Leveraging Deep Learning and Neural Networks for Real-Time Crop Monitoring in Smart Agricultural Systems
DOI:
https://doi.org/10.5281/zenodo.15973307Keywords:
Crop Detection, Segmentation, and Counting, High-level Feature Extraction for Crop Monitoring, Predictive Models for Crop Monitoring/Disease Mapping, Aerial/Satellite Imaging for Crop Detection and Monitoring, UAV Image-based Smart Crop Inspection, Weeds Detection, Snake Weeds Growth Detection, Planting Detection, Automatic Fall Disease Detection, Q&A Bots/Voice-based Smart Crop Detection and Monitoring, Crop Disease Detection using Deep Learning, Deep Learning Framework for Crop Plant and Yield Prediction, Hybrid Intelligent System for Smart Monitoring of Soil Health.Abstract
Rapid urbanization and globalization has posed a challenge of the food sustainability scenario but application of artificial intelligent systems can enhance productivity and improve food security. A deep learning algorithm based solution is proposed for the detection of diseases in crops is essential as the advancement of machinery was blazed away in Indian agriculture. India as an agrarian economy is highly dependent on agriculture since 57 percent of the population earns their livelihood out of it. With the late 1980s the agricultural system has seen a vexatious change with the introduction of the green revolution. However there remains a condition which threatens the entire backbone of agriculture scenario which is diseases. Chlorosis is one such disease which causes the yellowing of leaves and results in dropping of crops. The existing methods to detect chlorosis distance based methods or data mining based methods suffer from drawbacks of low accuracy and hence a solution is proposed using deep learning based approach. Environment affects world agricultural production due to alarming changes in climate. Acidity, temperature, salt, heavy metal contamination, herbicide or disease can cause osmotic stress in seeds
