TEJAS Journal of Technologies and Humanitarian Science

Shreyansh Gupta
Department of Computer Science & Engineering (AI & ML), KIPM College of Engineering and Technology
U.P., India

Author

Aakash Jaiswal
Department of Computer Science & Engineering (AI & ML), KIPM College of Engineering and Technology
U.P., India

Author

Prashant Kumar Nishad
Department of Computer Science & Engineering (AI & ML), KIPM College of Engineering and Technology
U.P., India

Author

Akarsh Yadav
Department of Computer Science & Engineering (AI & ML), KIPM College of Engineering and Technology
U.P., India

Author

📜 License:

This work is licensed under a Creative Commons Attribution 4.0 International License

• Share — Copy and Redistribute the material
• Adapt — Remix, Transform, and build upon the material
• The licensor cannot revoke these freedoms as long as you follow the license terms.

Abstract:

Agricultural decision-making in India continues to depend heavily on traditional manual assessment of soil properties, weather conditions, and visual inspection of plant health. These conventional methods often lead to inaccurate crop selection, delayed disease detection, inefficient fertilizer usage, and reduced yield outcomes. To address these limitations, this research introduces an AI-Powered Agriculture Crop Detection System that integrates machine learning and deep learning for automated, data-driven farm intelligence. The framework incorporates eight supervised learning algorithms for multi-crop recommendation, CNN-based leaf disease detection, and regression-driven rainfall and yield prediction. Soil parameters such as N, P, K, pH, temperature, humidity, and rainfall are analyzed to generate optimized crop suggestions, while a Convolutional Neural Network identifies disease symptoms from leaf imagery with high accuracy. Additional modules—including fertilizer recommendation based on nutrient deficiencies, real time weather integration, and predictive yield estimation— enhance the decision-making capabilities of the system. A diverse dataset constructed from open agricultural repositories and field collected samples enables robust training under varying environmental conditions. Experimental evaluation demonstrates a peak crop recommendation accuracy of 99.55%, while the disease detection model achieves strong generalization across multiple plant categories. The results highlight the potential of AI driven analytics to improve crop planning, reduce losses, optimize resource utilization, and support scalable smart-farming ecosystems suitable for real-world agricultural deployment.

References

[1] S. Pandey and R. Singh, “Machine Learning Techniques for Soil Fertility and Crop Recommendation,” International Journal of Agricultural Science and Technology, vol. 12, no. 3, pp. 45–52, 2021.
[2] M. Sharma, K. Rajput, and A. Verma, “Analysis of Soil Nutrients for Crop Prediction Using Supervised Learning,” Journal of Intelligent Computing, vol. 8, no. 2, pp. 110–118, 2022.
[3] S. Patel and D. Kumar, “CNN-Based Leaf Disease Detection for Smart Farming Applications,” IEEE Access, vol. 9, pp. 112900–112910, 2021.
[4] J. Li and H. Chen, “A Review of Ensemble Learning Methods in Classification,” Applied Soft Computing, vol. 108, pp. 1–14, 2021.
[5] P. Kaur and A. Dhillon, “Deep Learning Approaches for Plant Disease Classification: A Survey,” Computers and Electronics in Agriculture, vol. 178, pp. 105–115, 2020.
[6] R. Gupta and S. Tripathi, “Preprocessing Techniques for Improving Plant Leaf Image Classification,” Procedia Computer Science, vol. 167, pp. 240–247, 2020.
[7] A. Banerjee, S. Deb, and R. Bose, “Prediction of Suitable Crops Using Machine Learning Algorithms,” International Journal of Computer Applications, vol. 176, no. 32, pp. 22 29, 2020.
[8] S. Arora and P. Gupta, “Agricultural Decision Support Systems Based on Data Analytics,” Information Processing in Agriculture, vol. 8, no. 4, pp. 538–550, 2021.
[9] R. Mishra and A. Rai, “Regression Models for Crop Yield Prediction Using Meteorological Data,” Agricultural Informatics Journal, vol. 15, no. 2, pp. 30–38, 2022.
[10] V. Kumar and A. Jain, “A Robust Framework for Leaf Image Recognition and Disease Detection,” IEEE Conference on Image Processing, pp. 1250–1256, 2020.
[11] M. Chauhan, P. Singh, and R. Yadav, “Integrated Smart Agriculture Using IoT and Machine Learning,” IEEE Sensors Journal, vol. 22, no. 5, pp. 4120–4130, 2022.
[12] L. Zhang and Y. Wang, “A Hybrid ML Model for Predicting Crop Suitability Using Soil and Climate Variables,” Expert Systems with Applications, vol. 187, 2022.
[13] A. Hussain and M. Li, “Impact of Weather Variability on Crop Prediction Using Data-Driven Models,” Environmental Modelling & Software, vol. 141, 2021.
[14] K. Rout and S. Behera, “Improved Plant Disease Detection Using Convolutional Neural Networks with Data Augmentation,” Neural Computing and Applications, vol. 34, pp. 15579–15592, 2022.
[15] T. Singh and B. Thomas, “A Comparative Study on Crop Recommendation Using Machine Learning Techniques,” International Conference on Smart Agriculture and Systems Engineering, pp. 90–95, 2021.