Deformable ConvNet with Aspect Ratio Constrained NMS for Object Detection in Remote Sensing Imagery
نویسندگان
چکیده
Convolutional neural networks (CNNs) have demonstrated their ability object detection of very high resolution remote sensing images. However, CNNs have obvious limitations for modeling geometric variations in remote sensing targets. In this paper, we introduced a CNN structure, namely deformable ConvNet, to address geometric modeling in object recognition. By adding offsets to the convolution layers, feature mapping of CNN can be applied to unfixed locations, enhancing CNNs’ visual appearance understanding. In our work, a deformable region-based fully convolutional networks (R-FCN) was constructed by substituting the regular convolution layer with a deformable convolution layer. To efficiently use this deformable convolutional neural network (ConvNet), a training mechanism is developed in our work. We first set the pre-trained R-FCN natural image model as the default network parameters in deformable R-FCN. Then, this deformable ConvNet was fine-tuned on very high resolution (VHR) remote sensing images. To remedy the increase in lines like false region proposals, we developed aspect ratio constrained non maximum suppression (arcNMS). The precision of deformable ConvNet for detecting objects was then improved. An end-to-end approach was then developed by combining deformable R-FCN, a smart fine-tuning strategy and aspect ratio constrained NMS. The developed method was better than a state-of-the-art benchmark in object detection without data augmentation.
منابع مشابه
Low Cost UAV-based Remote Sensing for Autonomous Wildlife Monitoring
In recent years, developments in unmanned aerial vehicles, lightweight on-board computers, and low-cost thermal imaging sensors offer a new opportunity for wildlife monitoring. In contrast with traditional methods now surveying endangered species to obtain population and location has become more cost-effective and least time-consuming. In this paper, a low-cost UAV-based remote sensing platform...
متن کاملEnd-to-End Integration of a Convolution Network, Deformable Parts Model and Non-Maximum Suppression
Deformable Parts Models and Convolutional Networks each have achieved notable performance in object detection. Yet these two approaches find their strengths in complementary areas: DPMs are well-versed in object composition, modeling fine-grained spatial relationships between parts; likewise, ConvNets are adept at producing powerful image features, having been discriminatively trained directly ...
متن کاملSpatio-temporal distribution of off-shore ships in the Pars Special Economic Energy Zone based on satellite imagery
Special Economic Zones (SEZs) are areas controlled by specific legislations so as toattain economic prosperity. These zones are commonly established and controlled bygovernment officials and are primarily characterized by growing population and developingtransport infrastructure. One relevant case is the Pars Special Economic Energy Zone(PSEEZ) situated in the south of Iran, on the northern sho...
متن کاملA Total Ratio of Vegetation Index (TRVI) for Shrubs Sparse Cover Delineating in Open Woodland
Persian juniper and Pistachio are grown in low density in the rangelands of North-East of Iran. These rangelands are populated by evergreen conifers, which are widespread and present at low-density and sparse shrub of pistachio in Iran, that are not only environmentally but also genetically essential as seed sources for pistachio improvement in orchards. Rangelands offer excellent opportunities...
متن کاملA Convnet for Non-maximum Suppression
Non-maximum suppression (NMS) is used in virtually all state-of-the-art object detection pipelines. While essential object detection ingredients such as features, classifiers and proposal methods have been extensively researched it it surprising how little work has aimed to systematically address NMS. The de-facto standard for NMS is based on greedy clustering with a fixed distance threshold, w...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Remote Sensing
دوره 9 شماره
صفحات -
تاریخ انتشار 2017