METHOD OF ITERATIVE COMBINATION OF TV SIGNALS BASED ON LINEARIZATION FOR MACHINE VISION SYSTEMS

Abstract

The article describes a method for combining television signals for machine vision systems based on linearization. The relevance of the developed method involves reducing the error when combining signals in the presence of interference arising from changes in illumination, as well as ensuring high processing speed. It is shown that the idea of combination replaces the processing signals of trigonometric functions in the model with a Taylor series, and in adding two variables to the model – additive and multiplicative components. All matching parameters are evaluated by solving a system of linear equations, which is determined by decomposing the matched signal into a Taylor series. An experiment demonstrated the correct combination of television signals using the proposed method, and a comparison was made with the exhaustive search method in terms of measurement error and processing speed. It is noted that the work will be useful for developers of machine vision measurement systems with real time processing.