Machine vision already makes a crucial contribution to the manufacturing sector, primarily by providing automated inspection capabilities as a part of QC procedures. However, the planet of automation is becoming increasingly complex. Industry 4.0, the web of Things (IoT), cloud computing, AI, machine learning, and lots of other technologies present users and developers of vision systems with big challenges within the selection of the perfect system for his or her respective applications.

An Enabling Technology

With rapid developments in many various areas including imaging techniques; CMOS sensors; embedded vision; machine and deep learning; robot interfaces; data transmission standards and image processing capabilities, vision technology can benefit the manufacturing industry at many various levels. New imaging techniques have provided new application opportunities. For instance, hyperspectral imaging can provide information about the chemical composition of the materials being imaged. Computational imaging allows a series of images to be combined in several ways to reveal details that can’t be seen using conventional imaging techniques. Polarisation imaging can display stress patterns in materials. Other developments in machine vision technology cause enhanced performance, integration, and automation in the manufacturing industry. The degree of integration can range from manual assembly assistance through to finish integration into OEM equipment and on to the demanding requirements of Industry 4.0.

"Critically, Industry 4.0 requires a standard communication protocol for all sensor types so as to permit data transfer and sharing"

Aiding Manual Assembly

There are still huge numbers of products that are assembled manually and a ‘human assist’ camera is often wont to help to stop errors in such operations. The operator follows a group of assembly instructions loaded into the camera and displayed on a monitor. After every action, the system compares the result to the right stored image to make sure that it's been administered correctly and completely before the operator can advance to subsequent steps. If an action is incomplete or if an error is formed, it's showed the operator in order that it are often corrected. Each step completed is often verified and recorded to supply data which will be used for assembly work analysis and traceability.

Adding Vision to the assembly Line

Using vision inspection on a producing or packaging line may be a well-established practice. Systems range from single-point self-contained smart cameras that perform an inspection task and deliver a pass/fail result to the system, to PC-based systems which will feature multiple cameras and/or multiple inspection stations. Vision systems are often retrofitted to existing lines or designed into new ones. Vision inspection also can be utilized in conjunction with statistical process control methods to not only check critical measurements but to research trends in these measurements. during this way, interventions are often made to regulate the method before any out-of-tolerance product is produced. this is often probably the closest forerunner to the wants of Industry 4.0.

Vision-guided Robots

Industrial robots are already used extensively and with the emergence of collaborative robots and rapid developments in 3D image processing, they're getting used far more together, particularly for vision-guided robotics. The vision system identifies the precise location of the thing and these coordinates are transferred to the robot. Massive strides in vision-robot interfaces make this process much easier. one of the foremost popular uses for 3D robotic vision is in pick and place applications.

Embedded Vision

The availability of small, embedded processing boards, usually supported ARM architecture, offers great potential for the event of vision systems embedded into other equipment and manufacturing processes. Many of the leading image processing libraries and toolkits can now be ported to those platforms, offering a wider range of vision solutions during this format. Combining these processing capabilities with low-cost cameras, including board-level cameras, means vision systems might be incorporated into a good sort of products and processes with comparatively small cost overheads.

Machine and Deep Learning

There has been tons of hype about deep learning in machine vision, which uses convolutional neural networks (CNNs) to hold out classification tasks by identifying characteristics learned from a group of coaching images. However, the challenge remains that in industrial applications the amount of obtainable training images is restricted while the tools, training time and processor resources remain high. Other machine learning approaches are rapidly becoming recognized as a less expensive and simpler to implement an alternative to deep learning for industrial applications. this is often likely to seek out traction for high-performance, flexible vertical solutions that will even run on inexpensive embedded systems, making extremely cost-effective systems possible.

Onwards to Industry 4.0

The essence of the smart factory of the longer term is to optimize the method using big data analytics supported the feedback from many various sorts of sensors that are monitoring the method. These, of course, will include simple and smart vision sensors also as more sophisticated vision subsystems or systems. Critically, Industry 4.0 requires a standard communication protocol for all sensor types so as to permit data transfer and sharing. One standard which is proving popular during this area is that the OPC UA platform-independent, open standard for machine-to-machine communications. Recently the VDMA (the engineering Industry Association in Germany) has announced OPC UA Companion Specifications for Robotics and Machine Vision which can provide compatibility with this standard for robots and vision systems respectively. The building blocks are starting to close.