With the development of 3D data acquisition technology, point cloud wins the favor of researchers for it's simple but effective representation and it is widely used in the fields of remote sensing, scene reconstruction, 3D modeling, etc. Considering that the data acquisition process is easily disturbed by many factors such as equipment, environment and material, raw point cloud is often corrupted with noise and so it is of great significance to explore robust and efficient denoising algorithms. This paper firstly investigates the relevant research works of point cloud denoising and divides them into traditional algorithms based on the optimization idea and denoising algorithms based on the deep learning idea according to the implementation principles. Secondly, the research progress of each kind of algorithm is discussed and a detailed analysis of representative algorithms is presented. Thirdly, the data sets, the evaluation metrics and experimental results are summarized with an in-depth comparison. Finally, the problems and possible development directions and trends of point cloud denoising are prospected.

This review aims to guide the future development of related research in the field of learning path planning through the analysis of the current research status of learning path planning. Specifically, this review first introduces the definition of learning path planning and the commonly used parameters in learning path planning methods; then, it classifies in detail according to the algorithms used to generate learning path planning and summarizes the advantages and disadvantages of various learning path planning methods. In addition, the data set and evaluation method used by the learning path planning method is introduced. Finally, the challenges faced by the learning path planning method are summarized and the future development trend is predicted.

In recent years, semi-supervised learning (SSL) methods based on image augmentation and consistency regularization have been widely used and have achieved great success. However, little attention has been paid to pseudo-labeling (PL)-based semi-supervised learning methods because of the "confirmation bias" problem, i.e., errors in the model are accumulated by wrong pseudo-labels and thus difficult to be corrected. In this paper, we propose a feature refinement model based on the feature space graph. The model learns a graph attention model on the feature space mapped by the neural network. We apply this model to the feature space to make use of the information of the prototypes to refine the features. The pseudo-labels generated by the refined features are randomly and linearly combined with the pseudo-labels generated by the prototypes assignment to obtain new pseudo-labels. In this paper, we apply this module to two pseudo-labeling semi-supervised learning frameworks and achieve significant accuracy improvements in several CIFAR-10 and CIFAR-100 semi-supervised classification problems. In particular, we apply our feature refinement model to the pseudo-labeling semi-supervised learning framework PLCB and add the proposed mutual mix supervision techniques to achieve good results on this framework. By applying the proposed feature refinement module to several pseudo-labeling semi-supervised learning frameworks and conducting experiments on several datasets, the proposed algorithm is demonstrated to be universal and effective as an add-on module.

Automatic building extraction methods based on deep learning theory have the technical characteristics of high accuracy and speed,and are of great significance in industrial applications, such as urban planning,disaster prevention and mitigation. This paper introduces the deep learning modules and the traditional remote sensing validation section in the proposed building extraction method in high-resolution remote sensing imageries, forming an operational deep-learning-theory based building extraction technical system that integrates different backbone modules, UNet + + networks,and remote sensing authenticity verification modules. The basic network is transformed through the traditional convolutional network model backbones,such as VGG,ResNet, and Inception to improve the model operational efficiency,strengthen the model feature learning capabilities,verify the effectiveness and applicability of the algorithm through authenticity validation. Taking the Massachusetts building dataset published by Mnih as the data source,a comparative analysis was carried out with the traditional non-full convolutional network model and full convolutional network model. The results show that an increasing in the depth and width of the model can substantially improve the building extraction results. The InceptionV3-UNet + + backbone model has the best performance in recall rate,accuracy,CSI,F1 score,Kappa coefficients, and total accuracy,reaching 85. 14%,90. 50%,0. 781 6,0. 877 4,0. 850 4, and 95. 57%,respectively,and its robustness is also verified on the WHU datasets. This method has significantly improved the extraction accuracy and the details of the buildings extracted, especially on complex and irregular buildings, which will facilitate the building extraction applications in real, complex, and large scene of high-resolution remote sensing imageries.

Sunspot variations are the sun's symptoms of strong magnetic perturbations. In this paper, we use long short-term memory neural network and one-dimensional convolution neural network to detect sunspot variations. Here we use three different datasets, including the yearly mean sunspot number (YSSN) from 1700 to 2020, the monthly mean sunspot number (MSSN) from 1749 to 2021 and the monthly mean sunspot areas (MSSA) from 1874 to 2021. First, based on the YSSN dataset, we obtain YSSN for 2021 and the predicted YSSN in the 25th solar cycle appears at 2025 which equals 163.4; Then, based on the MSSN dataset, we obtain MSSN for June 2021 and the predicted YSSN in the 25th solar cycle appears in October 2024 which equals 245.9; Next, based on the MSSA dataset, the predicted MSSA for June 2021 is 73.1; Finally, the latitude is divided into 13 partitions to predict the butterfly diagram, and still, neural network can reconstruct the butterfly diagram. Therefore, neural network can provide a physical perspective for sunspot investigation.

