多粒度锚点嵌入判别式隐式低秩模糊表示聚类方法和系统

By using a multi-granularity anchor point embedding discriminative implicit low-rank fuzzy representation clustering method, the problem of insufficient noise and multi-scale feature capture in existing image segmentation techniques is solved, achieving high-precision and robust image segmentation results.

CN121788863BActive Publication Date: 2026-07-17BEIJING NORMAL UNIVERSITY
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIJING NORMAL UNIVERSITY
Filing Date
2026-02-24
Publication Date
2026-07-17

Smart Images

  • Figure CN121788863B_ABST
    Figure CN121788863B_ABST
Patent Text Reader

Abstract

本发明涉及基于聚类技术的图像分割技术领域，具体为多粒度锚点嵌入判别式隐式低秩模糊表示聚类方法和系统；为解决现有技术存在的图像分割效果差的技术问题，本发明方法首先通过构建融合双曲正切秩正则化潜在低秩表示、图拉普拉斯算子及模糊正则化项的初始目标函数，兼顾了数据的低秩结构、流形特性与模糊聚类的柔性划分需求；然后引入辅助变量并添加惩罚项构造增广拉格朗日函数，实现了目标函数中变量的完全分离，降低了求解复杂度；接着通过定义关键量梯度计算方式确定优化方向，迭代求解得到最优隶属度矩阵，保障了求解过程的稳定性与结果的准确性；最后基于最优隶属度矩阵完成图像分割，有效提升了图像分割的精度与鲁棒性。
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Image segmentation method and device based on improved entropy regularization fuzzy K-means

    CN117593519A

  • Image segmentation method based on multi-view migration multi-target rough fuzzy clustering

    CN117726816A