SVS-Vistek CameraLink svs16000, Руководство пользователя

Страница: 65 / 97

Users Manual Area Scan Cameras SVSGigE Page 65
Users Manual –svs11002 and svs16000
© SVS-VISTEK GmbH
Once an application is done with a camera it needs to close the connection and thus
allowing the camera container to release all resources that have been kept occupied in
meantime for the active application-camera relation.
Camera_closeConnection(Camera);
After a camera connection has been closed it can be re-opened again at any time as
long as the program is a valid client to the camera container.
Opening a camera requires the existence of a valid IP address in the camera. The
easiest way for making sure connectivity is to force valid network settings:
unsigned int IPAddress;
unsigned int SubnetMask;
SVGigERet = Camera_forceValidNetworkSettings(Camera, &IPAddress, &SubnetMask);
That function will do nothing in case the camera has already got a valid IP address. If
this is not the case the function will first force a free IP address in current subnet into
the camera. Subsequently that IP address may be assigned as a fixed IP address to
the camera if this is wanted:
Camera_setIPAddress(hCamera,IPAddress,NetMask);
When the available cameras in the attached network segment have changed it will be
necessary to run the
discovery
command again in order to synchronize the information
in the camera container with the actually available set of cameras in the network.
Afterwards one has to take care since a previous index of a specific camera might
have changed in meantime.
Alternatively to selecting a camera by index a search string can be provided to a
findCamera
function:
Camera = CameraContainer_findCamera(CameraClient, CameraItem);
In this case a string
CameraItem
will be provide to a camera container function which
tries to find a match between all discovered cameras where the given string is
contained in at least one of the MAC address, IP address, serial number and user
defined name items.
16.11.3 Identifying a camera
Every GigE camera maintains a set of information strings which allow for identifying a
camera in the network.
CString CameraName = Camera_getMacAddress(Camera);
CString CameraName = Camera_getIPAddress(Camera);
CString ModelName = Camera_getModelName(Camera);
CString Manufacturer = Camera_getManufacturerName(Camera);
CString SerialNumber = Camera_getSerialNumber(Camera);
CString DeviceVersion = Camera_getDeviceVersion(Camera);
CString CameraName
= Camera_getUserDefinedName(Camera);
An application could query those strings and could display one or a combination of
those information strings to a user for allowing a selection of a specific camera.
However, applications that know beforehand which camera to use for a given purpose
can use the
findCamera
function that was described before.
16.11.4 Determining maximal packet size
From network history of last decades, a packet size upto 1500 bytes is quite common
in most network infrastructures. This allows to transfer even huge files splitted into
small packets in an efficient way. At the same time it becomes obvious that the
overhead introduced by those relatively small packet sizes may limit the maximal
performance.
A way out of the performance bottleneck is the usage of bigger packet sizes, in
particular for fast streaming devices like digital cameras. Therefore current network
cards allow for switching “Jumbo frames” on with upto 16 Kbytes packet size. The
overall maximal usable network packet size, however, depends on all network system
components and will be equal to the lowest value that is supported by any network
card or switch on the network path.