Gesture recognition aims to recognize meaningful movements of human bodies, and is of utmost importance in intelligent human-computer/robot interactions. In this paper, we present a multimodal ...gesture recognition method based on 3-D convolution and convolutional long-short-term-memory (LSTM) networks. The proposed method first learns short-term spatiotemporal features of gestures through the 3-D convolutional neural network, and then learns long-term spatiotemporal features by convolutional LSTM networks based on the extracted short-term spatiotemporal features. In addition, fine-tuning among multimodal data is evaluated, and we find that it can be considered as an optional skill to prevent overfitting when no pre-trained models exist. The proposed method is verified on the ChaLearn LAP large-scale isolated gesture data set (IsoGD) and the Sheffield Kinect gesture (SKIG) data set. The results show that our proposed method can obtain the state-of-the-art recognition accuracy (51.02% on the validation set of IsoGD and 98.89% on SKIG).
Synthesis of inorganic shell microcapsules is very challenging in the field of self-healing cementitious materials. Here, we report successful fabrication of epoxy resins encapsulated into silica ...shell microcapsules with average particle size greater than 100 μm via interfacial polymerization. The silica shell microcapsules show good thermal stability. The nanoindentation test was used to determine the micromechanical properties of the microcapsules. The results show that silica shell microcapsules exhibit brittle rupture, high Young’s modulus and hardness of the shell, with a better deformation resistance. The observed micromechanical properties are attributed to the regular tetrahedral crystal structure of the silica shell materials.
Many clinical applications based on deep learning and pertaining to radiology have been proposed and studied in radiology for classification, risk assessment, segmentation tasks, diagnosis, ...prognosis, and even prediction of therapy responses. There are many other innovative applications of AI in various technical aspects of medical imaging, particularly applied to the acquisition of images, ranging from removing image artifacts, normalizing/harmonizing images, improving image quality, lowering radiation and contrast dose, and shortening the duration of imaging studies. This article will address this topic and will seek to present an overview of deep learning applied to neuroimaging techniques.
Continuous human action recognition (CHAR) is more practical in human-robot interactions. In this paper, an online CHAR algorithm is proposed based on skeletal data extracted from RGB-D images ...captured by Kinect sensors. Each human action is modeled by a sequence of key poses and atomic motions in a particular order. In order to extract key poses and atomic motions, feature sequences are divided into pose feature segments and motion feature segments, by use of the online segmentation method based on potential differences of features. Likelihood probabilities that each feature segment can be labeled as the extracted key poses or atomic motions, are computed in the online model matching process. An online classification method with variable-length maximal entropy Markov model (MEMM) is performed based on the likelihood probabilities, for recognizing continuous human actions. The variable-length MEMM method ensures the effectiveness and efficiency of the proposed CHAR method. Compared with the published CHAR methods, the proposed algorithm does not need to detect the start and end points of each human action in advance. The experimental results on public datasets show that the proposed algorithm is effective and highly-efficient for recognizing continuous human actions.
The synthesis of inorganic shell microcapsules is challenging in the field of self-healing cementitious materials. In this study, cement shell microcapsules were synthesized via hydration reaction of ...cement particles at the water-oil interface by using water-in-oil (W/O) Pickering emulsion method. The shell of the microcapsules is integrated with the cementitious materials, and incorporation of the microcapsules slightly effects the mechanical properties of cementitious materials.
Due to the development of the computer vision, machine learning, and deep learning technologies, the research community focuses not only on the traditional SLAM problems, such as geometric mapping ...and localization, but also on semantic SLAM. In this paper, we propose a Semantic SLAM system which builds the semantic maps with object-level entities, and it is integrated into the RGB-D SLAM framework. The system combines object detection module that is realized by the deep-learning method, and localization module with RGB-D SLAM seamlessly. In the proposed system, object detection module is used to perform object detection and recognition, and localization module is utilized to get the exact location of the camera. The two modules are integrated together to obtain the semantic maps of the environment. Furthermore, to improve the computational efficiency of the framework, an improved Octomap based on the Fast Line Rasterization Algorithm is constructed. Meanwhile, for the sake of accuracy and robustness of the semantic map, conditional random field is employed to do the optimization. Finally, we evaluate our Semantic SLAM through three different tasks, i.e., localization, object detection, and mapping. Specifically, the accuracy of localization and the mapping speed is evaluated on TUM data set. Compared with ORB-SLAM2 and original RGB-D SLAM, our system, respectively, got 72.9% and 91.2% improvements in dynamic environments localization evaluated by root-mean-square error. With the improved Octomap, the proposed Semantic SLAM is 66.5% faster than the original RGB-D SLAM. We also demonstrate the efficiency of object detection through quantitative evaluation in an automated inventory management task on a real-world data sets recorded over a realistic office.
