By streamlining creation of AI and machine learning models for edge hardware, Edge Impulse is working to enable devices to make decisions and provide insight where data is gathered.

SAN JOSE, Calif.—Demand for edge-capable AI software as a path to innovating factories and production lines has increased, according to Edge Impulse, with on-device computing enabling faster access to critical data insights, low latency, and more robust security and privacy compliance.

Edge Impulse, developer of a platform for building, refining, and deploying machine learning models to edge devices, recently introduced a new technology for unlocking visual anomaly detection on any edge device, the company said in a release. This includes devices ranging from NVIDIA GPUs to Arm MCUs, through the first model architecture of its kind: FOMO-AD (Faster Objects, More Objects – Anomaly Detection).

Visual anomaly detection, in particular, is an important use case for industrial AI. However, it is not widely used as it requires creating a library of known anomalous samples to train the model to spot deviations in industrial environments. Because companies cannot collect real-world samples for every anomaly, especially for unanticipated defects, this limits detection capabilities, the company said.

Edge Impulse’s FOMO-AD architecture, two years in development, reportedly offers the first widely accessible platform for visual anomaly detection on any edge device, from GPUs to MCUs. It is said to be the first scalable system capable of training models on an optimal state to detect and catalog anything outside that baseline as an anomaly in video and image data. This dramatically increases the productivity of visual inspection systems that will no longer have to be manually trained on anomalous samples before they can start generating real-time insights on-device, according to Edge Impulse.

“Virtually every industrial customer that wants to deploy computer vision really needs to know when something out of the ordinary happens,” said Jan Jongboom, co-founder and CTO at Edge Impulse, in the release. “Traditionally that’s been challenging with machine learning, as classification algorithms need examples of every potential fault state. FOMO-AD uniquely allows customers to build machine learning models by only providing ‘normal’ data.”

Most industrial camera systems capable of computer vision are powered by GPUs and CPUs, with a high installation cost that requires wiring and a power-hungry connection to mains electricity. Recent advancements from top-of-the-line silicon manufacturers, as well as novel edge model architectures, enable computer vision AI models to operate in either high- or low-power systems, giving businesses more choice. The benefits of low-power systems include the possibility of building battery-powered visual inspection systems, and lower production costs from using cost-effective hardware that can reduce the overall product form factor, the company said.

Edge Impulse said that in recent months, it has been testing FOMO-AD with customers, achieving proven results in industrial environments when proactively detecting irregularities in multiple production scenarios. Use of FOMO-AD is said to have led to marked improvements in machine performance and production line efficiencies for customers.

According to Edge Impulse, numerous use cases for visual anomaly detection exist across various manufacturing sectors, including general industrial, electronics, automotive, and medical. They are reported to include production line inspection, quality control monitoring, defect detection (industrial); integrated circuit inspection, PCB defect detection, and soldering inspection (silicon); part assembly quality control, crack detection, leak detection, EV battery inspection, painting, and surface defect detection (automotive); and medical device inspection, pill inspection, vial contamination inspection, and seal inspection.