Computer Engineering and Applications ›› 2024, Vol. 60 ›› Issue (7): 229-237.DOI: 10.3778/j.issn.1002-8331.2211-0211

• Graphics and Image Processing • Previous Articles     Next Articles

Camouflage Object Detection Algorithm Based on Edge Attention and Reverse Orientation

HE Wenhao, GE Haibo, CHENG Mengyang, AN Yu, MA Sai   

  1. School of Electronic Engineering, Xi’an University of Posts and Telecommunications, Xi’an 710121, China
  • Online:2024-04-01 Published:2024-04-01

边缘注意力及反向定位的伪装目标检测算法

何文昊,葛海波,程梦洋,安玉,马赛   

  1. 西安邮电大学 电子工程学院,西安 710121

Abstract: Camouflage object detection (COD) has important application value in many fields. The existing COD algorithm mainly focuses on the expression of the features extracted from the backbone network and the problem of feature fusion, ignoring the problems of focusing on the edge features of the object and inferring the real area of the object. Aiming at the above problems, a camouflaged object detection algorithm based on edge attention and reverse positioning is proposed. The algorithm consists of edge attention module (EAM), close integration module (CIM) and reverse positioning module (RPM). First, the EAM module is used in the feature encoding stage to enhance the expression of multi-level features extracted from the Res2Net-50 backbone network and highlight edge features. Then, the CIM module is used for the fusion of multi-level features to reduce the loss of feature information. Finally, the RPM module is used to process the rough prediction maps from different feature pyramids, reverse localize the real region of the object, and infer the real object. Experiments on 3 public datasets show that the proposed algorithm outperforms the other 8 state-of-the-art models. On the COD10K dataset, the mean absolute error (MAE) reaches 0.038.

Key words: camouflage object detection (COD), edge attention module (EAM), close integration module (CIM), reverse positioning module (RPM)

摘要: 伪装目标检测(camouflage object detection,COD)在众多领域中有着重要的应用前景。现有COD算法主要针对特征表达以及特征融合的问题进行研究,但是忽略了目标边缘特征的提取和推断目标真实区域的位置。针对上述问题,提出了基于边缘注意力及反向定位的伪装目标检测算法。算法由边缘注意力模块(edge attention module,EAM)、临近融合模块(close integration module,CIM)和反向定位模块(reverse positioning module,RPM)构成。EAM模块用于特征编码阶段,增强从Res2Net-50主干网络提取的多级特征的表达,突出边缘特征。CIM模块促进多层次特征的融合,减少特征信息丢失。使用RPM模块处理来自不同特征金字塔的粗糙预测图,反向定位目标的真实区域,推断出真实目标。在3个公开数据集上的实验表明,该算法优于其他8个最新模型。在COD10K数据集上,平均绝对误差(mean absolute error,MAE)达到了0.038。

关键词: 伪装目标检测, 边缘注意力模块, 临近融合模块, 反向定位模块