Texture analysis of digital images is a physics based concept. It aims to calculate physical features (called texture parameters) of images making possible extraction of image details that the human visual system cannot perceive.
I did my PhD in texture analysis of medical images, MRI specifically. You will find that texture is quite a big field, and as there is no unified definition of texture (although it is very intuitive!) there are different approaches to texture. If you want an introductory and comprehensive text to texture try this one:
We done a lot of work in image texture analysis of the carotid artery for distinguishing venerable plaque components that hey might be at risk to rupture. You may also have a look at our website at http://www.medinfo.cs.ucy.ac.cy/ where we have all our papers for download as well as code for texture image analysis incorporated to a GUI.
I also worked a little on such thing but it was not exactly what you are asking. However, can share your idea and let's see if something can be done on it.
I have worked in the field of remote sensing (satellite, airborne, and ship borne imaging) for 40 years and have investigated texture and high/low spatial frequency enhancement and detection. The last couple of years I have toyed around with the application of some of the methods I have developed to the medical imaging field. Below are links to a couple of papers that relate to this type of analyses of satellite images of the earth but are generally applicable to any digital image. Perhaps they may be of interest to you; if possible, perhaps you can make a couple of images available?
Pat
An automatic optimum kernel-size selection technique for edge enhancement (1982)