Search

As shield tunnels increase, managing shield muck strains construction and the environment. To mitigate this problem, shield muck replaced bentonite in silty clay to improve synchronous grouting slurry. Initially, the physical attributes and microstructural composition of shield muck were obtained, alongside an analysis of the effects of the muck content, particle size, and general influencing factors on the slurry properties through standardized tests and regression models. Subsequently, leveraging three-dimensional response surface methodology, admixture interactions and multiple factor impacts on the slurry were explored. Finally, utilizing the SQP optimization technique, an optimal slurry blend ratio tailored for actual project needs was derived for improved muck slurry. The findings reveal with the decreasing bleeding rates as the muck content rises, the particle size diminishes. An inverse relationship exists between the muck content and slurry fluidity. At soil–binder ratios below 0.6, a decrease in the soil–binder ratio intensifies the influence of the water–binder ratio on the slurry density, bleeding rate, and setting time. The fly flash–cement ratio inversely correlates with the slurry bleeding rate, while the ratio greater than 0.6 is positively correlated. For muck particle sizes under 0.2 mm, the fly flash–cement ratio inversely impacts the density, while over 0.2 mm, it correlates positively. The optimal proportion for silty clay stratum synchronous grouting slurry, substituting muck for bentonite, includes a water–binder ratio of 0.559, binder–sand ratio of 0.684, fly flash–cement ratio of 2.080, soil–binder ratio of 0.253, particle size under 0.075 mm, and water-reducing admixture of 0.06. [ABSTRACT FROM AUTHOR]

This work aims to address two issues that often exist in least square regression (LSR) models for classification tasks, which are (1) learning a compact projection matrix for feature selection and (2) adopting relaxed regression targets. To this end, we first propose a sparse regularized LSR framework for feature selection by introducing the ℓ 2 , 1 regularizer. Second, we utilize two different strategies to relax the strict regression targets based on the sparse framework. One way is to exploit the -dragging technique. Another strategy is to directly learn the labels from the inputs and constrain the distance between true and false classes simultaneously. Hence, more feasible regression schemes are constructed, and the models will be more flexible. Further, efficient iterative methods are derived to optimize the proposed models. Various experiments on image databases intend to manifest our proposed models have outstanding recognition capability compared with many state-of-the-art classifiers. [ABSTRACT FROM AUTHOR]

Due to the low computational complexity, the two-phase test sample representation method shows outstanding advantages in the field of real-time face recognition. However, its first phase does not fully consider the imbalance of the determined K-nearest training samples. This defect is not well reflected and will seriously affect the recognition accuracy of the second phase. In this paper, we explore the above issues and propose to incorporate the unselected training samples into the modeling process. The proposed method not only allows the unselected samples to perform the final representation but also makes the selected nearest training samples play a more significant role than the unselected ones in classification. It is efficiently solved with an analytical solution. The rationales and probability interpretation of the proposed method are presented to further guarantee its rationality and effectiveness. Extensive experiments on diverse face databases are conducted to demonstrate the superior recognition performance in comparison with other competing classifiers. [ABSTRACT FROM AUTHOR]

With the development of machine learning and computer vision, classification technology is becoming increasingly important. Due to the advantage in efficiency and effectiveness, collaborative representation-based classifiers (CRC) have been applied to many practical cognitive fields. In this paper, we propose a new neighborhood prior constrained collaborative representation model for pattern classification. Compared with the naive CRC models which approximate the test sample with all the training data globally, our proposed methods emphasize the guidance of the neighborhood priors in the coding process. Two different kinds of neighbor priors and the models' weighted extensions are explored from the view of sample representation ability and relationships between the samples. Consequently, the contributions of different samples can be distinguished adaptively and the obtained representations can be more discriminative for the recognition. Experimental results on several popular databases can verify the effectiveness of our proposed methods in comparison with other state-of-the-art classifiers. [ABSTRACT FROM AUTHOR]

Broad learning system (BLS), which emerges as a lightweight network paradigm, has recently attracted great attention for recognition problems due to its good balance between efficiency and accuracy. However, the supervision mechanism in BLS and its variants generally relies on the strict binary label matrix, which imposes limitations on approximation and fails to adequately align with the data distribution. To address this issue, in this article, two novel flexible label-induced BLS models with the manifold manner are proposed, whose notable characteristics are as follows. First, two proposed label relaxation strategies can both enlarge the margins between different categories and simultaneously enhance the diversity within labels. Second, the integration of manifold geometrical criterion enables the models to capture local feature structures, ensuring the obtained flexible labels align better with the similarity between samples. Third, the proposed models can be optimized efficiently with the alternating direction method of multipliers. Each iteration benefits from a closed-form solution, facilitating the optimization process. Extensive experiments and thorough theoretical analysis are intended to show the advantages of our proposed models compared to other state-of-the-art recognition algorithms.