Search

Sentinel-2 satellite remote sensing images have been widely used in various fields, such as change detection and resource monitoring. However, Sentinel-2 provides multispectral bands with inconsistent spatial resolutions (i.e., 60 m for three bands, 20 m for six bands, and 10 m for four bands), which has greatly limited the application values, especially for cooperative analysis or application on different bands at a unified resolution. In this article, we proposed a multistage progressive interactive fusion network to generate all 10-m high-resolution bands. Specifically, a refined multistage spatial resolution enhancement model in a multistage way is developed to progressively improve the low-resolution bands to preserve the spectral information of the enhanced bands. Moreover, an information interaction module is proposed for the three branches of high- (10 m), medium- (20 m), and low-resolution (60 m) bands to achieve effective information interaction. The experimental results show that our method is superior to other existing state-of-the-art methods, and it can be applied to the reconstruction of the high-resolution vegetation index.

Matching feature points from image pairs with significant visual changes and repetitive patterns remains a challenging problem due to the presence of mismatches in the putative corresponding point sets. In this paper, we propose a novel feature matching algorithm called Multi-scale Local Relative Motion Consistency (MLRMC) to accurately find matched feature points. The MLRMC algorithm exploits the multi-scale relative motion information between putative matches and their neighboring points to effectively remove outliers with accuracy and robustness. By clustering the relative motion, coincident relative motion among the matches is obtained. The similarity of each putative match is evaluated by considering both the consensus of relative motion and the consensus of neighborhood elements. To achieve accurate and robust feature point matching, a two-step outlier removal strategy is employed. Experimental evaluations are conducted using three datasets to verify the accuracy and robustness of our proposed method, MLRMC. The results demonstrate that our method outperforms six other classic feature matching methods.

This paper presents a compact wideband dual-polarized magnetoelectric dipole antenna suitable for 5G base stations, which can cover 5G NR n77/78/79 band. The proposed antenna is configured with a cross ME dipole loaded with a rectangular slot, and two L-shaped feed lines are coplanar with the vertical patch, which makes the transverse size of the antenna more compact. The prototype size is $0.57\,\,\lambda _{min} \times 0.57\,\,\lambda _{min}$ (length $\times $ width). According to the measured results, the operating band of the antenna is 3.06-5.28 GHz, and the impedance bandwidth is 53.2% of −10 dB. In the whole frequency range, the isolation between ports is less than −22 dB. Furthermore, metal baffles around the ground are used to stabilize the radiation pattern and improve the front-to-back ratio, achieving gains greater than 8dBi in the frequency band. However, the half power beam width (HPBW) exceeds 65°±5° for base station applications. The antenna has the advantages of broadband, high front-to-back ratio and low cross-polarization, and can be used as a good candidate antenna for 5G commercial small base stations.

The image/video retargeting (resolution adaptation) problem-adaption of source data (image/video) to different screen resolution-has garnered wide attention due to compelling applications in intelligent adaptive display. Most of the current image/video displaying application achieves the adaption through scaling source uniformly. Compared with traditional image/video retargeting, stereoscopic video retargeting poses new challenges that should preserve shape and depth of salient objects with temporal coherence. In this paper, we propose a depth trajectory-aware stereoscopic video retargeting method by imposing constraints to optimize the coordinates and depths simultaneously. To reduce visual discomfort induced by fast depth motion, our method uses a temporal depth distortion energy to optimize the depth trajectory in temporal direction. As a result, our solution can preserve shape, depth and temporal fidelity of salient objects simultaneously with comfortable visual experience.

Synthetic Aperture Radar (SAR) remote sensing images with all-day and all-weather advantages are attracting ever-increasing attention in various areas. However, the interpretation of SAR images is quite challenging and not adapted to the non-expert people. To improve the interpretation of SAR images, a SAR-to-optical image translation method based on a modified conditional generation adversarial network (cGAN) is proposed. In the proposed method, the characteristics of the SAR and optical images are comprehensively considered. To generate better results, a modified strongly constrained cGAN, with the generator and the discriminator networks, based on structure similarity index measure (SSIM) and L1 norm was constructed. The generator network aims to generate an artificial optical image by SAR, and the discriminator network aims to force the generated image to be close to the real optical image. The proposed method was verified and compared with several state-of-the-art methods, and the experimental results show the superiority of the proposed method.

