基于Contourlet变换与最小二乘支持向量机(least squares support vector machine,LSSVM),提出了一种玉米种子高精度识别算法。该算法首先对玉米种子图像进行多层Contourlet分解,结合指数函数和反正弦函数,提出了一种新型的阈值函数模...基于Contourlet变换与最小二乘支持向量机(least squares support vector machine,LSSVM),提出了一种玉米种子高精度识别算法。该算法首先对玉米种子图像进行多层Contourlet分解,结合指数函数和反正弦函数,提出了一种新型的阈值函数模型对高频分解系数进行去噪处理;其次,将低频分解系数与去噪后的高频分解系数进行重构,得到去噪后的玉米种子图像;最后采用LSSVM对去噪后的玉米种子图像进行识别,采用径向基函数模型作为LSSVM核函数模型。试验结果表明,对去噪后的图像进行LSSVM识别的精度优于直接对图像进行LSSVM、SVM识别的精度。展开更多
The flow behavior in porous media with threshold pressure gradient(TPG) is more complex than Darcy flow and the equations of motion, and outer boundary and inner boundary with TPG are also different from Darcy flow fo...The flow behavior in porous media with threshold pressure gradient(TPG) is more complex than Darcy flow and the equations of motion, and outer boundary and inner boundary with TPG are also different from Darcy flow for unsteady flow of a producing well in a reservoir. An analytic method to solve this kind of problem is in a need of reestablishment. The classical method of Green's function and Newman product principle in a new way are used to solve the unsteady state flow problems of various shapes of well and reservoir while considering the TPG. Four Green's functions of point, line, band and circle while considering the TPG are achieved. Then, two well models of vertical well and horizontal well are built and simultaneously the function to calculate the moving boundary of each well model is provided. The results show that when considering TPG the pressure field is much different, which has a sudden pressure change, with a moving boundary in it. And the moving boundary of each well model increases with time but slows down rapidly, especially when the TGP is large.展开更多
文摘基于Contourlet变换与最小二乘支持向量机(least squares support vector machine,LSSVM),提出了一种玉米种子高精度识别算法。该算法首先对玉米种子图像进行多层Contourlet分解,结合指数函数和反正弦函数,提出了一种新型的阈值函数模型对高频分解系数进行去噪处理;其次,将低频分解系数与去噪后的高频分解系数进行重构,得到去噪后的玉米种子图像;最后采用LSSVM对去噪后的玉米种子图像进行识别,采用径向基函数模型作为LSSVM核函数模型。试验结果表明,对去噪后的图像进行LSSVM识别的精度优于直接对图像进行LSSVM、SVM识别的精度。
基金Project(51304220) supported by the National Natural Science Foundation of ChinaProject(3144033) supported by the Beijing Natural Science Foundation,ChinaProject(20130007120014) supported by the Specialized Research Fund for the Doctoral Program of Higher Education of China
文摘The flow behavior in porous media with threshold pressure gradient(TPG) is more complex than Darcy flow and the equations of motion, and outer boundary and inner boundary with TPG are also different from Darcy flow for unsteady flow of a producing well in a reservoir. An analytic method to solve this kind of problem is in a need of reestablishment. The classical method of Green's function and Newman product principle in a new way are used to solve the unsteady state flow problems of various shapes of well and reservoir while considering the TPG. Four Green's functions of point, line, band and circle while considering the TPG are achieved. Then, two well models of vertical well and horizontal well are built and simultaneously the function to calculate the moving boundary of each well model is provided. The results show that when considering TPG the pressure field is much different, which has a sudden pressure change, with a moving boundary in it. And the moving boundary of each well model increases with time but slows down rapidly, especially when the TGP is large.
