Faster programing language can speed imaging processing

A new programming language for image-processing algorithms yields code that runs much faster, reports the Massachusetts Institute of Technology — and this could lead to much better in-camera performance in dedicated devices and smart phones.

Researchers at the Computer Science and Artificial Intelligence Laboratory developed a new programming language called Halide: using Halide to rewrite several common image-processing algorithms, the Halide versions were typically about one-third as long but offered significant performance gains — two-, three-, or even six-fold speedups, MIT reports.

Halide programs are also simpler to write than ordinary image-processing programs, and the code is easy to modify.

More information is here.