近年来,作为重要的多目标决策手段的轮廓查询逐渐得到学术界的重视,相继提出了基于不同支配关系的多种轮廓变体查询.首先,通过对实际应用需求进行分析,提出了基于元组对应数值间比例值大小的ρ-支配关系的定义,进而提出了ρ-支配轮廓查...近年来,作为重要的多目标决策手段的轮廓查询逐渐得到学术界的重视,相继提出了基于不同支配关系的多种轮廓变体查询.首先,通过对实际应用需求进行分析,提出了基于元组对应数值间比例值大小的ρ-支配关系的定义,进而提出了ρ-支配轮廓查询的概念.其次,对ρ-支配轮廓的基本性质进行了细致而深入的分析,在此基础上,提出了基于分支定界的ρ-支配轮廓查询算法(Branch and Boundρ-Dominant Skyline Algorithm,BBDS),避免了对R-树索引的多次访问,从而提高了ρ-支配轮廓查询的执行效率.最后,通过大量的仿真实验对ρ-支配轮廓查询的语义进行分析,并对BBDS算法的性能进行验证.实验结果表明,ρ-支配轮廓查询是轮廓查询语义的扩展和补充,而提出的BBDS算法则是求解ρ-支配轮廓查询的高效算法.展开更多
This paper introduces a novel method for estimating the skeleton proportions of a human figure from monocular data. The proposed system will first automatically extract the key frames and recover the perspective camer...This paper introduces a novel method for estimating the skeleton proportions of a human figure from monocular data. The proposed system will first automatically extract the key frames and recover the perspective camera model from the 2D data. The human skeleton proportions are then estimated from the key frames using the recovered camera model without posture re- construction. The proposed method is tested to be simple, fast and produce satisfactory results for the input data. The human model with estimated proportions can be used in future research involving human body modeling or human motion reconstruction.展开更多
The hard tissues of squid can provide important information for species identification. In this study, we used statolith and beak to identify three squid species including Uroteuthis duvaucelii, Loliolus beka, and U. ...The hard tissues of squid can provide important information for species identification. In this study, we used statolith and beak to identify three squid species including Uroteuthis duvaucelii, Loliolus beka, and U. edulis in the South China Sea. Because of the highly overlapping habitat and similar body morphology of the three squid species, we explored four different ways to identify them, by using statolith, upper beak, lower beak and a combination of statolith and beak. An outline geometric morphometric method and stepwise discriminant analysis were used to evaluate the most suitable method for the identification. We found that the combination of statolith and beak had the highest cross validation rate that was 75.0%, 87.5% and 88.7% for U. duvaucelii, L. beka and U. edulis, respectively. Using two beaks had similar results and the lowest cross validation rate was 60.0%, 50.0%, and 73.7% for the upper beak, 46.9%, 58.5% and 75.3% for the lower beak of U. duvaucelii, L. beka and U. edulis, respectively. Analyzing with the statolith had moderate cross validation which was 72.2%, 80.0%, and 87.7% for U. duvaucelii, L. beka and U. edulis, respectively. From the results it is suggested when the entire body of a squid is available, a combination of statolith and beak should be used for the identification. When only one hard tissue is available, species identification can be subjected to large errors.展开更多
A new method of elastic articulated objects (human bodies) modeling was presented based on a new conic curve. The model includes 3D object deformable curves which can represent the deformation of human occluding conto...A new method of elastic articulated objects (human bodies) modeling was presented based on a new conic curve. The model includes 3D object deformable curves which can represent the deformation of human occluding contours. The deformation of human occluding contour can be represented by adjusting only four deformation parameters for each limb. Then, the 3D deformation parameters are determined by corresponding 2D contours from a sequence of stereo images. The algorithm presented in this paper includes deformable conic curve parameters determination and the plane, 3D conic curve lying on, parameter determination.展开更多
MGAC (Motion Geometric Active Contours), a new variational framework of geometric active contours to track multiple nonrigid moving objects in the clutter background in image sequences is presented. This framework, in...MGAC (Motion Geometric Active Contours), a new variational framework of geometric active contours to track multiple nonrigid moving objects in the clutter background in image sequences is presented. This framework, incorporating with the motion edge information, consists of motion detection and tracking stages. At the motion detection stage, the motion edge map provides an approximate edge map of the moving objects. Then, a tracking stage, merely using the static edge information, is considered to improve the motion detection result. Force field regularization method is used to extend the capture range of the edge attraction force field in both stages. Experiments demonstrate that the proposed framework is valid for tracking multiple nonrigid objects in the clutter background.展开更多