Pneumatic muscle actuators are widely used in the manufacture of bionic robots and rehabilitation medical equipment. However, due to complicated inherent nonlinearities, time-varying characteristics ...and uncertainties, it is still a challenge to carry out the accurate dynamic modeling and controller design for PAM systems. To address above issues, we propose an error tracking-based neuro-adaptive iterative learning control scheme to get satisfactory non-uniform angle trajectory tracking performance. First, the error-tracking method is used to overcome the nonzero initial state error in iterative learning controller design for the PAM system. Second, a difference-learning neural network is utilized to compensate for unknown uncertainties in the PAM system dynamics. Moreover, a barrier Lyapunov function is applied to design controller so as to restrict the the difference between system out error and the desired error trajectory within the preset bound during each iteration. And the stability of the closed-loop system is proven theoretically by using Lyapunov synthesis. Finally, simulation results demonstrate the effectiveness of the proposed control scheme.
Abstract
Cluster of differentiation 24 (CD24), a mucin-like highly glycosylated molecule has been extensively studied as a cancer stem cell marker in a variety of solid cancers. The functional role ...of CD24 is either fulfilled by combining with ligands or participating in signal transduction, which mediate the initiation and progression of neoplasms. Recently, CD24 was also described as an innate immune checkpoint with apparent significance in several types of solid cancers. Herein, we review the current understanding of the molecular fundamentals of CD24, the role of CD24 in tumorigenesis and cancer progression, the possibility as a promising therapeutic target and summarized different therapeutic agents or strategies targeting CD24 in solid cancers.
•Uniformly dispersed GO/cyanate ester (CE)–epoxy (EP) composites were successfully synthesized by in situ polymerization.•The reaction mechanisms were discussed and were testified by FT-IR and ...XPS.•GO reinforced EP-CE composites demonstrated improved mechanical properties and better thermal stability than that of EP-CE matrices.
Uniformly dispersed graphene oxide (GO)/cyanate ester (CE)–epoxy (EP) composites were successfully synthesized by in situ polymerization. Both the results of FT-IR and XPS verified that epoxide groups on the GO sheets reacted with cyanate group (OCN) in the resin. These results could provide excellent dispersion of GO and strong interfacial interaction between GO and CE matrix. TEM confirmed that GO tended to be a single layer. XRD and SEM indicated that matrix molecules could be inserted into the interplanar spacing of GO. The mechanical properties and thermal behavior of the composites were investigated in detail. It is observed that GO reinforced EP-CE composites demonstrated improved mechanical properties and better thermal stability than that of EP-CE matrix, which make them suitable for use in aerospace applications and structural composites.
•The composites prepared by solution mixing showed the best thermoelectric properties.•The Seebeck coefficient slightly fluctuates with increasing MWNT content.•The electrical conductivity increases ...remarkably with increasing MWNT content.•The highest ZT of 8.71×10−4 was obtained with 80wt.% MWNT.
Different polythiophene (PTh)/multiwall carbon nanotube (MWNT) composites with 30wt.% and 50wt.% MWNT were prepared by mechanical ball milling, solution mixing and in situ composite, respectively. The composites prepared by solution mixing showed the best thermoelectric properties among these methods. Therefore, the morphology, internal structure and thermal stability of the composites by solution mixing were evaluated by SEM, XRD, FTIR and TGA. The results showed that the MWNT were uniformly dispersed in the polymer matrix and the composite materials exhibited good thermal stability under 200°C. The effect of MWNT content in the composites on thermoelectric properties, such as electrical conductivity, Seebeck coefficient and thermal conductivity were investigated. With increasing MWNT content, the Seebeck coefficient slightly fluctuates, varying from 27.7 to 22.7μV/K, and the thermal conductivity slightly increases, but the electrical conductivity increases remarkably, and thus leads to enhance the figure of merit (ZT) obviously. The highest ZT of 8.71×10−4 at 120°C was found in the composite with 80wt.% MWNT.