结合动态自适应调制和结构关系学习的细粒度图像分类

Fine Grained Image Classification Combining Dynamic Adaptive Modulation and Structural Relationship Learning

由于细粒度图像类间差异小,类内差异大的特点,因此细粒度图像分类任务关键在于寻找类别间细微差异.最近,基于Vision Transformer的网络大多侧重挖掘图像最显著判别区域特征.这存在两个问题:首先,网络忽略从其他判别区域挖掘分类线索,容易混淆相似类别;其次,忽略了图像的结构关系,导致提取的类别特征不准确.为解决上述问题,本文提出动态自适应调制和结构关系学习两个模块,通过动态自适应调制模块迫使网络寻找多个判别区域,再利用结构关系学习模块构建判别区域间结构关系;最后利用图卷积网络融合语义信息和结构信息得出预测分类结果.所提出的方法在CUB-200-2011数据集和NA-Birds数据集上测试准确率分别达到92.9%和93.0%,优于现有最先进网络.

Due to the small inter-class differences and large intra-class differences of fine-grained images,the key to finegrained image classification tasks is to find subtle differences between categories.Recently,Vision Transformer-based networks mostly focus on mining the most prominent discriminative region features in images.There are two problems with this.Firstly,the network ignores mining classification clues from other discriminative regions,which can easily confuse similar categories.secondly,the structural relationships of images are ignored,resulting in inaccurate extraction of category features.To solve the above problems,this study proposes two modules:dynamic adaptive modulation and structural relationship learning.The dynamic adaptive modulation module forces the network to search for multiple discriminative regions,and then the structural relationship learning module is used to construct structural relationships between discriminative regions.Finally,the graph convolutional network is used to fuse semantic and structural information to obtain predicted classification results.The proposed method achieves testing accuracy of 92.9%and 93.0%on the CUB-200-2011 dataset and NA-Birds dataset,respectively,which is superior to existing state-of-the-art networks.