Imaging a static dartboard using 2 cameras

I need a c++ library capable of real-time imaging of a standard dartboard, creating a pixel by pixel mapping of each scoring area, 1 for each of 2 cameras. This library will return the correct scoring area identifier (ie: double 20, single 1, or miss) based on a single pixel (x,y) and which camera provided it, passed in from external code. This library must correctly identify scoring areas on any standard dartboard with 99% accuracy and in under 100ms. The library must then relinquish control of the cameras to external code until the next iteration is called. You will be required to work with our developers on the interface and testing as you will not have access to the hardware directly. Fluent English will be required for this communication.