Rail fastener positioning method has high requirements for image quality and positioning accuracy.Therefore,a rail fastener positioning method based on gray mutation is proposed.Firstly,the image is denoised by an imp...Rail fastener positioning method has high requirements for image quality and positioning accuracy.Therefore,a rail fastener positioning method based on gray mutation is proposed.Firstly,the image is denoised by an improved median filter.Then,according to the characteristics of image frequency domain,the image is decomposed by wavelet transfrom to extract low-frequency component,and the low-frequency component is filtered and processed by gamma transformation to reduce the influence of natural environment factors on image quality.After that,according to the change rule of gray-scale values of different regions in the image,the gray-scale mutation in column and line directions of the image is statistically analyzed,and then the rail surface and the sleeper are located.Finally,the fastener area is accurately located by using the position relationship of the rail surface,the sleeper and the fastener in the image.The experimental results show that the positioning accuracy of the proposed method is 93.19%,which can quickly and effectively locate the fastener region,and has strong environmental adaptability,robustness and practicability.展开更多
To adjust the samples of filtering adaptively,an improved Gaussian particle filter algorithm based on Kullback-Leibler divergence(KLD)-sampling(KLGPF)is proposed in this paper.During the process of sampling,the algori...To adjust the samples of filtering adaptively,an improved Gaussian particle filter algorithm based on Kullback-Leibler divergence(KLD)-sampling(KLGPF)is proposed in this paper.During the process of sampling,the algorithm calculates the KLD to adjust the size of the particle set between the discrete probability density function of particles and the true posterior probability density function.KLGPF has significant effect when the noise obeys Gaussian distribution and the statistical characteristics of noise change abruptly.Simulation results show that KLGPF could maintain a good estimation effect when the noise statistics changes abruptly.Compared with the particle filter algorithm using KLD-sampling(KLPF),the speed of KLGPF increases by 28%under the same conditions.展开更多
基金National Natural Science Foundation of China(Nos.61616202,61461203)Ministry of Education Innovation Team Development Plan(No.IRT_16R36)Plateau Information Engineering and Control Key Practice Laboratory Open Project Fund of Gansu Province(No.201611105)。
文摘Rail fastener positioning method has high requirements for image quality and positioning accuracy.Therefore,a rail fastener positioning method based on gray mutation is proposed.Firstly,the image is denoised by an improved median filter.Then,according to the characteristics of image frequency domain,the image is decomposed by wavelet transfrom to extract low-frequency component,and the low-frequency component is filtered and processed by gamma transformation to reduce the influence of natural environment factors on image quality.After that,according to the change rule of gray-scale values of different regions in the image,the gray-scale mutation in column and line directions of the image is statistically analyzed,and then the rail surface and the sleeper are located.Finally,the fastener area is accurately located by using the position relationship of the rail surface,the sleeper and the fastener in the image.The experimental results show that the positioning accuracy of the proposed method is 93.19%,which can quickly and effectively locate the fastener region,and has strong environmental adaptability,robustness and practicability.
基金the China Postdoctoral Science Foundation(No.171980)the National Natural Science Foundation of China(Nos.61973160,51505221)Key Laboratory Fund of Science and Technology on Communication Networks(No.6142104180114).
文摘To adjust the samples of filtering adaptively,an improved Gaussian particle filter algorithm based on Kullback-Leibler divergence(KLD)-sampling(KLGPF)is proposed in this paper.During the process of sampling,the algorithm calculates the KLD to adjust the size of the particle set between the discrete probability density function of particles and the true posterior probability density function.KLGPF has significant effect when the noise obeys Gaussian distribution and the statistical characteristics of noise change abruptly.Simulation results show that KLGPF could maintain a good estimation effect when the noise statistics changes abruptly.Compared with the particle filter algorithm using KLD-sampling(KLPF),the speed of KLGPF increases by 28%under the same conditions.