— Wheat, widely cultivated cereal plays pivotal ensuring food
nutritional security. Despite being a crucial staple food, wheat crops are highly susceptible to various fungal and bacterial yellow
powdery mildew, lead to reductions ranging between 20% and 50%. Traditional disease detection approaches field
inspection agricultural
often inaccessible small-scale farmers.This study proposes an AI-powered wheat

for automated disease identification from leaf images captured through mobile devices. The system utilizes advanced architectures like EfficientNet and MobileNet, which provide superior accuracy and faster inference speeds while maintaining computational efficiency suitable for edge devices. The proposed model is integrated into a mobile application that incorporates regional language support using translation APIs, allowing farmers to receive disease names, causes, and treatment recommendations in their native languages (such as Marathi and Hindi).
Experimental evaluation on publicly available wheat leaf datasets demonstrates a classification accuracy exceeding 95%, confirming robustness in
environments. integration of AI-driven disease identification with multilingual output not only enhances usability and accessibility but also empowers farmers with timely, understandable, and actionable insights. The proposed framework represents a significant step toward precision agriculture, digital inclusivity, and sustainable crop management.
Keywords—
Deep EfficientNet, Mobile Application, Regional Language Translation, Image Processing, Smart Agriculture, Precision Farming.

Keywords—Wheat
EfficientNet, Application, Regional Language Translation, Image Processing, Precision Agriculture.