Computer Engineering and Applications ›› 2022, Vol. 58 ›› Issue (10): 224-230.DOI: 10.3778/j.issn.1002-8331.2010-0190

• Graphics and Image Processing • Previous Articles     Next Articles

Fine-Grained Ship Image Target Recognition Method Based on Multiple Feature Regions

XU Zhijing, SUN Jiuwu, HUO Yuhao   

  1. College of Information Engineering, Shanghai Maritime University, Shanghai 201306, China
  • Online:2022-05-15 Published:2022-05-15

多特征区域的细粒度船舶图像目标识别方法

徐志京,孙久武,霍煜豪   

  1. 上海海事大学 信息工程学院,上海 201306

Abstract: In order to solve the problem of low recognition accuracy in fine-grained ship images with a single feature, a ship target recognition method based on the fusion of recurrent attention convolutional neural network(RA-CNN) and multi-feature regions is proposed. This method introduces the scale-dependent pooling(SDP) algorithm in the VGG-19 network to solve the problem of excessive pooling of small targets in the network, and improves the recognition performance of small ships. The attention proposal network(APN) introduces joint clustering algorithm to form multiple independent feature regions, so that the whole model can make full use of global information and improve the accuracy of ship recognition. At the same time, a feature region optimization method is designed to reduce the overlap rate of multiple feature region and solve the problem of overfitting. By defining a new loss function to crosstrain the VGG-19 and APN, the convergence is accelerated. The proposed method is tested by using the open optoelectronic ship database, the recognition accuracy is up to 90.2%. Both the recognition rate and the robustness of the model are greatly improved compared with the single feature.

Key words: ship recognition, fine-grained image, multiple feature regions, recurrent attention convolutional neural network(RA-CNN)

摘要: 为解决单一特征细粒度船舶图像识别率低的问题,提出一种循环注意卷积神经网络(recurrent attention convolutional neural network,RA-CNN)与多特征区域融合的船舶目标识别方法。该方法通过在VGG-19网络中引入尺度依赖池化(scale-dependent pooling,SDP)算法解决小目标过度池化的问题,提升了小型船舶的识别性能;注意建议网络(attention proposal network,APN)加入联合聚类(joint clustering)算法,生成多个独立的特征区域,使整个模型充分利用全局信息,提高了船舶识别精度;同时设计特征区域优化方法降低多个特征区域的重叠率,解决了过拟合问题;通过定义新的损失函数来交叉训练VGG-19和APN,加快了收敛速度。利用公开的光电船舶数据集对该方法进行测试实验,识别准确率最高可达90.2%,无论是识别率还是模型的鲁棒性较单特征都有了很大的提升。

关键词: 船舶识别, 细粒度图像, 多特征区域, 循环注意卷积神经网络(RA-CNN)