针对3种典型的基于深度相机的同步定位与地图构建(SLAM)算法,包括RGB-D SLAM V2,RTAB-Map和DVO SLAM,介绍这3种SLAM算法的理论特点。采用两个公开的SLAM数据集,包括TUM数据集和ICL-NUIM数据集,进行SLAM算法的评测,评测指标包括SLAM算法...针对3种典型的基于深度相机的同步定位与地图构建(SLAM)算法,包括RGB-D SLAM V2,RTAB-Map和DVO SLAM,介绍这3种SLAM算法的理论特点。采用两个公开的SLAM数据集,包括TUM数据集和ICL-NUIM数据集,进行SLAM算法的评测,评测指标包括SLAM算法的精确度、运行性能以及鲁棒性。评测的实验结果表明,在选择基于深度相机的SLAM算法时:如果考虑精确度和鲁棒性优先于运行性能,则选择RGB-D SLAM V2;如果考虑运行性能和鲁棒性优先于精确度,则选择DVO SLAM;如果考虑精确度和运行性能优先于鲁棒性,则选择RTAB-Map。展开更多
针对协作过滤算法评测中普遍采用单一数据集,该文将传统的User-based(近邻数为20)、Item-based、Itemaverage、Item user average和Slope One 5种算法应用于MovieLens和Book-Crossing两种数据分布特征不同的数据集。结果显示,在Movielen...针对协作过滤算法评测中普遍采用单一数据集,该文将传统的User-based(近邻数为20)、Item-based、Itemaverage、Item user average和Slope One 5种算法应用于MovieLens和Book-Crossing两种数据分布特征不同的数据集。结果显示,在Movielens这种评分值相对比较稠密的数据集上,Slope One算法的预测精度最好;而在评分值相对比较稀疏的Book-Crossing数据集上,Item-based算法的预测精度最好,Slope One的预测精度最差。选择算法应根据用户和资源分布具体情况确定。展开更多
Poverty incidence is the key index that needs to be measured in the poverty exit examination and evaluation of 832 poverty-stricken counties and 128 000 poverty-stricken villages. In this paper, based on the statement...Poverty incidence is the key index that needs to be measured in the poverty exit examination and evaluation of 832 poverty-stricken counties and 128 000 poverty-stricken villages. In this paper, based on the statement of general concept and conventional calculation method of the poverty incidence, the calculation method of poverty incidence in the exit evaluation of poverty-stricken counties （also including poverty-stricken township and poverty-stricken villages） was investigated through the view of the third-party evaluation. In addition to considering the previous ＂number of planned poverty remaining population＂, the method also needed to give consideration to the exiting mistaken population, evaluation missing population. Based on the case in Yuanyang County, Honghe Hani and Yi Autonomous Prefecture, Yunnan Province, the poverty incidences of 10 exit planning villages by the end of 2017 in Yuanyang County were investigated and estimated, and suggestions were proposed to promote the precise poverty alleviation and poverty relief of the county.展开更多
Since there is lack of methodology to assess the performance of defogging algorithm and the existing assessment methods have some limitations,three new methods for assessing the defogging algorithm were proposed.One w...Since there is lack of methodology to assess the performance of defogging algorithm and the existing assessment methods have some limitations,three new methods for assessing the defogging algorithm were proposed.One was using synthetic foggy image simulated by image degradation model to assess the defogging algorithm in full-reference way.In this method,the absolute difference was computed between the synthetic image with and without fog.The other two were computing the fog density of gray level image or constructing assessment system of color image from human visual perception to assess the defogging algorithm in no-reference way.For these methods,an assessment function was defined to evaluate algorithm performance from the function value.Using the defogging algorithm comparison,the experimental results demonstrate the effectiveness and reliability of the proposed methods.展开更多
Sensory evaluation is the evaluation of signals that a human receives via its senses of sight, smell, taste, touch and hearing. In today’s industrial companies, sensory evaluation is widely used in quality inspection...Sensory evaluation is the evaluation of signals that a human receives via its senses of sight, smell, taste, touch and hearing. In today’s industrial companies, sensory evaluation is widely used in quality inspection of products, in marketing study and in many other fields such as risk evaluation, investment evaluation and safety evaluation. In practice, setting up a suitable mathematical formulation, an efficient working procedure and a pertinent computing method for sensory evaluation is quite difficult because of uncertainty and imprecision in sensory panels and their results involving linguistic expressions, non normalized data, data reliability, etc. At the present a prime problem of the practitioner is not the lack of useful methods but the lack of transparency in this area. In this tutorial lecture, we briefly describe some of the technology in the computational intelligence (CI) areas that has been developed for application to sensory evaluation and related fields. Moreover, we will illustrate the role of CI in sensory evaluation related applications from some recent publications.展开更多
