Semi-supervised learning uses underlying relationships in data with a scarcity of ground-truth labels. In this paper, we introduce an uncertainty quantification (UQ) method for graph-based semi-supervised multi-class classification problems. We not only predict the class label for each data point, but also provide a confidence score for the prediction. We adopt a Bayesian approach and propose a graphical multi-class probit model together with an effective Gibbs sampling procedure. Furthermore, we propose a confidence measure for each data point that correlates with the classification performance. We use the empirical properties of the proposed confidence measure to guide the design of a humanin-the-loop system. The uncertainty quantification algorithm and the human-in-the-loop system are successfully applied to classification problems in image processing and ego-motion analysis of body-worn videos.
The design of steering trapezoidal mechanism is one of the important aspects of the vehicle steering system. Every parameterin steering trapezoidal has significant influences on the steering performance, stability and tire service life of the vehicle.Based on the analysis of the relationship of the inside and outside wheel angle by analytic method, Matlab software can be used to design FSC racing steering trapezoidal mechanism. Considering the conditions of the automobile race,the corresponding parameter of the steering trapezoid is designed to make the relationship of the l wheel angle close to Ackermann geometry relationship, which reduces the wear of tires, ensuring good steering performance and holding the road so well.
In this paper, we have proposed a new way to detect lung nodules with image texture features. 104 cases of lung nodules, including 31 benign cases and 73 malignant cases, are collected, and the gray-scale correlation and texture heterogeneity are computed through CT imagings for all patients. We find that the gray correlation parameters are different between benign and malignant nodules. The heterogeneity parameters in malignant nodules are higher than that in benign noduals. The gray-scale texture correlation and heterogeneity parameters have diagnostic value in differentiating benign and malignant lung nodules. This study is an exploring study, which still needs further research.
In this paper, differences by using selected three typical specifications—DL/T 5141-2001 (Chinese), ASCE No.79 in the version of 1993(American) and ASCE No.79 in the version of 2012 (American)—in structural design of exposed steel penstock were explored. A practical example about exposed penstock laid on ground applied in hydropower station was also used to analyze specifications clearly. The result shows that the main differences between Chinese and American specifications are allowable stress and load combination. The former is direct factor of calculating exposed penstock shell thickness. Therefore, ASCE No.79 (2012) designs the minimum wall thickness, followed by DL/T 5141-2001 and the last is ASCE No.79 (1993), which is correspondingly contrary to sort by allowable stress. Basically, ASCE No.79 (2012) defines lower safety coefficient for exposed penstock, which is identical with authoritative rules of pressure vessels in the U.S.A and EU. The safety of DL/T 5141-2001 has been proved via rich engineering experience and this specification is recommended for Chinese projects. Besides, ASCE No.79 (2012) is recommended.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.