Detecting vehicles in aerial imagery plays an important role a wide range of applications. The current vehicle detection methods are mostly based on sliding-window search and handcrafted or shallow-learning-based features, having limited description capability heavy computational costs. Recently, due to the powerful feature representations, region convolutional neural networks (CNN) have achieved state-of-the-art performance computer vision, especially Faster R-CNN. However, directly using...

This work presents a robust graph mapping approach for the unsupervised heterogeneous change detection problem in remote sensing imagery. To address challenge that images cannot be directly compared due to different imaging mechanisms, we take advantage of fact share same structure information ground object, which is modality-invariant. The proposed method first constructs K-nearest neighbor represent each image, and then compares graphs within image domain by means calculate forward...

Vehicle detection with orientation estimation in aerial images has received widespread interest as it is important for intelligent traffic management. This a challenging task, not only because of the complex background and relatively small size target, but also various orientations vehicles captured from top view. The existing methods oriented vehicle need several post-processing steps to generate final results orientation, which are efficient enough. Moreover, they can get discrete...

Change detection (CD) of remote sensing (RS) images is one the important problems in earth observation, which has been extensively studied recent years. However, with development RS technology, specific characteristics remotely sensed images, including sensor characteristics, resolutions, noises, and distortions imagery, make CD more complex. In this article, we propose a structure consistency-based method for CD, detects changes by comparing structures two rather than pixel values images....

Change detection of heterogeneous multitemporal satellite images is an important and challenging topic in remote sensing. Since the imaging mechanisms sensors are different, it not possible to directly compare detect changes as homogeneous images. To address this challenge, we propose unsupervised image regression-based change method based on structure consistency. The proposed first adaptively constructs a similarity graph represent pre-event image, then uses translate domain post-event...

Fast and accurate vehicle detection in unmanned aerial (UAV) images remains a challenge, due to its very high spatial resolution few annotations. Although numerous methods exist, most of them cannot achieve real-time for different scenes. Recently, deep learning algorithms has achieved fantastic performance computer vision, especially regression based convolutional neural networks YOLOv2. It's good both at accuracy speed, outperforming other state-of-the-art methods. This paper the first...

As two different tools for earth observation, the optical and synthetic aperture radar (SAR) images can provide complementary information of same land types better cover classification. However, because imaging mechanisms SAR images, how to efficiently exploit becomes an interesting challenging problem. In this article, we propose a novel multimodal bilinear fusion network (MBFNet), which is used fuse features The MBFNet consists three components: feature extractor, second-order...

Change detection (CD) of remote sensing images is an important and challenging topic, which has found a wide range applications in many fields. In particular, one the main challenges to detect changes between heterogeneous images, where difference imaging mechanism makes it difficult carry out direct comparison. this article, we propose unsupervised CD framework based on patch similarity graph matrix (PSGM), assumes that structure each homogeneous or image consistent if no change occurs....

The coregistration of optical and synthetic aperture radar (SAR) imageries is the bottleneck in exploring complementary information from two multimodal datasets. difficulties lie not only complex radiometric relationship between them, but also distinct geometrical models SAR imaging systems, which cause it nontrivial to explicitly depict spatial corresponding image regions when elevation fluctuations exist. This article aims investigate flow technique for pixelwise dense registration...

Wireless sensor networks (WSNs) is always consist of stationary and mobile nodes, node deployment one the key topics addressed in researches WSNs, traditional virtual force(VF) algorithm presented. This paper proposes a method improved particle swarm optimization to solve problem. The simulation results show that has better performance on problem reduce network energy consumption increase whole coverage ratio.

Existing land cover classification methods mostly rely on either the optical or synthetic aperture radar (SAR) features alone, which ignore mutual complementary effects between and SAR sources. In this article, we compare distribution histograms of deep semantic extracted from modalities within categories, intuitively demonstrates that there are large potentials features. Therefore, propose a novel collaborative attention-based heterogeneous gated fusion network (CHGFNet), hierarchically...

Change detection (CD) of heterogeneous remote sensing images is a challenging topic, which plays an important role in natural disaster emergency response. Due to the different imaging mechanisms sensors, it hard directly compare images. To address this challenge, we explore unsupervised CD method based on adaptive local structure consistency (ALSC) between letter, constructs graph representing for each patch one image domain and then projects other measure change level. This exploits fact...

To fully explore the complementary information from optical and synthetic aperture radar (SAR) imageries, they need first to be coregistered with high accuracy. Due vast radiometric geometric disparity, problem match high-resolution SAR images is quite challenging. The present deep learning-based methods have shown advantages over traditional approaches, but performance increment not significant. In this article, we a better network framework for image matching three aspects. First, propose...

This paper provides a new strategy for the heterogeneous change detection (HCD) problem: solving HCD from perspective of graph signal processing (GSP). We construct to represent structure each image, and treat image as defined on graph. In this way, we convert into GSP comparison responses signals systems graphs, which attempts find structural differences due changes between images. Firstly, analyze vertex domain. show that once region has changed, local changes, <italic...

Multimodal change detection (MCD) is an increasingly interesting but very challenging topic in remote sensing, which due to the unavailability of detecting changes by directly comparing multimodal images from different domains. In this paper, we first analyze structural asymmetry between multitemporal and show their negative impact on previous MCD methods using image structures. Specifically, when there a asymmetry, structure based can only complete comparison or regression one direction...

Aiming at path planning of mobile robot under part dynamic unknown environment, there are some shortages in the aspects produce initial population and structure specific genetic operator current used algorithms. In this paper, using position feedback forecast moving direction obstacle, we present a new method based on improved algorithms combined with numerical potential field. The problem avoiding obstacles environment was resolved by re-planning. shape obstacle is not limited, research...