Effective Integration of Innovative Computer Vision Tools into Electronics Manufacturing Process

The success and accuracy of analysis and processing of raster images of electronic assemblies entering the machine vision algorithms and models within the electronics manufacturing process depend on parameters of these images. This necessitates the development of recording modules for computer vision systems using multifunctional cameras that allow not only capturing but also the maximum possible pre-processing of images. The paper analyzes the processing of raster images using machine vision methods and formulates the key requirements of computer vision algorithms and models for input images. A structural diagram of a computer vision system is presented. In accordance with this diagram, a recording module based on a 2-megapixel camera has been developed. Physical modeling of the module operation has been carried out using the example of processing an image of an electronic assembly. The proposed software and hardware version of the recording module solves the problem of effective implementation of innovative machine vision methods in electronics manufacturing process by pre-processing images on the camera side and their visualization in order to control the success of the system reconfiguration.