Arbitrary-Oriented Object Detection in Aerial Images with Dynamic Deformable Convolution and Self-Normalizing Channel Attention

Abstract

Objects in aerial images often have arbitrary orientations and variable shapes and sizes. As a result, accurate and robust object detection in aerial images is a challenging problem. In this paper, an arbitrary-oriented object detection method for aerial images, based on Dynamic Deformable Convolution (DDC) and Self-normalizing Channel Attention Mechanism (SCAM), is proposed; this method uses ReResNet-50 as the backbone network to extract rotation-equivariant features. First, DDC is proposed as a replacement for the conventional convolution operation in the Convolutional Neural Network (CNN) in order to cope with various shapes, sizes and arbitrary orientations of the objects. Second, SCAM embedded into the high layer of ReResNet-50, which allows the network to enhance the important feature channels and suppress the irrelevant ones. Finally, Rotation Regions of Interest (RRoI) are generated based on a Region Proposal Network (RPN) and a RoI Transformer (RT), and the RoI-wise classification and bounding box regression are realized by Rotation-invariant RoI Align (RiRoI Align). The proposed method is comprehensively evaluated on three publicly available benchmark datasets. The mean Average Precision (mAP) can reach 80.91%, 92.73% and 94.1% on DOTA-v1.0, DOTA-v1.5 and HRSC2016 datasets, respectively. The experimental results show that, when compared with the state-of-the-arts methods, the proposed method can achieve superior detection accuracy.

Improved Deformable Convolution Method for Aircraft Object Detection in Flight Based on Feature Separation in Remote Sensing Images

“…There are also methods that consider DCN from different aspects than this paper and use it for aircraft object detection. In reference [25], dynamic deformable convolution (DDC) was used to model rotating objects to improve the detection performance of rotating objects. Compared with DCN, adding multiple convolution and nonlinear elements in DDC will increase the number of model parameters and calculation operands.…”

Task-Aligned Oriented Object Detection in Remote Sensing Images

“…OASL [17] is another recently developed detection method, which introduced an orientation-aware structured information extraction module for capturing spatial contextual features. Furthermore, DDC-SCAM [18] is a novel method which integrates dynamic deformable convolution and self-normalizing channel attention for OOD in RSIs. Other recently proposed methods include KFIOU [19], RIDet [20], and CFCNet [21].…”

A Novel Strategy for Extracting Richer Semantic Information Based on Fault Detection in Power Transmission Lines

“…For example, in image classification tasks, channel attention can help the network to better distinguish feature differences between different categories [ 13 , 14 , 15 , 16 , 17 ]. In a target detection task, channel attention improves the network’s ability to accurately locate and recognize targets [ 18 , 19 , 20 , 21 , 22 ]. However, current channel attention modules simply and crudely use global pooling in the compression process, which may lose locally important information and lead to relatively limited modeling of complex semantic relationships.…”

“…For example, in image classification tasks, channel attention can help the network to better distinguish feature differences between different categories [13][14][15][16][17]. In a target detection task, channel attention improves the network's ability to accurately locate and recognize targets [18][19][20][21][22]. However, current channel attention modules simply and crudely use global pooling in the compression Since channel attention in the attention mechanism also has the use of category semantic information, we introduced the channel attention mechanism in the feature fusion phase, as shown in Figure 2, No.…”