Massive landslides over large regions can be triggered by heavy rainfalls or major seismic events. Mapping regional landslides quickly is important for disaster mitigation. In recent years, deep learning methods have been successfully applied in many fields, including landslide automatic identification. In this work, we proposed a deep learning approach, the ResU-Net, to map regional landslides automatically. This method and a baseline model (U-Net) were collectively tested in Tianshui city, Gansu province, where a heavy rainfall triggered more than 10,000 landslides in July 2013. All models were performed on a 3-band (near infrared, red, and green) GeoEye-1 image with a spatial resolution of 0.5 m. At such a fine spatial resolution, the study area is spatially heterogeneous. The tested study area is 128 km2, 80% of which was used to train models and the remaining 20% was used to validate accuracy of the models. This proposed ResU-Net achieved higher accuracy than the baseline U-Net model in this mountain region, where F1 improved by 0.09. Compared with the U-Net model, this proposed model (ResU-Net) performs better in discriminating landslides from bare floodplains along river valleys and unplanted terraces. By incorporating environmental information, this ResU-Net may also be applied to other landslide mapping, such as landslide susceptibility and hazard assessment.
Vehicle detection based on very high-resolution (VHR) remote sensing images is beneficial in many fields such as military surveillance, traffic control, and social/economic studies. However, intricate details about the vehicle and the surrounding background provided by VHR images require sophisticated analysis based on massive data samples, though the number of reliable labeled training data is limited. In practice, data augmentation is often leveraged to solve this conflict. The traditional data augmentation strategy uses a combination of rotation, scaling, and flipping transformations, etc., and has limited capabilities in capturing the essence of feature distribution and proving data diversity. In this study, we propose a learning method named Vehicle Synthesis Generative Adversarial Networks (VS-GANs) to generate annotated vehicles from remote sensing images. The proposed framework has one generator and two discriminators, which try to synthesize realistic vehicles and learn the background context simultaneously. The method can quickly generate high-quality annotated vehicle data samples and greatly helps in the training of vehicle detectors. Experimental results show that the proposed framework can synthesize vehicles and their background images with variations and different levels of details. Compared with traditional data augmentation methods, the proposed method significantly improves the generalization capability of vehicle detectors. Finally, the contribution of VS-GANs to vehicle detection in VHR remote sensing images was proved in experiments conducted on UCAS-AOD and NWPU VHR-10 datasets using up-to-date target detection frameworks.
This paper presents a nonlinear iterative magnetic equivalent circuit (MEC) model for a novel counter-rotation dual-rotor axial flux permanent magnet synchronous machine (CRDR-AFPMSM). The proposed machine mainly consists of two rotors with opposite rotation directions, two sets of concentrated windings and a stator core sandwiched in between the two rotors. The model takes into account of the nonlinear characteristics of the saturable permeance in the stator core, and the permeance matrix is updated by an iterative procedure to accurately illustrate its nonlinear feature. The air gap flux density, back electromotive force (EMF) and torque are predicted by the model. Based on the nonlinear model, the thickness of the stator yoke is determined. All of the results obtained by the proposed model match with finite element analysis (FEA) results closely, thus the validity of the proposed MEC model is verified.
Pedestrian detection is a critical challenge in the field of general object detection, the performance of object detection has advanced with the development of deep learning. However, considerable improvement is still required for pedestrian detection, considering the differences in pedestrian wears, action, and posture. In the driver assistance system, it is necessary to further improve the intelligent pedestrian detection ability. We present a method based on the combination of SSD and GAN to improve the performance of pedestrian detection. Firstly, we assess the impact of different kinds of methods which can detect pedestrians based on SSD and optimize the detection for pedestrian characteristics. Secondly, we propose a novel network architecture, namely data synthesis PS-GAN to generate diverse pedestrian data for verifying the effectiveness of massive training data to SSD detector. Experimental results show that the proposed manners can improve the performance of pedestrian detection to some extent. At last, we use the pedestrian detector to simulate a specific application of motor vehicle assisted driving which would make the detector focus on specific pedestrians according to the velocity of the vehicle. The results establish the validity of the approach.Pedestrian detection is a very popular topic in the field of computer vision, which has wide applications such as automatic driving, intelligent surveillance, human behavior analysis, and mobile robotics [1][2][3][4][5] . For now, more and more pedestrian detection systems are applied to automobiles to save people lives. In last few years, the significant role of computer vision systems is emphasized for accident prevention. The visual system must detect the front scene and warn the driver in advance if there is an unexpected situation. Absolutely, high quality speed and accurate systems that can perform detecting on all different scene is highly satisfactory, but until now, unfortunately, no such system exists. A few methods that detect body parts of people [6,7] have been elaborated. HOG ( Histogram of Oriented Gradients) algorithm [8] applies in pedestrian detection. Another approach to save the computational time in pedestrian detection is proposed [9] with the help of CBT ( Cluster Boosted Tree) framework based on edge features. A two-stage classifier was used for fast pedestrian detection [10] , features are prepared by analyzing HOG descriptors and are based on pixel orientation concept as well as multi-scaling levels. The results show a faster computation but still the occlusion detection and missing detection has a margin to improve. Bilgic B. [11] used rejection though Adaboost cascade framework approach for pedestrian identification. Feature selection is based on gradient direction histogram. They have attained an improved frame rate of 8 by using NVDIA
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.