There are 20 image segmentation algorithm modules, such as local threshold, standard threshold, double threshold segmentation, lag threshold, histogram segmentation, and dozens of regions, shapes and gray level characteristic extraction methods.
Based on the primacy of morphological opening, closing, corrosion, and expansion, it provides abundant extension operators, which effectively improves the efficiency of the algorithm in the process of image processing.
Including ten flexible matching algorithms based on gray level matching, shape matching, variable shape matching, key point description matching, even if the target changes, it can accurately locate the matching target.
The platform's rich neural network algorithm and powerful classifier provide a good guarantee for OCV, through the visual configuration window can be targeted training.
Code to identify
Including barcode and two-dimensional code, more than 20 kinds of code recognition operator modules are provided, which can quickly identify barcode or two-dimensional code in situations such as distortion, contamination, poor focusing, overexposure, etc.
Including 1D, 2D, 3D image measurement operator module, through the measurement type and configuration window, the accurate measurement results can be easily got.
Through camera calibration, high precision measurement results can be obtained, and the stable measurement accuracy of 1um can be obtained with good sub-pixel technology support.
The neural network
Relevant modules based on artificial neural networks provide useful tools for matching, classification, recognition, and recovery.