计算机工程与应用 ›› 2021, Vol. 57 ›› Issue (6): 108-116.DOI: 10.3778/j.issn.1002-8331.1912-0231

• 模式识别与人工智能 • 上一篇    下一篇

基于卷积神经网络的海面显著性目标检测

贺钰博,刘坤   

  1. 上海海事大学 信息工程学院,上海 201306
  • 出版日期:2021-03-15 发布日期:2021-03-12

Detection of Sea-Surface Saliency Object Based on Convolutional Neural Network

HE Yubo, LIU Kun   

  1. College of Information Engineering, Shanghai Maritime University, Shanghai 201306, China
  • Online:2021-03-15 Published:2021-03-12

摘要:

海面环境容易受到云雾等气象因素影响,采集到的海面图像对比度减小,噪声干扰较多,导致目标显著性提取时较难准确完整地获取显著性区域。针对以上问题,提出了一种基于改进的Deeplabv3网络的海面显著性目标检测方法。引用空洞卷积和全局注意力模块提取更多的特征信息。将不同空洞率特征矩阵进行并联,融合图像上下文特征信息。对二分类交叉熵损失函数添加约束项来对云雾遮挡的显著性特征进行约束。通过对大型数据集预训练及海面云雾遮挡数据集的训练后,保存其模型。实验结果表明:提出方法获取的受云雾遮挡干扰时显著性区域变化较小且能够较为完整地描述显著性目标。在遮挡程度为30、50、70情况下,该方法的F-measure值相比于其他几种对比算法平均提高了22.12%、15.83%、13.30%。

关键词: 海面目标, 卷积神经网络, 显著性检测, 深度学习, Deeplabv3, 全局注意力

Abstract:

The sea-surface environment is obscured to meteorological factors such as fog and the contrast of sea-surface images collected is reduced with more noise information interference, which makes it difficult to obtain the completed and accurate significance region when extracting the target significance. To solve the above problems, an improved algorithm is proposed for detecting sea-surface significance object in Deeplabv3 network. More feature information is extracted by using empty convolution and introducing global attention module. Context information of different void rates is connected by fusing the characteristic matrices. Then, the constraint term is added to the binary cross entropy loss function to constrain the significance of cloud occlusion. The model is saved after the training of the large data set and the training of the sea surface cloud shielding data set. Experimental results show that the significance region obtained by the method in this paper can describe the target region completely and the significance region changes undetermined when it is disturbed by the proposed method can describe traget region. The average F-measure value of the proposed method is 22.12%, 15.83% and 13.30% higher than that of other comparison algorithms when the occlusion degree is 30, 50 and 70.

Key words: sea-surface object, convolution neural network, saliency detecting, deep learning, Deeplabv3, global attention