I'm done with lens reviews

Microcontrast

Actually it is quite real. It can be measured and detecting micro contrast is the first step in most modern image process from fingerprint recognition to facial recognition. In 1976 Fairchild produced the first useful digital imaging sensor for intelligence, surveillance and reconnaissance (ISR ) applications. By 1985 when I got to FCI digital imaging was rapidly replacing film based systems although film based systems would have a resolution advantage into the mid 1990's. But digital was much more efficient. An aircraft can downlink the image in real time to the ground for interpretation.

The problem because - what to do with the volume so machines had become an integral part of the process. The building block of such processing is based on micro contrast. Technically micro contrast needs the concept of the "image gradient" for its definition. Like any gradient - the image gradient is an operation performed on the image to calculate how rapidly the contrast changes in a given direction. That is micro contrast is interrelated to the calculus of the image as viewed as a two dimensional function.

The out put of this operation is used to find edges in an image which are used to identify objects within the image. An edge that goes from minimal value to maximal value at the boundary of two pixels will produce the highest output value of the image gradient. Putting this together allows law enforcement to take an image of a finger print and it be characterized, the critical properties found and the identity of the person determined by a computer. It allows you touch ID to read your finger print and open you phone, iPad, etc. The image gradient is the first step in you iPhone unlocking using facial id. It all starts with the concept of micro contrast.

In the 1980's FCI and other companies expending large amounts of resources - some funded by the US Government and some internal exploring the use of the image geometry for practical purposes, read fingerprints, identify individuals, locations, identify objects in images, etc. The first step to all this was the gradient and higher order differentials of the image. The key to all that processing was the micro contrast within the image. FCI worked closely with the Courant Institute of Mathematical Sciences at NYP through a DARPA effort directed toward to the automatic target recognition problem brought on line the first fingerprint processing systems which after two years of evaluation was approved by the FBI It all starts with exploiting the calculus of micro contrast in the image. This effort was directed to the automatic target recognition problem - but had very immediate applications to forensics.

So yes micro contrast - it is real. It can be measured and quantified. Our eyes also perform many similar operations feeding the information to the brain that then helps us identify faces, objects, etc. Today it is taught in colleges around the world.

https://www.cs.toronto.edu/~mangas/teaching/320/slides/CSC320L05.pdf

https://pyimagesearch.com/2021/05/12/image-gradients-with-opencv-sobel-and-scharr/

https://evergreenllc2020.medium.com/fundamentals-of-image-gradients-and-edge-detection-b093662ade1b

https://www.cs.cmu.edu/~16385/s17/Slides/4.0_Image_Gradients_and_Gradient_Filtering.pdf

Interesting enough the one concept of an image that does not have a precise definition is the one most often spoken about - sharpness. Sharpness is a subjective concept. Micro contrast as measured directly by the image gradient or indirectly by the MTF is a real quantifiable concept.