基于改进孪生支持向量机的齿廓图像边缘失真分类研究 
下载： 563次
Classification of Edge Distortion of Tooth Profile Image Based on Improved Twin Support Vector Machine
1 沈阳工业大学 机械工程学院，沈阳 110870
2 辽宁科技学院 电气与信息工程学院，辽宁 本溪 117004
图 & 表
图 1.
Fig. 1. Geometric relationship of points on the involute
下载图片 查看原文
图 2.
Fig. 2. Segmentation of the tooth profile measurement area
下载图片 查看原文
图 3.
Fig. 3. Discrimination of edge distortion area & type
下载图片 查看原文
图 4.
Fig. 4. Schematic diagram of edge distortion type discrimination
下载图片 查看原文
图 5.
Fig. 5. Structure of gear vision measuring device
下载图片 查看原文
图 6.
Fig. 6. The working process of gear vision image acquisition
下载图片 查看原文
图 7.
Fig. 7. Eigenvector analysis
下载图片 查看原文
图 8.
Fig. 8. Test results on OCF-PBT-TWSVM algorithm
下载图片 查看原文
图 9.
Fig. 9. Test results on PBT-SVM algorithm
下载图片 查看原文
表 1不同归一化方式的对比
Table1. Comparison of different normalization methods
| Normalization method | Test accuracy | Optimal parameter selection of OCF-PBT-TWSVM |
|---|
| No normalization | 78% (39/50) | Population size N=20；Maximum number of iterations K=200; Best c1=46.52; c2=48.97; g=0.084 | | [-1,1]Normalization | 96% (48/50) | Population size N=20；Maximum number of iterations K=200; Best c1=6.544; c2=6.998; g=4.634 | | [0,1]Normalization | 96% (48/50) | Population size N=20；Maximum number of iterations K=200; Best c1=6.875; c2=6.529; g=16.004 |
|
查看原文
表 2不同核函数的对比
Table2. Comparison of different kernel functions
| Choice of kernel function | Test accuracy | Optimal parameter selection of OCF-PBT-TWSVM |
|---|
| Linear | 54% (27/50) | Population size N=20；Maximum number of iterations K=200; Best c1=9.93; c2=10.46; g=11.795 | | Polynomial | 92% (46/50) | Population size N=20；Maximum number of iterations K=200; Best c1=7.59; c2=8.92; g=14.234 | | Radial basis function | 96% (48/50) | Population size N=20；Maximum number of iterations K=200; Best c1=6.89; c2=7.42; g=13.931 | | Sigmoid | 42% (21/50) | Population size N=20；Maximum number of iterations K=200; Best c1=6.58; c2=7.39; g=17.242 |
|
查看原文
表 3不同算法的测试结果对比
Table3. Comparison of test results of different algorithms
| Algorithm | Test accuracy | Optimal parameter selection | Average test accuracy |
|---|
| OCF-PBT-TWSVM | 97.87% (46/47) | N=20; K=200; Best c1=15.38; c2=16.92; g=4.18 | 96.96% | | 97.22% (35/36) | N=20; K=200; Best c1=20.85; c2=22.02; g=20.69 | | 96% (48/50) | N=20; K=200; Best c1=6.89; c2=7.42; g=13.931 | | PBT-SVM | 95.75% (45/47) | Best c=5.66; g=4 | 94.06% | | 94.44% (34/36) | Best c=11.6; g=8 | | 92% (46/50) | Best c=8; g=16 |
|
查看原文
表 4齿廓边缘视觉测量结果实时统计
Table4. Real-time statistics of visual measurement results of tooth profile edges
| Normal edge signal | Ignore type of distorted signal | Eliminate type of distorted signal | Compensation type of distorted signal |
|---|
| 160 | 6 | 4 | 10 | | 158 | 7 | 3 | 12 | | 162 | 6 | 4 | 8 | | 156 | 8 | 4 | 12 | | 164 | 4 | 4 | 8 | | 153 | 9 | 5 | 13 |
|
查看原文
孙禾, 赵文珍, 赵文辉, 段振云. 基于改进孪生支持向量机的齿廓图像边缘失真分类研究[J]. 光子学报, 2020, 49(10): 1015002. He SUN, Wen-zhen ZHAO, Wen-hui ZHAO, Zhen-yun DUAN. Classification of Edge Distortion of Tooth Profile Image Based on Improved Twin Support Vector Machine[J]. ACTA PHOTONICA SINICA, 2020, 49(10): 1015002.
PDF全文
