Learning Adjustable Reduced Downsampling Network for Small Object Detection in Urban Environments
Learning Adjustable Reduced Downsampling Network for Small Object Detection in Urban Environments
复制标题
DOI：
10.3390/rs13183608
复制
发表时间：
2021-09
期刊：
影响因子：
--
通讯作者：
Huijie Zhang;Li An;V. Chu;D. Stow;Xiaobai Liu;Qinghua Ding
中科院分区：
文献类型：
--
作者：
Huijie Zhang;Li An;V. Chu;D. Stow;Xiaobai Liu;Qinghua Ding
检测小物体（例如，井盖、车牌和路边里程碑）是一个长期存在的挑战，主要是由于小物体的尺度和背景杂乱。尽管基于卷积神经网络（CNN）的方法在通用目标检测方面取得了重大进展并取得了令人印象深刻的结果，但小目标检测问题仍未解决。为了应对这一挑战，在这项研究中，我们开发了一个端到端的网络架构，有三个显着的特点相比，以前的作品。首先，我们设计了一个骨干网络模块，即减少下采样网络（RD-Net），以提取具有高空间分辨率的信息特征表示，并保留小对象的局部信息。其次，我们引入了一个可调节样本选择（ADSS）模块，该模块释放了Intersection-over-Union（IoU）阈值超参数，并根据生成的锚点和地面参考边界框之间的统计特征定义了正和负训练样本。第三，我们将广义交并（GIoU）损失用于边界框回归，这有效地弥合了基于距离的优化损失和基于区域的评估指标之间的差距。我们证明了我们的方法的有效性进行了广泛的实验上获得的公共城市元素检测（UED）数据集的移动的地图系统（MMS）。该方法的平均精度（AP）为81.71%，与流行的检测框架Faster R-CNN相比提高了1.2%。
Detecting small objects (e.g., manhole covers, license plates, and roadside milestones) in urban images is a long-standing challenge mainly due to the scale of small object and background clutter. Although convolution neural network (CNN)-based methods have made significant progress and achieved impressive results in generic object detection, the problem of small object detection remains unsolved. To address this challenge, in this study we developed an end-to-end network architecture that has three significant characteristics compared to previous works. First, we designed a backbone network module, namely Reduced Downsampling Network (RD-Net), to extract informative feature representations with high spatial resolutions and preserve local information for small objects. Second, we introduced an Adjustable Sample Selection (ADSS) module which frees the Intersection-over-Union (IoU) threshold hyperparameters and defines positive and negative training samples based on statistical characteristics between generated anchors and ground reference bounding boxes. Third, we incorporated the generalized Intersection-over-Union (GIoU) loss for bounding box regression, which efficiently bridges the gap between distance-based optimization loss and area-based evaluation metrics. We demonstrated the effectiveness of our method by performing extensive experiments on the public Urban Element Detection (UED) dataset acquired by Mobile Mapping Systems (MMS). The Average Precision (AP) of the proposed method was 81.71%, representing an improvement of 1.2% compared with the popular detection framework Faster R-CNN.