Search

Digital elevation models (DEMs), which can provide an accurate description of planetary surface elevation changes, play an important role in scientific tasks such as long-distance path planning, terrain analysis, and planetary surface reconstruction. However, generating high-precision planetary DEMs currently relies on expensive equipment together with complex remote sensing technology, thus increasing the cost and cycle of the task. Therefore, it is crucial to develop a cost-effective technology that can produce high-quality DEMs on the surfaces of planets. In this work, we propose a global attention-based DEM generation network (GADEM) to convert satellite imagery into DEMs. The network uses the global attention mechanism (GAM) together with a multi-order gradient loss function during training to recover precise terrain. The experimental analysis on lunar and Martian datasets not only demonstrated the effectiveness and accuracy of GADEM in bright regions, but also showed its promising reconstruction ability in shadowed regions. [ABSTRACT FROM AUTHOR]

Target detection technology of synthetic aperture radar (SAR) imageis widely used in the field of military reconnaissance and surveillance. The traditional SAR image target detection methods need to be provided a lot of empirical knowledge because the characteristics of SAR images in different configurations (attitude, pitch angle, imaging parameters, etc.) will change greatly,resulting in high generalization error. Currently, deep learning method has achieved great success in the field of image processing. Research shows that deep learning can achieve a more intrinsic description of the data, while the model has a stronger ability of modeling and generalization. In order to solve the problem of insufficient data in SAR data sets, an experimental system for acquiring SAR image data in real scenes was built. Then the transfer learning method and the improved convolution neural network algorithm (PCA+Faster R-CNN) are applied to improve the target detection precision. Finally, experimental results demonstrate the significant effectiveness of the proposed method.

Due to the complex imaging mechanism of SAR images and the lack of multi-angle and multi-parameter real scene SAR target data, the generalization performance of existing deep-learning-based synthetic aperture radar (SAR) image target detection methods are extremely limited. In this paper, we propose an unsupervised domain-adaptive SAR ship detection method based on cross-domain feature interaction and data contribution balance. First, we designed a new cross-domain image generation module called CycleGAN-SCA to narrow the gap between the source domain and the target domain. Second, to alleviate the influence of complex backgrounds on ship detection, a new backbone using a self-attention mechanism to tap the potential of feature representation was designed. Furthermore, aiming at the problems of low resolution, few features and easy information loss of small ships, a new lightweight feature fusion and feature enhancement neck was designed. Finally, to balance the influence of different quality samples on the model, a simple and efficient E 1 2 I o U Loss was constructed. Experimental results based on a self-built large-scale optical-SAR cross-domain target detection dataset show that compared with existing cross-domain methods, our method achieved optimal performance, with the mAP reaching 68.54%. Furthermore, our method achieved a 6.27% improvement compared to the baseline, even with only 5% of the target domain labeled data. [ABSTRACT FROM AUTHOR]

As an important representation of scenes in virtual reality and augmented reality, image stitching aims to generate a panoramic image with a natural field-of-view by stitching multiple images together, which are captured by different visual sensors. Existing deep-learning-based methods for image stitching only conduct a single deep homography to perform image alignment, which may produce inevitable alignment distortions. To address this issue, we propose a content-seam-preserving multi-alignment network (CSPM-Net) for visual-sensor-based image stitching, which could preserve the image content consistency and avoid seam distortions simultaneously. Firstly, a content-preserving deep homography estimation was designed to pre-align the input image pairs and reduce the content inconsistency. Secondly, an edge-assisted mesh warping was conducted to further align the image pairs, where the edge information is introduced to eliminate seam artifacts. Finally, in order to predict the final stitched image accurately, a content consistency loss was designed to preserve the geometric structure of overlapping regions between image pairs, and a seam smoothness loss is proposed to eliminate the edge distortions of image boundaries. Experimental results demonstrated that the proposed image-stitching method can provide favorable stitching results for visual-sensor-based images and outperform other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Weathered petroleum-contaminated soil was treated with Vetiveria zizanioides (Vetiver) and Cymbopogon nardus (Lemongrasss) to investigate the efficiency of phytoremediation. The initial total petroleum hydrocarbon (TPH) concentration of soil was 3000−8000 mg/kg, and after 6 months, the TPH concentrations were degraded by 50−75% under the action of soil native microbial. Planting vetiver and lemongrass stabilized soil pH and electrical conductivity, and it accelerated the decomposition of TPH in soil. Planting vetiver showed a better effect. After 6 months of planting, the TPH decomposition efficiency reached about 90%, and most of the easily decomposed TPH has been decomposed. The results of rhizosphere soil microbiota analysis also showed that planting vetiver increased the abundance of soil microbiota. [ABSTRACT FROM AUTHOR]

