Description: 傅立叶描述子是分析和识别物体形状的重要方法之一．利用基于曲线多边形近似的连续傅立叶变换方法 计算傅立叶描述子，并通过形状的主方向消除边界起始点相位影响的方法，定义了新的具有旋转、平移和尺度不变 性的归一化傅立叶描述子．与使用离散傅立叶变换和模归一化的传统傅立叶描述子相比，新的归一化傅立叶描述 子同时保留了模与相位特性，因此能够更好地识别物体的形状．实验表明这种新的归一化傅立叶描述子比传统的 傅立叶描述子能够更加高效、准确地识别物体的形状．-Fourier descriptor is to analyze and identify the important objects in the shape of one of the methods. The use of polygon-based curve near
Continuous Fourier transform method like Fourier descriptors, and through the main direction of the shape to eliminate the border from
Phase of the influence of starting point, define a new rotation, translation and scale invariance of the normalized Fourier
Ye descriptor. With the use of discrete Fourier transform and mode normalized compared with the traditional Fourier descriptors, the new
Normalized Fourier descriptors while maintaining the mold and phase characteristics, we are better able to identify the object shape
Like. Experiments show that the new normalized Fourier descriptors than the traditional Fourier descriptors to be more high-
Efficiency, accurately identify the shape of an object. Platform: |
Size: 6646784 |
Author:林传峰 |
Hits:
Search - MultiscaleFourierdescriptor