文摘近年来,作为重要的多目标决策手段的轮廓查询逐渐得到学术界的重视,相继提出了基于不同支配关系的多种轮廓变体查询.首先,通过对实际应用需求进行分析,提出了基于元组对应数值间比例值大小的ρ-支配关系的定义,进而提出了ρ-支配轮廓查询的概念.其次,对ρ-支配轮廓的基本性质进行了细致而深入的分析,在此基础上,提出了基于分支定界的ρ-支配轮廓查询算法(Branch and Boundρ-Dominant Skyline Algorithm,BBDS),避免了对R-树索引的多次访问,从而提高了ρ-支配轮廓查询的执行效率.最后,通过大量的仿真实验对ρ-支配轮廓查询的语义进行分析,并对BBDS算法的性能进行验证.实验结果表明,ρ-支配轮廓查询是轮廓查询语义的扩展和补充,而提出的BBDS算法则是求解ρ-支配轮廓查询的高效算法.
文摘This paper introduces a novel method for estimating the skeleton proportions of a human figure from monocular data. The proposed system will first automatically extract the key frames and recover the perspective camera model from the 2D data. The human skeleton proportions are then estimated from the key frames using the recovered camera model without posture re- construction. The proposed method is tested to be simple, fast and produce satisfactory results for the input data. The human model with estimated proportions can be used in future research involving human body modeling or human motion reconstruction.
基金the National Natural Science Foundation of China (No. NSFC41476129)the Shanghai Leading Academic Discipline Project (Fisheries Discipline)supported by Shanghai Ocean University International Center for Marine Studies and Shanghai 1000 Talents Program
文摘The hard tissues of squid can provide important information for species identification. In this study, we used statolith and beak to identify three squid species including Uroteuthis duvaucelii, Loliolus beka, and U. edulis in the South China Sea. Because of the highly overlapping habitat and similar body morphology of the three squid species, we explored four different ways to identify them, by using statolith, upper beak, lower beak and a combination of statolith and beak. An outline geometric morphometric method and stepwise discriminant analysis were used to evaluate the most suitable method for the identification. We found that the combination of statolith and beak had the highest cross validation rate that was 75.0%, 87.5% and 88.7% for U. duvaucelii, L. beka and U. edulis, respectively. Using two beaks had similar results and the lowest cross validation rate was 60.0%, 50.0%, and 73.7% for the upper beak, 46.9%, 58.5% and 75.3% for the lower beak of U. duvaucelii, L. beka and U. edulis, respectively. Analyzing with the statolith had moderate cross validation which was 72.2%, 80.0%, and 87.7% for U. duvaucelii, L. beka and U. edulis, respectively. From the results it is suggested when the entire body of a squid is available, a combination of statolith and beak should be used for the identification. When only one hard tissue is available, species identification can be subjected to large errors.
基金the Postdoctoral Science Foundation of China(Grant No.20070421018)
文摘A new method of elastic articulated objects (human bodies) modeling was presented based on a new conic curve. The model includes 3D object deformable curves which can represent the deformation of human occluding contours. The deformation of human occluding contour can be represented by adjusting only four deformation parameters for each limb. Then, the 3D deformation parameters are determined by corresponding 2D contours from a sequence of stereo images. The algorithm presented in this paper includes deformable conic curve parameters determination and the plane, 3D conic curve lying on, parameter determination.
文摘MGAC (Motion Geometric Active Contours), a new variational framework of geometric active contours to track multiple nonrigid moving objects in the clutter background in image sequences is presented. This framework, incorporating with the motion edge information, consists of motion detection and tracking stages. At the motion detection stage, the motion edge map provides an approximate edge map of the moving objects. Then, a tracking stage, merely using the static edge information, is considered to improve the motion detection result. Force field regularization method is used to extend the capture range of the edge attraction force field in both stages. Experiments demonstrate that the proposed framework is valid for tracking multiple nonrigid objects in the clutter background.
