Using the method of mathematical morphology,this paper fulfills filtration,segmentation and extraction of morphological features of the satellite cloud image.It also gives out the relative algorithms,which is realized...Using the method of mathematical morphology,this paper fulfills filtration,segmentation and extraction of morphological features of the satellite cloud image.It also gives out the relative algorithms,which is realized by parallel C programming based on Transputer networks.It has been successfully used to process the typhoon and the low tornado cloud image.And it will be used in weather forecast.展开更多
目前,基于深度学习的点云上采样方法缺失对局部区域特征关联性的关注和对全局特征的多尺度提取,导致输出的密集点云存在异常值过多、细粒度不高等问题。为解决上述问题,提出了嵌入注意力机制的并行多尺度点云上采样网络(Parallel Multi-...目前,基于深度学习的点云上采样方法缺失对局部区域特征关联性的关注和对全局特征的多尺度提取,导致输出的密集点云存在异常值过多、细粒度不高等问题。为解决上述问题,提出了嵌入注意力机制的并行多尺度点云上采样网络(Parallel Multi-scale with Attention mechanism for Point cloud Upsampling),网络由特征提取器、特征拓展器、坐标细化器和坐标重建器4个模块级联组成。首先给定一个N×3的稀疏点云作为输入,为了获得点云的全局和局部特征信息,设计了一个嵌入注意力机制的并行多尺度特征提取模块(PMA)用于将三维空间的点云映射到高维特征空间。其次使用边缘卷积特征拓展器拓展点云特征维度,得到高维点云特征,以更好地保留点云特征的边缘信息,将高维点云特征通过坐标重建器转换回三维空间中。最后使用坐标细化器精细调整输出点云细节。在合成数据集PU1K上的对比实验结果表明,PMA-PU生成的密集点云在倒角距离(CD)、豪斯多夫距离(HD)和点面距离(P2F)上都有显著提升,分别比性能次优的网络模型优化了7.863%,21.631%,14.686%。可视化结果证明了PMA-PU具有性能更好的特征提取器,能够生成细粒度更高、形状更接近真实值的密集点云。展开更多
Community Question Answering(CQA) in web forums, as a classic forum for user communication,provides a large number of high-quality useful answers in comparison with traditional question answering.Development of method...Community Question Answering(CQA) in web forums, as a classic forum for user communication,provides a large number of high-quality useful answers in comparison with traditional question answering.Development of methods to get good, honest answers according to user questions is a challenging task in natural language processing. Many answers are not associated with the actual problem or shift the subjects,and this usually occurs in relatively long answers. In this paper, we enhance answer selection in CQA using multidimensional feature combination and similarity order. We make full use of the information in answers to questions to determine the similarity between questions and answers, and use the text-based description of the answer to determine whether it is a reasonable one. Our work includes two subtasks:(a) classifying answers as good, bad, or potentially associated with a question, and(b) answering YES/NO based on a list of all answers to a question. The experimental results show that our approach is significantly more efficient than the baseline model, and its overall ranking is relatively high in comparison with that of other models.展开更多
In this paper, we propose a parallel computing technique for content-based image retrieval (CBIR) system. This technique is mainly used for single node with multi-core processor, which is different from those based ...In this paper, we propose a parallel computing technique for content-based image retrieval (CBIR) system. This technique is mainly used for single node with multi-core processor, which is different from those based on cluster or network computing architecture. Due to its specific applications (such as medical image processing) and the harsh terms of hardware resource requirement, the CBIR system has been prevented from being widely used. With the increasing volume of the image database, the widespread use of multi-core processors, and the requirement of the retrieval accuracy and speed, we need to achieve a retrieval strategy which is based on multi-core processor to make the retrieval faster and more convenient than before. Experimental results demonstrate that this parallel architecture can significantly improve the performance of retrieval system. In addition, we also propose an efficient parallel technique with the combinations of the cluster and the multi-core techniques, which is supposed to gear to the new trend of the cloud computing.展开更多
A new parallel architecture for quantified boolean formula(QBF)solving was proposed,and the prediction model based on machine learning technology was proposed for how sharing knowledge affects the solving performance ...A new parallel architecture for quantified boolean formula(QBF)solving was proposed,and the prediction model based on machine learning technology was proposed for how sharing knowledge affects the solving performance in QBF parallel solving system,and the experimental evaluation scheme was also designed.It shows that the characterization factor of clause and cube influence the solving performance markedly in our experiment.At the same time,the heuristic machine learning algorithm was applied,support vector machine was chosen to predict the performance of QBF parallel solving system based on clause sharing and cube sharing.The relative error of accuracy for prediction can be controlled in a reasonable range of 20%30%.The results show the important and complex role that knowledge sharing plays in any modern parallel solver.It shows that the parallel solver with machine learning reduces the quantity of knowledge sharing about 30%and saving computational resource but does not reduce the performance of solving system.展开更多
提出一种金属表面缺陷检测方法的改进模型.首先,基于YOLOv3(you only look once v3)目标检测模型,使用多尺度卷积并行结构,提取、融合多尺度特征;其次,使用高效下采样,在保留特征信息的同时减少特征升维的计算量;最后,使用空间可分离卷...提出一种金属表面缺陷检测方法的改进模型.首先,基于YOLOv3(you only look once v3)目标检测模型,使用多尺度卷积并行结构,提取、融合多尺度特征;其次,使用高效下采样,在保留特征信息的同时减少特征升维的计算量;最后,使用空间可分离卷积,在保持感受野不变的前提下增加模型的宽度与深度,从而得到模型参数量减少、同时提升了模型性能的改进模型YOLOv3I(you only look once v3 inception).改进模型提高了对复杂缺陷的特征提取能力,并进一步降低了对硬件配置的要求.实验结果表明,改进模型在精度与计算效率上均有明显提升.平均准确率在公开数据集上约提高5%,在企业提供的轴承数据集上约提高3%,模型参数量下降超过20%,两个数据集上模型浮点计算量分别减少1.6×10^(9)和1.2×10^(10)次.展开更多
文摘Using the method of mathematical morphology,this paper fulfills filtration,segmentation and extraction of morphological features of the satellite cloud image.It also gives out the relative algorithms,which is realized by parallel C programming based on Transputer networks.It has been successfully used to process the typhoon and the low tornado cloud image.And it will be used in weather forecast.
