OWA filters: A robust filtering method and its application to color images

Author(s):  
A. Basu ◽  
M. Nachtegael
2021 ◽  
Vol 237 ◽  
pp. 109544
Author(s):  
Gustavo E. Coelho ◽  
Maria Graça Neves ◽  
António Pascoal ◽  
Álvaro Ribeiro ◽  
Peter Frigaard

2013 ◽  
Vol 756-759 ◽  
pp. 344-348
Author(s):  
Ling Jing Meng ◽  
Hai Bo Liu

Wavelet-based robust filtering of process data is proposed in order to reduce the influence of the outliers and noise in Out-trajectory data. We utilize the moving median filtering method to reject outliers in the original data and then combine wavelet de-noising method with empirical Wiener threshold to suppress noise. Simulation calculation and real engineering application has shown that the novel algorithm reliably preserves the information encapsulated in a process signal corrupted with noise and outliers. The methodology has been proved to be reliable and robust.


IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 161141-161151
Author(s):  
Lirong Shen ◽  
Haifeng Sun ◽  
Xiaoping Li ◽  
Yanming Liu ◽  
Haiyan Fang ◽  
...  

Author(s):  
H. Yokoyama ◽  
H. Chikatsu

Recently, laser scanning has been receiving greater attention as a useful tool for real-time 3D data acquisition, and various applications such as city modelling, DTM generation and 3D modelling of cultural heritage sites have been proposed. And, former digital data processing were demanded in the past digital archive techniques for cultural heritage sites. However, robust filtering method for distinguishing on- and off-terrain points by terrestrial laser scanner still have many issues. In the past investigation, former digital data processing using air-bone laser scanner were reported. Though, efficient tree removal methods from terrain points for the cultural heritage are not considered. In this paper, authors describe a new robust filtering method for cultural heritage using terrestrial laser scanner with "the echo digital processing technology" as latest data processing techniques of terrestrial laser scanner.


Author(s):  
Anas Fouad Ahmed ◽  
Bilal R. Al-Kaseem ◽  
Zahraa Khduair Taha

2018 ◽  
Vol 32 (4) ◽  
pp. 427-441 ◽  
Author(s):  
Lidia Gurau ◽  
Nadir Ayrilmis

This study extensively investigated the surface roughness of injection molded wood plastic composites (WPCs) produced from different amounts of wood flour, polymer matrix, mineral filler, and other additives. A larger range of roughness parameters that used in the previous literature were obtained from nine different WPC compositions by using a robust filtering method (robust Gaussian regression filter) to have a better understanding of the overlaying quality of the WPC samples. Three melt flow index (MFI) of the polymer were tested (MFI 3.6, 12, and 25). It was found that WPCs produced with the polypropylene having a MFI value of 25 were the smoothest. It was noticed that not only the wood flour percentage matters but the combination of wood flour–mineral filler was also important. The WPC compositions with lower polymer amount in favor of more wood flour and mineral filler led to rougher surfaces. A decrease of wood flour in favor of increasing the mineral filler participation percentage had a surface smoothing effect. At the same wood flour content, a decrease in polymer combined with an increase in mineral filler led to rougher WPC surface. Among the tested WPC compositions, the smoothest surface was obtained in the specimens produced from 50% wood flour, 0% mineral filler, and around 40% polymer by weight. The results should be helpful to anticipate the effect on surface roughness of the percentage participation for each amount of the wood or mineral filler, polymer matrix, and additives in further development of WPC combinations.


2013 ◽  
Vol 411-414 ◽  
pp. 907-911
Author(s):  
She Sheng Gao ◽  
Yan Zhao ◽  
Wen Hui Wei

This paper presents a fuzzy anti-interference transfer alignment method for airborne strapdown inertial navigation system (SINS). This fuzzy anti-interference transfer alignment method takes the influence of systematic error into account for SINS transfer alignment. The fuzzy rules are constructed and incorporated in the filtering process to estimate the covariance matrices of observation vector and predicted state vector with random weight method. The experimental results demonstrate that the proposed method can resist the interferences caused by the airborne maneuvering process, thus improving the accuracy for SINS transfer alignment.


Sign in / Sign up

Export Citation Format

Share Document