In this article, we introduce a new method of image pixel classification. Our methodis a nonparametric classification method which uses combined evidence from the multiple hypothesis testings and minimum distance to carry out the classification. Our work is motivated by the test-based classification introduced by Liao and Akritas [2007].We focus on binary andmulticlass classification of image pixels taking into account of both equal and unequal prior probability of classes. Experiments show that our method works better in classifying image pixels in comparison with some of the standard classification methods such as linear discriminant analysis, quadratic discriminant analysis, classification tree, polyclass method, and Liao and Akritas’s method. We apply our classifier to perform image segmentation. Experiments show that our test-based segmentation has excellent edge detection and texture preservationproperty for both grey scale and color images.
展开▼
机译：在本文中，我们介绍了一种新的图像像素分类方法。我们的方法是一种非参数分类方法，它使用来自多个假设检验和最小距离的组合证据来进行分类。我们的工作是由廖和Akritas [2007]引入的基于测试的分类启发的。我们着重考虑图像类的先验概率相等和不相等的情况，对图像像素进行二进制和多类分类。实验表明，与某些标准分类方法（例如线性判别分析，二次判别分析，分类树，多类方法以及Liao and Akritas方法）相比，我们的方法在对图像像素进行分类时效果更好。我们应用分类器执行图像分割。实验表明，我们基于测试的分割对于灰度和彩色图像都具有出色的边缘检测和纹理保留特性。
展开▼
