Coreference resolution plays an important role in text understanding. In the literature, various neural approaches have been proposed and achieved considerable success. However, structural information, which has been proven useful in coreference resolution, has been largely ignored in previous neural approaches. In this paper, we focus on effectively incorporating structural information to neural coreference resolution from three aspects. Firstly, nodes in the parse trees are employed as a constraint to filter out impossible text spans (i.e., mention candidates) in reducing the computational complexity. Secondly, contextual information is encoded in the traversal node sequence instead of the word sequence to better capture hierarchical information for text span representation. Lastly, additional structural features (e.g., the path, siblings, degrees, category of the current node) are encoded to enhance the mention representation. Experimentation on the data-set of the CoNLL 2012 Shared Task shows the effectiveness of our proposed approach in incorporating structural information into neural coreference resolution.
Tactile sensing plays a crucial role in robot manipulation, robot interaction, and health monitoring. Because of high sensitivity, simple structure, and superior interference immunity, optical tactile sensors based on optical imaging or optical conduction have been one of the most active research. Herein, a novel liquid lens-based optical sensor (LLOS) is presented. Different with existed optical tactile sensors, the main body of the proposed sensor belongs to a variable-focus optical lens with a liquid-membrane structure, and its focal length is changed with the contact force, thereby changing the propagation direction of light and affecting the perceived light intensity of the photosensitive element. By conducting some testing experiments, the LLOS demonstrates fast response (about 0.021s), stable dynamic response characteristics, and good linearity (R-squared is about 0.99), repeated measurement accuracy(<0.006V), and measurement accuracy (< 0.2N). Hence, the LLOS provides a new and promising method to measure tactile and has potential application in robotics nondestructive grasping and interactive input devices.
Phase contrast computed tomography (PCCT) provides an effective non-destructive testing tool for weak absorption objects. Limited by the phase stepping principle and radiation dose requirement, sparse-view sampling is usually performed in PCCT, introducing severe artifacts in reconstruction. In this paper, we report a dual-domain (i.e., the projection sinogram domain and image domain) enhancement framework based on deep learning (DL) for PCCT with sparse-view projections. It consists of two convolutional neural networks (CNN) in dual domains and the phase contrast Radon inversion layer (PCRIL) to connect them. PCRIL can achieve PCCT reconstruction, and it allows the gradients to backpropagate from the image domain to the projection sinogram domain while training. Therefore, parameters of CNNs in dual domains are updated simultaneously. It could overcome the limitations that the enhancement in the image domain causes blurred images and the enhancement in the projection sinogram domain introduces unpredictable artifacts. Considering the grating-based PCCT as an example, the proposed framework is validated and demonstrated with experiments of the simulated datasets and experimental datasets. This work can generate high-quality PCCT images with given incomplete projections and has the potential to push the applications of PCCT techniques in the field of composite imaging and biomedical imaging.
The distribution of ranging errors of time of arrival techniques fails to satisfy zero means and equal variances. It is one of the major causations of position error of least square-based localization algorithm. The optimization of time of arrival ranging is defined as a nonlinear programming problem. Then, time of arrival ranging error model and geometric constraints are used to define the initial values, objective functions, and constraints of nonlinear programming, as well as to detect line of sight and nonline of sight. A three-dimensional localization algorithm of an indoor time of arrival-based positioning is proposed based on least square and the optimization algorithm. The performance of the ranging and localization accuracies is evaluated by simulation and field testing. Results show that the optimized ranging error successfully satisfies zero mean value and equal variances. Furthermore, the ranging and localization accuracies are significantly improved.
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.