为解决基于随机森林的3D人体姿态估计算法容易出现的误分类问题,提出一种基于自适应融合特征提取和误分类处理机制的改进算法.该算法利用自适应融合特征提取方法自适应提取深度融合特征,此特征可表达图像距离信息和部位尺寸信息,增强特...为解决基于随机森林的3D人体姿态估计算法容易出现的误分类问题,提出一种基于自适应融合特征提取和误分类处理机制的改进算法.该算法利用自适应融合特征提取方法自适应提取深度融合特征,此特征可表达图像距离信息和部位尺寸信息,增强特征的表征能力;针对识别部位误分类问题,分别从识别部位误分点聚集情况和迭代整合思想出发,提出误分类处理机制,改善部位识别结果;最后提出可进一步处理误分点的改进主方向分析(Principal direction analysis,PDA)算法,自适应计算出部位主方向向量,实现3D人体姿态估计.结果表明,该算法能有效去除部位误分点,并显著改善了3D人体姿态估计.展开更多
To analyze the errors of processing data, the testing principle for jet elements is introduced and the property of testing system is theoretically and experimentally studied. On the basis of the above, the method of p...To analyze the errors of processing data, the testing principle for jet elements is introduced and the property of testing system is theoretically and experimentally studied. On the basis of the above, the method of processing data is presented and the error formulae, which are the functions of the testing system property, are derived. Finally, the methods of reducing the errors are provided. The measured results are in correspondence with the theoretical conclusion.展开更多
The mathematical modeling for evaluation of the sphericity error is proposed with minimum radial separation center. To obtain the minimum sphericity error from the form data, a geometric approximation technique was de...The mathematical modeling for evaluation of the sphericity error is proposed with minimum radial separation center. To obtain the minimum sphericity error from the form data, a geometric approximation technique was devised. The technique regarded the least square sphere center as the initial center of the concentric spheres containing all measurement points, and then the center was moved gradually to reduce the radial separation till the minimum radial separation center was got where the constructed concentric spheres conformed to the minimum zone condition. The method was modeled firstly, then the geometric approximation process was analyzed, and finally,the software for data processing was programmed. As evaluation example, five steel balls were measured and the measurement data were processed with the developed program. The average iteration times of the approximation technique is 4.2, and on average the obtained sphericity error is 0. 529μm smaller than the least square solution,with accuracy increased by 7. 696%.展开更多
In many image resolution enhancement applications,the blurring process of the imaging system is unknown.This paper discusses the problem of single image blind resolution enhancement without estimating the point spread...In many image resolution enhancement applications,the blurring process of the imaging system is unknown.This paper discusses the problem of single image blind resolution enhancement without estimating the point spread function(PSF).A regularization model is constructed for image enhancement based on the prior informaton of image error and image gray value,which does not need any prior information of PSF.Moreover,through the solution of Euler equations,an anisotropic nonlinear diffusion equation are obtained,which can avoid the high computational cost of regularization model.Furthermore,in order to get sub-pixel superresolved image,the regularization model for image enhancement is extended to the enlargement of image,which can enlarge and enhance image at the same time.Last,to get clearer edges,a high frequency enhancement filter is used on the superresolved image.Numerical results show that the new model can get much clearer super-resolution images than traditional methods,and the peak signal to noise ratios (PSNRs) are also higher than traditional methods.展开更多
Batch processes are important in chemical industry,in which operators usually play a major role and hazards may arise by their inadvertent acts.In this paper,based on hazard and operability study and concept of qualit...Batch processes are important in chemical industry,in which operators usually play a major role and hazards may arise by their inadvertent acts.In this paper,based on hazard and operability study and concept of qualitative simulation,an automatic method for adverse consequence identification for potential maloperation is proposed.The qualitative model for production process is expressed by a novel directed graph.Possible operation deviations from normal operating procedure are identified systematically by using a group of guidewords.The proposed algorithm is used for qualitative simulation of batch processes to identify the effects of maloperations.The method is illustrated with a simple batch process and a batch reaction process.The results show that batch processes can be simulated qualitatively and hazards can be identified for operating procedures including maloperations.After analysis for possible plant maloperations,some measures can be taken to avoid maloperations or reduce losses resulted from maloperations.展开更多
