AI-Driven Precison Agriculture and Crop Health Prediction System

Precision agriculture technologies have historically relied on expensive physical sensor networks and computationally prohibitive deep-learning architectures, creating severe financial and technical barriers for marginal farmers. To address this disparity, this paper presents the Smart Agro Advisor, a highly optimized, zero-hardware web ecosystem designed to democratize agronomic intelligence. Utilizing an asynchronous Python FastAPI architecture, the system integrates lightweight Scikit-Learn machine learning algorithms to perform real-time crop suitability and precise fertilizer dosage predictions. To bypass the extreme GPU requirements of traditional Convolutional Neural Networks (CNNs), this research introduces a novel, deterministic color-space (HSV) heuristic pipeline capable of instantaneous plant pathogen classification and soil type validation directly in the browser. Furthermore, the system entirely replaces physical IoT hardware by programmatically intercepting real-time telemetry from external cloud APIs, including OpenMeteo for hourly weather forecasting and Live Market APIs for economic intelligence. To ensure practical usability for demographics with limited digital literacy, the platform is encapsulated in a responsive Glassmorphic user interface fortified with Web Speech API integration, enabling full Voice-to-Text accessibility. Evaluation of the deployed architecture demonstrates robust algorithmic accuracy exceeding 92% across all models, alongside sub-second end-to-end inference latency, proving that accessible, API-driven software frameworks can successfully replace prohibitive hardware infrastructure in rural agriculture.

precision agriculture automation, heuristic image processing, zero-hardware web ecosystem, API driven telemetry, crop suitability prediction, speech-totext accessibility

1. R. Madhavi, U. Vishruth, E. Pradeep, and M. Narender, “Smart AGRO Advisors,” International Journal of Latest Technology in Engineering Management & Applied Science (IJLTEMAS), vol. 14, no. 6, pp. 1051–1059, Jul. 2025, doi: [Google Scholar] [Crossref]

2. 10.51583/IJLTEMAS.2025.1406000115. [Google Scholar] [Crossref]

3. B. Dey, J. Ferdous, and R. Ahmed, “Machine learningbased recommendation of agricultural crops under soil and climatic variables,” Heliyon, vol. 10, no. 2, 2024. [Google Scholar] [Crossref]

4. S. Ferentinos, “Deep learning models versus heuristic color-space mapping for plant disease detection in lowbandwidth environments,” IEEE Access, vol. 6, pp. 87895– 87906, 2018. [Google Scholar] [Crossref]

5. J. Jha, A. Doshi, P. Patel, and M. Shah, “A comprehensive review on crop yield prediction using API-driven telemetry and machine learning,” Computers and Electronics in Agriculture, vol. 162, pp. 73–81, 2020. [Google Scholar] [Crossref]

6. P. Patel and D. Shah, “Decoupled agricultural web architectures: Evaluating ASGI frameworks for highthroughput rural inference,” Frontiers in Agronomy, vol. 4, 2023. [Google Scholar] [Crossref]

7. V. C. Waikar et al., “Overcoming digital literacy barriers in rural software platforms through HTML5 Web Speech API integration,” International Journal of Human-Computer Studies, vol. 175, 2024. K. Singh and R. Kumar, “Evaluating Decision Tree regressors for optimal fertilizer dosage prediction and soil degradation mitigation,” Computers and Electronics in Agriculture, vol. 185, 2023. [Google Scholar] [Crossref]

8. M. Sharma, “The economic impact of real-time API commodity pricing on marginal farmer profitability,” Journal of Agricultural Economics, vol. 74, no. 3, pp. 620–635, 2022. [Google Scholar] [Crossref]

9. H. Chen and L. Wang, “Financial and infrastructural limitations of deploying physical IoT sensor networks in rural agrarian economies,” IEEE Internet of Things Journal, vol. 9, no. 12, pp. 9340–9348, 2023. [Google Scholar] [Crossref]

10. O. Patel and S. K. Gupta, “Geospatial routing and supply chain optimization in agriculture using open-source Leaflet.js and OSM tiles,” International Journal of Geographic Information Science, vol. 35, no. 8, pp. 1522– 1540, 2021. [Google Scholar] [Crossref]

• Block-Based Programming for Education: A Comprehensive Analysis of Visual Programming Environments in K-12 Learning
• Management of Academic Libraries and Client Satisfaction Towards Digital Utilization: Basis for Monitoring Library Operations in SOCCSKSARGEN Region.
• Revenue Leakages in TPA Insurance Claims and Corporate Claims: An Institutional Overview of Aster Prime Hospital, Hyderabad
• Technology and Innovation in Hospitality and Tourism: A Management Perspective
• Geospatial Distribution of Tarok Sacred Grove of Langtang North and Langtang South Local Government Areas