Deep capture, machine vision software in deep learning allows the quality identification of wood planks in planer output.

The system ensures the detection and classification of several anomalies (black knots, cracks, lack of materials, etc.).

As in other machine vision applications, faults are subject to interpretation. Our deep learning system allows to differentiate the results according to the faults location on the board (acceptable faults on the edges of the boards).

The acquisition operates in real time at a scrolling rate of 30cm / sec.