As an active microwave device, synthetic aperture radar (SAR) uses the backscatter of objects for imaging. SAR image ship targets are characterized by unclear contour information, a complex background and strong scattering. Existing deep learning detection algorithms derived from anchor-based methods mostly rely on expert experience to set a series of hyperparameters, and it is difficult to characterize the unique characteristics of SAR image ship targets, which greatly limits detection accuracy and speed. Therefore, this paper proposes a new lightweight position-enhanced anchor-free SAR ship detection algorithm called LPEDet. First, to resolve unclear SAR target contours and multiscale performance problems, we used YOLOX as the benchmark framework and redesigned the lightweight multiscale backbone, called NLCNet, which balances detection speed and accuracy. Second, for the strong scattering characteristics of the SAR target, we designed a new position-enhanced attention strategy, which suppresses background clutter by adding position information to the channel attention that highlights the target information to more accurately identify and locate the target. The experimental results for two large-scale SAR target detection datasets, SSDD and HRSID, show that our method achieves a higher detection accuracy and a faster detection speed than state-of-the-art SAR target detection methods. [ABSTRACT FROM AUTHOR]

Unlike optical images, synthetic aperture radar (SAR) images have unique characteristics, such as few samples, strong scattering, sparseness, multiple scales, complex interference and background, and inconspicuous target edge contour information. Current SAR target detection algorithms have difficulty in balancing accuracy and speed, and the performance of these algorithms is relatively limited, thus making it difficult to deploy practical applications. To this end, this article proposes AFSar, an innovative anchor-free SAR target detection algorithm based on multiscale enhancement representation learning. First, we introduce the latest anchor-free architecture YOLOX as the basic framework. Second, to reduce the computational complexity of the model and to improve the ability of multiscale feature extraction, we redesigned the lightweight backbone, namely, MobileNetV2S. Furthermore, we propose an attention enhancement PAN module, called CSEMPAN, which highlights the unique strong scattering characteristics of SAR targets by integrating channel and spatial attention mechanisms. Finally, in view of the multiscale and strong sparse characteristics of SAR targets, we propose a new target detection head, namely, ESPHead. ESPHead extracts the features of targets with different scales by using dilated convolution with different dilated rates, so as to enhance the detection ability of the model for targets with different scales. The results of ablation experiments on the SSDD dataset show that the mAP of our algorithm reaches 0.977, while the Flops is only 9.86 G, achieving state of the art. [ABSTRACT FROM AUTHOR]

This study aimed to analyze the application of the diagnostic model based on deep learning technology in the evaluation of thyroid contrast-enhanced ultrasound images and to provide a reference for the evaluation of benign and malignant thyroid. A diagnosis model of ultrasound images based on long- and short-term memory neural network (LSTM), C-LSTM, was proposed. The diagnostic method was compared with that based on support vector machine (SVM) and manual feature (MF), and it was applied to the diagnosis of thyroid contrast-enhanced ultrasound images. The results showed that the sensitivity, specificity, and accuracy of the C-LSTM model were greatly higher than those of SVM and MF, and the differences were considerable (P < 0.05). The number of parameters and the calculation amount of the C-LSTM model were greatly lower than those of SVM- and MF-based diagnosis methods (P < 0.05). The sensitivity, specificity, and accuracy of the C-LSTM model were greatly greater than those of the C-LSTM-0 model, while the amounts of parameters and calculations were greatly lesser than those of the C-LSTM-0 model (P < 0.05). The numbers of benign tumors with contrast-enhanced ultrasound modes of no enhancement, no enhancement at early stage, and low enhancement were more than those of malignant tumors, while the numbers of high-enhancement tumors were greatly less than those of malignant tumors (P < 0.05). The diagnostic area under the curve (AUC) of rise time (RT) ratio, time to peak (TTP) ratio, and mean transit time (mTT) ratio for malignant masses were large, which were 0.856, 0.794, and 0.761, respectively. RT ratio, TTP ratio, and mTT ratio were of high diagnostic sensitivity and specificity for malignant masses, while RT, TTP, and mTT were of low diagnostic sensitivity and specificity. In summary, the contrast-enhanced ultrasound images based on the deep learning C-LSTM model can effectively improve the diagnostic effect of benign and malignant thyroid masses. The image feature parameters RT ratio, TTP ratio, and mTT ratio were of good efficiency in diagnosing benign and malignant thyroid masses. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicle (UAV) is one of the main means of information warfare, such as in battlefield cruises, reconnaissance, and military strikes. Rapid detection and accurate recognition of key targets in UAV images are the basis of subsequent military tasks. The UAV image has characteristics of high resolution and small target size, and in practical application, the detection speed is often required to be fast. Existing algorithms are not able to achieve an effective trade-off between detection accuracy and speed. Therefore, this paper proposes a parallel ensemble deep learning framework for unmanned aerial vehicle video multi-target detection, which is a global and local joint detection strategy. It combines a deep learning target detection algorithm with template matching to make full use of image information. It also integrates multi-process and multi-threading mechanisms to speed up processing. Experiments show that the system has high detection accuracy for targets with focal lengths varying from one to ten times. At the same time, the real-time and stable display of detection results is realized by aiming at the moving UAV video image. [ABSTRACT FROM AUTHOR]

