News |  Sitemap |  Contact
PDF Export  | 

Easy Article Figures Encapsulating original data and processing steps.

Abstract

In this paper, we present a new ImageJ plugin dedicated to scientific article figures preparation.

Figures are crucial components of research communications. Beyond the field of life sciences, the data presented in figures is displayed to visually support and summarize the findings discussed in the article text. The motivation for including images resides in the strength of the visual perception. Building a convincing figure is a demanding task that covers very different steps ranging from figure content acquisition to figure assembly in an image editing software. Notions of image processing are often required when it comes to even simple tasks such as cropping or resizing images and assembling them in a single figure of a unique resolution.

Scientific images are typically well handeled in dedicated software suites but very poorly supported in software used for laying out the final version of a figure for submission to peer review process.

The proposed new software tries and make the figure creation process easier for the end-user:

  • Based on ImageJ, it seemlessly supports most current scientific image data formats handled by stock ImageJ.
  • A constrained treemap-derived layout ensures a clean and editable pixel-precise panel arrangement.
  • Each panel displays a choosen region from a source dataset view, and tracks it’s datasource, preprocessing steps and display setting.
  • High quality geometrical transformation of image data is achieved using Erik Meijering's imagescience library, and allows accurate scaling and rotation of images.
  • Because original pixel scaling is preserved during the whole process of figure creation, common overlay elements like scale bars can accurately be placed at the end of the figure preparation process.
  • Figures can be annotated using ImageJ built-in overlay functions, saved in an open format for further editing, or exported in standard image formats, together with a human readable text file describing image processing steps for all panels.

The focus is given on image data, but examples are provided on how to include other types of panel data, like high quality plots from the R statistical package or from MS-Excel, or 3D molecular visualizations from PyMOL.

In preliminary tests, complex figures creation time shifts from hours to minutes.

Keywords

article figures, metadata, processing history

Administrative data

Presenting author: Jerome Mutterer
Organisation: CNRS

co-authors: Edda Zinck

© Luxembourg Institute of Science and Technology | Legal Notice