It is still challenging to effectively detect ship objects in optical remote-sensing images with complex backgrounds. Many current CNN-based one-stage and two-stage detection methods usually first predefine a series of anchors with various scales, aspect ratios and angles, and then the detection results can be outputted by performing once or twice classification and bounding box regression for predefined anchors. However, most of the defined anchors have relatively low accuracy, and are useless for the following classification and regression. In addition, the preset anchors are not robust to produce good performance for other different detection datasets. To avoid the above problems, in this paper we design a paired semantic segmentation network to generate more accurate rotated anchors with smaller numbers. Specifically, the paired segmentation network predicts four parts (i.e., top-left, bottom-right, top-right, and bottom-left parts) of ships. By combining paired top-left and bottom-right parts (or top-right and bottom-left parts), we can take the minimum bounding box of these two parts as the rotated anchor. This way can be more robust to different ship datasets, and the generated anchors are more accurate and have fewer numbers. Furthermore, to effectively use fine-scale detail information and coarse-scale semantic information, we use the magnified convolutional features to classify and regress the generated rotated anchors. Meanwhile, the horizontal minimum bounding box of the rotated anchor is also used to combine more context information. We compare the proposed algorithm with state-of-the-art object-detection methods for natural images and ship-detection methods, and demonstrate the superiority of our method.

With the successful application of the convolutional neural network (CNN), significant progress has been made by CNN-based ship detection methods. However, they often face considerable difficulties when applied to a new domain where the imaging condition changes significantly. Although training with the two domains together can solve this problem to some extent, the large domain shift will lead to sub-optimal feature representations, and thus weaken the generalization ability on both domains. In this paper, a domain adaptive ship detection method is proposed to better detect ships between different domains. Specifically, the proposed method minimizes the domain discrepancies via both image-level adaption and instance-level adaption. In image-level adaption, we use multiple receptive field integration and channel domain attention to enhance the feature’s resistance to scale and environmental changes, respectively. Moreover, a novel boundary regression module is proposed in instance-level adaption to correct the localization deviation of the ship proposals caused by the domain shift. Compared with conventional regression approaches, the proposed boundary regression module is able to make more accurate predictions via the effective extreme point features. The two adaption components are implemented by learning the corresponding domain classifiers respectively in an adversarial training way, thereby obtaining a robust model suitable for both of the two domains. Experiments on both supervised and unsupervised domain adaption scenarios are conducted to verify the effectiveness of the proposed method.

No abstract

Infrared and visible images (multi-sensor or multi-band images) have many complementary features which can effectively boost the performance of object detection. Recently, convolutional neural networks (CNNs) have seen frequent use to perform object detection in multi-band images. However, it is very difficult for CNNs to extract complementary features from infrared and visible images. In order to solve this problem, a difference maximum loss function is proposed in this paper. The loss function can guide the learning directions of two base CNNs and maximize the difference between features from the two base CNNs, so as to extract complementary and diverse features. In addition, we design a focused feature-enhancement module to make features in the shallow convolutional layer more significant. In this way, the detection performance of small objects can be effectively improved while not increasing the computational cost in the testing stage. Furthermore, since the actual receptive field is usually much smaller than the theoretical receptive field, the deep convolutional layer would not have sufficient semantic features for accurate detection of large objects. To overcome this drawback, a cascaded semantic extension module is added to the deep layer. Through simple multi-branch convolutional layers and dilated convolutions with different dilation rates, the cascaded semantic extension module can effectively enlarge the actual receptive field and increase the detection accuracy of large objects. We compare our detection network with five other state-of-the-art infrared and visible image object detection networks. Qualitative and quantitative experimental results prove the superiority of the proposed detection network.

No abstract