Crescera - AI Infused Irrigation with Visual Insights Geared Towards the Farmer

Abstract

Stage 1 of this research, published in IJRASET Vol. 13 Issue XI (November 2025), demonstrated the feasibility of an integrated Artificial Intelligence (AI) and Internet of Things (IoT) irrigation platform — achieving 91% irrigation decision accuracy, 95% CNN-based disease classification accuracy, and a 25-30% reduction in water consumption. Stage 2 advances the system across four principal dimensions: (i) hardware consolidation through migration from NodeMCU (ESP8266) to ESP32 dual-core microcontroller with expanded sensor support; (ii) an enhanced Water Need Index (WNI) incorporating two additional parameters — soil temperature and soil pH; (iii) a deepened CNN architecture employing ResNet-34 with Convolutional Block Attention Module (CBAM) and an expanded eight-class disease taxonomy covering Powdery Mildew, Bacterial Blight, Anthracnose, and Early Blight in addition to the original four classes; and (iv) a Progressive Web Application (PWA) frontend replacing the static dashboard, providing offline capability via IndexedDB and real-time Chart.js analytics. Field trials across rice, wheat, and sugarcane over 45 days validate irrigation decision accuracy of 94.3%, disease classification accuracy of 97.1%, average system response latency of 1.2 seconds, and 35% water-use reduction relative to conventional irrigation. The Stage 2 system constitutes a production-ready precision agriculture framework deployable in low-connectivity rural environments.

Introduction

Agriculture consumes nearly 70% of global freshwater, yet traditional irrigation systems are only 35–40% efficient, leading to significant water wastage. This research presents Stage 2 of Crescera, an AI- and IoT-based smart irrigation and crop disease detection system designed to improve water management and agricultural productivity.

Stage 1 successfully introduced an AI-IoT platform featuring a three-parameter Water Need Index (WNI), CNN-based leaf disease detection, dual AI/manual operation, and a multilingual interface. However, it faced limitations such as processing bottlenecks on the ESP8266, limited sensor inputs, support for only four grape diseases, and the lack of offline functionality.

To address these issues, Stage 2 introduced several major enhancements:

• Upgraded hardware from ESP8266 to a dual-core ESP32 with FreeRTOS for better multitasking.
• Expanded the WNI from three to five parameters by adding soil temperature and soil pH measurements.
• Replaced the basic CNN with a ResNet-34 + CBAM attention model, increasing disease detection from 4 to 8 crop diseases.
• Developed a Progressive Web Application (PWA) with offline support, real-time visualization, and multilingual access.
• Conducted a 45-day field trial on rice, wheat, and sugarcane to validate system performance.

The enhanced system uses a five-parameter WNI (soil moisture, ambient temperature, humidity, soil temperature, and soil pH) to make intelligent irrigation decisions. A hysteresis control mechanism reduces unnecessary pump switching, improving system stability and energy efficiency.

For disease detection, the ResNet-34 + CBAM model was trained on 12,400 leaf images from public datasets and field-collected samples, enabling accurate recognition of eight plant diseases across multiple crops.

Key Results

• Irrigation accuracy: 94.3% (up from ~91% in Stage 1).
• Disease detection accuracy: 97.1% (up from 95%).
• Water savings: 35% compared with conventional irrigation.
• Response latency: Reduced to 1.2 seconds (40% faster than Stage 1).
• Pump cycling: Reduced by 41% using hysteresis control.
• Offline functionality: PWA supports 72-hour local data storage.
• Field validation: Successfully tested over 45 days on rice, wheat, and sugarcane.

Conclusion

This paper presented Stage 2 of the Crescera AI-Infused Irrigation system — a comprehensive advancement over the Stage 1 platform published in IJRASET Vol. 13 Issue XI, November 2025. Four principal contributions were delivered: (i) ESP32 dual-core hardware consolidation enabling concurrent FreeRTOS task execution; (ii) a five-parameter WNI with soil temperature and pH integration for crop-adaptive irrigation control with hysteresis; (iii) a ResNet-34 + CBAM CNN extending disease classification to eight classes with 97.1% accuracy; and (iv) a PWA frontend delivering offline caching and real-time Chart.js analytics. Field validation across rice, wheat, and sugarcane over 45 days confirmed irrigation decision accuracy of 94.3%, water consumption reduction of 35%, average system response latency of 1.2 seconds, and pump cycling reduction of 41% — representing consistent improvements across all key metrics relative to Stage 1. The system achieves production-readiness for deployment in small-to-medium scale, connectivity-constrained rural agricultural environments.

References

[1] Y.Tace, et al., \"IoT and Machine Learning-Based Smart Irrigation Framework for Sustainable Agriculture,\" International Journal of Emerging Technology and Advanced Engineering, vol. 14, no. 2, pp. 52-59, 2025. [2] M. Razak, et al., \"Agriculture 5.0 and Explainable AI for Smart Agriculture: A Scoping Review,\" IEEE Access, vol. 13, pp. 24412-24427, 2025. [3] H. Wei, et al., \"Irrigation with Artificial Intelligence: Problems, Premises, and Promises,\" Computers and Electronics in Agriculture, vol. 214, no. 8, pp. 321-337, 2024. [4] G. E. Oguzturk, \"AI-Driven Irrigation Systems for Sustainable Water Management: Systematic Review and Meta-Analysis,\" Elsevier Journal of Agricultural Informatics, vol. 12, no. 1, pp. 17-30, 2025. [5] S. Gupta, \"Smart Agriculture Using IoT for Automated Irrigation, Water Conservation, and Crop Monitoring,\" IJRASET, vol. 13, no. 4, pp. 401-406, 2025. [6] Z. Zeng, \"AI-Driven Smart Agriculture Using Hybrid Transformer-CNN Models for Disease Detection,\" IEEE Access, vol. 12, pp. 121344-121356, 2025. [7] P. Rastogi, \"Early Disease Detection in Plants Using CNN,\" International Journal of Computer Applications, vol. 186, no. 29, pp. 9-15, 2024. [8] A. Y. Ashurov, \"Enhancing Plant Disease Detection Through Deep Learning,\" Journal of Intelligent Systems, vol. 30, no. 7, pp. 192-199, 2025. [9] X. Liu, \"Intelligent and Automatic Irrigation System Based on IoT Using Fuzzy Control Technology,\" IEEE Internet of Things Journal, vol. 11, no. 3, pp. 1516-1524, 2025. [10] XenonStack Research Team, \"Smart Irrigation: Leveraging Sensor Data and AI for Sustainable Water Management,\" Industry White Paper on Precision Agriculture, XenonStack, 2024.

Copyright

Copyright © 2026 Siddharth Mohite, Rushikesh Patil, Siddharth Jadhav, Madhuri Choudhari. This 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.