JColloids: Image analysis for video-microscopy studies of colloidal suspensions

Abstract

We present an updated version of JChainsAnalyzer [1], a Java and ImageJ-based software for the analysis of video-microscopy images regarding the aggregation dynamics of super-paramagnetic particles in magneto-rheological fluids [2,3]. This new version of the software has been adapted for a general use in image-based experiments of microprobes in suspension, and, consequently, has been renamed as JColloids. In this new version, the number of available options has been reduced, and the image filtering is virtually automatic, depending only on one numeric factor. JColloids has been recently used in the image analysis for micro-rheological studies using sedimented micro-particles in suspension [4] and for tracking microbeads trapped by optical tweezers. New version program summaryProgram Title: JColloids (JChainsAnalyser v.2.0)Program Files doi:http://dx.doi.org/10.17632/jbw8cdt2s5.1Licensing provisions: GPLv3Programming language: JavaSupplementary material: A new document Help.pdf includes basic running instructions for the new code.Journal reference of previous version: Computer Physics Communications, 80 (2009) 1956–1960Does the new version supersede the previous version?: Yes. This new version provides the same image analysis functionality, but the options available in the graphical user interface (GUI), the number of input parameters and the secondary calculations have been substantially simplified.Nature of problem: The previous version of this software was focused on its application to magneto-rheological fluids or aggregated chains of magnetic particles in colloidal dispersions. However, video-microscopy set-ups capturing objects in the micrometer scale are apt to be used for an extended variety of different video-microscopy experiments regarding colloidal science. Our aim is to modify the JChainsAnalyzer software for it can be used in the detection of colloidal objects by image analysis and the subsequent record of the temporal evolution of their center of masses.Solution method: We modify JChainsAnalyzer to be adapted to the analysis of more general experiments in colloidal science. JColloids is an updated and simplified version of JChainsAnalyzer, which uses the same methodology and image analysis algorithms as the previous version. However, this new software is less complicated for the user and it provides the same general functionalities.Reasons for the new version: The original code needed an update since its publication for its application beyond the aggregation of beads in magnetic colloids. Indeed, the new version, JColloids, can be successfully applied to the image analysis of different but similar video-microscopy experiments in colloidal physics, such as micro-rheology using optical tweezers of the study or the micro-structure of a sedimented suspension of micro-particles, as shown in Fig. 1.Summary of revisions: The external libraries read by the software have been updated to their latest available versions. In this new version of JChainsAnalyzer, the GUI has been notably simplified in comparison with the previous one. The GUI reads the external parameters needed for the image filtering, which have been reduced to only three. Only one of the input parameters is a numeric number, whose modification is not usually needed, with the purpose of multiplying the threshold value calculated from the adaptive algorithms in the image analysis. There are another two external parameters, which may be useful in particular situations: one allows to erode the detected particle’s contour, while the other is used to separate joined particles. Fig. 1 shows several examples of the filtering of images of micro-particles in water in different experimental configurations (see figure’s caption). All the binary images are obtained using the same input parameters, making the process almost automatic. Note that in Fig. 1 (b), the upper-left particle is slightly out of focus compared with its location in Fig. 1 (a), and the adaptative algorithm is still detecting correctly the contour of the particle. The filtering in Fig. 1 (c) corresponds to a different experimental setup [4] and zoom configuration, and the detection is also properly performed.Finally, we have to mention that the code has been notably simplified and it is more readable since we have refined the internal calculations related with technical aspects in the research of the scaling in the aggregation of magnetic fluids [2], assuming that any necessary calculation can be performed afterwards using the output file. In this sense, the output file, which contains the (x,y) positions of the centers of the detected objects, has been designed to be compatible with the input needed with tracking routines in IDL, which are standard in the analysis of micro-rheological experiments [5].Additional comments including Restrictions and Unusual features: The software can be run without GUI, but a long run command indicating all the external libraries is needed (how to do this is indicated in the help file). The software does not perform additional calculations, only the main parameters which characterize the objects are detected from the image analysis.AcknowledgmentsThe authors acknowledge MINECO for its financial support through projects FIS2013-47350-C5-5-R (P.D.G, M.P and M.A.R.) and CTQ-2016-78895-R (F.O.). P.D.G. also acknowledges J. C. Gómez-Sáez for her proofreading of the texts.