This paper proposes an estimation method of fracture resistance of oil ＆ gas pipes, based on the results of full scale hydraulic pressure pipe testing with the artificial notch, made by the authors at 2002-2006. It s...This paper proposes an estimation method of fracture resistance of oil ＆ gas pipes, based on the results of full scale hydraulic pressure pipe testing with the artificial notch, made by the authors at 2002-2006. It shows that the fracture process at the stage of stripping （initialization） of crack can be characterized by Critical Fracture Diagram （CFD）. Calculating and experimental way to construct CFD is suggested and discussed. The fracture resistance requirements for oil ＆ gas pipes in terms of fracture mechanics （Kc, Lc） are formulated and proved. The results of full scale hydraulic pressure pipe testing, based on metal testing data of the standard Charpy specimens （KCV） and pipe＇s specification requirements are interrelated. The way to estimate the admissible temperature operation conditions of oil ＆ gas pipes by means of based metal ＂KCV-temperature＂ dependence is proposed.展开更多
Numerous edge-chasing deadlock detection algonthms were developed lor the cycle detection in distributed systems, but their detections had the n steps speed limitation and n （ n- 1） overhead limitation to detect a c...Numerous edge-chasing deadlock detection algonthms were developed lor the cycle detection in distributed systems, but their detections had the n steps speed limitation and n （ n- 1） overhead limitation to detect a cycle of size n under the one-resource request model. Since fast deadlock detection is critical, this paper proposed a new algorithm to speed up the detection process. In our algorithm, when the running of a transaction node is blocked, the being requested resource nodes reply it with the waiting or being waited message simultaneously, so the blocked node knows both its predecessors and successors, which helps it detecting a cycle of size 2 directly and locally. For the cycle of size n （ n 〉 2）, a special probe is produced which has the predecessors information of its originator, so the being detected nodes know their indirect predecessors and direct successors, and can detect the cycle within n - 2 steps. The proposed algorithm is formally proved to be correct by the invariant verification method. Performance evaluation shows that the message overhead of our detection is （ n^2 - n - 2）/2, hence both the detection speed and message cost of the proposed algorithm are better than that of the existing al gorithms.展开更多
Lucas-Kanade(LK) algorithm, usually used in optical flow filed, has recently received increasing attention from PIV community due to its advanced calculation efficiency by GPU acceleration. Although applications of th...Lucas-Kanade(LK) algorithm, usually used in optical flow filed, has recently received increasing attention from PIV community due to its advanced calculation efficiency by GPU acceleration. Although applications of this algorithm are continuously emerging,a systematic performance evaluation is still lacking. This forms the primary aim of the present work. Three warping schemes in the family of LK algorithm: forward/inverse/symmetric warping, are evaluated in a prototype flow of a hierarchy of multiple two-dimensional vortices. Second-order Newton descent is also considered here. The accuracy & efficiency of all these LK variants are investigated under a large domain of various influential parameters. It is found that the constant displacement constraint, which is a necessary building block for GPU acceleration, is the most critical issue in affecting LK algorithm's accuracy, which can be somehow ameliorated by using second-order Newton descent. Moreover, symmetric warping outbids the other two warping schemes in accuracy level, robustness to noise, convergence speed and tolerance to displacement gradient, and might be the first choice when applying LK algorithm to PIV measurement.展开更多
文摘针对3种典型的基于深度相机的同步定位与地图构建(SLAM)算法,包括RGB-D SLAM V2,RTAB-Map和DVO SLAM,介绍这3种SLAM算法的理论特点。采用两个公开的SLAM数据集,包括TUM数据集和ICL-NUIM数据集,进行SLAM算法的评测,评测指标包括SLAM算法的精确度、运行性能以及鲁棒性。评测的实验结果表明,在选择基于深度相机的SLAM算法时:如果考虑精确度和鲁棒性优先于运行性能,则选择RGB-D SLAM V2;如果考虑运行性能和鲁棒性优先于精确度,则选择DVO SLAM;如果考虑精确度和运行性能优先于鲁棒性,则选择RTAB-Map。
文摘针对协作过滤算法评测中普遍采用单一数据集,该文将传统的User-based(近邻数为20)、Item-based、Itemaverage、Item user average和Slope One 5种算法应用于MovieLens和Book-Crossing两种数据分布特征不同的数据集。结果显示,在Movielens这种评分值相对比较稠密的数据集上,Slope One算法的预测精度最好;而在评分值相对比较稀疏的Book-Crossing数据集上,Item-based算法的预测精度最好,Slope One的预测精度最差。选择算法应根据用户和资源分布具体情况确定。
文摘Poverty incidence is the key index that needs to be measured in the poverty exit examination and evaluation of 832 poverty-stricken counties and 128 000 poverty-stricken villages. In this paper, based on the statement of general concept and conventional calculation method of the poverty incidence, the calculation method of poverty incidence in the exit evaluation of poverty-stricken counties （also including poverty-stricken township and poverty-stricken villages） was investigated through the view of the third-party evaluation. In addition to considering the previous ＂number of planned poverty remaining population＂, the method also needed to give consideration to the exiting mistaken population, evaluation missing population. Based on the case in Yuanyang County, Honghe Hani and Yi Autonomous Prefecture, Yunnan Province, the poverty incidences of 10 exit planning villages by the end of 2017 in Yuanyang County were investigated and estimated, and suggestions were proposed to promote the precise poverty alleviation and poverty relief of the county.
