custom CNN architecture and transfer learning models (ResNet50 and VGG16) is presented to increase accuracy and robustness of coal recognition. The proposed pipeline consists of preprocessing, resizing, normalization, augmentation, and per-class evaluation to handle the texture, lighting, and background variation problems. The automatic extraction of discriminative visual features of coal images and classification into specific classes differs from traditional manual inspection of coal images. This experiment aims to compare the performance of three models under the same training and validation conditions in order to find out which one is the most suitable. Accuracy, F1-score, confusion matrix, and explainability using Grad-CAM are selected as criteria for performance evaluation. The methodology can be further developed for industrial application in coal sorting and inspection.},
        keywords = {Coal categorization, deep learning, CNN, transfer learning, ResNet50, VGG16, image preprocessing},
        month = {October},
        }