This paper presents a content retargeting and depth adaptation method for the stereoscopic image. Our method allows the user to specify the retargeting scenarios to enhance the user's visual experience. Toward this end, we proposed a warping framework that takes content retargeting, depth adaptation, and interactive editing into account simultaneously for stereoscopic images. From the viewpoint of accuracy, comfortability, and controllability, we exploit the complementary relationship among image quality, depth quality, and important content and propose a grid optimization framework to fuse the three indictors. The experimental results demonstrate that our method achieves a preferable tradeoff among accuracy, comfortability, and controllability of information presentation in retargeting, obtaining satisfactory visual experience for users.

In this paper, we develop a reliability-based transit assignment model with capacity constraints that adopt a new stochastic overload delay formulation. The in-vehicle travel times and waiting times are set to random variables. We adopt the statistical characteristics of the normal distribution to model the stochastic overload delay. The stochastic overload delay gradually varies as the passenger flow changes. A method is developed based on the method of successive averages algorithm to solve the reliability-based transit assignment model. Moreover, numerical experiments are conducted to illustrate the properties of the proposed model and the efficiency of the proposed algorithm.

To enhance the depth perception for stereoscopic image retargeting, this paper presents a depth-adjustable stereoscopic image retargeting method via pixel fusion technique. The traditional methods enforce the depth-preserving constraint on the left and right images with poor depth adaption. To solve this issue, we try to address two critical properties that how to change the widths of unmatched region to control the depth perception, and how to change the alignment of left and right seams to satisfy visual comfort limits, and thus propose a seam manipulator tool to optimize the process for scaling factor determination of stereoscopic seams. The important contribution of our method is to take scaling preservation energy and disparity consistency energy into optimization. The experimental results demonstrate that our method can achieve promising performances on depth perception and visual comfort adjustment for better visual experience.

Depth-image-based rendering (DIBR) techniques can be used to generate virtual views for free-viewpoint video application. However, the DIBR algorithms will introduce geometric distortions that mainly distribute at the disoccluded regions in the synthesized views. It has been demonstrated that conventional 2-D quality metrics are not suitable for the synthesized views. In this paper, we propose a new quality model for 3-D synthesized images by measuring the block-wise texture similarity and color contrast similarity in critical areas, and the global gradient magnitude deviation. A critical area detection module is first employed using a warping method with morphological operation. Then, the critical areas are partitioned into blocks, which are classified as edge blocks, texture blocks, and smooth blocks by computing discrete cosine transform coefficient values. Block-wise texture similarity and color contrast similarity in the corresponding areas are calculated, which are weighted by the size of critical areas. Furthermore, gradient magnitude deviation is measured to quantify global sharpness. Finally, the two scores are pooled to obtain the overall quality. The experimental results on the IRCCyN/IVC, IETR, and MCL-3-D DIBR image databases indicate that our method achieves higher quality prediction accuracy than the state-of-the-art quality metrics.

The high-quality pansharpened image with both high spatial resolution and high spectral fidelity is highly desirable in various applications. However, existing pansharpening methods may lead to spatial distortion and spectral distortion. To measure the degrees of distortion caused by the pansharpening methods, we conduct in-deep studies on the subjective and objective quality assessment of pansharpened images. We built a subjective database consisting of 360 images generated from 20 couples of panchromatic (PAN)/multispectral (MS) images using 18 pansharpening methods. Based on the database, we proposed a no-reference quality assessment method to blindly predict the quality of pansharpened images via opinion-unaware learning. The proposed method first extracted features from the MS images' spectral bands and typical information indexes which comprehensively reflect spatial distortion, spectral distortion, and the effects of pansharpening on applications. Based on the features extracted from the pristine MS image training dataset, a benchmark multivariate Gaussian (MVG) model is learned. The distance between the benchmark MVG and the MVG fitted on the test image is calculated to measure the quality. The experimental results show the superiority of our method on our database.