Computer Assisted Language Learning （CALL） has been in use as a learner centered method since 1960. In this paper an attempt is made to evolve a Remedial Grammar （RG） which enables the teacher to develop a reperto...Computer Assisted Language Learning （CALL） has been in use as a learner centered method since 1960. In this paper an attempt is made to evolve a Remedial Grammar （RG） which enables the teacher to develop a repertoire of remedial activities based on the errors identified following Error Analysis （EA） for repeated presentation. The error types are considered as teaching/learning points. The present paper argues that CALL is the right model for effective remedial learning. Remedial Gmaster （RGm）, an authoring package is introduced. This is an exclusive mode of error correction in which interlingual, intralingual, developmental, and global errors are dealt with. The syntactical problems due to the subtle difference between simple past and perfect tenses in English language use are discussed and CALL authoring package presented.展开更多
Validated satellite-derived sea surface temperatures (SSTs) are widely used for climate monitoring and ocean data assimilation systems. In this study, the Fengyun-3A (FY-3A) SST experimental product is evaluated using...Validated satellite-derived sea surface temperatures (SSTs) are widely used for climate monitoring and ocean data assimilation systems. In this study, the Fengyun-3A (FY-3A) SST experimental product is evaluated using Advanced Very High Resolution Radiometer (AVHRR)-merged and in situ SSTs. A comparison of AVHRR-merged SSTs reveals a negative bias of more than 2K in FY-3A SSTs in most of the tropical Pacific and low-latitude Indian and Atlantic Oceans. The error variance of FY-3A SSTs is estimated using three-way error analysis. FY-3A SSTs show regional error variance in global oceans with a maximum error variance of 2.2 K in the Pacific Ocean. In addition, a significant seasonal variation of error variance is present in FY-3A SSTs, which indicates that the quality of FY-3A SST could be improved by adjusting the parameters in the SST retrieval algorithm and by applying regional and seasonal algorithms, particularly in key areas such as the tropical Pacific Ocean. An objective analysis method is used to merge FY-3A SSTs with the drifter buoy data. The errors of FY-3A SSTs are decreased to-0.45K comparing with SST observations from GTSPP.展开更多
To improve frame synchronization precision, a scheme named training symbol correlation （TSC） is presented for orthogonal frequency division multiplexing （ OFDM ） system. Based on the solution from Schmidl and Cox,...To improve frame synchronization precision, a scheme named training symbol correlation （TSC） is presented for orthogonal frequency division multiplexing （ OFDM ） system. Based on the solution from Schmidl and Cox, a timing metric related to TSC scheme is put forward and examined. The specific method to select a threshold value provides more precise detection results, which can be shown by performance comparison between the two schemes through Monte Carlo simulation. Taking IEEE 802.1 la WLAN standard as an example, the proposed approach is superior to the most popular Schmidl scheme in terms of BER.展开更多
文摘为解决基于随机森林的3D人体姿态估计算法容易出现的误分类问题,提出一种基于自适应融合特征提取和误分类处理机制的改进算法.该算法利用自适应融合特征提取方法自适应提取深度融合特征,此特征可表达图像距离信息和部位尺寸信息,增强特征的表征能力;针对识别部位误分类问题,分别从识别部位误分点聚集情况和迭代整合思想出发,提出误分类处理机制,改善部位识别结果;最后提出可进一步处理误分点的改进主方向分析(Principal direction analysis,PDA)算法,自适应计算出部位主方向向量,实现3D人体姿态估计.结果表明,该算法能有效去除部位误分点,并显著改善了3D人体姿态估计.
文摘To analyze the errors of processing data, the testing principle for jet elements is introduced and the property of testing system is theoretically and experimentally studied. On the basis of the above, the method of processing data is presented and the error formulae, which are the functions of the testing system property, are derived. Finally, the methods of reducing the errors are provided. The measured results are in correspondence with the theoretical conclusion.
基金Supported by National Natural Science Foundation of China(No.50175081) andTianjin Municipal Science and Technology Commission (No.0431835116).
文摘The mathematical modeling for evaluation of the sphericity error is proposed with minimum radial separation center. To obtain the minimum sphericity error from the form data, a geometric approximation technique was devised. The technique regarded the least square sphere center as the initial center of the concentric spheres containing all measurement points, and then the center was moved gradually to reduce the radial separation till the minimum radial separation center was got where the constructed concentric spheres conformed to the minimum zone condition. The method was modeled firstly, then the geometric approximation process was analyzed, and finally,the software for data processing was programmed. As evaluation example, five steel balls were measured and the measurement data were processed with the developed program. The average iteration times of the approximation technique is 4.2, and on average the obtained sphericity error is 0. 529μm smaller than the least square solution,with accuracy increased by 7. 696%.
