19MANUFACTURING TECHNOLOGY INSIGHTSFEBRUARY - 2023CXO INSIGHTSThe promise of machine vision has really taken off in the 10 years that it. From self-driving cars all the way to facial recognition doorbells ­ the applications have really captured the imagination of the public. There was a massive amount of work on the embedded infrastructure needed to get these solutions right. It is exciting for me, as a developer of products, to see the industry continuously learn, and evolve in a scalable way as the demand for machine vision has grown. I would like to share three trends: Levels of recognition. Often, we will get requests to "recognize" an object or people. Recognition has a wide span of meanings. There is the deep learning/machine learning level of recognition driven by the real-time needs of self-driving, facial-based identity, and instant awareness of a large number of objects. Figure 1's upper right quadrant represents this well and is an area that many are calling ubiquitous in commercial and industrial applications.But in most cases, we have found that lower levels of recognition are good enough for the application rest. There is feature-level recognition (see Figure 2); which is looking for the existence of features in an image. And there is also basic object detection. In both cases, there has been a growth in interesting options for developers. Referring to Figure 1 again, these interesting options are represented by the upper left and lower right quadrants. · The upper left quadrant represents the sort of new options available using powerful edge compute devices but with lower-end optics and sensors. The optical scanner used for fingerprint identification on smartphones is an example. Accurate results can be taken as low as 500dpi optical sensor.· The lower right quadrant represents the options where a more embedded autonomous compute is paired with higher-THE FUTURE OF MACHINE VISION AND HOW TO BE READY FOR THAT FUTUREBy Paul Chen, Sr. Director, Electronic Design, MattelPaul Chen
< Page 9 | Page 11 >