Log Analysis Methodology to Understanding Detailed IIIF Image Usage

Abstract

It is important for libraries and museums to understand how digital collections and their contents have been used for many reasons, e.g., accountability for stakeholders. In these years, a lot of libraries and museums have adopted IIIF. In IIIF-compatible digital collections, an image is fetched via IIIF Image API. Every time the image is zoomed and panned on an image viewer, the image is called via IIIF Image APIs with varying the value of the region. Thus, it is possible to investigate the detailed image usage by examining which regions of images have been requested. In our presentation, we show a method to analyze the image usage and to visualize the analysis result. The method is comprised of the two steps: 1) measure the number of accesses to each pixel and 2) generate heat map in which the color of each pixel represents the number of accesses to the corresponding pixel. Since a pixel is the smallest unit that composes an image, we enable a fine-grained analysis. Heat maps visualize which regions of an image get more and less accesses. The generated heat maps are displayed over the corresponding target images using the layer function of Mirador. As a result of the analysis, we observe the tendency that the regions close to the center get more accesses than other regions. One of the reasons is that the regions close to the center have higher probabilities to be accessed, when users zoom and pan an image on an image viewer. Therefore, we theoretically compute the probability to be accessed for each pixel and examine how to adjust the number of accesses to each pixel depending on the position of the pixels. In the presentation, we discuss adjustment methods that use the computed probabilities. Finally, we present possible applications including collaborative research platform and transcription platform.