文摘目前,基于深度学习的点云上采样方法缺失对局部区域特征关联性的关注和对全局特征的多尺度提取,导致输出的密集点云存在异常值过多、细粒度不高等问题。为解决上述问题,提出了嵌入注意力机制的并行多尺度点云上采样网络(Parallel Multi-scale with Attention mechanism for Point cloud Upsampling),网络由特征提取器、特征拓展器、坐标细化器和坐标重建器4个模块级联组成。首先给定一个N×3的稀疏点云作为输入,为了获得点云的全局和局部特征信息,设计了一个嵌入注意力机制的并行多尺度特征提取模块(PMA)用于将三维空间的点云映射到高维特征空间。其次使用边缘卷积特征拓展器拓展点云特征维度,得到高维点云特征,以更好地保留点云特征的边缘信息,将高维点云特征通过坐标重建器转换回三维空间中。最后使用坐标细化器精细调整输出点云细节。在合成数据集PU1K上的对比实验结果表明,PMA-PU生成的密集点云在倒角距离(CD)、豪斯多夫距离(HD)和点面距离(P2F)上都有显著提升,分别比性能次优的网络模型优化了7.863%,21.631%,14.686%。可视化结果证明了PMA-PU具有性能更好的特征提取器,能够生成细粒度更高、形状更接近真实值的密集点云。
基金developed by the NLP601 group at School of Electronics Engineering and Computer Science, Peking University, within the National Natural Science Foundation of China (No. 61672046)
文摘Community Question Answering(CQA) in web forums, as a classic forum for user communication,provides a large number of high-quality useful answers in comparison with traditional question answering.Development of methods to get good, honest answers according to user questions is a challenging task in natural language processing. Many answers are not associated with the actual problem or shift the subjects,and this usually occurs in relatively long answers. In this paper, we enhance answer selection in CQA using multidimensional feature combination and similarity order. We make full use of the information in answers to questions to determine the similarity between questions and answers, and use the text-based description of the answer to determine whether it is a reasonable one. Our work includes two subtasks:(a) classifying answers as good, bad, or potentially associated with a question, and(b) answering YES/NO based on a list of all answers to a question. The experimental results show that our approach is significantly more efficient than the baseline model, and its overall ranking is relatively high in comparison with that of other models.
基金supported by the Natural Science Foundation of Shanghai (Grant No.08ZR1408200)the Shanghai Leading Academic Discipline Project (Grant No.J50103)the Open Project Program of the National Laboratory of Pattern Recognition
文摘In this paper, we propose a parallel computing technique for content-based image retrieval (CBIR) system. This technique is mainly used for single node with multi-core processor, which is different from those based on cluster or network computing architecture. Due to its specific applications (such as medical image processing) and the harsh terms of hardware resource requirement, the CBIR system has been prevented from being widely used. With the increasing volume of the image database, the widespread use of multi-core processors, and the requirement of the retrieval accuracy and speed, we need to achieve a retrieval strategy which is based on multi-core processor to make the retrieval faster and more convenient than before. Experimental results demonstrate that this parallel architecture can significantly improve the performance of retrieval system. In addition, we also propose an efficient parallel technique with the combinations of the cluster and the multi-core techniques, which is supposed to gear to the new trend of the cloud computing.
基金Project(61171141)supported by the National Natural Science Foundation of China
文摘A new parallel architecture for quantified boolean formula(QBF)solving was proposed,and the prediction model based on machine learning technology was proposed for how sharing knowledge affects the solving performance in QBF parallel solving system,and the experimental evaluation scheme was also designed.It shows that the characterization factor of clause and cube influence the solving performance markedly in our experiment.At the same time,the heuristic machine learning algorithm was applied,support vector machine was chosen to predict the performance of QBF parallel solving system based on clause sharing and cube sharing.The relative error of accuracy for prediction can be controlled in a reasonable range of 20%30%.The results show the important and complex role that knowledge sharing plays in any modern parallel solver.It shows that the parallel solver with machine learning reduces the quantity of knowledge sharing about 30%and saving computational resource but does not reduce the performance of solving system.
文摘提出一种金属表面缺陷检测方法的改进模型.首先,基于YOLOv3(you only look once v3)目标检测模型,使用多尺度卷积并行结构,提取、融合多尺度特征;其次,使用高效下采样,在保留特征信息的同时减少特征升维的计算量;最后,使用空间可分离卷积,在保持感受野不变的前提下增加模型的宽度与深度,从而得到模型参数量减少、同时提升了模型性能的改进模型YOLOv3I(you only look once v3 inception).改进模型提高了对复杂缺陷的特征提取能力,并进一步降低了对硬件配置的要求.实验结果表明,改进模型在精度与计算效率上均有明显提升.平均准确率在公开数据集上约提高5%,在企业提供的轴承数据集上约提高3%,模型参数量下降超过20%,两个数据集上模型浮点计算量分别减少1.6×10^(9)和1.2×10^(10)次.