Most research on anomaly detection has focused on event that is different from its spatial-temporal neighboring events.It is still a significant challenge to detect anomalies that involve multiple normal events intera...Most research on anomaly detection has focused on event that is different from its spatial-temporal neighboring events.It is still a significant challenge to detect anomalies that involve multiple normal events interacting in an unusual pattern.In this work,a novel unsupervised method based on sparse topic model was proposed to capture motion patterns and detect anomalies in traffic surveillance.scale-invariant feature transform(SIFT)flow was used to improve the dense trajectory in order to extract interest points and the corresponding descriptors with less interference.For the purpose of strengthening the relationship of interest points on the same trajectory,the fisher kernel method was applied to obtain the representation of trajectory which was quantized into visual word.Then the sparse topic model was proposed to explore the latent motion patterns and achieve a sparse representation for the video scene.Finally,two anomaly detection algorithms were compared based on video clip detection and visual word analysis respectively.Experiments were conducted on QMUL Junction dataset and AVSS dataset.The results demonstrated the superior efficiency of the proposed method.展开更多
A new method for interaction recognition based on sparse representation of feature covariance matrices was presented.Firstly,the dense trajectories(DT)extracted from the video were clustered into different groups to e...A new method for interaction recognition based on sparse representation of feature covariance matrices was presented.Firstly,the dense trajectories(DT)extracted from the video were clustered into different groups to eliminate the irrelevant trajectories,which could greatly reduce the noise influence on feature extraction.Then,the trajectory tunnels were characterized by means of feature covariance matrices.In this way,the discriminative descriptors could be extracted,which was also an effective solution to the problem that the description of the feature second-order statistics is insufficient.After that,an over-complete dictionary was learned with the descriptors and all the descriptors were encoded using sparse coding(SC).Classification was achieved using multiple instance learning(MIL),which was more suitable for complex environments.The proposed method was tested and evaluated on the WEB Interaction dataset and the UT interaction dataset.The experimental results demonstrated the superior efficiency.展开更多
The dispersion characteristics of fuel particles in the dense phase zone in circulating fluidized bed(CFB)boilers have an important influence on bed temperature distribution and pollutant emissions.However,previous re...The dispersion characteristics of fuel particles in the dense phase zone in circulating fluidized bed(CFB)boilers have an important influence on bed temperature distribution and pollutant emissions.However,previous research in literature was mostly on small-scale apparatus,whose results could not be applied directly to large-scale CFB with multiple dispersion sources.To help solve this problem,we proposed a novel method to estimate the lateral dispersion coefficient(Dx)of fuel particles under partial coal cut-off condition in a 35o MW supercritical CFB boiler based on combustion and dispersion models.Meanwhile,we carried out experiments to obtain the Dx in the range of 0.1218-0.1406 m2/s.Numerical simulations were performed and the influence of operating conditions and furnace structure on fuel dispersion characteristics was investigated,the simulation value of Dx was validated against experimental data.Results revealed that the distribution of bed temperature caused by the fuel dispersion was mainly formed by char combustion.Because of the presence of intermediate water-cooled partition wall,the mixing and dispersion of fuel and bed material particles between the left and right sides of the furnace were hindered,increasing the non-uniformity of the bed temperature near furnace front wall.展开更多
为改善牛粪燃烧性能,本文选用牛粪和青稞秸秆颗粒燃料为对象,首先分析了牛粪颗粒燃料在不同升温速率(10℃/min、15℃/min、30℃/min)下的燃烧特性,在此基础上研究了混合不同比例(青稞秸秆含量为25%、50%、75%)的青稞秸秆对牛粪燃烧性能...为改善牛粪燃烧性能,本文选用牛粪和青稞秸秆颗粒燃料为对象,首先分析了牛粪颗粒燃料在不同升温速率(10℃/min、15℃/min、30℃/min)下的燃烧特性,在此基础上研究了混合不同比例(青稞秸秆含量为25%、50%、75%)的青稞秸秆对牛粪燃烧性能造成的影响。结果表明:(1)从着火温度、燃尽温度、最大燃烧速率、平均燃烧速率、可燃特性指数和综合燃烧特性指数综合分析,牛粪颗粒燃料的最佳升温速率为15℃/min;(2)青稞秸秆颗粒含量≧50%时,对牛粪颗粒的燃烧性能均有不同程度的改善。(3)采用一级反应动力学模型计算牛粪和3种混合燃料挥发分燃烧段和固定碳燃烧段的活化能,牛粪颗粒燃料受青稞秸秆颗粒影响较小,随青稞秸秆含量增加,燃料活化能分别为24.58 k J·mol^(-1)、27.05 k J·mol^(-1)和27.67 k J·mol^(-1)。该研究为进一步探索牛粪致密成型燃料的燃烧特性及相关设备的研究提供了理论和实验依据。展开更多
