Metadata for YOLOv5_FPC Detected Images

The CSV and Excel file contain metadata for YOLOv5_FPC-detected images, including URLs, Index, FileName, Year, Month, Day, YearMonthDay, and NewspaperSource. These were identified from the Chosun Ilbo Newspaper Digital Archives covering the period from 1920 to 1940, from a total of 47,777 JPG image files, accessed from https://newslibrary.chosun.com. The image extraction process was automated using the YOLOv5_FPC model, a fine-tuned iteration of the YOLOv5 model. In the URLs column, there are direct links to the official Chosun Ilbo website where the Four-panel-cartoon (FPC) is published. The Index column consists of numbers from 1 to 1035, representing the FPC files discovered using YOLOv5_FPC, totaling 1040 FPC image objects. The FileName column contains the image file names, while the Year column indicates the publication year of the FPC, with Month and Day columns denoting the respective month and day of publication. The NewspaperSource column provides information about the newspaper, in this case, Chosun Ilbo. With these metadata, any researchers can easily access the corresponding URL addresses for specific publication dates and study the 1040 FPC objects, including previously undiscovered FPCs. Our dataset includes FPCs that have not been discovered by previous researchers, to the best of our knowledge. We hope that this dataset will foster further global interdisciplinary discourse on Korean culture and history, during 1920 to 1940, shedding light on socio-cultural phenomena in modern Korea as well.