SDK for polarised-sensor machine vision modules
Sony announces details of industry’s first SDK for polarised-sensor machine vision modules
Significantly cuts time and cost of developing polarised applications using Sony’s XCG-CP510
Sony Europe’s Image Sensing Solutions has announced the industry’s first software development kit for camera modules based on Pregius IMX250MZR polarised image sensor.
Created to run on its new XCG-CP510 polarised module, the XPL-SDKW was previewed at Vision and is available immediately, enabling development teams to cut polarised machine vision application design time and cost significantly
The IMX250MZR sensor, the industry’s first to integrate polarisation functionality at the pixel level can filter light in 4 planes - 0o, 90o, 45o, and 135o - with pixels assigned a plane in a 2x2 calculation unit.
Sony’s Stephane Clauss said: “We’ve worked in close consultation with our customers to identify the key functions for the SDK, and develop its highly- optimised algorithms. And we will continue to add more over the next year.
“Depending on the dev team and application, a standard polarised-camera application would typically take between 6 to 24 months. Through the SDK, and its image processing library, the dev time for the same application can be cut to around 6-12 weeks.”
The use of polarised camera modules will prove vital in a wide range of applications and Sony has developed and validated software models for the major subset of these, including: weakness detection (eg for glass, PET and display manufacturing); inspection (eg for manufacturing / ITS / pharmaceuticals); scratch identification.
The set of functions available with the SDK have been developed to run on a standard PC*.
Support functions are the first level supplied in the SDK, which includes for example demosaic and raw extraction.
The ‘Cosine fit’ allows the developer to define a virtual polarizer angle for the whole image and the ‘Average’ function creates a non-polarized image from the raw data to simultaneously export what a standard machine vision camera would see for comparison.
Pre-processing functions calculate various polarization specific information like the ‘Degree of Polarisation’, ‘Stokkes Vector’ and the ‘Surface Normal Vector’.
At the higher-end level, ‘Applications-Oriented’ functions have been implemented to manage reflections and measure stress.
These functions’ algorithms have been optimised for accuracy and high computing to enable its use in real-time inspection.
* The choice of CPU / GPU architecture will be dictated by the processing functions as well as resolution and frame rate.