Object detection is a challenging task in computer vision. Now many detection networks can get a good detection result when applying large training dataset. However, annotating sufficient amount of data for training i...
详细信息
Object detection is a challenging task in computer vision. Now many detection networks can get a good detection result when applying large training dataset. However, annotating sufficient amount of data for training is often time-consuming. To address this problem, a semi-supervised learning based method is proposed in this paper. Semi-supervised learning trains detection networks with few annotated data and massive amount of unannotated data. In the proposed method, Generative Adversarial Network is applied to extract data distribution from unannotated data. The extracted information is then applied to improve the performance of detection network. Experiment shows that the method in this paper greatly improves the detection performance compared with supervised learning using only few annotated data. The results prove that it is possible to achieve acceptable detection result when only few target object is annotated in the training dataset.
Single image super-resolution (SR) has been widely studied in recent years as a crucial technique for remote sensing applications. This paper proposes a SR method for remote sensing images based on a transferred gener...
详细信息
Single image super-resolution (SR) has been widely studied in recent years as a crucial technique for remote sensing applications. This paper proposes a SR method for remote sensing images based on a transferred generative adversarial network (TGAN). Different from the previous GAN-based SR approaches, the novelty of our method mainly reflects from two aspects. First, the batch normalization layers are removed to reduce the memory consumption and the computational burden, as well as raising the accuracy. Second, our model is trained in a transfer-learning fashion to cope with the insufficiency of training data, which is the crux of applying deep learning methods to remote sensing applications. The model is firstly trained on an external dataset DIV2K and further fine-tuned with the remote sensing dataset. Our experimental results demonstrate that the proposed method is superior to SRCNN and SRGAN in terms of both the objective evaluation and the subjective perspective.
Synthetic aperture radar (SAR) and optical imaging are different remote sensing methods. Given a SAR image, is it possible to predict what the observed scene looks like in an optical image? Transfer between SAR data a...
详细信息
Synthetic aperture radar (SAR) and optical imaging are different remote sensing methods. Given a SAR image, is it possible to predict what the observed scene looks like in an optical image? Transfer between SAR data and optical data seems to be impossible. However, this article shows examples that by applying deep learning techniques on high resolution airborne SAR images and GoogleEarth optical images, the SAR images and optical images can transfer with each other. The transferring help us to better understand the relationship between SAR and optical image, and can be potentially used to transfer detection or classification algorithms for optical image straightforwardly to be applied on SAR image.
China is a flood disaster-prone country, floods occur almost every year, especially in July and August. Rapid detection and assessment for floods affected areas are of great significance. The Chinese GF-3 SAR satellit...
详细信息
China is a flood disaster-prone country, floods occur almost every year, especially in July and August. Rapid detection and assessment for floods affected areas are of great significance. The Chinese GF-3 SAR satellite, which uses active ground observation technology, has obvious advantages in flood disaster monitoring owing to its all-day, all-weather imaging characteristics. For the purpose of rapid water detection in flooding area, an automatic detection method of flood area based on GF-3 single-polarization SAR data is proposed. The proposed method consists of image preprocessing and water extraction. The experimental results show that the proposed method can realize rapid and accurate extraction of waters in flood disaster area.
Inshore ship detection in SAR image faces difficulties on correctly identifying near-shore ships and onshore objects. This article proposes a multi-scale full convolutional network (MS-FCN) based sea-land segmentation...
详细信息
Inshore ship detection in SAR image faces difficulties on correctly identifying near-shore ships and onshore objects. This article proposes a multi-scale full convolutional network (MS-FCN) based sea-land segmentation method and applies a rotatable bounding box based object detection method (DR-Box) to solve the inshore ship detection problem. The sea region and land region are separated by MS-FCN then DR-Box is applied on sea region. The proposed method combines global information and local information of SAR image to achieve high accuracy. The networks are trained with Chinese Gaofen-3 satellite images. Experiments on the testing image show most inshore ships are successfully located by the proposed method.
Target classification is an important part in automatic target recognition (ATR) systems. Deep learning methods get state of the art performance in SAR target classification. Simulation is a useful data augmentation m...
详细信息
Target classification is an important part in automatic target recognition (ATR) systems. Deep learning methods get state of the art performance in SAR target classification. Simulation is a useful data augmentation method when the numbers of real samples for training is not sufficient. This article discusses how to release the full potential of simulated samples which is used to improve performance of SAR target classifier. The proposed method is based on cycle adversarial network (CycleGAN), which can transfer simulated samples to be more similar with real samples in image domain. Experiments show that adding simulated samples straightforward into training dataset is not helpful to improve the performance. However, adding the transferred simulated samples for training results in about 10% increase in accuracy in the designed SAR airplane classification experiment, compared with training without data augmentation.
The conventional shape similarity measurements of remote sensing data face problems in the situation of noise interference, partial information occlusion and missing. A method of shape similarity measurement based on ...
The conventional shape similarity measurements of remote sensing data face problems in the situation of noise interference, partial information occlusion and missing. A method of shape similarity measurement based on principal curvature enhancement distance transformation is proposed. The distance transformation is carried out to extend the range of the shape contour, improving the robustness of the similarity measure. Besides, to ensure the accuracy of measurement results, the distance map is enhanced by the principal curvature of the shape contour, improving the response of contours with rich information. application experiments of road vectors with GPS data and optical remote sensing images show that the method is effective in practical application.
Person re-identification is a crucial task of identifying pedestrians of interest across multiple surveillance camera views. For person re-identification, a pedestrian is usually represented with features extracted fr...
详细信息
The theoretical modeling and analysis of SAR location error play an important role in SAR system design and error source budget. Existing SAR geolocation error models are mainly implicit, which are not easy to do anal...
详细信息
The Super-Low Frequency (SLF) electromag- netic prospecting technique, adopted as a non-imaging remote sensing tool for depth sounding, is systematically proposed for subsurface geological survey. In this paper, we ...
详细信息
The Super-Low Frequency (SLF) electromag- netic prospecting technique, adopted as a non-imaging remote sensing tool for depth sounding, is systematically proposed for subsurface geological survey. In this paper, we propose and theoretically illustrate natural source magnetic amplitudes as SLF responses for the first step. In order to directly calculate multi-dimensional theoretical SLF responses, modeling algorithms were developed and evaluated using the finite difference method. The theore- tical results of three-dimensional (3-D) models show that the average normalized SLF magnetic amplitude responses were numerically stable and appropriate for practical interpretation. To explore the depth resolution, three-layer models were configured. The modeling results prove that the SLF technique is more sensitive to conductive objective layers than high resistive ones, with the SLF responses of conductive objective layers obviously show- ing uprising amplitudes in the low frequency range. Afterwards, we proposed an improved Frequency-Depth transformation based on Bostick inversion to realize the depth sounding by empirically adjusting two parameters. The SLF technique has already been successfully applied in geothermal exploration and coalbed methane (CBM) reservoir interpretation, which demonstrates that the proposed methodology is effective in revealing low resistive distributions. Furthermore, it siginificantly contributes to reservoir identification with electromagnetic radiation anomaly extraction. Meanwhile, the SLF inter- pretation results are in accordance with dynamic production status of CBM reservoirs, which means it could provide an economical, convenient and promising method for exploring and monitoring subsurface geo-objects.
暂无评论