The paper presents a study aimed at optimizing the reading and evaluating process of multiple-choice exam sheets (optical forms) as a smart campus application at Izmir Bakircay University, significantly reducing the time and cost involved. Traditional manual evaluation methods are time-consuming and expensive. The proposed solution leverages digital transformation and image processing technologies. The algorithm developed processes scanned images of optical forms, generates scores, and provides statistical reports without requiring specialized knowledge. The methodology is divided into three main steps: detecting sections of the optical form, cropping sections, reading marked circles based on coordinates, reporting exam results, and statistical analysis. Testing with previously used optical forms showed 100% accuracy at a resolution of 600 \(\,\times \,\) 600 dpi. Errors caused by user handling were anticipated and the system was designed to be robust against such errors. The results of implementing this automated system include a 98.15% reduction in processing time and a 99.95% reduction in cost. The new evaluation time is around 5 min per exam, costing 347 TL. The process now requires only one person instead of four. This study exemplifies the potential of digital transformation and image processing to enhance operational efficiency and cost-effectiveness in academic settings.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Automating Exam Assessment Through Digital Transformation and Image Processing

  • Onur Dogan

摘要

The paper presents a study aimed at optimizing the reading and evaluating process of multiple-choice exam sheets (optical forms) as a smart campus application at Izmir Bakircay University, significantly reducing the time and cost involved. Traditional manual evaluation methods are time-consuming and expensive. The proposed solution leverages digital transformation and image processing technologies. The algorithm developed processes scanned images of optical forms, generates scores, and provides statistical reports without requiring specialized knowledge. The methodology is divided into three main steps: detecting sections of the optical form, cropping sections, reading marked circles based on coordinates, reporting exam results, and statistical analysis. Testing with previously used optical forms showed 100% accuracy at a resolution of 600 \(\,\times \,\) 600 dpi. Errors caused by user handling were anticipated and the system was designed to be robust against such errors. The results of implementing this automated system include a 98.15% reduction in processing time and a 99.95% reduction in cost. The new evaluation time is around 5 min per exam, costing 347 TL. The process now requires only one person instead of four. This study exemplifies the potential of digital transformation and image processing to enhance operational efficiency and cost-effectiveness in academic settings.