Smart Cameras – What are they?

Smart cameras are becoming more prevalent. For those who don’t know what they are Evaluation Engineering provides a good overview. Essentially, they are stand alone, self-contained (not connected to a PC) imaging systems, which generally use CCD or CMOS technology for sensing and have built-in connectivity such as Ethernet.

Vision sensors is another term for smart cameras and sometimes refers to low-end smart cameras in that they can take the place of other kinds of sensors – like bar code readers.

Customers are interested in smart cameras because they provide:
Lower cost by replacing complete PC-based imaging systems
Greater reliability
Distributed processing rather than central processing
Adaptation to a diverse set of sensors (light, optical, etc).

Example applications include 2D/3D, bar code, gauging, and pattern recognition. Some real-world industry examples include Nissan putting smart cameras into their cars to determine if an airbag should be deployed based on the car’s position relative to oncoming objects.

Johns Hopkins uses smart cameras for video surveillance in response to the death of two students. The cameras allow the university to monitor the campus without putting a guard on every corner.

One drawback to some smart cameras is that they can run slower. The processor is weighed down with additional algorithms and tasks which limits the performance of the device. Another potential drawback is lowered flexibility. As the camera becomes “smart” it becomes more dedicated to a specific set of tasks and is not as flexible as it once was.

If you are working with Smart Cameras, I would like to hear from you. Please email me at hall.martin@ni.com.

Best regards,
Hall T. Martin