形态识别包括：局部特征点及其方向，弯曲的结构特征，整体描述参数以及拓扑特征。
The shape identification includes the local character point and its direction, the curve structure, the global parameter and topology trait.
根据像素点的梯度方向和直线局部特性提出了一种新的亚像素直线特征检测方法。
A new method is proposed according to the pixels gradient direction and the lines local property.
首先提出了局部图像的梯度估计方法，称之为BEG(梯度的最佳估计)算法，用来提取图像的纹理方向特征参数。
First, a method to estimate the local image gradient, called the BEG (best estimation of gradient) algorithm was proposed to extract the texture orientation parameters.
而且为了保持图像特征，利用图像的方向导数局部地调整非线性扩散系数。
To preserve image features, the nonlinear diffusion coefficients are locally adjusted according to the directional derivatives of the image.
为了保持图像的特征和细节，文章利用图像的方向导数局部地调整正则系数。
To preserve image features and fine part, the regular coefficient is locally adjusted according to the directional derivatives of the image.
指纹图像的周期性和局部方向性是指纹图像的重要结构特征，充分利用这两个结构特征是提高指纹增强算法的重要方法。
Periods and local direction of fingerprint image are the important structure characters. The two characters of structure are important in the fingerprint enhancement.
为了更有效地利用图像的局部特征恢复被噪声感染的图像，基于图像局部纹理方向概率统计模型，提出一种针对混合噪声的非线性滤波算法。
In order to more effectively make use of local features to restore the noise-infected image, a nonlinear filtering algorithm based on local texture direction probability statistic model was proposed.
构建了一种新的局部梯度方向直方图，同时定义了特征点的主方向，从而提出了一种具有旋转不变性的图像配准算法。
A novel gradient orientation histogram was built using a Gaussian-weighted circular window and a major orientation to each feature point was assigned based on local image properties.
构建了一种新的局部梯度方向直方图，同时定义了特征点的主方向，从而提出了一种具有旋转不变性的图像配准算法。
A novel gradient orientation histogram was built using a Gaussian-weighted circular window and a major orientation to each feature point was assigned based on local image properties.
应用推荐