最后应用神经网络，反复对边缘特征点进行拟合训练，得到平滑的边缘拟合曲线。
Finally, a smoothing fitting curves will be obtained by training of the edge feature using of neural networks.
该算法充分的利用了B -样条函数在边缘拟合上的优势以及小波方法的多尺度优势，算法的处理结果令人满意。
This algorithm makes good use of the B-Spline function's well performance in edge estimation and wavelet 's multi-scales character. The experiment results are satisfactory.
该算法充分的利用了B -样条函数在边缘拟合上的优势以及小波方法的多尺度优势，算法的处理结果令人满意。
This algorithm makes good use of the B-Spline functions well performance in edge estimation and wavelets multi-scales character. The experimental results are satisfactory.
应用推荐