A Single Shot Framework with Multi-Scale Feature Fusion for Geospatial Object Detection

Abstract

With the rapid advances in remote-sensing technologies and the larger number of satellite images, fast and effective object detection plays an important role in understanding and analyzing image information, which could be further applied to civilian and military fields. Recently object detection methods with region-based convolutional neural network have shown excellent performance. However, these two-stage methods contain region proposal generation and object detection procedures, resulting in low computation speed. Because of the expensive manual costs, the quantity of well-annotated aerial images is scarce, which also limits the progress of geospatial object detection in remote sensing. In this paper, on the one hand, we construct and release a large-scale remote-sensing dataset for geospatial object detection (RSD-GOD) that consists of 5 different categories with 18,187 annotated images and 40,990 instances. On the other hand, we design a single shot detection framework with multi-scale feature fusion. The feature maps from different layers are fused together through the up-sampling and concatenation blocks to predict the detection results. High-level features with semantic information and low-level features with fine details are fully explored for detection tasks, especially for small objects. Meanwhile, a soft non-maximum suppression strategy is put into practice to select the final detection results. Extensive experiments have been conducted on two datasets to evaluate the designed network. Results show that the proposed approach achieves a good detection performance and obtains the mean average precision value of 89.0% on a newly constructed RSD-GOD dataset and 83.8% on the Northwestern Polytechnical University very high spatial resolution-10 (NWPU VHR-10) dataset at 18 frames per second (FPS) on a NVIDIA GTX-1080Ti GPU.

Multi-Scale Remote Sensing Semantic Analysis Based on a Global Perspective

“…The atrous filter enlarges the FOV without increasing the number of parameters to be calculated, and thus saves computational resources. Additionally, high-level features containing semantic information and low level features containing fine details are fused together through up-sampling and concatenation [28] so that different layers of feature are both considered for detection tasks, especially for small object detection. In a similar model [29], the deep residual network (ResNet) is used as the encoder, and high level features are combined with corresponding low-level features as the up-sampling stage decoder.…”

“…The multi-scale concept in this study refers to the relationship between the local and the global, that is, a small local area and a large area within a certain neighborhood. However, the current multi-scale research focuses on the multi-scale feature extraction and fusion of the same training sample image [24][25][26][27][28][29][30][31]. There are few studies on how to consider the semantic analysis of variable regions in different spatial extents [32][33][34][35].…”

A Novel Effectively Optimized One-Stage Network for Object Detection in Remote Sensing Imagery

“…On this basis, object detection in remote sensing imagery has been widely studied in recent years [27][28][29][30][31][32]. In the field of remote sensing, many researchers have made great efforts to object detection methods based on CNN [33][34][35][36][37][38][39].…”

“…In addition, the final detection results are obtained by the way of making a decision fusion on the results of the three sub-networks. Based on the YOLOv2 [23], a single-shot geospatial object detection framework based on multi-scale feature fusion modules has been proposed in [33]. Note that detectors in this model are used in conjunction with multi-scale feature fusion modules.…”

“…However, there are still some challenges [5,33] to be solved in geospatial object detection even though abovementioned excellent methods have been put forward because of the characteristic of remote sensing image as follows.…”