I am currently using a FLIR A50 camera with the CVB C# SDK. The camera is properly configured to output a Mono16 image, but when I inspect the image planes in my code, the data type indicates it is 8 bits per pixel (8bpp).
I am struggling to figure out why this is happening, especially since the CVB GenICam Browser clearly shows that the stream is running in Mono16 when I start the acquisition.
Could you help me understand what underlying process might be forcing the buffer to 8-bit, and how I can properly retrieve the 16-bit radiometric data in C#?
You are probably using the older acquisition stack (VIN-Stack) loading the GenICam.vin driver.
With this the default CVB Color Format for pixel format conversion is “Auto Detect” which means that a Mono16 Bit image is converted to 8Bit for display purposes.
In the configuration for your configured device, you can change the “CVB Color Format” from “Auto Detect” to “Mono 16Bit” to get the Mono16 image when loading the driver.
With the new acquisition stack (GenTL-Stack) using the ImageStream, you would receive an image directly with the Mono16 data without the need of changing the configuration.
For more information on how to implement the GenTL Stack Acquisition, take a look into the current documentation in the chapter “Image Acquisition”: