Level of interest sensing in spoken dialog using decision-level fusion of acoustic and lexical evidence

Automatic detection of a user's interest in spoken dialog plays an important role in many applications, such as tutoring systems and customer service systems. In this study, we propose a decision-level fusion approach using acoustic and lexical information to accurately sense a user's interest at the utterance level. Our system consists of three parts: acoustic/prosodic model, lexical model, and a model that combines their decisions for the final output. We use two different regression algorithms to complement each other for the acoustic model. For lexical information, in addition to the bag-of-words model, we propose new features including a level-of-interest value for each word, length information using the number of words, estimated speaking rate, silence in the utterance, and similarity with other utterances. We also investigate the effectiveness of using more automatic speech recognition (ASR) hypotheses (n-best lists) to extract lexical features. The outputs from the acoustic and lexical models are combined at the decision level. Our experiments show that combining acoustic evidence with lexical information improves level-of-interest detection performance, even when lexical features are extracted from ASR output with high word error rate.     

Personalized identification of abdominal wall hernia meshes on computed tomography

An abdominal wall hernia is a protrusion of the intestine through an opening or area of weakness in the abdominal wall. Correct pre-operative identification of abdominal wall hernia meshes could help surgeons adjust the surgical plan to meet the expected difficulty and morbidity of operating through or removing the previous mesh. First, we present herein for the first time the application of image analysis for automated identification of hernia meshes. Second, we discuss the novel development of a new entropy-based image texture feature using geostatistics and indicator kriging. Third, we seek to enhance the hernia mesh identification by combining the new texture feature with the gray-level co-occurrence matrix feature of the image. The two features can characterize complementary information of anatomic details of the abdominal hernia wall and its mesh on computed tomography. Experimental results have demonstrated the effectiveness of the proposed study. The new computational tool has potential for personalized mesh identification which can assist surgeons in the diagnosis and repair of complex abdominal wall hernias.     

一种具有发展前景的饲料磷酸盐——脱氟磷酸三钙

脱氟磷酸三钙由于含钙量高、钙磷比适合于喂养家禽，是良好的饲料磷酸盐。它是由磷矿和其它添加物团粒，经高温焙烧后制成。具有工艺流程简单、生产成本低等优点。可结合本地情况，发展生产。     

硬质PVC挤出发泡材料的塑化性能研究

采用Haake转矩流变仪测定了挤出发泡PVC混合料的恒温、升温熔融塑化性能，结果表明：随着PVC树脂聚合度增加，塑化时间和加工转矩增加；随着ACR加工助剂用量和分子质量的增加，塑化时间缩短，转矩增加；ACR抗冲改性剂具有类似ACR加工助剂的塑化改良行为；添加填充剂碳酸钙促进PVC熔融塑化；ADC发泡剂可延缓PVC混合料的塑化速率、提高熔体粘度，添加发泡剂NaHCO3将大大延缓混合料的塑化速率。     

基于自适应分块的动态场景HDR图像合成算法

通过提出基于梯度上升优化处理的自适应分块算法与基于图像序列梯度域运动检测与去除相结合的方法,自适应调整分块大小及动态区域块的融合权值,最终达到混淆去除的目的.同时,利用Gaussian中心函数窗口滤波,去除在分块融合过程中引入的块边缘不连续性痕迹.实验结果表明,该方法能快速有效地增强多曝光图像融合生成HDRI并去除混淆问题.     

局部颜色特征与工艺参数融合的铜浮选入矿类型识别

针对铜浮选过程中矿源变化频繁而造成的浮选工况难以识别、生产过程不稳定等问题,在分析铜浮选流程特点的基础上,提出了一种局部颜色特征与关键工艺参数融合的铜浮选入矿类型识别方法.首先采用多元图像处理方法提取泡沫图像局部颜色特征,然后采用混合滤波、时间配准算法将其与关键工艺参数进行数据融合,最后用融合后的混合特征进行模糊聚类,得到隶属度矩阵和聚类中心.工业数据仿真结果验证了本文所提方法的有效性.     

结合小波变换的梯度场图像渐进融合算法

基于梯度场的图像融合算法只适用于尺度差异不大于1∶4的多光谱图像与全色图像。针对尺度差异为1∶8的北京一号卫星多光谱图像及高分辨率全色图像的融合问题,提出一种结合小波变换的梯度场图像尺度渐进融合算法。利用小波变换方法将多光谱图像与高分辨率全色图像尺度差异倍数缩小,得到基于小波变换的初级融合,再进行基于梯度场的Poisson图像融合。实验结果表明,渐进融合图像与多光谱图像的平均颜色差异值为23.5,与高分辨率全色图像的平均梯度差异值为2.1,多尺度纹理特征值差异值分别为3.98、10.2、18.9,渐进融合图像与高分辨率全色图像的空间细节和纹理细节吻合程度更好。     

Bioactive IL7-diphtheria fusion toxin secreted by mammalian cells

A number of targeted cytotoxic agents have been developed that selectively kill malignant or otherwise pathological cells. These engineered proteins consist of a potent cytotoxic element connected to a ligand domain that binds to specific molecules on the surface of the target cell. Several of these agents have shown promise in clinical trials and one is currently administered to patients. A significant technical obstacle that has impeded the development of some of these toxins is the difficulty of preparing certain recombinant proteins in properly folded forms. These fusion proteins have generally been produced in bacteria requiring them to be denatured and renatured in vitro. For some proteins this is an efficient process whereas for others it is not. We describe here a system to produce fusion toxins rapidly and efficiently by engineering mammalian cells to secrete them as properly folded molecules which can be purified in native form from cell culture medium. We have used this system to produce highly active preparations of DAB(389)-IL7, a molecule consisting of the catalytic and transmembrane domains of diphtheria toxin fused to interleukin 7. This system is generalizable and can be used to produce and evaluate rapidly fusion toxins incorporating novel or uncharacterized ligands.     

基于无线传感网的智能环境监控系统设计

针对传统环境监控系统存在的缺点,设计了一种新型无线传感器网络的环境监控系统。系统采用分布式融合结构,通过终端节点的多种传感器采集环境信息,终端对信息预处理后实时将信息传输到协调器。协调器融合各终端传感器信息,提取精确信息上传至融合中心,融合中心通过分析数据发布命令给控制中心,控制中心利用单片机驱动电机可以自动或手动控制风机和水泵对环境进行调节和控制。同时,系统还具有自动预警功能。通过验证,该系统稳定、可靠,能广泛应用于不同领域的分布式多点环境监控。