IDS peak 2.10.0 / uEye+ firmware 3.34

IDS Peak comfortSDK, genericSDK, IPL, and AFL developer manuals are external documents. Please contact us if you need them.

CNN means convolutional neural network. It is a special type of an artificial neuronal network. It has several folding layers which gave the CNN its name. Because of its design and structure, a CNN is especially suitable for machine learning in the field of image recognition, because CNNs are excellent for extracting features from images. They can handle very different and variable objects and backgrounds, which is often the case when classifying objects in industrial environments.

CNNs are suitable to perform qualitative classification tasks that a trained employee could make within a second. Such as a component is missing here, the apple is rotten, the packaging is damaged. In addition, damages or insufficient apples can also be located with the help of CNNs. The localization of objects in the image is called object detection.

CNNs are not suitable for quantitative tasks and statements such as "This crack is 2 mm long". Simply put, these are tasks for which an employee needs additional tools for decision-making, e.g. a caliper gauge.