
AI kit based uses Sparse Modelling for easy training
Sparse Modelling requires much less training data than traditional AI techniques to make accurate predictions. It can begin with 50 or fewer images compared to the 1,000 or more images required for traditional AI. Hacarus’ Sparse Modelling Kit can be used stand-alone or as an add-on to existing inspection systems.
The new embedded AI kit combines congatec hardware and Hacarus software. It can quickly be deployed and tested in any GigE and USB 3.x environment. In a COM form-factor, the system measures only 173 x 88 x 21.7 mm. It uses the latest Intel Atom and Celeron processors ( Apollo Lake) that are all available for series production today. The system features a rich set of I/Os, for flexibility, including 2 x GbE application ready for GigE Vison, one USB3.0/2.0, four USB2.0 and 1 x UART (RS-232). The kit is extendable using the Mini-PCIe with USIM socket, the mSATA socket and 16-bit programmable GPIO.
More information
Related news
Arrow and Basler partner for industrial embedded vision solutions
IP67 protected embedded vision
Smartek Vision will become FRAMOS Embedded Engineering
Chronocam: Event-based sensing enables machine vision solutions
