scholarly journals VIRTUAL DIMENSIONALITY ESTIMATION IN HYPERSPECTRAL IMAGERY BASED ON UNSUPERVISED FEATURE SELECTION

2016 ◽  
Vol III-7 ◽  
pp. 17-23 ◽  
Author(s):  
M. Ghamary Asl ◽  
B. Mojaradi
Keyword(s):  
Feature Selection ◽  
High Speed ◽  
Hyperspectral Imagery ◽  
Current Paper ◽  
Estimation Methods ◽  
Tuning Parameters ◽  
Spectral Signatures ◽  
Virtual Dimensionality ◽  
Image Partition ◽  
Partition Space

Virtual Dimensionality (VD) is a concept developed to estimate the number of distinct spectral signatures in hyperspectral imagery. Intuitively, detecting the number of spectrally distinct signatures depends on determining the number of distinct bands of the data. Considering this idea, the current paper aims at estimating the VD based on finding independent bands in the image partition space. Eventually, the number of independent selected bands is accepted as the VD estimate. The proposed method is automatic and distribution-free. In addition, no tuning parameters and noise estimation processes are needed. This method is compared with three well-known VD estimation methods using synthetic and real datasets. Experimental results show high speed and reliability in the performance of the proposed method.

scholarly journals VIRTUAL DIMENSIONALITY ESTIMATION IN HYPERSPECTRAL IMAGERY BASED ON UNSUPERVISED FEATURE SELECTION

2016 ◽  
Vol III-7 ◽  
pp. 17-23
Author(s):  
M. Ghamary Asl ◽  
B. Mojaradi
Keyword(s):  
Feature Selection ◽  
High Speed ◽  
Hyperspectral Imagery ◽  
Current Paper ◽  
Estimation Methods ◽  
Tuning Parameters ◽  
Spectral Signatures ◽  
Virtual Dimensionality ◽  
Image Partition ◽  
Partition Space

Virtual Dimensionality (VD) is a concept developed to estimate the number of distinct spectral signatures in hyperspectral imagery. Intuitively, detecting the number of spectrally distinct signatures depends on determining the number of distinct bands of the data. Considering this idea, the current paper aims at estimating the VD based on finding independent bands in the image partition space. Eventually, the number of independent selected bands is accepted as the VD estimate. The proposed method is automatic and distribution-free. In addition, no tuning parameters and noise estimation processes are needed. This method is compared with three well-known VD estimation methods using synthetic and real datasets. Experimental results show high speed and reliability in the performance of the proposed method.

The Impact of Structural Reforms and Regulations on the Demand Side in the Railway Industry

2019 ◽  
Vol 18 (1) ◽  
pp. 1-33
Author(s):  
Fumitoshi Mizutani
Keyword(s):  
Market Structure ◽  
Vertical Integration ◽  
High Speed ◽  
Fixed Effect Model ◽  
Structural Reform ◽  
Estimation Methods ◽  
Public Ownership ◽  
Estimation Model ◽  
Passenger Demand ◽  
The Impact

Abstract The main purpose of this study is to evaluate factors affecting passenger rail demand, with special attention to the effects of structural reform/regulation and competition. In order to do this, we use data obtained from 30 OECD countries for the 24 years from 1990 to 2013. As structural reform/regulation and competition variables, we take the OECD’s five kinds of regulatory indices: (i) overall, (ii) entry, (iii) public ownership, (iv) vertical integration, and (v) market structure; and for competition variables, we take (vi) rail passenger-freight ratio, (vii) rail share, and (viii) high-speed train ratio. As estimation methods, both the fixed effect model and the Hausman-Taylor estimation model are used. The major findings are as follows. First, competition as competitiveness (i.e. the share of rail, passenger over freight ratio) increases passenger demand. And the existence of high-speed trains increases passenger demand. Second, overall, entry regulation, and market structure have no significant effect on demand. Third, public ownership affects passenger demand positively. Last, vertical integration reduces passenger demand.

Experimental Research on the Surf-Riding Region for High Speed Vessels

2015 ◽  
Author(s):  
Atsuo Maki ◽  
Yoshiki Miyauchi
Keyword(s):  
Experimental Research ◽  
High Speed ◽  
Safety Assessment ◽  
Estimation Methods ◽  
Model Experiments ◽  
Size Limitation ◽  
Analytical Estimation ◽  
Free Running ◽  
Surface Combatant ◽  
Surf Riding

It is well known that surf-riding phenomenon is the prerequisite of the broaching-to in following and quartering conditions. For the safety assessment of the fast vessel such as surface combatant sand patrol crafts, the estimation of the surf-riding condition is important. Therefore, so far several experimental efforts have been made. However, in these previous researches, the free running model experiments in high speed region, i.e.up to Froude number of 0.6 or 0.7, have not been conducted because of tank size limitation. As shown in this paper, there are occurrence and disappearance boundaries of surf-riding in lower and faster region, respectively. In our study, free running model experiments are carried out in high speed region, and then both boundaries are experimentally obtained. By using obtained results, the analytical estimation methods proposed by the authors can be well validated. Furthermore, the free running model experiments in irregular seas are also conducted. Then, surf-riding phenomenon in irregular seas is also discussed.

Sign in / Sign up

Export Citation Format

Share Document