What is a smart camera? Or perhaps a better way to phrase this is, what makes a smart camera “smart?”
Used ubiquitously in marketing copy for products in home automation, phones, TVs, cars, and even cities, the word “smart” has become an identifiable if not common buzzword. In each case, the word “smart” means something slightly different. If used in the traditional sense, I might say, “I am grateful that our audience of engineers and integrators—all of whom are smarter than I—are willing to read my editorials.” But I digress.
In the machine vision sector, smart cameras typically combine lighting, image sensors, software, and I/O to address industrial inspection and process control applications. Like a few terms within the market that can sometimes lead to confusion (“Embedded vision” comes to mind), the definition of a smart camera might mean slightly different things to different people. This is especially true right now as new types of cameras have emerged as of late that offer capabilities and features like the “traditional” smart camera, but that come in different packages, specifications, and form factors.
For example, most smart cameras in the past came equipped with image sensors ranging in the VGA to 2 MPixel range, for the most part. Today, there are numerous smart cameras on the market with image sensors well past this benchmark and into the 5, 12, and even 20 MPixel range. All of these cameras with large image sensors serve the new and expanding needs of customers deploying machine vision systems.
Similarly, a few companies have introduced new smart cameras that may identify more as an “embedded vision” camera than a smart camera, yet are able to tackle similar tasks. In some cases these cameras can accomplish new imaging tasks that traditional smart cameras cannot, such as deep learning and artificial intelligence-related tasks.
In our September 2019 issue, our product focus article on smart cameras highlights some of the latest advancements and available products in the world of smart cameras, including larger image sensors and embedded smart cameras that target several different types of machine vision applications, from quality inspection to deep learning to barcode reading.
Elsewhere in the issue we have a number of articles covering novel machine vision and imaging applications including a computer vision smartphone application that creates personalized skin care routine suggestions, an update on the official adoption of the OPC Machine Vision specification, a look at novel machine vision illumination methods, and a machine vision system that measures paint coating thickness on steel products.
If there are certain types of applications or technologies you’d like to see covered, or covered more often, we’d like to hear from you. Please feel free to reach out at any time with comments, suggestions, or ideas. But please only send the “smart” ones, of course.
Thanks for reading, and hope to hear from you soon.
