Mathematica is generally easier to use for image processing than is MATLAB, and it very widely used. However, it all depends on the level at which you are working (both in terms of control and flexibility). That said, as MATLAB isn't really designed for DIP anymore than Mathematica is, it sounds as if you might find it more useful (as opposed to professional software designed for image processing).
Apart from Matlab o mathematica, you can also use some other programming languages, such as C++, and use the libraries created for image processing. As example, openCV is very used for image processing. If you need to work with 3D images, ITK is a good option to do that.
Actually I would say that programming languages like C++ aren't just other options but in many ways superior. I'm not familiar with ITK, so naturally I'll trust Jose Dolz on this, but as far as languages like C++ are concerned they are far more powerful than anything like MATLAB. After all, statistical and mathematical software packages/environments were written to carry out statistical tests, build models, etc., while more generic languages (particularly those that rival assembler in their ability to control exactly what operations are performed yet don't require hundreds of lines of code where a language like C++ and even Java would need a few dozen) were designed to be all-purpose solutions, whether you wanted to program your own operating system or write code for fractal analysis.
That said, there is the matter of ease-of-use. If one is familiar with programming languages like C, C#, C++, Java, etc. then these offer greater flexibility and more power. If not, the learning period required is probably not worth it.
If this helps, I will tell you that I completely agree with Andrew Messing regarding all the arguments he gave for the use of MATLAB and/or C++ or other high level programming languages.