Manufacturing – “Smart Factory”
The whole manufacturing industry is currently overthrowing traditional production, quality control and surveillance practices.
In such environments the DeepEyes algorithm on standard cameras (as a data capturing device, sensor) makes working environments smarter, more precise and more secure. Traditional control centres where specialists watch computer screens around the clock so that they can intervene or inform if there is a disruption will be a thing of the past. DeepEyes will alert any defined anomaly in real time to the relevant stakeholder, on any device.
For customer specific requirements the DeepEyes core algorithm is always fine tuned to clearly defined recognition goals.
Detect and alert when individuals enter a restricted or hazardous area.
Make sure that individuals wear security apparel like helmets, goggles, caps etc.
Monitor individuals performing necessary activities, gestures, head movements etc. when operating a machine.
DeepEyes traces activities or processes in order to ensure that quality standards are met.
For example, DeepEyes recognizes if
manual adjustments in the production process are executed in the correct manner.
necessary hand movement during maintenance are in place
activities important for worker safety are executed
have parts been correctly assembled?
have the correct variants of parts been assembled?
are all parts in place / in the right position?
Quality checks become more effective with DeepEyes and can replace maintenance-prone special cameras or sensors which often do not return results in real time. As an example, with two 1080p cameras DeepEyes detects anomalies in objects of up to 0.01 mm in 3D at 1 m distance, even if they are moving on an assembly line.
Depending on specific requirements DeepEyes will be customized to meet context-aware requirements and can alert appliances and systems of all kinds to dynamically adapt their behavior at run time.
Data Security / Privacy
DeepEyes only reads and analyzes specific sets of pixels which are relevant for recognizing clearly defined anomalies, outliers or deviations. We do not store any customer data nor do we use biometrical unique identifiers. They are not relevant for our analysis.
DeepEyes operates autonomously on device and does not need the cloud.