A Selective Multiclass Support Vector Machine Ensemble Classifier for Engineering Surface Classification Using High Definition Metrology

Abstract

The surface appearance is sensitive to change in the manufacturing process and is one o f the most important product quality characteristics. The classification o f workpiece sur face patterns is critical fo r quality control, because it can provide feedback on the manu facturing process. In this study, a novel classification approach fo r engineering surfaces is proposed by combining dual-tree complex wavelet transform (DT-CWT) and selective ensemble classifiers called modified matching pursuit optimization with multiclass sup port vector machines ensemble (MPO-SVME), which adopts support vector machine (SVM) as basic classifiers. The dual-tree wavelet transform is used to decompose threedimensional (3D) workpiece surfaces, and the features o f workpiece surface are extracted from wavelet sub-bands o f each level. Then MPO-SVME is developed to classify different workpiece surfaces based on the extracted features and the performance o f the proposed approach is evaluated by computing its classification accuracy. The performance o f MPO-SVME is validated in case study, and the results demonstrate that MPO-SVME can increase the classification accuracy with only a handful o f selected classifiers.

Co-Kriging Method for Form Error Estimation Incorporating Condition Variable Measurements

“…From Fig. 7, some findings are obtained (1) The proposed CK based method performs significantly better than OK when systematic machining errors exist. For instance, in Case 1 the flatness error estimation assessment ratio is between 0.7 and 0.75 for OK while between 0.85 and 0.95 for CK.…”

“…Since TIN and IDW methods do not have estimation variance error, criteria MSPE and RMSPE cannot be obtained for them. From Table 4, some findings are obtained (1) The unmeasured point can be estimated by nearest measured points using traditional interpolation methods like TIN and IDW, but the result of estimation is less precise than the Kriging methods. The indices of ME and MSE for TIN and IDW are larger than that for the Kriging methods.…”

“…Surface form is predominantly considered as one of the most important features of practical product surfaces due to its crucial influence on the functional behavior of a machined part [1][2][3][4][5][6][7][8]. The form error estimation under various machining conditions is an essential step in the assessment of part surface quality generated in machining processes [9].…”

Case study of surface micro-waves in ultra-precision raster fly cutting

“…They found that chip serration causes microwaves on machined surfaces, which increases machined surface roughness. In addition, Du and Fei conducted a series of studies on surface topography classification and measurement in conventional milling [14][15][16][17]. However, there has been no relevant research involving UPRFC.…”

Binary tree optimization using genetic algorithm for multiclass support vector machine

“…SVM classifies a sample as a positive or negative Vapnik, 2000). However, most problems in the real world are usually multiclass problems (Du, Liu, & Xi, 2015;Huang, Zhang, Zeng, & Bushel, 2013;Pathak & Sunkaria, 2014;YJlmaz & KJlJkçJer, 2013). For multiclass problems, there are more than two classes, and a sample is classified as one class among many classes.…”