中文摘要：

为了解决模式识别应用中传统的不变量特征之间的相关性问题,基于Zernike矩提出一种构造其完备的相似变换不变量集的新方法.首先,根据图像的Zernike矩与径向矩之间的关系,以径向矩为中间桥梁,建立原图像的Zernike矩和旋转缩放后图像的Zernike矩之间的关系,然后由原图像的同阶和低阶Zernike矩线性组合即可得到完备的Zernike矩旋转和缩放不变量集.类似地,可以构造完备的Zernike矩平移不变量.并将两者结合最终得到Zernike矩的相似变换不变量完备集.图像分类实验结果表明,与现有的一些方法相比,所提出的方法在分类正确率和运算时间方面的效果更好,具有较强的噪声鲁棒性.

英文摘要：

To resolve the dependence problem of traditional invariant features in pattern recognition,a new method to derive a complete set of Zernike moments similarity invariants is presented.Firstly,based on the connection between Zernike moments and radial moments,the relationship between Zernike moments of the original image and those of the images having the same shape but distinct scale and orientation is established.Then through a linear combination of original Zernike moments with same and lower order a complete set of scale and rotation invariants can be formed.By the same method,a complete set of translation invariants can be derived.Finally,the complete set of Zernike moment similarity invariants（rotation,scale and translation,RST） can be obtained.Experimental results demonstrate that the proposed method performs better than the existing methods in both classification accuracy rate and computational time.