Different from natural image quality assessment methods, satellite stereo images have different requirements on quality in different application scenarios, which poses a huge challenge to establish a suitable objective evaluation model. In this paper, we focus on the quality evaluation of high resolution panchromatic (satellite stereo) images in specific application scenarios of building detection. First, we build a new satellite stereo image database (SSID), which consists of 400 distorted source satellite stereo images (SSIs) generated from the 20-source SSIs with two distortion types and 10-distortion strengths. We use detection accuracy scores to represent the quality of the SSIs, which is obtained through building detection, not subjective testing. We then propose an objective evaluation model based on joint dictionary learning. In the training phase, we bridge the features of the SSIs and the corresponding detection accuracy scores through joint dictionary learning. In the testing phase, we used sparse coding to get the quality of the testing image. The experimental results demonstrate the effectiveness of the proposed method.

In the open pit mine production systems, a certain number of trucks transport mine and rock between the power shovel and the unloading point. Due to the mining truck has characteristics of high height, long width and big size, it has a large blind zone and a long braking distance. Therefore, the probability of accidents in mining trucks is high, which results in huge loss of manpower, material resources and financial resources. In this paper, tiny-yolov3 is used to detect obstacles in the mine, its real-time performance is high enough, but the detection accuracy is not ideal. Therefore, this paper proposes an improved target detection model based on tiny-yolov3. The residual network structure based on convolutional neural network is added to the tiny-yolov3 structure, and the accuracy of obstacle detection is improved under the condition of real-time detection. The experimental results show that compared with tiny-yolov3, the detect precision of tiny-yolov3 with residual structure is improved, and the detection speed is reduced slightly, there is no particular impact on the real-time nature of the entire algorithm.

Hyperlens consisted by period plasmonic metamaterials is an interesting technology for the optical imaging of arbitrary objects beyond the diffraction limit in the far field, which has great potential applications in nanophotonics. However, the functionality of previously presented hyperlens is limited to image two-dimensional (2-D) objects and work at a certain wavelength due to the dispersion of metal, dramatically impose restrictions on its practical application. In this paper, we proposed and theoretically demonstrated a multiwavelength tunable hyperlens, with the ability of 3-D imaging by focal plane tuning achieved by electro-optical modulation. The simulation results indicate that full-width half-maximum (FWHM) of minimum imaging spot can be reduced down to the level of less than half-wavelength of the incident in a wide wavelength range real-timely, down to the level of ~200 nm using visible light illumination. In addition, the focal position of the hyperlens can also be tuned in large spatial scope, which greatly facilitates the application of the hyperlens in optical signal acquisition and processing.

In remote medical diagnosis, the percentage of poor-quality fundus images is very high, which requires automated quality assessment of fundus images in the acquisition stage to reduce the retransmission cost. In this paper, we propose a fundus image quality classifier via the analysis of illumination, naturalness, and structure, which use three effective secondary indices (or 5-D feature set) and different classification methods to determine the recommendation indexes of fundus images for further diagnosis. We construct a fundus image database including `accept' and `reject' classes based on the definition of illumination, naturalness, and structure. The model can achieve a sensitivity of 94.69%, specificity of 92.29%, and accuracy of 93.60% for the classifying of the fundus images.

Since the goal of image retargeting is to adapt source images on target displays with different sizes and aspect ratios, how to objectively evaluate the quality of retargeted images is particularly important to optimize the retargeting operations. In this paper, we proposed a new image retargeting quality assessment metric, which constructs the metric using both hand-crafted features and deep-learned features. To enhance the reliability and accuracy of the proposed method: 1) we use similarity transformation as local descriptor to extract hand-craft features, and measure structure distortion and content loss from the hand-craft features and 2) we use deep learning architecture to construct encoders and extract deep-learned features, and measure texture similarity and semantic similarity from the deep-learned features. We conduct experiments on two databases: RetargetMe and CUHK. Experimental results show that our method can achieve superior performance to the state-of-the-art metrics.

Quality assessment of 3-D images encounters more challenges in better understanding of human visual system. In this paper, we propose a perceptual quality assessment approach for stereoscopic images by modeling visual properties of the primary visual cortex. For this purpose, we obtain a new feature encoding approach for the visual information, and define a new similarity measure approach to match the feature encoding to give more reliable and accurate quality assessment. Experimental results on three symmetrically, asymmetrically, and multiple distorted stereoscopic image quality assessment databases demonstrate that our method has high consistency with subjective assessment.