基于(I,O_(s))-模糊粗糙集的图像边缘提取和特征选择应用

Semi-overlap Function Based on Fuzzy Rough Set and its Application in Image Processing and Feature Selection

首先,利用半重叠函数和模糊粗糙集,构造了(I,O_(s))-模糊粗糙集模型,提高了模型的灵活性和适应性;其次,将该模型与模糊C-均值算法相结合,提出了一种新的图像边缘提取算法,该算法在较低引入率的情况下,能够提取到完整的图像边缘;最后,基于(I,O_(s))-模糊粗糙集模型,提出了一种新的特征选择算法,相较于传统的特征选择算法,该算法在保持高分类精度的情况下,可以选择更少的属性条件.实验证明,基于(I,O_(s))-模糊粗糙集模型的算法具有可行性和有效性,可以为相关研究和领域提供支持.

In this paper,we innovatively propose the(I,O_(s))-fuzzy rough set model by combining the semi-overlap function and fuzzy rough set.Theoretically,we utilize the semi-overlap function with looser constraints to construct the upper approximation operator of the model,which makes the model have higher flexibility and adaptability.In practice,we combine the model with the fuzzy C-mean algorithm to propose a new image edge extraction algorithm.The algorithm is able to accurately extract complete image edges at a lower noise introduction rate.Meanwhile,we also propose a new feature selection algorithm based on the(I,O_(s))-fuzzy rough set model,which is able to select fewer attribute conditions compared to the traditional algorithm while maintaining high classification accuracy.Experiments prove that our proposed model and algorithm are feasible and effective,and provide strong support for research and application in related fields.