基金Supported by the National Natural Science Foundation of China(No.60272013) and National Excellent Doctoral DissertationFund of China(No.200140)
文摘In many image resolution enhancement applications,the blurring process of the imaging system is unknown.This paper discusses the problem of single image blind resolution enhancement without estimating the point spread function(PSF).A regularization model is constructed for image enhancement based on the prior informaton of image error and image gray value,which does not need any prior information of PSF.Moreover,through the solution of Euler equations,an anisotropic nonlinear diffusion equation are obtained,which can avoid the high computational cost of regularization model.Furthermore,in order to get sub-pixel superresolved image,the regularization model for image enhancement is extended to the enlargement of image,which can enlarge and enhance image at the same time.Last,to get clearer edges,a high frequency enhancement filter is used on the superresolved image.Numerical results show that the new model can get much clearer super-resolution images than traditional methods,and the peak signal to noise ratios (PSNRs) are also higher than traditional methods.
文摘Batch processes are important in chemical industry,in which operators usually play a major role and hazards may arise by their inadvertent acts.In this paper,based on hazard and operability study and concept of qualitative simulation,an automatic method for adverse consequence identification for potential maloperation is proposed.The qualitative model for production process is expressed by a novel directed graph.Possible operation deviations from normal operating procedure are identified systematically by using a group of guidewords.The proposed algorithm is used for qualitative simulation of batch processes to identify the effects of maloperations.The method is illustrated with a simple batch process and a batch reaction process.The results show that batch processes can be simulated qualitatively and hazards can be identified for operating procedures including maloperations.After analysis for possible plant maloperations,some measures can be taken to avoid maloperations or reduce losses resulted from maloperations.
文摘Computer Assisted Language Learning （CALL） has been in use as a learner centered method since 1960. In this paper an attempt is made to evolve a Remedial Grammar （RG） which enables the teacher to develop a repertoire of remedial activities based on the errors identified following Error Analysis （EA） for repeated presentation. The error types are considered as teaching/learning points. The present paper argues that CALL is the right model for effective remedial learning. Remedial Gmaster （RGm）, an authoring package is introduced. This is an exclusive mode of error correction in which interlingual, intralingual, developmental, and global errors are dealt with. The syntactical problems due to the subtle difference between simple past and perfect tenses in English language use are discussed and CALL authoring package presented.
基金supported by the National Basic Research Program of China(973 Program,Grant Nos.2010CB951902 and 2011CB403505)the National Key Technologies R&D Program of China(Grant No.2009BAC51B03)the National Natural Science Foundation of China(Grant No.41106003)
文摘Validated satellite-derived sea surface temperatures (SSTs) are widely used for climate monitoring and ocean data assimilation systems. In this study, the Fengyun-3A (FY-3A) SST experimental product is evaluated using Advanced Very High Resolution Radiometer (AVHRR)-merged and in situ SSTs. A comparison of AVHRR-merged SSTs reveals a negative bias of more than 2K in FY-3A SSTs in most of the tropical Pacific and low-latitude Indian and Atlantic Oceans. The error variance of FY-3A SSTs is estimated using three-way error analysis. FY-3A SSTs show regional error variance in global oceans with a maximum error variance of 2.2 K in the Pacific Ocean. In addition, a significant seasonal variation of error variance is present in FY-3A SSTs, which indicates that the quality of FY-3A SST could be improved by adjusting the parameters in the SST retrieval algorithm and by applying regional and seasonal algorithms, particularly in key areas such as the tropical Pacific Ocean. An objective analysis method is used to merge FY-3A SSTs with the drifter buoy data. The errors of FY-3A SSTs are decreased to-0.45K comparing with SST observations from GTSPP.
文摘To improve frame synchronization precision, a scheme named training symbol correlation （TSC） is presented for orthogonal frequency division multiplexing （ OFDM ） system. Based on the solution from Schmidl and Cox, a timing metric related to TSC scheme is put forward and examined. The specific method to select a threshold value provides more precise detection results, which can be shown by performance comparison between the two schemes through Monte Carlo simulation. Taking IEEE 802.1 la WLAN standard as an example, the proposed approach is superior to the most popular Schmidl scheme in terms of BER.
