GPU-based image processing

Accelerated processes due to GPU-based image processing

New EnsensoSDK release

Users of the Ensenso X series with high-resolution 5 MP cameras benefit from an improved quality of the 3D point cloud, but also have to be prepared for larger amounts of data and therefore longer processing times. The new EnsensoSDK 2.2 offers a solution that guarantees very fast processes and high clock rates – and is also very interesting for Ensenso X systems with 1.3 MP cameras.

So far, 3D data processing was CPU based. With the current EnsensoSDK version, stereo matching can now be supported by a graphics card. This feature can specifically be used with Nvidia GPUs. Depending on the parameterization of the system, the additional computing capacity of the GPU accelerates the processing by about five times. GPU-based stereo matching and the resulting speed advantages can be used with all Ensenso cameras.

GPU-based image processing

The new SDK release also offers other useful features. For example, the NxProfiler GUI tool visualizes timelines of NxLib actions and user debug blocks which enables developers to evaluate and optimize processes. In addition, the NxLib can now create so-called “Ensenso FileCameras" from saved original data which users then can use to process data offline.