In this work we apply the differential transformation method or DTM for solving some classes of Lane-Emden type equations as a model for the dimensionless density distribution in an isothermal gas sphere and as a stud...In this work we apply the differential transformation method or DTM for solving some classes of Lane-Emden type equations as a model for the dimensionless density distribution in an isothermal gas sphere and as a study of the gravitational potential of (white-dwarf) stars , which are nonlinear ordinary differential equations on the semi-infinite domain [1] [2]. The efficiency of the DTM is illustrated by investigating the convergence results for this type of the Lane-Emden equations. The numerical results show the reliability and accuracy of this method.展开更多
Objective To determine the positions of marking in the presence of distracting shadows, highlight, pavement cracks, etc. Methods RGB color space is transformed into I 1 I 2 I 3 color space and I 2 ...Objective To determine the positions of marking in the presence of distracting shadows, highlight, pavement cracks, etc. Methods RGB color space is transformed into I 1 I 2 I 3 color space and I 2 component was used to form a new image with less effect of the clutter. Using an improved edge detection operator, an edge strength map was produced, and binarilized by adaptive thresholds. The binary image was labeled and circularity of all connected components is calculated. The Self Organizing Mapping is adopted to extract regions which imply potential marking. Finally the position of marking was obtained by curve fitting. Results Color information was utilized fully, all thresholds were set adaptively and lane marking could be detected in challenging images with shadows, highlight or other cars. Conclusion The method based on circularity of connected components shows its outstanding robustness to lane marking detection and has a wide variety of applications in the areas of vehicle autonomous navigation and driver assistance system.展开更多
基于Transformer的车道预测LSTR(Lane Shape Prediction with Transformers)算法在检测车道线时存在缺少捕捉局部特征的能力和多头注意力机制中头数多余的问题.本文提出了改进LSTR算法的车道线检测方法,首先在最后一个编码器中前馈网络...基于Transformer的车道预测LSTR(Lane Shape Prediction with Transformers)算法在检测车道线时存在缺少捕捉局部特征的能力和多头注意力机制中头数多余的问题.本文提出了改进LSTR算法的车道线检测方法,首先在最后一个编码器中前馈网络的后面引入CBAM(Convolutional Block Attention Module)注意力机制模块,充分利用通道和空间上的信息,捕捉特征图中更多的细节;然后对解码器中的掩码多头注意力机制进行剪枝,使用掩码单头注意力机制来进行替换,以便更多关注前一时刻的车道线信息.改进后的LSTR算法在TuSimple数据集上准确度为96.31%,明显高于PolyLaneNet(Lane Estimation via Deep Polynomial Regression)等算法,在CULane数据集上比原始算法的F1评分上升了2.11%.展开更多
文摘In this work we apply the differential transformation method or DTM for solving some classes of Lane-Emden type equations as a model for the dimensionless density distribution in an isothermal gas sphere and as a study of the gravitational potential of (white-dwarf) stars , which are nonlinear ordinary differential equations on the semi-infinite domain [1] [2]. The efficiency of the DTM is illustrated by investigating the convergence results for this type of the Lane-Emden equations. The numerical results show the reliability and accuracy of this method.
文摘Objective To determine the positions of marking in the presence of distracting shadows, highlight, pavement cracks, etc. Methods RGB color space is transformed into I 1 I 2 I 3 color space and I 2 component was used to form a new image with less effect of the clutter. Using an improved edge detection operator, an edge strength map was produced, and binarilized by adaptive thresholds. The binary image was labeled and circularity of all connected components is calculated. The Self Organizing Mapping is adopted to extract regions which imply potential marking. Finally the position of marking was obtained by curve fitting. Results Color information was utilized fully, all thresholds were set adaptively and lane marking could be detected in challenging images with shadows, highlight or other cars. Conclusion The method based on circularity of connected components shows its outstanding robustness to lane marking detection and has a wide variety of applications in the areas of vehicle autonomous navigation and driver assistance system.
文摘基于Transformer的车道预测LSTR(Lane Shape Prediction with Transformers)算法在检测车道线时存在缺少捕捉局部特征的能力和多头注意力机制中头数多余的问题.本文提出了改进LSTR算法的车道线检测方法,首先在最后一个编码器中前馈网络的后面引入CBAM(Convolutional Block Attention Module)注意力机制模块,充分利用通道和空间上的信息,捕捉特征图中更多的细节;然后对解码器中的掩码多头注意力机制进行剪枝,使用掩码单头注意力机制来进行替换,以便更多关注前一时刻的车道线信息.改进后的LSTR算法在TuSimple数据集上准确度为96.31%,明显高于PolyLaneNet(Lane Estimation via Deep Polynomial Regression)等算法,在CULane数据集上比原始算法的F1评分上升了2.11%.