考虑类间差异损失的中华白海豚个体识别

Identification of Sousa Chinensis Considering the Inter-Class Loss

为了提升中华白海豚个体识别的准确率,提出基于类间差异损失的细粒度识别模型.首先基于不同类别图像的特征之间具有差异性的事实,设计类间差异损失函数;其次在VGG16的基础上,根据类别数调整部分卷积层的通道数,并设计2种全连接方式用于识别不同规模的个体;最后将所提损失函数与已有损失进行组合,激励网络学习到更具区分度的特征.选取在厦门湾拍摄的2 177幅中华白海豚图像,人工标注为30头个体作为数据集进行实验的结果表明,所提损失函数可以将准确率提升1.05个百分点,达到98.65%,且比主流的细粒度识别算法至少高出0.9个百分点.

To improve the accuracy of individual identification of Sousa Chinensis,a fine-grained identification model based on the loss of inter-class difference is proposed.Firstly,based on the fact that the features of dif-ferent types of images are different,a loss function of inter-class difference is designed;Secondly,the channel number of several convolution layers about VGG16 is adjusted according to the number of categories.And two full connection modes are designed for identification of large-scale and small-scale individuals respectively;Finally,the proposed loss and the existing loss are combined to stimulate the network to learn more distin-guishing features.2177 images of Sousa Chinensis taken in Xiamen Bay are selected and manually labeled as 30 individuals.Experiments on this dataset show that the proposed loss function can improve the accuracy rate by 1.05 percentage points to 98.65%,which is at least 0.9 percentage points higher than the mainstream fine-grained identification algorithms.