ABSTRACT
We present the development of a comprehensive model that was undertaken to determine the relationships between the components of an image and the light intensity values present in the image of the microvessels of translucent tissues such as the bulbar conjunctiva. Experiments were conducted during the modeling process by use of a cylindrical microvessel embedded in a diffuse medium (phantom) on a reflecting background to affirm model components and simulations. The three-dimensional model was reduced to a single illumination plane with four regions of interest and modeled as Lambertian radiators and surfaces. The modeling showed that the top of the cylinder and its immediate vicinity are diffuse reflectors of light from the source plus light reflected from the background. The limbus of the cylinder is a diffuse reflector of the source and background illumination and a specular reflector of background reflections that achieve a high grazing angle with the cylinder. The immediate vicinity of the cylinder receives direct illumination from the source, but the light is partially obscured by the cylinder. The region beyond the shadow of the cylinder is a diffuse reflector of the overhead light. The diffuse medium additionally reflects the source and also attenuates the illumination reaching the other compo- rents of the scene. The direct and reflected illumination at each region of the model was calculated by use of specific geometric relationships. To verify those calculations, we analyzed a video simulation for the effects of different illumination conditions and their contributing elements. Intensity values were calculated from the relative reflectivity data determined from the video signals. The illumination values at the points along the line at the meridian of the cylinder were due to its reflectivity and also that of the medium. Similarly, the values of points distant from the shadow of the cylinder were due to the reflectivity of the background and the medium. The excellent agreement between the model and the phantom provides a foundation for the detection and precise measurement of microvessel dimensions within a diffuse medium. The additional ability to compute relative depth, from a single view, also permits discrimination between neighboring microvessels in complex images.
