Perhaps yo could use the same technique used to interpolate some spatial  terrain characteristics in oil reservoir análisis. You van obtain information searching with the term "Kriging" in google.

Popoff, C.C. Computing reserves of Mineral deposits: principles and

conventional methods. Washinton , Bureau of Mines. 113p.

Zhang, R., Myers, D. E., y Warrick, A. W., (1992) Estimation of the

Spatial Distribution of Soil Chemical Using Pseudo Cross-Variograms, Soil

Science Society of America Journal, Vol. 56, No. 5, p.1444-1452. Dehdari.

Seems a bit like a surface fitting problem, have a rectangular surface on the grid (say a B-spline surface), each pressure could be considered a height on the surface, fit a surface through the heights then read off the surface height at any desired intermediate point. e.g. http://www.geometrictools.com/Documentation/BSplineSurfaceLeastSquaresFit.pdf

Hi Prabodhi,

I am not exactly clear on the question. But I feel you can increase the number of columns by increasing the resolution of individual range an mapping the colors accordingly. 

If you are trying to predict the intermediate value between 2 different sensor reading, I am not sure how are the different sensors related. It all depends on their relation. You can try to have a running average slope of your data and predict a mid-point between 2 readings. You can even try exponential weighted average (predVal = alpha*currVal + (1-alpha)*prevVal). Of course you may need to figure out the value of alpha depending on your curve.

Abhijit

Hopefully these will be helpful:

http://tinyurl.com/oby5hk3

http://tinyurl.com/nqd6c5s

http://tinyurl.com/opwqkb7

http://tinyurl.com/orj8n66

http://tinyurl.com/pf8axh6