Vehicle detection in the aerial image is an essential and challenging task widely used in industry and agriculture. Deep learning technology has recently achieved rapid development and good object detection results. However, the background of aerial images is complex; targets are densely distributed, and some of them are occluded. For densely distributed targets, we need to predict at each feature point. In the case of complex background and target occlusion, it is often difficult to determine whether a location contains a target if the model only focuses on the local information. Therefore, we need a global perspective and contextual information to help train the model. This letter proposes a new anchor-free small object detection algorithm, which improves feature extraction by fusing contextual semantic information. In addition, a dynamic activation function (DAF) is also used in our network, which helps us calculate the activation function value for each point from a global perspective. Moreover, we also use the channel attention module and the transformer as the spatial attention module to help the network efficiently obtain global information. We evaluate the effectiveness of our method on the public dataset DLR-3K and vehicle detection in aerial imagery dataset (VEDAI), and the average precision (AP) achieves 0.896 and 0.875.

中文翻译：

---

基于上下文信息和Transformer的航拍图像无锚车辆检测算法


航空图像中的车辆检测是工业和农业中广泛应用的一项重要且具有挑战性的任务。深度学习技术近年来取得了快速发展并取得了良好的物体检测效果。但航拍图像背景复杂，目标分布密集，部分目标被遮挡。对于密集分布的目标，我们需要在每个特征点进行预测。在背景复杂、目标遮挡的情况下，如果模型只关注局部信息，往往很难判断某个位置是否包含目标。因此，我们需要全局视角和上下文信息来帮助训练模型。这封信提出了一种新的无锚小目标检测算法，该算法通过融合上下文语义信息来改进特征提取。此外，我们的网络中还使用了动态激活函数（DAF），这有助于我们从全局角度计算每个点的激活函数值。此外，我们还使用通道注意力模块和变压器作为空间注意力模块来帮助网络有效地获取全局信息。我们在公共数据集 DLR-3K 和航空影像数据集车辆检测（VEDAI）上评估了我们的方法的有效性，平均精度（AP）达到 0.896 和 0.875。