Automatic annotation of tennis games: An integration of audio,vision, and learning

Fully automatic annotation of tennis game using broadcast video is a task with a great potential but with enormous challenges. In this paper we describe our approach to this task, which integrates computer vision, machine listening, and machine learning. At the low level processing, we improve upon our previously proposed state-of-the-art tennis ball tracking algorithm and employ audio signal processing techniques to detect key events and construct features for classifying the events. At high level analysis, we model event classification as a sequence labelling problem, and investigate four machine learning techniques using simulated event sequences. Finally, we evaluate our proposed approach on three real world tennis games, and discuss the interplay between audio, vision and learning. To the best of our knowledge, our system is the only one that can annotate tennis game at such a detailed level.     

Experimental Study of an EMG-Controlled 5-DOF Anthropomorphic Prosthetic Hand for Motion Restoration

In this paper, we attempted to evaluate the performance of an electromyography (EMG)-controlled 5-DOF prosthetic hand on ten transradial amputees. The proposed prosthesis is composed of a five-fingered hand, a passive wrist, and a customized socket for each subject. The EMG control methods included both a commonly used pattern recognition-based scheme (DD-SVM) and a novel digital encoding strategy (double-channel template matching (DCTM)). A virtual 3D hand platform was developed for training the subjects and rapidly testing the control methods. For each subject, the performance of the EMG control methods was firstly measured by off-line classification accuracy; then, according to the accuracy, a particular control method was selected and embedded in the EMG controller for further validation on ordinary daily life activities. Our experiments were conducted to test not only the hand’s grasp ability but also other multifinger cooperation skills. The result indicated that the subjects of rich control experience can accomplish several intuitive motion control over their hands. However, the kinds of the motions and their relative recognition accuracy may depend on some individual differences, such as the amputation level, the activity of the residual nerve-muscle system, and the richness of control experience. Meanwhile, the proposed digital encoding method, DCTM, which only utilized two channels of EMG, was necessary for those amputees with few available control signals. This paper suggested that the EMG control method should be differently considered according to the particular condition of each subject.     

Toward seed-insensitive solutions to local community detection

Local community detection aims at finding a community structure starting from a seed which is a given vertex in a network without global information, such as online social networks that are too large and dynamic to ever be known fully. Nonetheless, the existing approaches to local community detection are usually sensitive to seeds, i.e., some seeds may lead to missing of some true communities. In this paper, we present a seed-insensitive method called GMAC and its variation iGMAC for local community detection. They estimate the similarity among vertices by investigating vertices’ neighborhoods, and reveal a local community by maximizing its internal similarity and minimizing its external similarity simultaneously. Extensive experimental results on both synthetic and real-world data sets verify the effectiveness of our algorithms.     

过氧乙酸的合成与杀生作用研究

以乙酸和过氧化氢为原料合成了新型水处理杀生剂——过氧乙酸．并考察了过氧乙酸的贮存稳定性及稳定剂8-羟基喹啉的影响。通过细菌测试瓶法测细菌总数，研究了过氧乙酸杀生效果与投加浓度、pH值和作用时间之间的关系。结果表明：过氧乙酸作为水处理剂具有杀生速度快、杀生效果好的特点，是一种具有良好应用前景的水处理杀生剂。     

填料对反-1,4-聚异戊二烯性能的影响

研究了填料对反－1，4－聚异戊二烯（TPI）硫化胶和混炼胶的物理机械性能和结晶性能的影响。结果表明，在TPI中加入填料后，其硫化胶的300％定伸应力随填料用量的增加而增加，拉伸强度则在填料用量为20份时达到最大值；加入填料同时能降低TPI混炼胶的结晶度，填料用量增加，TPI混炼胶的结晶下降；且填料增强效果越好，结晶度下降越大。     

溶胶-凝胶法制备多孔陶瓷的研究进展

本文综合论述了溶胶-凝胶法制备多孔陶瓷的工艺研究进展以及影响溶胶-凝胶法制备多孔陶瓷的主要因素，并讨论了今后溶胶-凝胶法制备多孔陶瓷的主要研究问题及其发展前景。     

Relative image similarity learning with contextual information for Internet cross-media retrieval

With the fast explosive rate of the amount of image data on the Internet, how to efficiently utilize them in the cross-media scenario becomes an urgent problem. Images are usually accompanied with contextual textual information. These two heterogeneous modalities are mutually reinforcing to make the Internet content more informative. In most cases, visual information can be regarded as an enhanced content of the textual document. To make image-to-image similarity being more consistent with document-to-document similarity, this paper proposes a method to learn image similarities according to the relations of the accompanied textual documents. More specifically, instead of using the static quantitative relations, rank-based learning procedure by employing structural SVM is adopted in this paper, and the ranking structure is established by comparing the relative relations of textual information. The learning results are in more accordance with the human’s recognition. The proposed method in this paper can be used not only for the image-to-image retrieval, but also for cross-modality multimedia, where a query expansion framework is proposed to get more satisfactory results. Extensive experimental evaluations on large scale Internet dataset validate the performance of the proposed methods.     

Memristor crossbar-based unsupervised image learning

This letter presents a new memristor crossbar array system and demonstrates its applications in image learning. The controlled pulse and image overlay technique are introduced for the programming of memristor crossbars and promising a better performance for noise reduction. The time-slot technique is helpful for improving the processing speed of image. Simulink and numerical simulations have been employed to demonstrate the useful applications of the proposed circuit structure in image learning.     

改性低毒脲醛树脂的合成及性能研究

采用六亚甲基四胺做催化剂，加入聚乙烯醇（PVA）、三聚氰胺和硫脲作为改性剂，通过降低甲醛与尿素物质的量比、采用尿素分批加入的合成方法，制备出改性低毒的脲醛树脂（UF），使游离甲醛质量分数低于0．30％；采用捏合和模压成型方法，将改性和普通UF树脂分别与剑麻纤维（SF）制备UF／SF共混复合材料，研究了复合材料的力学性能和吸水性能。结果表明，甲醛和尿素物质的量比为1．1时，游离甲醛质量分数降低到0．16％；甲醛和尿素物质的量比为1．2时，三聚氰胺、PVA及硫脲加入量分别为尿素总量的2．0％，1．5％，1．0％（质量分数）时，游离甲醛质量分数降低到了0．10％，粘度达到0．41Pa·S，贮存期大于80d；改性UF／SF复合材料的性能比普通UF／SF复合材料的冲击性能和耐磨性能分别提高了114％和261％。