In the current times, Embedded Vision is one of the most exciting fields in Technology. Equipping machines with the ability to see, sense and immediately respond to the world, this technology has created tremendous opportunities across the manufacturing space. Integrated with disparate sub-systems such as video and vision I/O with multiple image processing pipelines, embedded-vision systems can perform complex vision-based tasks in real-time in a resource constraint embedded environment.
Building Vision Systems to Enable Factory Automation
A slight deviation in part’s position / orientation may cause assembly failures in manufacturing setup. That’s why automated systems in manufacturing work in a very fast and accurate manner. As humans with their vision capability and brain understand can navigate through the process thoroughly, Vision-based Industrial Automation systems are capable of delivering higher quality with lesser errors. Such systems are backed by cameras, vision processors, and software algorithms. They adapt to evolving manufacturing line circumstances and extend the benefits to other areas of the supply chain, e.g., piece parts and finished goods inventory tracking.
Conventionally, such Vision-based technology is found in complex and expensive systems. Nowadays, however, as Technology advances, Cost, Performance, and power-intensive computing has become available in lower price ranges, reinforcing the proliferation of vision into several mainstream automated manufacturing platforms.
In this post, we’re going to discuss the primary embedded-vision products used in factory automation and some application areas for industrial automation.
Smart Sensors – A single unit designed to perform a single machine-vision related task. It doesn’t require any configuration and has limited on-board processing. Oftentimes, a lens and lighting are integrated into the unit.
Smart Cameras – Smart Cams comprise of a machine-vision camera, a processor and I/O in a very compact enclosure. Such cameras are configurable with built-in LED and a facility to change the lens.
Compact Vision Systems – Known as a complete Vision System, it incorporates one or two cameras and a processor module. One of the more distinguished features of compact vision systems is that they have smart cameras to enable information gathering from different cams in the group. This comes across as a cost-effective solution when an application requires multiple images from different locations / angles.
Machine Vision Cameras – These devices convert an optical image into an analog/digital signal. These signals may be stored in the RAM, however, may not be processed within the device.
Frame Grabber – A type of PCB card for interfacing the video output from a camera with a system. Frame Grabbers are also known as video capturing boards/cards. Their functionality varies from bringing up a simple interface to a more complex one that can handle a variety of functions like triggering, exposure rates, shutter speeds, and complex signal processing.
Machine Vision Software – All software that is sold as a final product, specifically crafted for machine-vision applications. These generally consist of library and system software.
Application Areas in the Manufacturing Space
Inventory Handling & Tracking – Embedded-Vision systems can improve product tracking in production lines with an enhanced storage efficiency. On the other hand, bar codes and radio-frequency identification tags can also help track and route materials. As image sensor technologies continue to advance and other vision processing components become increasingly integrated, product tracking and handling will lay the foundation for the next generation of inventory management systems. Though high-resolution cameras can provide detailed images of work material and inventory tags, real-time software is required to analyze the images, identify objects within them, recognize ID tags associated with these objects and to perform quality checks. Vision Systems, hence, devised for inventory tracking and management can deliver robust capabilities without exceeding acceptable infrastructure costs by integrating multiple real-time video analytics extracted from a single video stream for a combined, intelligent output
Automated Assembly Lines – When it comes to factory production floors, embedded vision works in raw material handling. Cameras can be used in acquiring images of parts and destinations. Subsequent vision processing algorithms send data to a robot, which further enable the system to perform functions, e.g., picking up and placing a component. Manufacturing units can also use Vision Systems in a variety of applications which require high-precision assembly, with cameras to click the image of the components after they are picked up. With vision systems, picking parts from a bin has also become easier. Some latest vision processors can now easily handle immense datasets by working on complex algorithms, developed to extract depth information and rapidly make decisions.
Automated Inspection Process – Vision systems can detect, predict and prevent gridlock and other unwanted outcomes. In addition to this, vision-based software performs tasks such as checking for the presence of components, reading text and bar codes, checking orientation, measuring dimensions and locating defects and patterns. Conventionally, quality assurance was performed by randomly selecting samples from the production line. Then, using statistical analysis, the results can be extrapolated for the larger manufacturing run. In the robotic era, automated inspection can deliver 100 percent quality assurance results, by inspecting each and every part going through the product / assembly lines. Interestingly, with recent advancements in vision processing performance, automated inspection is no longer a hindrance in the throughput value of manufacturing.
Indubitably, embedded vision opens up limitless opportunities with the potential to bring transformational changes. In the manufacturing space, capable vision systems, equipped with depth-discerning sensors and software algorithms, are bringing all forms of autonomy and productivity, making machines smarter & more intelligent. Such systems are bringing long-standing autonomous and adaptive industrial automation ideas to life. Most of the leading companies have already adopted this ground-breaking technology to have a custom system in place that perfectly suits their application(s) and provides return on investment.
So, how ready are you to jump on to the factory automation bandwagon, to harness the significant benefits of improving quality, precision, and throughput of your production line? At KritiKal, we build smart products using Embedded Vision technology to enable automation at the heart of the manufacturing units. We’d love to hear about your manufacturing hurdles, to work together and eliminate them using smart automation.
by Taniya Arya