Machine Vision

According to the Automated Imaging Association (AIA) , machine vision (MV) encompasses all industrial and non-industrial applications in which a combination of hardware and software provides operational guidance to devices in the execution of their functions based on the capture and processing of images. MV systems rely on digital sensors protected inside industrial cameras with specialized optics to acquire images so that computer hardware and software can process, analyze, and measure various characteristics for decision making.

While industrial MV uses many of the same algorithms and approaches as academic, governmental, and military applications, the constraints differ. These systems demand greater robustness, reliability, and stability compared to academic MV systems, and typically cost much less than those used in governmental or military applications. The ideal industrial MV solution supports low cost, low accuracy, high bandwidth, and high reliability.

Challenges on the Edge

As edge computing becomes more prevalent, IIoT enterprises utilizing MV would likely require a larger number of network endpoints (e.g., MV sensors) outside of DMZs, VPNs, firewalls, and other legacy technologies built to secure business applications in the cloud. While digital transformation produces increased connectivity and convergence between OT & IT environments, the attack surface is widened. Since legacy data network and security architectures are not equipped to handle the MV security needs at the edge of the network, enterprises need to address this gap with alternative technologies.

Xaptum’s Solution

The three most notable outcomes of using the ENF to secure MV applications are as follows:

Endpoint-Agnostic

As an edge-computing friendly, overlay security solution, the ENF can meet bandwidth and latency constraints for MV implementations. Xaptum’s solution offers the flexibility of incorporating best-suited ancillary components – hardware, software, and communication interfaces can be commercial-off-the-shelf (COTS), modular, and usable across various use cases. To learn more about the solution’s interoperability, please refer to FAQs on the topic.

Scalable Security on a Flexible Network

Customers can collate vision traffic from dispersed deployments securely on Xaptum’s transaction layer, given its Traffic Obfuscation and Anonymity to Advanced Threats . The ENF uses IPv6 addresses as persistent identities for MV assets such as specialized IR sensors in providing them global IP mobility. It not only allows the devices to switch safely between last-mile networks but also facilitates a seamless handoff with session persistence.

Over-the-Air Programming

The customer can leverage context-aware routing between Virtual Edge LANs, where sensors/cameras are deployed, and hybrid cloud environments where data is analyzed in real time. Since today’s machine learning relies on continuous learning, there is a need for model updates to be sent frequently over-the-air. The ENF provides a secure, user-friendly platform for securely programming MV assets OTA.