The biggest difference from other imaging software is that IMAL is oriented toward scientific and technical applications. This requires a lot of additional programming to ensure that the numeric values in the image are accurate, and that information is not lost for the sake of convenience.
Histogram of image produced by IMAL (A) and a well-known Unix image viewer (B).
For example, the figure above shows a histogram of the same 16-bit image which was opened and then saved in IMAL (left) and another, well-known Unix image viewer (right). In this other program, the ``Normal Size'' box was checked and every other precaution was taken to ensure that the image was saved correctly. The visual appearance of the two images on the screen was identical. Yet the graph shows that the other program had cut corners and saved only 55 of the original 65,536 different intensity values. The other 99.9% of the information was thrown away. Similar, although not as extreme, situations can happen for color images as well - information not required for displaying the image is discarded without your knowledge for the sake of processing speed, because the program is only designed to display images, not to analyze them numerically.
Another example occurs in densitometry. Because of this concern for maintaining data integrity, results obtained in IMAL can be significantly more accurate than those obtained from other programs. Part of this is also due to the algorithms used in IMAL, some of which were invented for this express purpose. Of course, for some applications, the accuracy doesn't matter. However, the user is usually not aware that the other software takes shortcuts for the sake of speed. This can often produce inaccurate or unusable results.
For example, I was once asked to analyze an image of a Western blot that had very faint bands compared to the background. The person could not repeat the experiment, since these were irreplaceable autopsy samples, and had first tried to analyze the image with another well-known Macintosh-based image analysis program. Unfortunately, it turned out that the other program was skipping some crucial calculations in its calculation of the background level, resulting in errors of 50% with each measurement. Even after taking multiple measurements and discarding those which gave negative numbers, it was impossible to get credible numbers from the other program. Increasing the image contrast merely increased the error by the same factor. Using IMAL, the background could be determined to several decimal points, and clear-cut results were easily obtained.
Because IMAL was written by a research biochemist, the examples in this manual tend to be oriented toward biochemistry. However, the algorithms in this program were designed to be flexible enough to accommodate the requirements for technical image analysis in any field. If you have an application that is not handled adequately by IMAL, or if you have a suggestion for new features or would like to contribute to the IMAL project, we would like to hear from you.