Expertise-Free Annotation System for Korean Cultural Heritage Big Data (HCIK 2018, KOCCA)


Goal

New annotation system suggestion for meaningful data extraction and accurate data convergence of Cultural Heritage formative element -> Annotation data accumulation and big data analysis for diverse applications



Procedure

  • Annotation system design to annotate ceramic ware formative element by icons

  • SUS based usability test and result analysis --> 11 people (Male 9, Female 2; No knowledge in Cultural Heritage, design, formative element and system usability 9)


Result

  • Complexity, Simplicity and Convenience: No unnecessary, complicated, or cumbersome elements, and the annotation operation confirms low workload. Fig1

  • Technical assistance unnecessary and Learning unnecessary: Users may need technical help when using the system for the first time, but after learning how to use it, it is easy to use. Fig2

  • Unnecessary of Design expertise, Expertise and Cultural Heritage expertise: Even without expertise, this system can be used to perform annotation tasks. Fig3


Fig1.


Fig2.

Fig3.

  • Annotation expected workload and task persistence with annotation system

- Annotation estimated workload: minimum 3 images, maximum 700 images(average 149.5 (SD = 218.6))

- Annotation estimated task persistence: minimum 1hour, maximum 20hours (average 3.4 hours (SD=5.6))

- Each one image takes about 1 minute for annotation work → 150 images are processed efficiently within 150 minutes