Contact: Jack Ridge

Dept. of Geology

x72494

jack.ridge@tufts.edu

 

Analysis Software for Images of  Geologic Lake Sediment Cores


   This project is the design of a stand-alone program that performs repeated, in-sequence, thickness measurements on overlapping JPG  images collected along the axis of a core of lake sediment.  A program of this type is used in the Geology Dept. to construct records of annual variation in sediment thickness that serves as a chronologic and correlation tool as well as a proxy variable for climate.
    Currently we use a program first written back in 1996 by a Geology undergraduate as a plug-in to UTHSCSA ImageTool 2.0.  Last winter Shuai Yuan, an MS graduate student in ECE, revised the plug-in for operation with Windows XP and an updated version of UTHSCSA ImageTool ( 3.0  available at http://ddsdx.uthscsa.edu/dig/itdesc.html).  Shuai did an excellent job of getting us up and running over last winter break.  This program accepts images (BMP, JPG, and other formats) captured from a old video camera we have in our department.  The images are sequential along the length of a core (bottom to top) and have a low resolution given their size of 800 X 600 pixels. The program assists in measurements by recording marker lines on the images (posted with mouse clicks) and collecting data in a table corresponding to the vertical distances between horizontal lines identified as various types.  The program  will allow measurements to be corrected and creates a revised image set showing the positions of all measurements and annotations placed on the images. Unfortunately the combination of the plug-in and ImageTool 3.0 does not work with high resolution JPG images from our more recently acquired digital cameras that would give better resolution to our measurements even on their lowest image size.  Measurements come out misplaced on high resolution images because the program will not display the images in such a way that the slide bars on the image margins will work without registering false data.  The program seems to have trouble indexing the images.  The program has also behaved differently on different computers.  These problems seem to be a bug with ImageTool.
   We are dodging a bullet every time a new version of ImageTool or Windows comes out or we use a different computer.  This is why it seems like an appropriate time to construct a stand alone program that would be universal.  With the consent of Prof. Stafford the new program can be constructed in any language that seems appropriate and can be revised at a later date.  I would be more than happy to show you how the current program operates and I also have some ideas as to how the program might be improved or made more general, allowing it to have application beyond my specific needs.  A measurement program of this type would likely be useful to any one studying sequences that require repeated distance measurements.  In Geology this includes making repeated measurements of the thickness of stacked annual and non-annual sediment layers in images of cores or field exposures and in tree ring analysis. There are probably many other applications outside of geology as well.