A neural statistical approach to the reconstruction of novel viewpoint image us ing general regression neural networks(GRNN) is presented. Different color value will be obtained by watching the same surface point of a...A neural statistical approach to the reconstruction of novel viewpoint image us ing general regression neural networks(GRNN) is presented. Different color value will be obtained by watching the same surface point of an object from different viewpoints due to specular reflection, and the difference is related to the pos ition of viewpoint. The relationship between the position of viewpoint and the c olor of image is non linear, neural network is introduced to make curve fitting , where the inputs of neural network are only a few calibrated images with obvio us specular reflection. By training the neural network, network model is obtaine d. By inputing an arbitrary virtual viewpoint to the model, the image of the vir tual viewpoint can be computed. By using the method presented here, novel viewpo int image with photo realistic property can be obtained, especially images with obvious specular reflection can accurately be generated. The method is an image based rendering method, geometric model of the scene and position of lighting are not needed.展开更多
Vector-to-raster conversion is a process accompanied with errors.The errors are classified into predicted errors before rasterization and actual errors after that.Accurate prediction of the errors is beneficial to dev...Vector-to-raster conversion is a process accompanied with errors.The errors are classified into predicted errors before rasterization and actual errors after that.Accurate prediction of the errors is beneficial to developing reasonable rasterization technical schemes and to making products of high quality.Analyzing and establishing a quantitative relationship between the error and its affecting factors is the key to error prediction.In this study,land cover data of China at a scale of 1:250 000 were taken as an example for analyzing the relationship between rasterization errors and the density of arc length(DA),the density of polygon(DP) and the size of grid cells(SG).Significant correlations were found between the errors and DA,DP and SG.The correlation coefficient(R2) of a model established based on samples collected in a small region(Beijing) reaches 0.95,and the value of R2 is equal to 0.91 while the model was validated with samples from the whole nation.On the other hand,the R2 of a model established based on nationwide samples reaches 0.96,and R2 is equal to 0.91 while it was validated with the samples in Beijing.These models depict well the relationships between rasterization errors and their affecting factors(DA,DP and SG).The analyzing method established in this study can be applied to effectively predicting rasterization errors in other cases as well.展开更多
A transition diagram is used to describe the behavior of systems. Birth-death equations were derived from transition diagram depicting the state of the birth-death processes. Queue models and characteristics of queue ...A transition diagram is used to describe the behavior of systems. Birth-death equations were derived from transition diagram depicting the state of the birth-death processes. Queue models and characteristics of queue models are also derivable from birth-death processes. These queue models consist of mathematical formulas and relationships that can be used to determine the operating characteristics （performance measures） for a waiting line. Schematic and transition diagrams of different single server queue models were shown. Relationships between birth-death processes, waiting lines （queues） and transition diagrams were given. While M/M/I/K queue model states was limited by K customers and had （K＋I） states, M/M/1/1 queue model had only two states. M/G/1/∝/∝ and M/M/1/∝/∝ shared similar characteristics. Many ideal queuing situations employ M/M/1 queueing model.展开更多
文摘A neural statistical approach to the reconstruction of novel viewpoint image us ing general regression neural networks(GRNN) is presented. Different color value will be obtained by watching the same surface point of an object from different viewpoints due to specular reflection, and the difference is related to the pos ition of viewpoint. The relationship between the position of viewpoint and the c olor of image is non linear, neural network is introduced to make curve fitting , where the inputs of neural network are only a few calibrated images with obvio us specular reflection. By training the neural network, network model is obtaine d. By inputing an arbitrary virtual viewpoint to the model, the image of the vir tual viewpoint can be computed. By using the method presented here, novel viewpo int image with photo realistic property can be obtained, especially images with obvious specular reflection can accurately be generated. The method is an image based rendering method, geometric model of the scene and position of lighting are not needed.
基金Under the auspices of Strategic Priority Research Program of Chinese Academy of Sciences(No.XDA05050000)Special Program for Informatization of Chinese Academy of Sciences(No.INF0-115-C01-SDB3-02)
文摘Vector-to-raster conversion is a process accompanied with errors.The errors are classified into predicted errors before rasterization and actual errors after that.Accurate prediction of the errors is beneficial to developing reasonable rasterization technical schemes and to making products of high quality.Analyzing and establishing a quantitative relationship between the error and its affecting factors is the key to error prediction.In this study,land cover data of China at a scale of 1:250 000 were taken as an example for analyzing the relationship between rasterization errors and the density of arc length(DA),the density of polygon(DP) and the size of grid cells(SG).Significant correlations were found between the errors and DA,DP and SG.The correlation coefficient(R2) of a model established based on samples collected in a small region(Beijing) reaches 0.95,and the value of R2 is equal to 0.91 while the model was validated with samples from the whole nation.On the other hand,the R2 of a model established based on nationwide samples reaches 0.96,and R2 is equal to 0.91 while it was validated with the samples in Beijing.These models depict well the relationships between rasterization errors and their affecting factors(DA,DP and SG).The analyzing method established in this study can be applied to effectively predicting rasterization errors in other cases as well.
文摘A transition diagram is used to describe the behavior of systems. Birth-death equations were derived from transition diagram depicting the state of the birth-death processes. Queue models and characteristics of queue models are also derivable from birth-death processes. These queue models consist of mathematical formulas and relationships that can be used to determine the operating characteristics （performance measures） for a waiting line. Schematic and transition diagrams of different single server queue models were shown. Relationships between birth-death processes, waiting lines （queues） and transition diagrams were given. While M/M/I/K queue model states was limited by K customers and had （K＋I） states, M/M/1/1 queue model had only two states. M/G/1/∝/∝ and M/M/1/∝/∝ shared similar characteristics. Many ideal queuing situations employ M/M/1 queueing model.
