To improve the accuracy of detected parameters at curves by an in-use track geometry measurement (TGM) system, a real-time railway plane alignment (RPA) discrimination method based on linear discriminant analysis of an Lp norm with sparsity constraint (LDA-Lp-SC) is proposed. A standard database to identify the radius of a track curve is established, using the relevant ledger and track TGM data. Multi-dimensional feature extraction is performed on the TGM data, and the parameters of the method are optimized to improve the accuracy of the model classification. Finally, the method is embedded in a TGM system combined with the original RPA discrimination method, so that the key parameters of RPA discrimination can be automatically switched in real time according to the classification result of the track curve radius. Compared with the original method, the method in this study greatly improves the recognition accuracy of track curve with a radius in the range of 150–20,000 m, while ensuring real-time performance.
ChenZ. Railway curve track (in Chinese). Beijing, China: China Railway Publishing House, 2016, p. 245.
2.
ChenFLiang-YaoGUYangY, et al.Optimum method for horizontal curve re-surveying of the existing railway. J Railway Sci Eng2012; 9: 90–95.
3.
SánchezABravoJLGonzálezA. Estimating the accuracy of track-surveying trolley measurements for railway maintenance planning. J Surv Eng2017; 143: 05016008.
4.
WestonPRobertsCYeoG, et al.Perspectives on railway track geometry condition monitoring from in-service railway vehicles. Vehicle Syst Dyn2015; 53: 1063–1091.
5.
HaoXYangJYangF, et al.Track geometry estimation from vehicle–body acceleration for high-speed railway using deep learning technique. Vehicle Syst Dyn2022.
6.
Infrastructure Inspection Research Institute CAoRS. Dynamic inspection for railway infrastructure—track geometry measurement system. Beijing, China: China State Railway Group Co., Ltd, 2020, p. 12.
7.
YuNWeiSLiY, et al.Detection algorithm of railway curve feature point based on convolutional neural network. Railway Eng2021; 61: 133–137.
8.
MohantyNJohnAL-SManmathaR, et al.Chapter 10 - shape-based image classification and retrieval. In: RaoCRGovindarajuV (eds) Handbook of statistics. Amsterdam, The Netherlands: Elsevier, 2013, pp. 249–267.
9.
BuiTDQuachKGDuongCN, et al.Lp norm relaxation approach for large scale data analysis: a review. In: CampilhoAKarrayFter Haar RomenyB (eds) Image analysis and recognition. Cham, Germany: Springer International Publishing, 2018, pp. 285–292.
10.
WangYWangPWangX, et al.Position synchronization for track geometry inspection data via big-data fusion and incremental learning. Transp Res C Emerg Tech2018; 93: 544–565.
11.
XuPSunQLiuR, et al.Optimizing the alignment of inspection data from track geometry cars. Comp Aid Civil Infrastruct Eng2015; 30: 19–35.
12.
KhosraviMSoleimanmeigouniIAhmadiA, et al.Reducing the positional errors of railway track geometry measurements using alignment methods: a comparative case study. Measurement2021; 178: 109383.
13.
EklöfKNwichi-HoldsworthAEklöfJ. Novel algorithm for mutual alignment of railway track geometry measurements. Transp Res Rec2021; 2675: 995–1004.
14.
LiYWangHHouZ, et al.Design and development of on-board track inspection system based on CAN bus. China Railway Sci2020; 41: 163–170.
15.
WeiSLiYZhaoY, et al.Design and development of GJ-6 track detection system. Railway Eng2012; 2: 97–100.
16.
GeltnerPB. General Rayleigh quotient iteration. SIAM J Numer Anal1981; 18: 839–843.
17.
ChuDThyeGS. A new and fast implementation for null space based linear discriminant analysis. Pattern Recogn2010; 43: 1373–1379.
18.
ZhangWHeLDengY, et al.Time–frequency cepstral features and heteroscedastic linear discriminant analysis for language recognition. IEEE Trans Audio Speech Lang Process2011; 19: 266–276.
19.
MinerGDElderJFastA, et al.Feature selection and dimensionality reduction. In: MinerGDelenDElderJ, et al. (eds) Practical text mining and statistical analysis for non-structured text data applications. Boston, MA: Academic Press, 2012, pp. 929–934.
20.
TanXDengLYangY, et al.Optimized regularized linear discriminant analysis for feature extraction in face recognition. Evol Intellig2019; 12: 73–82.
21.
TheodoridisSKoutroumbasK. Pattern recognition. 4th ed.Amsterdam, The Netherlands: Elservier Inc., 2009, pp. 280–281.
22.
Hiriart-UrrutyJ-BLemaréchalC. Abstract duality for practitioners. In: Hiriart-UrrutyJ-BLemaréchalC (eds). Convex analysis and minimization algorithms II: advanced theory and bundle methods. Berlin, Germany: Springer Berlin Heidelberg, 1993, pp. 137–193.
23.
ZhangHWangYChangX, et al.L1/2 regularization. SCIENTIA SINICA Inform2010; 40: 412–422.
24.
XuZ. Data modeling: visual psychology approach andL1/2 regularization theory. Proc Int Congress Mathematicians2010; 4: 3151–3184.
25.
XuZChangXXuF, et al. L1/2 regularization: a thresholding representation theory and a fast solver. IEEE Trans Neural Networks Learn Syst2012; 23: 1013–1027.
26.
LyonsRG. Understanding digital signal processing. 3rd ed.London, UK: Pearson, 2010, pp. 540–541.
27.
ProakisJGManolakisDG. Digital signal processing: principles, algorithms, and applications. 4th ed.London, UK: Pearson, 2013, pp. 664–671.
28.
ZhangYLiuYYangG, et al.SSIT: a sample selection-based incremental model training method for image recognition. Neural Comput Appl2022; 34: 3117–3134.
29.
LindnerC. Chapter 1 - automated image interpretation using statistical shape models. In: ZhengGLiSSzékelyG (eds) Statistical shape and deformation analysis. Boston, MA: Academic Press, 2017, pp. 3–32.
30.
ArlotSGenuerR. Comments on: a random forest guided tour. TEST2016; 25: 228–238.
