Driver fatigue detection based on computer vision is considered as one of the most hopeful applications of image recognition technology. The key issue is to extract and select useful features from the driver images. In this work, we use the properties of image sequences to describe states of drivers. In addition, we introduce a kernelized fuzzy rough sets based technique to evaluate quality of candidate features and select the useful subset. Fuzzy rough sets are widely discussed in dealing with uncertainty in data analysis. We construct an algorithm for feature evaluation and selection based on fuzzy rough set model. Two classification algorithms are introduced to validate the selected features. The experimental results show the effectiveness of the proposed techniques.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
A numerical study based on an optimal heat line simulation method is proposed to calculate the steady-state temperature distribution of underground power cables surrounded by sand backfill and soil. Virtual heat lines are used to replace the heat in the cable conductors in this method. The heat and location of these virtual heat lines are determined by solving Laplace’s and Poisson’s equations with boundary conditions satisfied based on the genetic algorithm. An experiment was carried out to verify the validity of the proposed method. Through comparison with the experimental results, the proposed model is proved to be effective to calculate the temperature distribution of buried power cables. Furthermore, the ampacity of buried power cables is calculated by iteration method.
PL
W artykule zaprezentowano metodę symulacji nagrzewania podziemnego kabla. Wykonano obliczenia wykorzystując równania Laplace i Poissona przy warunkach granicznych bazujących na algorytmie genetycznym. Określono też przeciążalność prądową kabla.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Nitrous oxide (N[2]O) was one of the major atmospheric greenhouse gases. Its budget was poorly understood in alpine meadow, a dominant vegetation type on the Tibetan Plateau. To characterize a Kobresia humilis meadow on the plateau, N[2]O emission rates were monitored from June 2003 to June 2006 in the study area located at 3280 m a.s.l. Nine plots with 1 m x 1 m each were divided into three treatments, i.e. intact herbaceous community (HCK), removal of aboveground plant biomass (CBK), and removal of both above and belowground plant biomass (BSK), to estimate contribution of plants, r hizosphere and bulk soil to the total N[2]O emission. N[2]O emission from plant aboveground biomass was calculated by flux difference between HCK and CBK, denoted as F[(HCK-CBK)], from rhizosphere by F[(CBK-BSK)], and from bulk soil was the flux in BSK treatment. Static chambers (height 50 cm, area 0.5 x 0.5 m2 )were used for gas collection. N[2]O emission rate was significantly correlated with soil temperature at 5 cm depth in both HCK and BSK (P <0.001). Both treatments demonstrated a seasonal peak rate in growing season and minimum rate in dormancy period. The mean emission rates in the three years were 39.7[plus or minus]2.9 and 30.6[plus or minus]2.5 [mu]g m[^-2] h[^-1] in HCK and BSK, respectively, with the former significantly higher than the latter (P <0.05). In CBK, however, the emission rate did not show consistent correlation with soil temperature, especially in growing season. Its three-year mean emission rate was 36.2[plus or minus]3.3 [mu]g m[^-2] h[^-1]. In the K. humilis meadow, bulk soil contributed much more than plants and rhizosphere. The mean emission rate was 3.5[plus or minus]2.9, 5.7[plus or minus]3.8, and 30.6[plus or minus]2.5 [mu]g m[^-2] h[^-1] (P <0.001) from plants, rhizosphere and bulk soil, and these accounted for 9, 14 and 77%, separately. Our results implied that N[2]O emission rate decreased little with grazing as indicated by the difference between HCK and CBK in K. humilis meadow (P <0.05). N2O emission from alpine meadow could not be ignored in addressing regional greenhouse gases budget on the Tibetan Plateau, considering the vast area and much higher radiative forcing of N[2]O.
4
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Field measured N2O emissions in two years were used to parameterize and validate a process-based model, DNDC, for an alpine Kobresia humilis meadow on the Tibetan Plateau in China. Although this model failed to capture the N2O fluxes in some time periods in the spring or autumn, the modeled results showed overall a good performance in terms of simulating the seasonal variation of N2O fluxes and quantifying the annual total emissions. The relative deviation on the annual basis was about 12.4% and -15.9% for the two years, respectively. The modeled data showed that nitrification contributed about 53% of total N2O production, slightly higher than denitrification. The modeled fluxes were sensitive to soil organic content (SOC), pH, and temperature, but less sensitive to variation of precipitation, soil ammonium and nitrate contents. Further modifications for the model were suggested to focus on the process of soil freezing and thawing as well as the crop growth sub-model that would improve the model's performance for quantifying N2O emission from the alpine meadow.