Cyclopentenones are important synthetic building blocks and as motifs appear in bioactive molecules and natural products. We applied density functional theory (DFT) calculations to gain insight into the modular strategy involved in the palladium-catalyzed synthesis of cyclopentenone from α,β-unsaturated acid chlorides and alkynes in the presence of hydrosilane. The study unveils that the transformation proceeds via the sequence:the disassembly of α,β-unsaturated acid chloride into vinyl, carbonyl, and Cl fragments with the palladium catalyst; carbon monoxide release; coupling of alkyne with vinyl group; carbon monoxide re-coordination and migratory insertion to form another C-C bond with alkyne, ring-closure via C=C bond insertion, transmetalation with hydrosilane, C, H-reductive elimination to release the product. Different from the mechanism proposed by the experimentalists, the CO group is involved in the reaction via separate liberation and re-coordination in the solvent cage, rather than persistent coordination with palladium. The transmetalation for H/Cl exchange takes place at the late stage and is a bottleneck of the transformation, instead of at early disassembly stage.

Multiple high-order APSK modulation schemes are added in the DVB-S2X standard on the basis of the DVB-S2 standard, and an 8+8APSK constellation is added based on the original 16APSK. In this paper, characteristics of a 16-APSK constellation are mainly analyzed, and a new soft demodulation optimization algorithm is proposed. Compared with the LLR algorithm and the Max-log-MAP algorithm, the proposed algorithm has the significantly reduced computational complexity. Simulation results of the three algorithms show that the optimization algorithm is small in performance loss and has good practicability in engineering application.

Urban eco-efficiency is an effective indicator to show the relationship between regional economic development and its impact on resources and environment. The spatiotemporal evolution of the eco-efficiency of 285 cities above the prefecture level in China from 2005 to 2017 was calculated by the super-efficiency SBM model containing the undesirable outputs. The panel regression models were further constructed to analyze the factors that significantly influenced the urban eco-efficiency. The conclusions are as follows. 1) The average urban eco-efficiency across China fluctuated slightly and showed a rising trend. 2) Even though the gap of eco-efficiency between cities decreased gradually from 2005 to 2017, the regional disparity in the whole country was still obvious. 3) The spatiotemporal pattern of urban eco-efficiency among different regions and cities has evolved significantly. The cities with higher ecological efficiency in 2005 were mainly distributed in the Pearl River Delta, the Yangtze River Delta, and Shandong Province. By contrast, the territorial ranges of cities with higher urban eco-efficiency in 2017 were greatly expanded. 4) Results from the regression models showed that factors as urbanization rate, economic development level, openness to foreign economies, science and technology investments, and industrial transition were important determinants promoting the urban eco-efficiency in China. Therefore, high priorities should be given to upgrade industrial structure, increase investments in science and technology, and reduce resources consumption and pollutants emissions when authorities make policies towards regional sustainable development.

Large well building is important remote sensing object, and the research on object detection of large well buildings is of great significance to national defense. At the data level, due to the small number of large well buildings samples, there is currently no valid data set available for the object detection. Building effective datasets is of great value for the research in related fields. At the algorithm level,the different resolutions of the remote sensing images result in multi-scale characteristics of the large well buildings, which is one of the difficulties in solving the object detection problem. Based on the above analysis, firstly, this article built the first large well buildings object detection dataset using Google Earth. Then an effective detection model was designed for large well building object detection task. The model in this paper fully integrates the object's multi-scale features and contextual information, and detects the object through the multi-stage cascade network. The model can effectively detect large well buildings, and the detection effect is better than the results of the current mainstream algorithms.

Based on a high-speed photographic system, experiments on three-dimensional free jets of liquid metal in the absence of a magnetic field and in a horizontal magnetic field have been carried out to observe the process of liquid GaInSn jet breakup and droplet formation in an oxygen-free environment with a maximum We of 400 and a maximum Ha of 30. From the results on jet morphology, surface disturbance, and breakup length, we analyzed the characteristics of jet breakup. In the absence of magnetic field, the jet shows nine different morphologies, and the surface disturbance shows two forms:expansion wave and sinusoidal wave; with the increase of We, the disturbance amplitude first decreases and then increases, and the breakup length first increases and then decreases. When a horizontal magnetic field is imposed, the jet shows four typical morphologies, with the leading edge of the jet being flattened in the direction of the vertical magnetic field line and elliptical along the magnetic field line. As the Ha number increases, the jet break length tends to increase overall, but decreases in some operating conditions. The results of this paper have enriched the phenomenon of liquid metal jets under magnetohydrodynamic effects.

