基于斜切分段模型的蚕茧顶端展开算法研究  

Research on the algorithm for unfolding the top of silkworm cocoons based on oblique segmentation model

在线阅读下载全文

作  者:崔文宇 梁曼 CUI Wenyu;LIANG Man(College of Mechanical and Electrical Engieering,China Jiliang University,Hangzhou 310018,China)

机构地区:[1]中国计量大学机电工程学院,浙江杭州310018

出  处:《中国计量大学学报》2024年第4期664-672,共9页Journal of China University of Metrology

基  金:浙江省基础公益研究计划项目(No.LGG20E050014)。

摘  要:目的:针对蚕茧顶端直接采集图像存在曲面信息压缩的问题,提出一种斜切分段展开方法,用于获取蚕茧顶端信息。方法:首先,将蚕茧顶端投影图像还原至顶端球面图像,对球面进行切分及斜切分操作,实现顶端投影图像到斜切分段模型的转化;其次,对相应模型建立坐标系,推导出单一斜切分段的投影图像与展开图像间的坐标变换关系;最后,对单一斜切分段展开图像的底边归一化,实现蚕茧完整顶端图像的展开。结果:斜切分段展开在水平方向误差均值为2.714像素,垂直方向误差均值为3.786像素。结论:斜切分段模型能够有效实现蚕茧顶端图像的展开,进而获取顶端疵点信息,为获取蚕茧完整表面疵点提供良好基础。Aims:Aiming at the problem of surface information compression in directly collecting images at the top of silkworm cocoons,an oblique segmentation model was proposed to accurately obtain information at the top of silkworm cocoons.Methods:Firstly,the image of the top of the silkworm cocoon was restored to the top spherical model;and segmentation and oblique segmentation operations were performed on the spherical surface to achieve transformation from the top image to the oblique segmented model.Secondly,a coordinate system for the corresponding model was established;and the coordinate transformation relationship between the projection image of a single oblique segment and the unfolded image was derived.Finally,the bottom edge of a single oblique segmented unfolded image was normalized to achieve the complete unfolding of the top image of the silkworm cocoon.Results:The average horizontal error of the oblique segmented unfolding model was 2.714 pixels;and the average vertical error was 3.786 pixels.Conclusions:The oblique segmentation model can effectively unfold the top image of silkworm cocoons,thereby obtaining information on top defects and providing a good foundation for obtaining complete surface defects of silkworm cocoons.

关 键 词:蚕茧分选 图像处理 球面斜切分段模型 数学建模 微积分 

分 类 号:TP391[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象