1) I am trying to use GrFingerXClass in Visual Studio 2005, but this message is showed: “Failed to import the ActiveX control. Please ensure it is properly registered”
SOLUTION: Do the following:
- Close the project.
- Copy the AxInterop.GrFingerXLib.dll from a VS.2003 project (our VS.2003 sample, for example) to the "obj/debug" folder of your project.
- Open your project.
- Try to open the library again.
IMPORTANT: this problem doesn't happen in VS.2003 projects.
2) My GrFinger is not working in .NET
SOLUTION: Check if the Stdole.dll (file which is necessary for GrFinger) is in your PC. This file comes with .NET 1.0, but doesn't come with 1.1 or 1.2 versions.
3) I'm having a -107 error when I try to access BiometricDisplay method.
SOLUTION: The last parameter of that function is "matchContext". If you haven't processed an identification, then this parameter must be GR_NO_CONTEXT. Only use a valid context number, such as GR_DEFAULT_CONTEXT, if you have already called Identify or Verify.
4) The identification is very slow. What can I do?
SOLUTION: To improve the identification speed check if you are doing the following:
- Use "Identify" instead of "Verify". "Identify" is much faster than "Verify" for searchings in database. Remove the "Verify" from your program.
- Check if the IdentifyPrepare is inside the loop. If is, put it out. You need to pass by IdentifyPrepare only once.
- Another possibility is an implementation of more than one server. Using (and distributing data to) two or more servers can avoid problems (for example, if one server has problems, there are other servers to replace it) and increase the matching speed.
- Clusters also are recommended to increase the performance.
- If your PC has a dual-core processor, try using contexts. Distributing the operations in multiple threads, the operations can be executed simultaneoulsy and the performance will increase even more.
Note that the identification speed is from 5000/s to 8000/s in normal conditions. High speed taxes are got in ideal conditions.
5) How to reduce the size of the template?
SOLUTION: By GrFinger, it is not possible to reduce the size of the template, unless you enroll medium or bad fingerprints. What you can do is use a compression algorithm. For more details, check in http://en.wikipedia.org/wiki/Data_compression