This capstone project proposes a performance-based comparative analysis of Image Processing and Optical Character Recognition (OCR) services in the application of sports card attribute identification for efficient inventory. The analysis focuses on two web-based image processing services, Google Vision AI and ChatGPT 4.0. The performance analysis serves to determine the most accurate and overall higher-performing OCR service in the domain of attribute extraction from sports cards. Each service will be given image samples from a controlled source and an experimental source. This will provide variation concerning card features and image quality. These images will be used to generate OCR service responses that will be used to derive performance metrics. These metrics will represent the capabilities and accuracy of text detection extraction for each service to highlight the service with superior performance.

Recommended Citation: Hernandez T., Layman L., Modaresnezhad M., Morago B., Guinn C., (2024). Comparative Analysis of OCR Services in Automated Text Extraction for Efficient Sports Card Inventory Management. UNCW MS CSIS Proceedings. V. 18 , N. 10 .