SmartDoc: AI Based Mobile Application for Document Scanning, Verification and Categorization

Management of academic documents in higher education institutions often depends on manual verification processes that are time-consuming, difficult to scale, and susceptible to human error. This paper presents SmartDoc, a mobile-based system designed to automate document scanning, classification, and verification within academic environments. The proposed system integrates image preprocessing, Optical Character Recognition (OCR), and rule-based validation to ensure accurate and structured extraction of document content. The system integrates OCR with a hybrid validation approach combining rule-based checks and Gemini AI for intelligent document verification. The preprocessing module performs orientation correction, noise reduction, contrast enhancement, and user-assisted cropping to improve recognition reliability under practical capture conditions. Google ML Kit is used for on-device OCR processing, while extracted data is validated against predefined institutional rules to detect inconsistencies and incomplete information. The system also incorporates Gemini AI integration to support intelligent categorization, structured interpretation of extracted text, and automated verification feedback. SmartDoc is implemented using a cross-platform mobile framework with cloud-based backend services for secure storage, authentication, and role-based access control, enabling controlled document lifecycle management for students, faculty, and administrators. Experimental observations indicate improved consistency in document handling and reduced administrative workload. The proposed system provides a structured and secure approach to digital academic document management.