With the development of display technologies, image retargeting plays a significant role in computer vision and pattern recognition communities currently. Image retargeting aims to display an image on a series of appliances with different resolutions and target aspect ratios. During the last decade, representative algorithms for image retargeting have been presented in the literature and achieved state-of-the-art performance. In this survey, we provide a comprehensive review of image retargeting, covering a wide variety of pioneering works for 2D image retargeting and stereoscopic image retargeting. 2D image retargeting focuses on preserving interesting regions when modifying the original image with arbitrary resolutions appropriately. Different from 2D image retargeting, stereoscopic image retargeting needs to preserve both the shape structure of salient objects and the depth consistency of 3D scenes simultaneously. In this survey, we start the first attempt to analyze the trends of 2D image retargeting and then summarize different types of stereoscopic image retargeting. Secondly, image retargeting quality assessment metrics for 2D images and stereoscopic images are introduced to evaluate retargeted images. Thirdly, we also investigate the evaluation datasets, and give the comparison results and analysis between different representative methods. Finally, the promising future research is thoroughly discussed to further improve the performance of 2D and stereoscopic image retargeting. [ABSTRACT FROM AUTHOR]

• Experimental method of pedestrian wayfinding based on virtual reality. • Wayfinding model considering both spatial knowledge and visual information. • Analysis based on mathematical statistics analysis and deep learning framework. • Comparing the impact of dimensional differences on pedestrian wayfinding behavior. • Incorporating spatial knowledge at multiple scales. Understanding pedestrian wayfinding behavior is crucial for traffic management and building design. The use of virtual reality technology presents an efficient approach for investigating pedestrian wayfinding behavior in large public spaces, offering numerous advantages for data collection. However, the impact of different scenario dimensions on pedestrian wayfinding behavior in large public spaces remains unclear. Additionally, the selection of virtual experiment scenario dimensions currently relies primarily on researchers' experience and practical conditions, lacking sufficient evidence to support their rational. Another challenge is the limited focus on spatial knowledge's effect on wayfinding behavior, with insufficient analysis of the utility of pedestrian visual information and a lack of precise methods to quantify visual field information accurately. This study addresses these gaps by incorporating spatial knowledge at multiple scales and pedestrian visual field information as influencing factors in the analysis of wayfinding behavior. Furthermore, it distinguishes between three-dimensional and two-dimensional scenarios to compare the impact of dimensional differences on pedestrian wayfinding behavior. By analyzing behavior data from non-immersive wayfinding experiments, this research employs statistical analysis methods and a deep learning framework to derive results regarding the factors influencing wayfinding behavior. The findings demonstrate that considering both spatial and visual field information effectively enhances the predictive ability of the wayfinding model. Additionally, dimensional differences significantly influence the pedestrian wayfinding process. These results offer empirical evidence to guide researchers in selecting experimental scenarios of pedestrian behavior and provide insights for public space layout, signage design, and improving pedestrian efficiency. [ABSTRACT FROM AUTHOR]