The Regional Detection of 2D Barcode in Complicated Backgrounds of Metal Parts
The traditional detection algorithms fall into non-machine learning method, which utilize geometrical characteristic of 2D code, introduce digital image analysis method,marginal detection and geometry detection, etc. while these algorithms are no more than elementary methods, they are limited to printed paper, and not applicable to other material surface which Data Matrix is punched on. To solve the drawbacks mentioned above, this paper presents machine methods integrated into cascade filter methods to position the region of 2D code, then employ clustering growth method. Our experiments reveals, compared with traditional method, our methods have achieved higher rate of detection with good robustness.
KeywordsLocal Binary Pattern Connected Region Cluster Growth License Plate Recognition Cascade Filter
Unable to display preview. Download preview PDF.
- 1.Arnould, S., Awcock, G.J., Thomas, R.: Remote barcode localization using mathematical morphology. In: IEE Conference, pp. 642–646 (1999)Google Scholar
- 2.Viard-Gaudin, C., Normand, N., Barba, D.: Bar code location algorithm using a two-dimensional approach. In: Proceedings of the 2nd International Conference on Document Analysis and Recognition, pp. 45–48 (1993)Google Scholar
- 3.Jain, A.K., Chen, Y.: Bar code localization using texture analysis. In: Proceedings of the 2nd International Conference on Document Analysis and Recognition, pp. 41–44 (1993)Google Scholar
- 5.Viola, P., Jones, M.: Fast and robust classification using asymmetric adaboost and a detector cascade. In: Proceedings from NIPS 2001 (2001)Google Scholar
- 7.Levi, K., Weiss, Y.: Learning Object Detection from a Small Number of Examples:The Importance of Good Features. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2004), pp. 53–60 (2004)Google Scholar