IndustryTap has reported on neuromorphic vision sensors that help machines see better, modeling and designing vision technology based on the dragonfly eye, and systems that help robots see around corners.
Machine Vision (MV) is anything but simple with lenses, illumination, cameras, frame grabbers, and bus systems along with data structures, image enhancements and transformations, segmentation, feature extraction, morphology, template matching, and more.
Complete Machine Vision Development
Some companies offer comprehensive software solutions including integrated development environments (IDEs). MVTec Software GmbH’s HALCON, for example, helps engineers design, build, and operate complete MV systems. A typical machine vision system is illustrated below.
The world of MV is fascinating for both newcomers and experienced engineers with increasing availability of educational programs worldwide. Microsoft, for example, is now offering open-source machine learning,
Machine Vision in Automotive Manufacturing & Beyond
The key to creating MV systems that help in automobile manufacturing, for example, is cost effectively integrating machine vision with programmable logic controllers (PLCs). This requires sophisticated machine vision algorithms and applications.
Moving beyond industrial applications, Google Android recently teamed with Movidius to integrate real-time facial recognition into smartphones. And new narrow chips combined with MV could dramatically improve the capabilities of these devices and robots as well.
The following is a short introduction to machine vision by Microscan.
The following video by VDMAonline defines what a successful machine vision project is.