Abstract

Connected component labelling is an important part of identifying regions and performing feature extraction. In a realtime industrial environment where online image analysis is required, it is highly desirable to have a labelling algorithm that can be implemented at pixel rate in parallel with a raster scan of the image. Using such algorithms based on a fixed-size local window, relevant feature data are stored online and updated as each pixel is labelled. As the number of labels increases with the complexity and size of the image, the hardware feature memory (FM), fixed a priori, may overflow. Any realtime industrial application such as a visual inspection system must deal with this problem. Two solutions to the problem which involve reusing the online memory are proposed in this paper. The relationship between the FM size and the size of the image is discussed. It is shown that the proposed algorithm makes optimal use of the online feature memory and delays purging the FM and storing data offline as long as possible. Also derived are the maximum number of purging phases needed in terms of the size of the FM, N and M, where N × M is the size of the image.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.