考虑变换域融合方法的刀具磨损区域三维重构
发布时间：2018-08-08 20:28
【摘要】：针对数控机床加工领域刀具磨损区域的三维重构精度提高的问题,提出了变换域融合方法。将非下采样轮廓波变换和小波变换相结合,对非下采样轮廓波变换得到的低频系数采用小波变换方法进行融合,对得到的高频系数构造区域方差进行融合,突出刀具磨损图像的整体灰度信息和边缘细节信息,从而获得更为清晰的二维刀具图像,以及更为准确的磨痕形貌及磨痕体积。对基于非下采样轮廓波变换、小波变换方法和变换域融合方法得到磨损区域三维形貌和磨损体积与显微镜的观察结果进行了对比分析,发现提出的方法得到的三维形貌与显微镜结果最接近,且计算出的磨损体积平均相对误差小于10%。分析和实验数据表明,所提方法无论在图像的清晰度和磨损体积的准确性方面都有一定的优势,该方法对于刀具磨损区域三维重构精度较高,优于单独采用非下采样轮廓波变换和小波变换方法得到的结果。
[Abstract]:In order to improve the precision of 3D reconstruction of tool wear area in the machining field of CNC machine tools, a transform domain fusion method is proposed. Combining the non sampling contour wave transform and the wavelet transform, the low frequency coefficients obtained by the non down sampling contour wave transform are fused by the wavelet transform method, and the high frequency coefficients are constructed for the regional square. The difference is fused to highlight the overall grayscale information and edge details of the tool wear image, so as to obtain a clearer two-dimensional tool image, as well as a more accurate grinding mark and wear volume. The results are compared with the observation results of the microscope. It is found that the proposed method is closest to the results of the microscope, and the average relative error of the calculated wear volume is less than the 10%. analysis and the experimental data. The proposed method has some advantages in the image definition and the accuracy of the wear volume. This method has higher accuracy for 3D reconstruction of tool wear area, and is superior to the results obtained by non-downsampling contour wave transform and wavelet transform alone.
【作者单位】： 西安交通大学现代设计及转子轴承系统教育部重点实验室;
【基金】：陕西省自然科学基金资助项目(2016JM5054)
【分类号】：TG659;TP391.41
本文编号：2172957
[Abstract]:In order to improve the precision of 3D reconstruction of tool wear area in the machining field of CNC machine tools, a transform domain fusion method is proposed. Combining the non sampling contour wave transform and the wavelet transform, the low frequency coefficients obtained by the non down sampling contour wave transform are fused by the wavelet transform method, and the high frequency coefficients are constructed for the regional square. The difference is fused to highlight the overall grayscale information and edge details of the tool wear image, so as to obtain a clearer two-dimensional tool image, as well as a more accurate grinding mark and wear volume. The results are compared with the observation results of the microscope. It is found that the proposed method is closest to the results of the microscope, and the average relative error of the calculated wear volume is less than the 10%. analysis and the experimental data. The proposed method has some advantages in the image definition and the accuracy of the wear volume. This method has higher accuracy for 3D reconstruction of tool wear area, and is superior to the results obtained by non-downsampling contour wave transform and wavelet transform alone.
【作者单位】： 西安交通大学现代设计及转子轴承系统教育部重点实验室;
【基金】：陕西省自然科学基金资助项目(2016JM5054)
【分类号】：TG659;TP391.41
【相似文献】
相关期刊论文 前5条
1 骆良顺;王新;苏彦庆;李新中;郭景杰;傅恒志;;基于连续切片的三维重构技术在材料凝固组织研究中的应用[J];材料导报;2010年09期
2 刘坤;徐磊;李双明;;三维重构技术在金属材料微观结构研究中的应用[J];材料导报;2012年19期
3 朱会;杨湘杰;郭洪民;;A356铝合金近球晶组织三维重构及空间结构表征[J];热加工工艺;2013年17期
4 赵辉煌;周德俭;黄春跃;张少华;;基于SFS原理的SMT焊点表面三维重构技术研究[J];计算机科学;2009年07期
5 隗汉兴;;刀具磨损区域的三维重构[J];金属加工(冷加工);2008年22期
相关硕士学位论文 前3条
1 吴保亮;铸件微观组织结构三维重构技术研究[D];南昌大学;2015年
2 朱会;铝合金流变铸造组织三维重构及空间结构表征[D];南昌大学;2013年
3 张萌;基于标识点的三维重构技术及其在铸件尺寸检测中应用的研究[D];华中科技大学;2011年
，本文编号：2172957
本文链接：https://www.wllwen.com/kejilunwen/jiagonggongyi/2172957.html
