Scribd
Upload
47 views2 pages

2.3.2.2 Embedded Smart Cameras

An embedded smart camera is a camera embedded inside another device like a mobile phone. The camera's artificial system intelligence processing (ASIP) functionality is enabled by a dedicated processor or the device's main processor. Embedded smart cameras are commonly used in mobile devices to enable new applications. Stand-alone smart cameras have dedicated embedded processors and algorithms to provide ASIP functionality, and look like typical security or industrial cameras. Compact-system smart cameras connect an external but dedicated image processing unit to the camera via cable, with the camera performing some preprocessing and the unit providing remaining ASIP functionality. Distributed smart cameras involve jointly processing images from networked cameras to achieve ASIP goals like collaborative video analysis across multiple perspectives.

Uploaded by

hanu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
47 views2 pages

2.3.2.2 Embedded Smart Cameras

An embedded smart camera is a camera embedded inside another device like a mobile phone. The camera's artificial system intelligence processing (ASIP) functionality is enabled by a dedicated processor or the device's main processor. Embedded smart cameras are commonly used in mobile devices to enable new applications. Stand-alone smart cameras have dedicated embedded processors and algorithms to provide ASIP functionality, and look like typical security or industrial cameras. Compact-system smart cameras connect an external but dedicated image processing unit to the camera via cable, with the camera performing some preprocessing and the unit providing remaining ASIP functionality. Distributed smart cameras involve jointly processing images from networked cameras to achieve ASIP goals like collaborative video analysis across multiple perspectives.

Uploaded by

hanu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

2.3.2.

2 Embedded Smart Cameras


An embedded smart camera is a camera embedded inside another
device, for example,a mobile phone. Sometimes the camera is
completely hidden inside anotherdevice and people may be unaware of
its existence. Examples are vision-based optical mice, vision-based
fingerprint readers, cameras used in some robotic and automobile
applications. Some mobile phones come with a camera that can read and
recognize barcodes or other similar codes of a product or a company and
then direct users to the web pages of the product or company, which are
displayed on the phone screen. In this kind of smart camera, the ASIP
functionality is made possible by either a dedicated processor or the
processor of the device in which the camera is embedded. Embedded
smart cameras are often enablers for new applications and novel
products, especially in mobile devices [290].
2.3.2.3 Stand-Alone Smart Cameras
Stand-alone smart cameras are probably the most natural and common
smart cameras, especially in industrial machine vision where smart
cameras have reached a certain level of maturity and established a solid
market presence. They look like normal cameras, much like CCTV
cameras or general-purpose industry cameras. The ASIP functionality is
provided by dedicated embedded processors and intelligent algorithms
running on the processors. Many cameras also run a real-time operating
system, to simplify the programming of the cameras and improve the
user interface.An example of stand-alone smart camera is shown in Fig.
2.2.2 Smart Cameras: Fundamentals and Classification 33
2.3.2.4 Compact-System Smart Cameras
Typically, a compact-system smart camera is a normal camera connected
through dedicated cables or communication interfaces to a separate,
external but dedicated image processing unit nearby. The camera
performs image capture and sometimes some ASIP functionality, such as
pre-processing to reduce the amount of data, or feature extraction. The
remaining system ASIP is performed by the external unit. The advantage
of this type of smart camera is that the camera can be standardand is
usually not expensive and easy to be replaced or upgraded. The external

processing unit can afford more processing power due to availability of


more memory,storage, and other resources. There are many examples of
this type of smart camera in video surveillance applications for either
security or traffic flow analysis.In industrial machine vision, this type of
smart camera is called compact vision systems (CVS). Examples of the
CVS are the NI CVS-145x camera systems by National Instruments.4
Some so-called compact-system smart cameras are in fact PC-hosted
smart cameras in that the external processing unit is usually a dedicated
image processing card or cards plugged into the internal extension slots
of a PC or of a computer. This type of smart camera seems to be popular
within academic circles. In fact, the often-cited smart camera project by
W. Wolf et al. at the Princeton University [571] was a PC-hosted smart
camera, designed and developed for real-time gesture recognition for
humancomputer interface and surveillance applications.Their camera
system consisted of a couple of cameras, each connected to a Philips
TriMedia video processor card inserted into the PCI slot of a host
computer.PC-hosted smart cameras offer a high level of flexibility and
better user interfaces.
2.3.2.5 Distributed Smart Cameras
Thanks to the progress in networking, sensor networks, and wireless
communication technologies, distributed or networked smart cameras
have recently attracted significant interest from both academia and
industry. The participating cameras may or may not have overlapping
fields of view but the images captured by these cameras are often
processed jointly to achieve planned ASIP functionality [448].
Sometimes this network of distributed smart cameras can be thought of
as a single virtual smart camera, especially when video analysis or ASIP
is performed collaboratively across the cameras in a distributed fashion.
In these cases, the camera network can achieve better performance than
each camera independently together. This distributed vision system
presents a novel and powerful computing platform that holds promise of
solving many tough problems encountered with single smart cameras.
For example, carefully deployed multiple cameras can help to resolve
problems such as occlusion of view, both static and dynamic [448],
depth information about foreground objects,object tracking, and the

You might also like