The quadratic assignment problem (QAP) is one of the NP-hard combinatorial optimization problems and is known for its diverse applications in real life. The current relatively mature heuristic algorithms are usually problem-oriented to design customized algorithms and lack the ability to transfer and generalize. In order to provide a unified QAP solution strategy, this paper abstracts the flow matrix and distance matrix of QAP problem into two undirected complete graphs and constructs corresponding correlation graphs, thus transforming the assignment task of facilities and locations into node selection task on the association graph. Based on actor-critic framework, this paper proposes a new algorithm ACQAP(actor-critic for QAP). Firstly, the model uses a multi-headed attention mechanism to construct a policy network to process the node representation vectors from the graph convolutional neural network; Then, the actor-critic algorithm is used to predict the probability of each node being output as the optimal node. Finally, the model outputs an action decision sequence that satisfies the objective reward function within a feasible time. The algorithm is free from manual design and is more flexible and reliable as it is applicable to different sizes of inputs. The experimental results show that on QAPLIB instances, the algorithm has stronger transfer and generalization ability under the premise that the accuracy is comparable to the traditional heuristic algorithm, while the assignment cost for solving is less compared to the latest learning-based algorithms such as NGM, and the deviation is less than 20% in most instances.

In this paper, we calculate the norm of the Hausdorff operator $\mathscr{H}_Φ$ defined on the Morrey space with power weights L^p,λ($\mathbb{R}^n$,|x|^αdx) and the homogeneous central Morrey space with power weights $\dot B$^p,λ($\mathbb{R}^n$,|x|^αdx), respectively. We also extend our results to the product Hausdorff operator $\mathscr{H}$_Φ^m.

Space radiation interference especially the single event upset (SEU) effect can make a great impact on the normal and stable operation of the neural network chips, which can lead to the random bit flips of the weight parameters stored in the SRAM, and thus change the values of the neural weight parameters and the accuracy of the neural network chip outputs. In this paper, we analyze the commonly used radiation resistant methods, and try to overcome the problem of hardware consumption, recovery time, and processing speed. After the software simulation of the neural network accuracy with the different ratios of weight parameter errors, we adopt the dropout algorithm to construct the novel network framework to avoid the decline of accuracy. The algorithm can mask the neurons affected by SEU with some probability. The simulation experiment results show that this method can promote the accuracy of the neural network affected by SEU.

The northern Pakistan is the central area of the Gandhara culture in history, where a large number of Hinduist, Buddhist and Islamic cultural heritage sites are located. Based on multisource data such as historical sources, DEM data, and field survey, the spatial and temporal distribution characteristics and preservation of Hinduist, Buddhist, and Islamic cultural heritage sites have been studied by using spatial analysis methods of GIS, and the reasonable protection countermeasures have been suggested. The research results indicate that Northern Pakistan has witnessed the coming, rising, and falling of Hinduist, Buddhist, and Islamic cultures. The kernel density estimate shows that these three types of cultural heritage sites have different distribution ranges and aggregation degree, but they all gather around Islamabad with the highest aggregation degree. High resolution DEM analysis shows that more than 90% of these three types of cultural heritage sites are no farther than 800 m away from the river, which confirms the great significance of the river in history when these culture sites were selected. It also can be known from DEM that the location of these three types of sites is mainly no higher than 800 m, and the slope range is mainly between 0° and 10°, along rivers and main roads. Of the statistical data, the number of Islamic sites is the largest, followed by Buddhist sites and Hinduist sites respectively, reflecting the change of the number of preserved sites over time. Due to nature factors and human activities, at present, many sites such as Bhera and Vijhi are in a bad condition, which demands more efforts and better strategies to preserve. As to the current preservation and risks of the cultural heritage sites in Northern Pakistan, the corresponding protection strategies are put forward.

