Deep learning machine vision software for perfume bottles analysis
Incised inspection and detection machine on the bottles bottoms

Industrial vision in deep learning for glass and other packaging

Our machine vision inspection machine allows the detection and sorting of incisions, broths and calcines on the bottoms of glass perfume bottles or decanters of spirits.

Operating thanks to our Deep Capture software developed in deep learning, the means allows to detect faults that no conventional machine visionsystem was previously able to see.

Our machine, working with 8 cameras streaming video at 25FPS each, is capable of working at speeds on over 120 bottles / hour. The bottles are handled by a conveyor system with adjustable wedging. The bottles ejection is managed by blowing on an intermediate ramp

Our developers teams, specializing in deep learning, take care of the various development phases, from labeling photos, configuring the neural network to on-site integration of the system or machine.

Consumer PC

INTEL CORE I9 7940X
32GB RAM
2x GEFORCE RTX 2080TI-11G
WINDOWS 10

SVS VISTEK camera

Configuration of the deep learning model compatible with Google’s open source platform: TensorFlow.

Results detection and communication on Deep Capture software.

Taking photos and labeling on deep vision software