While the capabilities of autonomous driving have advanced rapidly, merging into dense traffic remains a significant challenge, many motion planning methods for this scenario have been proposed but it is hard to evaluate them. Most existing closed-loop simulators rely on rule-based controls for other vehicles, which results in a lack of diversity and randomness, thus failing to accurately assess the motion planning capabilities in highly interactive scenarios. Moreover, traditional evaluation metrics are insufficient for comprehensively evaluating the performance of merging in dense traffic. In response, we proposed a closed-loop evaluation benchmark for assessing motion planning capabilities in merging scenarios. Our approach involves other vehicles trained in large scale datasets with micro-behavioral characteristics that significantly enhance the complexity and diversity. Additionally, we have restructured the evaluation mechanism by leveraging Large Language Models (LLMs) to assess each autonomous vehicle merging onto the main lane. Extensive experiments and test-vehicle deployment have demonstrated the progressiveness of this benchmark. Through this benchmark, we have obtained an evaluation of existing methods and identified common issues. The simulation environment and evaluation process can be accessed at https://github.com/WZM5853/Bench4Merge.

Optical remote sensing techniques, particularly those relying on visible optical sensors, are critical for Earth observations. However, atmospheric haze severely degrades image quality owing to light scattering effects. These effects often lead to clouds in the observed images, resulting in partial image degradation. To address the challenge of non-homogeneous image dehazing, we propose a model-inspired deep unfolding network (MiDUNet). The architecture of MiDUNet is strictly derived and inspired by a novel alternating minimization algorithm. During algorithm design, traditional image regularization terms are replaced by several nonlinear functions to align with neural network processing, increasing flexibility and applicability. Afterward, a Multi-scale Spatial frequency Residual Block (MSRB) is used to capture complex scene structures by fusing multi-scale spatial information and frequency information. With these spatial-frequency features and a Multi-path Feature Enhancement Block (MFEB), local and global information are obtained, which increases the robustness to non-homogeneous dehazing and image texture preservation. Numerical experiments with remote sensing images and natural images demonstrate the competitive advantages of MiDUNet in non-homogeneous dense haze scenarios.

Cloud and edge computing paradigms are increasingly being applied to data processing for Internet of Things (IoT) sensors. Meanwhile, unmanned aerial vehicles (UAVs) can assist these sensor systems in data acquisition, especially in smart applications such as environmental monitoring and smart agriculture, where direct network connectivity for sensors is limited due to remote deployment. In this work, the Age of Information (AoI) is introduced for wireless sensor networks to measure the freshness of data information. We also develop a hierarchical UAV-assisted data processing framework to minimize AoI, where the multi-UAVs hover over the sensor clusters to collect data and conduct computing offloading by flexibly using the computation resources of edge server or cloud. Then, we innovatively propose a joint service association, trajectory scheduling and computing offloading mechanism for UAVs oriented by AoI. Specifically, we design a sensor clustering and sensor-hovering point (HP) association management scheme to improve the efficiency of data collection, and then propose an HP clustering model to establish the HP-UAV association. Further, a multi-objective optimization model is solved by devising a learning-based trajectory scheduling scheme. Simulation results show that the proposed ATSCO can not only converge well and improve the freshness of data information, but also realize superior performance than the mainstream schemes in various situations.

Abstract This article presents a new DIRECT-type SCABALL (scattering balls) algorithm with a new partition method for derivation-free optimization problems. It does not focus on dividing the region of interest into specific geometric shapes, but rather scatters several balls to cover it. In SCABALL, several potential optimal regions are selected at each iteration, and they are covered by smaller balls sequentially. In this way, the SCABALL ensures the everywhere dense convergence. The center points and radii of the scattered balls significantly influence the efficiency of SCABALL; therefore, the minimax designs are used in the initial and sequential stages to obtain better coverage. The SCABALL parameters, including the number of balls and their radii, were analyzed by numerical investigation. We provided the empirical choices for those parameters and found that the balls’ radii can be contracted to balance efficiency and global convergence. Numerical experiments show that the SCABALL algorithm is locally biased and robust.

Accurate and efficient image alignment is the core problem in the research of panoramic stitching nowadays. This paper proposes a local-adaptive image alignment method based on triangular facet approximation, which directly manipulates the matching data in the camera coordinates, and therefore rises superior to the imaging model of cameras. A more robust planar transformation model is proposed and extended to be local-adaptive via combining it with two weighting strategies. By approximating the scene as a combination of adjacent triangular facets, the planar and spherical triangulation strategies are introduced to more efficiently align normal and fisheye images respectively. The efficiency of the proposed method are verified through the comparative experiments on several challenging cases both qualitatively and quantitatively.

