Quantifying degree of Coral Bleaching in ImageJ

[Callis]

Hello Everyone!

I'm currently working with the effects of pesticides on non target species, i.e. corals (Acropora).

One of my endpoints are degree of bleaching and I have 840 photos to help me quantify the bleaching degree as a function of divergent pesticide concentration.

The photos are taken at similar light conditions, with the same camera at a fixed distance and settings.
In all the images (attaching two views to examply) a color chart and a scale is present.
Each coral fragment (n=35) has been photographed four times (four perpendicular views).

35 * 4 = 140 images per photo session. I have six photo sessions to analyze, 840 images.

I want to measure the light intensity for a fixed area. This is no problem. However, I would be thrilled if I could do this automatically instead of defining individual ROIs manually for each fragment, which I do now.

The problem I'm having is that the corals differ in shape (branshed, bent etc) and the construction which the fragments are attached to are present in some photographs but not in other (due to the altered perpendicular views). In addition, I only want to analyze pixels above the cable tie. I have had a lot of difficulty to write a macro in ImageJ which met the stated conditions since the image itself differs (with regard to coral morphology and fixation point) depending on perpendicular views.  
 
Is it possible to write a macro which can exclude parts of the image in some pictures but not in others?

Would be very greatful if macro skilled users could help me fascilitate the quantification.

Best Wishes!

Callis Amid
Department of System Ecology
Stockholm University, Sweden