You can start with a program called NASAVIEW, available for use without fee or licence from Nasa/JPL . You do need to register to obtain this but the procedure is not onerous.

NASAVIEW lets you read the native image format of many different space missions and imagers, so it's something any serious dabbler should have. If nothing else, it lets you output standard graphics formats from the proprietary internal NASA ones, and it lets you read the embedded labels.

Robert - you already have this so we go to step 2

Each image consists of a set of "data numbers" (usually 0-255 but sometimes larger byte size). These data numbers are mapped by a scaling curve to real-world numbers such as IR intensity or brightness temperature. Strictly speaking, you need to find the calibration tables for the instrument to translate the data numbers to real measurements.

For many purposes, we can deal in relative terms and skip the formal calibration (though we end up not knowing exactly what temperatures we're dealing with, which is one of your fundamental questions!)

Let's look at image I02026006 as an example - one of those you found in your search for high brightness temperatures. You can view a version of this image here.

We see craters of various sizes with sun-facing rims lit up and shaded rims dark. There is a gradation of the background intensity down the image, being brighter at the bottom (not neccesarily south!), where there are also many old fissures.

You can download the brightness temperature (BTR) image here.

In an html browser it's mostly gibberish, but you can read the header information OK. The image was taken at ~3:50 pm (local martian time) at Ls=20 (midsummer) at 39 degrees north. A nice warm time of year, especially for dry dusty rocks.

The BTR is an IR band chosen to be free of most atmospheric effects and one which responds well to ground temperatures as found on Mars. To a first approximation, the data numbers range linearly from the minimum BTR (here 302.213 K) to the maximum (322.192 K) {not a very wide temperature range, btw, compared to other images}

Now, RObert, the question you keep posing is "where are the highest temperatures?" Well that's real easy to answer. Download the BTR image to your hard drive. Open it with NASAVIEW and use the "options" "view Histogram" controls to bring up the histogram of data distribution.

In this case, we have a sort of two-humped camel with a long neck. The hottest areas are the ones with the highest data numbers. In the default Palette (Greyscale) these are the brightest pixels. You can emphasise these by changing colour palettes.

A scan of the image will already tell you 90% of the answer - the hottest areas are on the sun-facing ramparts of craters and fissure walls. (you have to close the histogram window to be able to scroll the image)

Now try focussing sharper - use the histogram tool again. This time play with the "minimum", "maximum" and median sliders to see how to highlight the hottest areas. (Best way is to leave min and max alone and dial median down to about 50 - then you can see exactly what colour each "hot" pixel is). Result is the same. Hot areas are the sunny crater walls and fissure edges.

No sign of local hot spots...

Have fun with the toolbox, and tell us if you find something interesting - if you do tell us what and where, how you processed things, and preferably attach a snapshot of your results.