The Visual/Kinesthetic Effects of Melodic Contour in Musical Notation as it Affects Vocal Timbre in Singers of Classical and Music Theater Repertoire

The purpose of this investigation was to obtain information from professional singers active in performing of both classical and music theater repertoire with regard to the visual/kinesthetic effect of melodic contour in musical notation as it affects vocal timbre. The evaluation of data gathered during the study indicates that there is reason to investigate the resulting postural shifts of the head and neck because they are guided by visual maps of melodic contour and its inversion. Significantly, it was discovered that the tone quality produced when a singer's head and neck postures followed the natural melodic contour of the melody was in no case considered to be the most pleasing of the conditions studied and that, in many cases, the tone quality produced when a singer's head and neck postures followed a map of the inversion of the melodic contour was judged to be much improved. The results of the study also indicate that the development of new technology for the general teaching studio, designed to address these effects, may be useful to these singers.     

苹果霉心病光谱在线检测的摆放姿态及建模方法优化研究

苹果营养丰富、口味酸甜,是深受大众喜爱的一种水果。苹果霉心病是一种真菌侵染果实病害,隐蔽性极强,一般在近成熟期果实内部发生霉变,肉眼从外观观察难以分辨,市面上大多数品种的苹果都受其影响。霉心病病果重量变轻、口感变差,严重的甚至不能食用,对经济效益的影响巨大。采用可见近红外光谱分析技术,使用微型光谱仪在线无损检测苹果霉心病,针对4种苹果在线输送时摆放姿态（竖放柄朝上、竖放柄朝下、横放柄朝输送方向和横放柄垂直输送方向）的判别效果进行了优化分析。首先使用主成分分析对600～900 nm波段的透射光谱提取主成分后分别建立线性判别分析（LDA）、马氏距离（MD）和K近邻法（KNN）模型并对校正集和预测集的判别准确率进行对比;其次对600～900 nm波段中心化预处理后建立偏最小二乘判别分析（PLS-DA）模型并给出4种摆放姿态的判别效果;最后使用两种机器学习算法极限学习机（ELM）和支持向量机（SVM）分别建立霉心病判别模型进行预测。对比上述所有6种判别模型,通过观察4种摆放姿态整体的判别效果得到最佳的建模方法为PLS-DA,其中竖放柄朝上和竖放柄朝下摆放的判别准确率都为93.75%,其他2种摆放姿态的判别准确率也都超过85%,再根据PLS-DA模型波段变量投影重要性指标得分值分布提取特征波段690～720 nm重新建立模型,对比4种摆放姿态效果最好的是竖放柄朝上摆放,其预测集的判别准确率达到93.75%,并且对病果的判别效果最佳。研究结果表明PLS-DA可以作为判别苹果霉心病一种有效方法,竖放柄朝上摆放可以作为苹果霉心病在线检测时一种有效姿态。     

Effects of Inhalatory Abdominal Wall Movement on Vertical Laryngeal Position During Phonation

The configuration of the body resulting from inhalatory behavior is sometimes considered a factor of relevance to voice production in singing and speaking pedagogy and in clinical voice therapy. The present investigation compares two different inhalatory behaviors: (1) with a "paradoxical" inward movement of the abdominal wall, and (2) with an expansion of the abdominal wall, both with regard to the effect on vertical laryngeal position during the subsequent phonation. Seventeen male and 17 female healthy, vocally untrained subjects participated. No instructions were given regarding movements of the rib cage. Inhaled air volume as measured by respiratory inductive plethysmography, was controlled to reach 70% inspiratory capacity. Vertical laryngeal position was recorded by two-channel electroglottography during the subsequent vowel production. A significant effect was found; the abdomen-out condition was associated with a higher laryngeal position than the abdomen-in condition. This result apparently contradicted a hypothesis that an expansion of the abdominal wall would allow the diaphragm to descend deeper in the torso, thereby increasing the tracheal pull, which would result in a lower laryngeal position. In a post-hoc experiment including 6 of the subjects, body posture was studied by digital video recordings, revealing that the two inhalatory modes were clearly associated with postural changes affecting laryngeal position. The "paradoxical" inward movement of the abdominal wall was associated with a recession of the chin toward the neck, such that the larynx appeared in a lower position in the neck, for reasons of a postural change. The results suggest that the laryngeal position can be affected by the inhalatory behavior if no attention is paid to posture, implying that instructions from clinicians and pedagogues regarding breathing behavior must be carefully formulated and adjusted in order to ensure that the intended goals are reached.     

基于激光测距仪全局匹配扫描的SLAM算法研究

针对传统的SLAM(Simultaneous Localization and Mapping)算法构建地图时容易受环境因素和外界条件的的影响,在非线性系统状态下误差修正能力不足,且当机器人位姿都处于未知状态时,移动机器人位姿获取不精确,地图构建SLAM技术特征量的获取比较繁琐、不准确等问题。以电力巡检机器人为平台,研究了基于全局匹配的扫描算法,摒弃传统的栅格地图模型的插值方法,采用双线性滤波的插值方法,保证子栅格单元的精确性,估算栅格占用函数的概率和导数。最后采用此算法解决了SLAM地图构建的问题,并分别在室内室外环境进行实验。实验结果表明：基于激光测距仪的全局匹配扫描的SALM算法,在室内室外两种不同环境下,不受复杂背景的影响,准确地进行机器人位姿定位,以及环境地图的构建