通过自动识别自然环境下获取果实图像中的未成熟果实,以实现自动化果实估产的目的。该文以番茄为对象,根据视觉显著性的特点,提出了使用基于密集和稀疏重构(dense and sparse reconstruction,DSR)的显著性检测方法检测未成熟番茄果实图...通过自动识别自然环境下获取果实图像中的未成熟果实,以实现自动化果实估产的目的。该文以番茄为对象,根据视觉显著性的特点,提出了使用基于密集和稀疏重构(dense and sparse reconstruction,DSR)的显著性检测方法检测未成熟番茄果实图像,该方法首先计算密集和稀疏重构误差;其次使用基于上下文的重构误差传播机制平滑重构误差和提亮显著性区域;再通过多尺度重构误差融合与偏目标高斯细化;最后通过贝叶斯算法融合显著图得到DSR显著灰度图。番茄DSR灰度图再经过OTSU算法进行分割和去噪处理,最终使用该文提出的改进随机Hough变换(randomized hough transform,RHT)圆检测方法识别番茄果簇中的单果。结果显示,该文方法对未成熟番茄果实的正确识别率能达到77.6%。同时,该文方法与人工测量的圆心和半径的相关系数也分别达到0.98和0.76,研究结果为估产机器人的多种果实自动化识别提供参考。展开更多
基金Project(50808025)supported by the National Natural Science Foundation of ChinaProject(20090162110057)supported by the Doctoral Fund of Ministry of Education,China
文摘Most research on anomaly detection has focused on event that is different from its spatial-temporal neighboring events.It is still a significant challenge to detect anomalies that involve multiple normal events interacting in an unusual pattern.In this work,a novel unsupervised method based on sparse topic model was proposed to capture motion patterns and detect anomalies in traffic surveillance.scale-invariant feature transform(SIFT)flow was used to improve the dense trajectory in order to extract interest points and the corresponding descriptors with less interference.For the purpose of strengthening the relationship of interest points on the same trajectory,the fisher kernel method was applied to obtain the representation of trajectory which was quantized into visual word.Then the sparse topic model was proposed to explore the latent motion patterns and achieve a sparse representation for the video scene.Finally,two anomaly detection algorithms were compared based on video clip detection and visual word analysis respectively.Experiments were conducted on QMUL Junction dataset and AVSS dataset.The results demonstrated the superior efficiency of the proposed method.
基金Project(51678075) supported by the National Natural Science Foundation of ChinaProject(2017GK2271) supported by the Science and Technology Project of Hunan Province,China
文摘A new method for interaction recognition based on sparse representation of feature covariance matrices was presented.Firstly,the dense trajectories(DT)extracted from the video were clustered into different groups to eliminate the irrelevant trajectories,which could greatly reduce the noise influence on feature extraction.Then,the trajectory tunnels were characterized by means of feature covariance matrices.In this way,the discriminative descriptors could be extracted,which was also an effective solution to the problem that the description of the feature second-order statistics is insufficient.After that,an over-complete dictionary was learned with the descriptors and all the descriptors were encoded using sparse coding(SC).Classification was achieved using multiple instance learning(MIL),which was more suitable for complex environments.The proposed method was tested and evaluated on the WEB Interaction dataset and the UT interaction dataset.The experimental results demonstrated the superior efficiency.
基金supported by the National Natural Science Foundation of China(grant No.52176101).
文摘The dispersion characteristics of fuel particles in the dense phase zone in circulating fluidized bed(CFB)boilers have an important influence on bed temperature distribution and pollutant emissions.However,previous research in literature was mostly on small-scale apparatus,whose results could not be applied directly to large-scale CFB with multiple dispersion sources.To help solve this problem,we proposed a novel method to estimate the lateral dispersion coefficient(Dx)of fuel particles under partial coal cut-off condition in a 35o MW supercritical CFB boiler based on combustion and dispersion models.Meanwhile,we carried out experiments to obtain the Dx in the range of 0.1218-0.1406 m2/s.Numerical simulations were performed and the influence of operating conditions and furnace structure on fuel dispersion characteristics was investigated,the simulation value of Dx was validated against experimental data.Results revealed that the distribution of bed temperature caused by the fuel dispersion was mainly formed by char combustion.Because of the presence of intermediate water-cooled partition wall,the mixing and dispersion of fuel and bed material particles between the left and right sides of the furnace were hindered,increasing the non-uniformity of the bed temperature near furnace front wall.
文摘为改善牛粪燃烧性能,本文选用牛粪和青稞秸秆颗粒燃料为对象,首先分析了牛粪颗粒燃料在不同升温速率(10℃/min、15℃/min、30℃/min)下的燃烧特性,在此基础上研究了混合不同比例(青稞秸秆含量为25%、50%、75%)的青稞秸秆对牛粪燃烧性能造成的影响。结果表明:(1)从着火温度、燃尽温度、最大燃烧速率、平均燃烧速率、可燃特性指数和综合燃烧特性指数综合分析,牛粪颗粒燃料的最佳升温速率为15℃/min;(2)青稞秸秆颗粒含量≧50%时,对牛粪颗粒的燃烧性能均有不同程度的改善。(3)采用一级反应动力学模型计算牛粪和3种混合燃料挥发分燃烧段和固定碳燃烧段的活化能,牛粪颗粒燃料受青稞秸秆颗粒影响较小,随青稞秸秆含量增加,燃料活化能分别为24.58 k J·mol^(-1)、27.05 k J·mol^(-1)和27.67 k J·mol^(-1)。该研究为进一步探索牛粪致密成型燃料的燃烧特性及相关设备的研究提供了理论和实验依据。
