Smart System Using Deep Learning for Detection of Paddy Crop Diseases and Automated Pesticide Spraying

Abstract

Paddy a staple food crop that supports more than half of the world’s population. However, several situations, including rice blast, brown spot, bacterial scab, and scald, which can lower yield and result in significant losses, have a significant impact on its civilisation. Prior and accurate discovery of these conditions is vital for proper crop operation. The requirement for an automated outcome is highlighted by the fact that traditional methods that rely on manual evaluation are frequently laborious, time-consuming, and prone to errors. This research describes an AI-based approach that uses computer vision and deep learning to automatically detect paddy crop issues and target germicide dispersal. The system utilizes a Convolutional Neural Network (CNN), a type of computer model designed to dissect images, to examine images of paddy leaves and identify symptoms of complaint. A camera continuously captures videotape of the crop field, and frames are taken at regular intervals for processing. These frames undergo a preprocessing method similar to resizing, normalization, and data augmentation to enhance the delicacy and Mitigate Possible spelling mistakes. The trained CNN classifies the images as either healthy or diseased. When an issue is detected, a signal is transmitted to a Jeer Pi (a small, affordable computer), which activates a DC diaphragm pump to spot pesticides directly onto the affected area of the factory. This targeted system reduces the use of chemicals and lowers environmental impact. Overall, the proposed system enhances discovery delicacy, reduces manual labor, and supports precision agriculture by enabling real-time monitoring and effective control operations in paddy cultivation.