A Novel Mixed Denoising Approach for Eengineering Drawings Based on Mathematical Morphology
Mathematical morphology denoising has been used abroadly in image processing, due to it overcomes the disadvantage of image edge illegible caused by traditional image denoising. But this denoising method caused image information distortion because it filtered the image s geometry information less than structural elements togethor with the noise. An improved denoising methods of mathematical morphology is introduced in this paper. Combined using arithmetic operation and template filter in image mathematical morphology denoising, the fine thread and narrow space, which has been filtered or filled incorrectly in the open- close operation of mathematical morphology operation, can be recovered. Experiments on practical engineering drawings have ivalidated that: the improved approach retained all information of original image, and can preserve the integrity and accuracy of images needed to be process. The approach has the great practicality in the engineering drawing processing.
mathematical morphology arithmetic operation mixed denoising structural element engineering drawings
Hui Dong Lili Dong Runling Yang Weiguo Liu
School of Information and Control Engineering, Xian University of Architecture and Technology Xi School of Information and Control Engineering, Xian University of Architecture and Technology Xi School of Automation, Northwestern Polytechnical University Xian, China
国际会议
太原
英文
264-267
2011-02-26（万方平台首次上网日期，不代表论文的发表时间）