Abstract
Object detection is facing various challenges as an important aspect in the field of remote sensing—especially in large scenes due to the increase of satellite image resolution and the complexity of land covers. Because of the diversity of the appearance of track and fields, the complexity of the background and the variety between satellite images, even superior deep learning methods have difficulty extracting accurate characteristics of track and field from large complex scenes, such as the whole of China. Taking track and field as a study case, we propose a stable and accurate method for target detection. Firstly, we add the “deconvolution” and “concat” module to the structure of the original Single Shot MultiBox Detector (SSD), where Visual Geometry Group 16 (VGG16) is served as a basic network, followed by multiple convolution layers. The two modules are used to sample the high-level feature map and connect it with the low-level feature map to form a new network structure multi-scale-fused SSD (abbreviated as MSF_SSD). MSF-SSD can enrich the semantic information of the low-level feature, which is especially effective for small targets in large scenes. In addition, a large number of track and fields are collected as samples for the whole China and a series of parameters are designed to optimize the MSF_SSD network through the deep analysis of sample characteristics. Finally, by using MSF_SSD network, we achieve the rapid and automatic detection of meter-level track and fields in the country for the first time. The proposed MSF_SSD model achieves 97.9% mean average precision (mAP) on validation set which is superior to the 88.4% mAP of the original SSD. Apart from this, the model can achieve an accuracy of 94.3% while keeping the recall rate in a high level (98.8%) in the nationally distributed test set, outperforming the original SSD method.
Highlights
Remote-sensing information extraction has encountered unprecedented opportunities and challenges
Experts and scholars have proposed many detection methods, from the earliest target matching method based on template matching, the method based on prior knowledge to the target detection method based on object image analysis [1]
The results showed that deep learning-based methods is far better than traditional algorithm in target detection [33,34,35]
Summary
Remote-sensing information extraction has encountered unprecedented opportunities and challenges. Since the 1990s, machine learning methods, such as support vector machine [12], CRF [13], K-NN [14], boosting [15] and random Forest [16], can be applied to a small area by automatically constructing information extraction models based on a small number of sample data set through training and learning [17,18,19,20], and become the mainstream algorithms for remote sensing target detection. With the continuous improvements on its network models, breakthrough progress has been made in target detection, and the accuracy in multiple tasks has exceeded that of manual identification These algorithms comprise Region-CNN (R-CNN) [23], Spatial Pyramid Pooling Convolutional Networks (SPP-Net) [24], Fast R-CNN [25], Faster R-CNN [26], You Only Look Once (YOLO) [27], SSD [28], etc.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
