Automatic image annotation has been an active topic of research in computer vision and pattern recognition for decades.A two stage automatic image annotation method based on Gaussian mixture model(GMM) and random walk...Automatic image annotation has been an active topic of research in computer vision and pattern recognition for decades.A two stage automatic image annotation method based on Gaussian mixture model(GMM) and random walk model(abbreviated as GMM-RW) is presented.To start with,GMM fitted by the rival penalized expectation maximization(RPEM) algorithm is employed to estimate the posterior probabilities of each annotation keyword.Subsequently,a random walk process over the constructed label similarity graph is implemented to further mine the potential correlations of the candidate annotations so as to capture the refining results,which plays a crucial role in semantic based image retrieval.The contributions exhibited in this work are multifold.First,GMM is exploited to capture the initial semantic annotations,especially the RPEM algorithm is utilized to train the model that can determine the number of components in GMM automatically.Second,a label similarity graph is constructed by a weighted linear combination of label similarity and visual similarity of images associated with the corresponding labels,which is able to avoid the phenomena of polysemy and synonym efficiently during the image annotation process.Third,the random walk is implemented over the constructed label graph to further refine the candidate set of annotations generated by GMM.Conducted experiments on the standard Corel5 k demonstrate that GMM-RW is significantly more effective than several state-of-the-arts regarding their effectiveness and efficiency in the task of automatic image annotation.展开更多
In real traffic,any vehicle must give lane-changing signal(i.e.the turn signal) before changing lanes;at this time,the vehicles behind the lane-changing vehicle will be hindered and may form "plugs" due to t...In real traffic,any vehicle must give lane-changing signal(i.e.the turn signal) before changing lanes;at this time,the vehicles behind the lane-changing vehicle will be hindered and may form "plugs" due to the turn signal effect.However,few studies focus on exploring the effect.In this paper,the turn signal effect was taken into account by proposing a new symmetric two-lane cellular automaton(T-STCA) model,and the new model was set to compare with the STCA,H-STCA and A-STCA models.Numerical results show that using the T-STCA model to describe lane-changing or overtaking,the process appeared in several consecutive time steps;while using the other three models,the process appeared only in one time step.In addition,the T-STCA model could describe the mixed traffic flow more realistically and the turn signal effect could help the plugs to dissolve more quickly.展开更多
文摘混合元胞自动机(Mixed-cell cellular automata,MCCA)模型改进了传统的元胞自动机(Cellular automata,CA)模型,基于现实复杂土地结构引入混合元胞,实现了从定性、静态模拟到定量、动态模拟的跨越。本文首先探究MCCA模型在黑河中游甘临高地区(甘州区、临泽县和高台县)的适用性;之后分别采用多目标线性规划(Multiple-objective programming,MOP)模型、普通线性回归模型预测得到2035年可持续发展(Sustainable development,SUD)情景、基本发展(Basic development,BAD)情景中不同地类面积数值,然后将面积输入MCCA模型中进行不同情景的土地利用空间结构可视化,并开展对比研究。结果表明:各项精度评价指标均表明MCCA模型的模拟精度较高,Kappa系数、混合元胞质量系数(Mixed-cell figure of merit,mcFoM)和平均相对熵(Relative entropy,RE)分别为0.886、0.261和0.508,优于基于纯净元胞的斑块生成土地利用变化模拟(Patch-generating land use simulation model,PLUS)模型,因此MCCA模型适用于研究区土地利用结构模拟。2035年SUD情景中林地范围明显高于BAD情景,生态效益较BAD情景增速快,建设用地和耕地适度扩张,综合效益增速较快。该结果表明耦合MOP和MCCA模型模拟的土地利用优化配置方案能够更好地协调经济与环境的关系,既有利于经济快速发展,又能保护生态环境和维持社会稳定。
基金Supported by the National Basic Research Program of China(No.2013CB329502)the National Natural Science Foundation of China(No.61202212)+1 种基金the Special Research Project of the Educational Department of Shaanxi Province of China(No.15JK1038)the Key Research Project of Baoji University of Arts and Sciences(No.ZK16047)
文摘Automatic image annotation has been an active topic of research in computer vision and pattern recognition for decades.A two stage automatic image annotation method based on Gaussian mixture model(GMM) and random walk model(abbreviated as GMM-RW) is presented.To start with,GMM fitted by the rival penalized expectation maximization(RPEM) algorithm is employed to estimate the posterior probabilities of each annotation keyword.Subsequently,a random walk process over the constructed label similarity graph is implemented to further mine the potential correlations of the candidate annotations so as to capture the refining results,which plays a crucial role in semantic based image retrieval.The contributions exhibited in this work are multifold.First,GMM is exploited to capture the initial semantic annotations,especially the RPEM algorithm is utilized to train the model that can determine the number of components in GMM automatically.Second,a label similarity graph is constructed by a weighted linear combination of label similarity and visual similarity of images associated with the corresponding labels,which is able to avoid the phenomena of polysemy and synonym efficiently during the image annotation process.Third,the random walk is implemented over the constructed label graph to further refine the candidate set of annotations generated by GMM.Conducted experiments on the standard Corel5 k demonstrate that GMM-RW is significantly more effective than several state-of-the-arts regarding their effectiveness and efficiency in the task of automatic image annotation.
基金supported by the National Natural Science Foundation of China (Grant No.71101098)the Beijing Municipal Education Commission Foundation of China (Grant Nos. SM201210038008 and 00791154430107)the Funding Project for Academic Human Resources Development in Institutions of Higher Learning under the Jurisdiction of Beijing Municipality (Grant No.PHR201007117)
文摘In real traffic,any vehicle must give lane-changing signal(i.e.the turn signal) before changing lanes;at this time,the vehicles behind the lane-changing vehicle will be hindered and may form "plugs" due to the turn signal effect.However,few studies focus on exploring the effect.In this paper,the turn signal effect was taken into account by proposing a new symmetric two-lane cellular automaton(T-STCA) model,and the new model was set to compare with the STCA,H-STCA and A-STCA models.Numerical results show that using the T-STCA model to describe lane-changing or overtaking,the process appeared in several consecutive time steps;while using the other three models,the process appeared only in one time step.In addition,the T-STCA model could describe the mixed traffic flow more realistically and the turn signal effect could help the plugs to dissolve more quickly.
