Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68386
Volume:
10 |
10 |
Oct 2026
Pages:
1--
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department of Information Technology
Sandip Institute of Technology and Research Centre (SITRC)
Department of Information Technology
Sandip Institute of Technology and Research Centre (SITRC)
Department of Information Technology
Sandip Institute of Technology and Research Centre (SITRC)
Department of Information Technology
Sandip Institute of Technology and Research Centre (SITRC)
Department of Information Technology
Sandip Institute of Technology and Research Centre (SITRC)
Department of Information Technology
Sandip Institute of Technology and Research Centre (SITRC)
Department of Information Technology
Sandip Institute of Technology and Research Centre (SITRC)
While deep learning has significantly advanced au-tomated crop disease detection, existing models frequently suffer from severe performance degradation when deployed in unstruc-tured field environments. Furthermore, current computer vision applications in agriculture typically function in isolation, evalu-ating visual symptoms without incorporating the critical micro-environmental context necessary for practical decision-making. This study proposes an integrated, multimodal smart farming support system specifically engineered for tomato crop manage-ment, a high-value but disease-susceptible crop. The architec-ture utilizes a dual-model deep learning pipeline: a YOLOv8-based object detection model for real-time fruit localization and maturity grading, coupled with a lightweight MobileNetV2 Convolutional Neural Network (CNN) incorporating depthwise separable convolutions for leaf and fruit disease classification. To bridge the gap between visual diagnosis and actionable agronomy, the system fuses the visual AI outputs with manually inputted soil parameters (pH, NPK) and automatically fetched geolocation-based meteorological data. A contextual Python-based rule engine processes this fused data to generate highly specific organic bio-fertilizer recommendations and automated seed-count requirements based on farm area. Addressing rural technological barriers, the web-based prototype, supported by a scalable Flask and MySQL backend, incorporates Web Speech APIs for voice-assisted outputs specifically in Marathi, ensuring accessibility for local illiterate farmers. The system’s predictive robustness is evaluated by comparing classification accuracy on controlled laboratory datasets against unstructured field-captured images, providing a scalable, edge-compatible frame-work for low-resource agricultural deployment.
KEYWORDS
This study presented a scalable, edge-compatible, multi-modal smart farming system designed to address the diagnostic vacuum in tomato cultivation. By deploying a dual-model computer vision pipeline via a robust Flask API backend, the system decoupled maturity grading from disease classi-fication. Integrating a Python-based rule engine allowed the system to synthesize visual predictions with meteorological data and localized soil parameters, generating precise, context-aware organic interventions. Furthermore, the inclusion of an automated seed calculator, market-price fetching, and Marathi voice-assisted outputs transformed an isolated diagnostic tool into a comprehensive economic and resource-planning plat-form. Future iterations incorporating offline edge inference will further solidify this architecture as a transformative so-lution for resource-constrained agriculture.
[1] FAO, “The State of Food and Agriculture 2023,” Food and Agriculture Organization of the United Nations, 2023. [2] A. Kamilaris and F. X. Prenafeta-Boldu´, “Deep learning in agriculture: A survey,” Computers and Electronics in Agriculture, vol. 147, pp. 70-90, 2018. [3] S. Sladojevic, M. Arsenovic, A. Anderla, D. Culibrk, and D. Stefanovic, “Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification,” Computational Intelligence and Neuroscience, vol. 2016, 2016. [4] D. Singh, N. Jain, P. Jain, P. Kayal, S. Kumawat, and N. Batra, “PlantDoc: A Dataset for Visual Plant Disease Detection,” Proceedings of the 7th ACM IKDD CoDS, pp. 249-253, 2020. [5] K. P. Ferentinos, “Deep learning models for plant disease detection and diagnosis,” Computers and Electronics in Agriculture, vol. 145, pp. 311-318, 2018. [6] J. Chen, J. Chen, D. Zhang, Y. Sun, and Y. A. Nanehkaran, “Using deep transfer learning for image-based plant disease identification,” Computers and Electronics in Agriculture, vol. 173, p. 105393, 2020. [7] G. Liu, J. C. Nouaze, P. L. Touko Mbouembe, and J. H. Kim, “YOLO-Tomato: A Robust Algorithm for Tomato Detection Based on YOLOv3,” Sensors, vol. 20, no. 7, p. 2145, 2020. [8] S. R. Chettri and S. B. Bera, “A comprehensive survey on deep learning in smart agriculture,” Artificial Intelligence in Agriculture, vol. 9, pp. 1-25, 2023.
Publication Date
10/27/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
3
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...