Abstract: An adequate solution for computer vision applications is arrived at more efficiently and, with more automation, enables users with limited or no special image processing and pattern recognition knowledge to create reliable vision systems for their applications. Computer rendering of CAD models is used to automate the dataset acquisition process and labeling process. In order to speed up the training data preparation while maintaining the data quality, a number of processed samples are generated from one or a few seed images.

Abstract: An adequate solution for computer vision applications is arrived at more efficiently and, with more automation, enables users with limited or no special image processing and pattern recognition knowledge to create reliable vision systems for their applications. Computer rendering of CAD models is used to automate the dataset acquisition process and labeling process. In order to speed up the training data preparation while maintaining the data quality, a number of processed samples are generated from one or a few seed images.

Abstract: An automated printout inspection system identifies glyphs in an image by calculating a connectedness score for each foreground pixel, and comparing this score with a specified threshold. The system further generates training images by simulating printouts from an impact printer, including the specifying of specific error types and their magnitudes. The simulated printouts are combined with scan images of real-world printout to train an automated printout inspection system. The inspection results of the automated system are compared with inspection results from human inspectors, and test parameters of the automated system are adjusted so that it renders inspection results within a specified range of the average human inspector.

Abstract: An automated printout inspection system identifies glyphs in an image by calculating a connectedness score for each foreground pixel, and comparing this score with a specified threshold. The system further generates training images by simulating printouts from an impact printer, including the specifying of specific error types and their magnitudes. The simulated printouts are combined with scan images of real-world printout to train an automated printout inspection system. The inspection results of the automated system are compared with inspection results from human inspectors, and test parameters of the automated system are adjusted so that it renders inspection results within a specified range of the average human inspector.