Image stitching aims at generating high-quality panoramas with the lowest computational cost. In this paper, we propose a parallax-tolerant image stitching method based on robust elastic warping, which could achieve accurate alignment and efficient processing simultaneously. Given a group of point matches between images, an analytical warping function is constructed to eliminate the parallax errors. Then, the input images are warped according to the computed deformations over the meshed image plane. The seamless panorama is composed by directly reprojecting the warped images. As an important complement to the proposed method, a Bayesian model of feature refinement is proposed to adaptively remove the incorrect local matches. This ensures a more robust alignment than existing approaches. Moreover, our warp is highly compatible with different transformation types. A flexible strategy of combining it with the global similarity transformation is provided as an example. The performance of the proposed approach is demonstrated using several challenging cases.

In computer vision, video stitching is a very challenging problem. In this paper, we proposed an efficient and effective wide-view video stitching method based on fast structure deformation that is capable of simultaneously achieving quality stitching and computational efficiency. For a group of synchronized frames, firstly, an effective double-seam selection scheme is designed to search two distinct but structurally corresponding seams in the two original images. The seam location of the previous frame is further considered to preserve the interframe consistency. Secondly, along the double seams, 1-D feature detection and matching is performed to capture the structural relationship between the two adjacent views. Thirdly, after feature matching, we propose an efficient algorithm to linearly propagate the deformation vectors to eliminate structure misalignment. At last, image intensity misalignment is corrected by rapid gradient fusion based on the successive over relaxation iteration (SORI) solver. A principled solution to the initialization of the SORI significantly reduced the number of iterations required. We have compared favorably our method with seven state-of-the-art image and video stitching algorithms as well as traditional ones. Experimental results show that our method outperforms the existing ones compared in terms of overall stitching quality and computational efficiency.

In image processing, nonconvex regularization has the ability to smooth homogeneous regions and sharpen edges but leads to challenging computation. We propose some iterative schemes to minimize the energy function with nonconvex edge-preserving potential. The schemes are derived from the duality-based algorithm proposed by Bermúdez and Moreno and the fixed point iteration. The convergence is proved for the convex energy function with nonconvex potential and the linear convergence rate is given. Applying the proposed schemes to Perona and Malik's nonconvex regularization, we present some efficient algorithms based on our schemes, and show the approximate convergence behavior for nonconvex energy function. Experimental results are presented, which show the efficiency of our algorithms, including better denoised performance of nonconvex regularization, faster convergence speed, higher calculation precision, lower calculation cost under the same number of iterations, and less implementation time under the same peak signal noise ratio level.

A novel fast and adaptive method for synthetic aperture radar (SAR) superresolution imaging is developed. Based on the point scattering model in the phase history domain, a dictionary is constructed so that the superresolution imaging process can be converted to a problem of sparse parameter estimation. The approximate orthogonality of this dictionary is exploited by theoretical derivation and experimental verification. Based on the orthogonality of the dictionary, we propose a fast algorithm for basis selection. Meanwhile, a threshold for obtaining the number and positions of the scattering centers is determined automatically from the inner product curves of the bases and observed data. Furthermore, the sensitivity of the threshold on estimation performance is analyzed. To reduce the burden of mass calculation and memory, a simplified superresolution imaging process is designed according to the characteristics of the imaging parameters. The experimental results of the simulated images and an MSTAR image illustrate the validity of this method and its robustness in the case of the high noise level. Compared with the traditional regularization method with the sparsity constraint, our proposed method suffers less computation complexity and has better adaptability.

In many image resolution enhancement applications, the blurring process of the imaging system is unknown. In this paper we discuss the problem of image sharpening without estimating the point spread function (PSF). A regularization model is constructed based on the prior information of image error and image gray value, which does not need any prior of PSF. The regularization model is translated to an anisotropic nonlinear diffusion equation by solving its Euler equation, which can avoid the high computational cost of regularization model. Furthermore, to get clearer edges, a high frequency enhancement filter is used on the image superresolved by nonlinear diffusion equation. Numerical results show that the new model can get much clearer superresolution images than traditional methods, and the peak signal to noise ratios (PSNRs) are also higher than traditional methods.

This paper puts forward a new edge-directed forward and backward anisotropic diffusion model for texture enhancement and edge sharpening of image. Since forward and backward diffusion cannot hold flow-like edges, and coherence enhancement diffusion cannot sharpen edges, our new model gets a new diffusivity, which combine the merits of these two methods. The new model adapts an image structure tensor like coherence enhancement diffusion for edge orientation, and defines eigenvalues based on edge orientation such that the new tensor has positive diffusivity in edge direction and negative diffusivity in gradient direction. Both theory analysis and numerical results show that the new model has better enhancement result than coherence enhancing diffusion and forward and backward diffusion.