基金Projects(91220301,61175064,61273314)supported by the National Natural Science Foundation of ChinaProject(126648)supported by the Postdoctoral Science Foundation of Central South University,ChinaProject(2012170301)supported by the New Teacher Fund for School of Information Science and Engineering,Central South University,China
文摘Since there is lack of methodology to assess the performance of defogging algorithm and the existing assessment methods have some limitations,three new methods for assessing the defogging algorithm were proposed.One was using synthetic foggy image simulated by image degradation model to assess the defogging algorithm in full-reference way.In this method,the absolute difference was computed between the synthetic image with and without fog.The other two were computing the fog density of gray level image or constructing assessment system of color image from human visual perception to assess the defogging algorithm in no-reference way.For these methods,an assessment function was defined to evaluate algorithm performance from the function value.Using the defogging algorithm comparison,the experimental results demonstrate the effectiveness and reliability of the proposed methods.
文摘Sensory evaluation is the evaluation of signals that a human receives via its senses of sight, smell, taste, touch and hearing. In today’s industrial companies, sensory evaluation is widely used in quality inspection of products, in marketing study and in many other fields such as risk evaluation, investment evaluation and safety evaluation. In practice, setting up a suitable mathematical formulation, an efficient working procedure and a pertinent computing method for sensory evaluation is quite difficult because of uncertainty and imprecision in sensory panels and their results involving linguistic expressions, non normalized data, data reliability, etc. At the present a prime problem of the practitioner is not the lack of useful methods but the lack of transparency in this area. In this tutorial lecture, we briefly describe some of the technology in the computational intelligence (CI) areas that has been developed for application to sensory evaluation and related fields. Moreover, we will illustrate the role of CI in sensory evaluation related applications from some recent publications.
文摘This paper proposes an estimation method of fracture resistance of oil ＆ gas pipes, based on the results of full scale hydraulic pressure pipe testing with the artificial notch, made by the authors at 2002-2006. It shows that the fracture process at the stage of stripping （initialization） of crack can be characterized by Critical Fracture Diagram （CFD）. Calculating and experimental way to construct CFD is suggested and discussed. The fracture resistance requirements for oil ＆ gas pipes in terms of fracture mechanics （Kc, Lc） are formulated and proved. The results of full scale hydraulic pressure pipe testing, based on metal testing data of the standard Charpy specimens （KCV） and pipe＇s specification requirements are interrelated. The way to estimate the admissible temperature operation conditions of oil ＆ gas pipes by means of based metal ＂KCV-temperature＂ dependence is proposed.
文摘Numerous edge-chasing deadlock detection algonthms were developed lor the cycle detection in distributed systems, but their detections had the n steps speed limitation and n （ n- 1） overhead limitation to detect a cycle of size n under the one-resource request model. Since fast deadlock detection is critical, this paper proposed a new algorithm to speed up the detection process. In our algorithm, when the running of a transaction node is blocked, the being requested resource nodes reply it with the waiting or being waited message simultaneously, so the blocked node knows both its predecessors and successors, which helps it detecting a cycle of size 2 directly and locally. For the cycle of size n （ n 〉 2）, a special probe is produced which has the predecessors information of its originator, so the being detected nodes know their indirect predecessors and direct successors, and can detect the cycle within n - 2 steps. The proposed algorithm is formally proved to be correct by the invariant verification method. Performance evaluation shows that the message overhead of our detection is （ n^2 - n - 2）/2, hence both the detection speed and message cost of the proposed algorithm are better than that of the existing al gorithms.
基金supported by the National Natural Science Foundation of China(Grant Nos.11372001 and 11490552)
文摘Lucas-Kanade(LK) algorithm, usually used in optical flow filed, has recently received increasing attention from PIV community due to its advanced calculation efficiency by GPU acceleration. Although applications of this algorithm are continuously emerging,a systematic performance evaluation is still lacking. This forms the primary aim of the present work. Three warping schemes in the family of LK algorithm: forward/inverse/symmetric warping, are evaluated in a prototype flow of a hierarchy of multiple two-dimensional vortices. Second-order Newton descent is also considered here. The accuracy & efficiency of all these LK variants are investigated under a large domain of various influential parameters. It is found that the constant displacement constraint, which is a necessary building block for GPU acceleration, is the most critical issue in affecting LK algorithm's accuracy, which can be somehow ameliorated by using second-order Newton descent. Moreover, symmetric warping outbids the other two warping schemes in accuracy level, robustness to noise, convergence speed and tolerance to displacement gradient, and might be the first choice when applying LK algorithm to PIV measurement.
