充分利用双树复小波变换的旋转不变性、良好的方向选择性以及有限的冗余等优点，将其有效地应用于纹理特征提取过程中。
The excellent characteristics of rotation-invariance, good orientation selection and finite redundancy are fully utilized, and applied in texture feature extraction.
特征提取是手写体汉字识别的关键，目前四方向网格特征已被实验证实是一种较好的手写体汉字特征。
Feature extraction is the key part of handwritten Chinese character recognition. It is found that the directional feature dug by elastic mesh is suitable for Chinese character recognition.
基于多方向纹理边缘检测的特征提取方法利用了纹理的位置、灰度、大小、方向、相关性等多种结构特征，因此提取出来的可区分性特征使虹膜识别的准确性得到大幅度的提高。
The method based on the multidirectional edge detection uses the position, gray, size, direction and relativity of texture, so the dividing features ensure to increase the precision much more.
因此，特征提取线的偏差受限于在任何方向的全跳动公差，而不是在任何方向的径向跳动（3.10）公差。
Therefore the deviations of the generator line of the toleranced feature are limited by the total run-out tolerance in any direction, but not by the circular run-out tolerance in any direction (3.10).
因此，特征提取线的偏差受限于在任何方向的全跳动公差，而不是在任何方向的径向跳动（3.10）公差。
Therefore the deviations of the generator line of the toleranced feature are limited by the total run-out tolerance in any direction, but not by the circular run-out tolerance in any direction (3.10).
应用推荐