Image processing

The microscopist has a wide range of imaging systems and analysis software to his/her disposal. In the AIF we have at the moment 8 different microscope types (11 different microscopes) and we use different software packages to analyse our digital images.

However, a digital image is not just a nice picture, IT IS DATA! To extract the data we are interested in, we often have to do some form of image processing. We use filters, deconvolution algorithms, background subtractions, intensity adjustments, gamma corrections, etc. Are we always aware what we are doing or if what we do is allowed?

Unfortunatly, there is a "culture of beautification" as Cromey puts it in his recent publication. The number of cases of fraudulent image manipulation in manuscripts is increasing and be not surprised if a reviewer of your article asks to see the original image. That is one reason why all users of our imaging facilities are told to store their data in the original file format.

Other reasons to store data in its original format are a) different analysis can be performed on the same data set, maybe not now but in the future, b) meta-data (information about the acquisition settings) are stored in the original file format, c) there are indications that in the future all image data used in published research has to be made available online. That is also why it is important that we setup an imaging storage and data base system.

Guidelines for image processing are difficult to give as it often depends on the type of research and what would be allowed in one case would be questionable in an other situation. We can say that the first step is to use the proper imaging system, acquisition settings, standard calibration and controls for your experiments. Secondly, report besides the acquisition system and settings, the different processing steps in your manuscript, which software is used and which software version. This means that you have to document the different steps of image processing and software used at the moment you are analysing your data, not when you write the manuscript! Too often steps are missed in the manuscript and I would advice to add a text file with processing information to each directory with image data you analyse. It makes life much easier later on and you can give answers without problems when questions are asked.

Have a look at these online Guidelines for Best Practices for image processing.

Further reading:
Cromey DW (2010) Avoiding Twisted Pixels: Ethical Guidelines for the Appropriate Use and Manipulation of Scientific Digital Images. Sci Eng Ethics 16:639–667.
Benos DJ and Vollmer SH (2010) Generalizing on Best Practices in Image Processing: A Model for Promoting Research Integrity. Commentary on: Avoiding Twisted Pixels: Ethical Guidelines for the Appropriate Use and Manipulation of Scientific Digital Images. Sci Eng Ethics 16:669–673.
Waters JC (2009) Accuracy and precision in quantitative fluorescence microscopy. JCB 185: 1135-1148.
Rossner M and Yamada KM (2004) What's in a picture? The temptation of image manipulation. JCB 166: 11-15.

Share this page: