The concept of computational imaging may be new to many, but the value of this already somewhat mature technology is far-reaching. Applications involving multi-image and multi-light-imaging input with computed output images based on algorithms have expanded well beyond the original realm of R&D. This technology has found its way into industrial automated inspection, where the creative use of illumination components has emerged as an enabling technology providing valuable imaging capabilities.
Broadly stated, the term “computational imaging” may be applied to varying techniques that employ algorithms to create a single image from more than a single optical acquisition. Research on computational imaging involves lens-less, single-pixel, and even flat cameras. Machine vision implementations of computational imaging are much more mature, and components and software for specific application areas easier to use than ever. In two common use cases, illumination devices and controls contribute to key imaging capabilities. These use cases are photometric stereo (or shape-from-shading) imaging, used to highlight geometric features, and high-resolution color imaging.