Browsing by Author "Chang, Ming-Wei"

Now showing 1 - 2 of 2

A new vehicle detection approach in traffic jam conditions
(2007)
Wu, Bing-Fei
;
Juang, Jhy-Hong
;
Tsai, Ping-Tsung
;
Chang, Ming-Wei
;
Fan, Chung-Jui
The well-known vehicle detectors utilize the background extraction methods to segment the moving objects. The back-ground updating concept is applied to overcome the luminance variation which results in the error detection. These systems will meet a challenge when detecting the vehicles in the traffic jam conditions at sunset. The vehicles will cover the road surface so that the background information cannot be smoothly updated. Once the traffic is released, the existing background is not suitable for the moving segmentation. The main contribution of this paper is that an efficient vehicle detection approach is proposed to improve the detection accuracy in traffic jam conditions. The land mask decision gives the land information and the merged boundary box rule is presented to realize vehicle detection. The signed square normalized correlation coefficient calculation is addressed, and it is applied to vehicle tracking. The experimental results show that this approach works well in highway and urban area with high accuracy.
Establishing an invar leveling calibration system
(2008-07-01)
Chen, Chun-Sung
;
Wu, Chien-Ting
;
Chang, Ming-Wei
;
Chang, Wei-Chen
;
土木工程學系
Precise height systems are widely utilized in civil engineering projects Such as high-speed railroads, and highways and bridge construction. However, highly precise height data is obtained using precise leveling. An extremely precise invar leveling rod is necessary for obtaining precise results. In this study, an automatic calibration system for invar leveling has been established by the National Measurement Laboratory (NML). The content belongs to the long scales calibration system of D17 of the NML. This system utilizes a steady frequency laser interferometer, a CCD microscopic monitor, and a precise moving mechanism. With this system, an invar rod and a mark invar leveling rod are applied to estimate uncertainty with a 95% confidence level. Uncertainty results are 22 mu m and 21 mu m for the invar leveling rods and mark invar leveling rods, respectively.