Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] adob_v1_0.zip(1333 Kbytes)
Manuscript Title: DataScan: an extensible program for image analysis in Java.
Authors: K.A. Ritley, M. Schlestein, H. Dosch
Program title: DataScan
Catalogue identifier: ADOB_v1_0
Distribution format: zip
Journal reference: Comput. Phys. Commun. 137(2001)300
Programming language: Java.
Computer: PC 266MHz Pentium-II.
Operating system: Windows95/98/NT, Tru64 Unix (v.4.0F).
RAM: 2M words
Word size: 8
Keywords: Image analysis, Linescans, Convolution, Filtering, Scanning probe microscopy, Atomic force microscopy, Scanning tunneling microscopy, X-ray/neutron diffraction, Graphical user interface, Java, Graphics.
Classification: 14.

Nature of problem:
DataScan provides a graphical user interface for performing image analysis tasks and mathematical/logical operations on large (typically 500 x 500 or 1000 x 1000) arrays of double precision numbers. It includes basic image analysis tasks, sequential processing of hundreds images, plus suites of routines useful for data reduction in X-ray/neutron diffraction and scanning probe microscopy. New routines may be easily included by the user.

Solution method:
The internal program section which defines the graphical user interface (GUI) is clearly separated from program section which performs image analysis tasks, to simplify adding new computational extensions to the program. The use of the Java programming language ensures portability between all operating systems and computers which offer a Java Virtual Machine (JVM).

Restrictions:
The number of images which can be loaded depends on available machine memory, typically 8 bytes per pixel.

Unusual features:
The program is self-contained and portable. The image analysis features are incorporated whenever possible in a library of independent, standalone subroutines, so they may be re-used in other programs and applications.

Running time:
Execution times vary upon desired image processing tasks and size of the image. A linescan requires less than 1 s for computation. Convolution of a 1000 x 1000 pixel image requires about 1 s on a typical PC with 266 MHz Pentium-II processor.