基于深度特征聚类的商标检索方法

Trademark Retrieval Method Based on Deep Feature Clustering

商标的注册需要规避已注册相同或较为相似的商标,如何从海量的商标数据当中快速准确地检索出与商标数据集中相似度较高的商标,就变得尤为重要。在传统的商标检索方法中改进了对商标特征的聚类算法,使其得到的中心点中的数据平均化,相似度较大的商标划分到同一个中心点且每个中心点对应的数据量较为平均,在相似中心点中进行相似度对比执行商标检索任务,当其应用在庞大的商标库上时,能够达到较好的检索性能。

The registration of trademarks needs to avoid the registration of identical or relatively similar trademarks.How to quickly and accurately retrieve trademarks with high similarity to the trademark dataset from the massive trademark data has become particularly important.In traditional trademark retrieval methods,the clustering algorithm for trademark features has been improved to average the data in the center points obtained.Trademarks with high similarity are divided into the same center point and each center point corresponds to an average amount of data.Comparing the similarity among similar center points to perform trademark retrieval tasks can achieve good retrieval performance when applied to a large trademark database.