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{195233,
        author = {Varshitha Kottagolla and Shireen and K. Soni and V. Anitha},
        title = {Farmers Digital Brain - An Integrated Advisory Tool Using Deep Learning Models},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {7513-7519},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195233},
        abstract = {Agriculture is an important sector for the security of the world’s food supply. Besides, it is an important sector for economic development. In traditional farming, decision-making is usually based on experience rather than the outcome of data analysis. This may lead to improper selection of crops, hence inefficiency in agriculture. The need to develop intelligent decision support systems has thus become critical. In the present research work, the data related to agriculture, such as soil images, plant leaf images, and environmental parameters in tabular form, was used to develop a framework for integrated agricultural analysis. In the data preprocessing stage, the data was resized, normalized, scaled, and split before applying the models to the data for analysis. Deep learning models such as ResNet18, EfficientNet-B0, Vision Transformer (ViT), and a Convolutional Neural Network (CNN) were implemented to classify the soil types. For the purpose of crop recommendation, machine learning models such as Light GBM, XG Boost, Multi-Layer Per-ceptron (MLP), and TabNet were implemented on the environmental parameters such as Nitrogen, Phosphorus, Potassium, temperature, humidity, rainfall, and soil pH. For plant disease detection, the MobileNetV2 model was implemented on the deep learning model, and crop yield prediction was done on the deep learning regression model on the data related to agricultural production. In the soil classification module, ResNet18 and efficient model architectures have the highest test accuracy of 100%, while the vision transformer model has a test accuracy of 98.24%. In the crop recommendation module, the lightgbm model has the highest accuracy of 98.64% over other machine learning models like xgboost (98.24%), MLP (97.27%), and TabNet (92.73%). The crop yield prediction model has a high coefficient of determination, i.e., R² = 0.9808. The proposed system will be able to make accurate predictions about classifying the soil, crop recommendation system, plant disease detection system, crop yield estimation system, etc. The system will support in decision-making in the improvement of the crop planning system in the agriculture field.},
        keywords = {Precision Agriculture; Deep Learning; Crop Recommendation; Soil Classification; Plant Disease Detection; Machine Learning; Crop Yield Prediction; Computer Vision.},
        month = {March},
        }