Lindane, as one of persistent organic pollutants (POPs), has high toxicity, lipophilicity, chemical stability, and bioaccumulation. The use of lindane seriously threatens ecological environment and biosafety in China. Lindane bioaccumulation can cause liver function damage, but its toxic effects on liver have not been fully understood. In this study, we have investigated effects of lindane-induced oxidative stress on inflammation and autophagy in HepG2 cells. HepG2 cells were exposed to lindane for 24 h. Cell viability, intracellular reactive oxygen species (ROS), and autophagy were determined by thiazolyl blue (MTT) assay, 2,7-dichlorofluorescein (DCF) and immunofluorescence, respectively. The expression levels of IL-6, IL-8, IL-1β, TNF-α, NF-κB, Beclin1, ULK1 mRNA and P-P65, IL-1β, P62, LC3 protein were detected by real-time quantitative PCR and Western blot, respectively. We found that high-concentration lindane significantly decreased the cell viability, triggered an increase in both intracellular ROS and lactate dehydrogenase (LDH) levels, and decreased the catalase (CAT) activity. Moreover, the expression levels of TNF-α, IL-6, IL-1β, IL-8, NF-κB mRNA and P-p65, IL-1β, P62, LC3, and NF-κB protein were significantly up-regulated after lindane exposure. These results indicated that the lindane exposure causes oxidative stress and autophagy as well as inflammatory through activating NF-κB signaling pathway in HepG2 cells.

In recent years, suburban mountain areas have become a good choice for urban residents to go outing. Intensive tourist outings and villagers' production activities bring fire safety hazards to mountains and forests. And road information is vital information for forest fire prevention emergency. However, due to the problems of occlusion, shadow, narrow and multiple branches in suburban mountainous roads, conventional urban road extraction algorithms have poor performance in suburban mountain areas. This paper proposes a road semantic segmentation model and a training method that transforms the binary into a multi-class classification problem, forcing the model to focus on learning spatial distance information to generate road results with better spatial continuity. Then, experiments were carried out on the Yajishan road dataset made by ourselves and the Massachusetts public road dataset respectively to verify the effectiveness of our model and training method. In addition, it is verified that the training method is also applicable to other commonly used semantic segmentation models such as U-Net and DeepLabV3. Finally, this paper also conducts post-processing research based on the above road extraction results to output road surface, road centerline vector data with road width information, and conducts fire truck traffic analysis application in Beijing Yaji Mountain. The research results have alleviated the problem of insufficient road information for commercial electronic maps in the suburban mountain areas with few people, and provided information technology support for forest fire emergency rescue.

The collision between the Indian and Asian plates started about 60 Ma ago. It has continued to shorten the Asian side by at least 1 000 km, and thus formed the huge Qinghai-Tibetan Plateau. One important lithospheric dynamical issue is what kind of force could overcome such huge resistance from the Asian plate and from the Qinghai-Tibetan Plateau and drive the Indian plate continuously moving northward for such a long time. We performed 3D numerical modeling to reveal the role of mantle flow for the dynamics of the India-Asia collision by considering the global temperature structure converted from the seismic tomographic model S20RTS as initial temperature and the absolute plate motion velocities as initial surficial velocity boundary conditions. Our results suggest that drags from the moving asthenospheric mantle could be the main cause for transporting Indian plate northward.

Aiming at solving the difficult problem of tracking moving ships, which is caused by their great position variation in long time interval sequential SAR images, a matching tracking method combining track initiation and image feature is proposed. Firstly, the ships are detected from the SAR image to obtain the image slices and extract the ships' image features and spatial position information. Then, ships at different times are correlated to form multiple candidate trajectories according to the constraints of velocity and acceleration. Finally, ship trajectories are screened out based on the minimum difference principle of feature matching to realize moving ship tracking. The effectiveness of the proposed method is verified by simulation experiments and airborne sequential SAR images processing analysis.

The GF-4 satellite is a geostationary-orbit optical remote sensing satellite with the highest resolution in the world. The satellite can incessantly observe the designated area with high temporal resolution, high spatial resolution, and large imaging width. In this paper, a multiple moving ships detection method for GF-4 satellite is proposed. Firstly, the median filtering denoising and non-linear gray-scale stretching are applied to the remote sensing image. Secondly, the saliency map is extracted by the spectral residual (SR) method. Finally, a fusion processing is further performed by adopting the image fusion method based on weighted Dempster-Shafer evidence theory (WDS) to detect ships. Evaluated by the experiment based on real data of GF-4 satellite, it is shown that the proposed method can detect the multiple moving ships accurately and efficiently by the GF-4 satellite image.

With the rapid development of remote sensing technology, the remote sensing data become valuable in many practical applications. Among them, target detection has always been an important topic. However, most of the target detection algorithms in remote sensing images merely concentrate on single-temporal data, and there are few algorithms for multi-temporal data. In the field of target detection in multi-temporal remote sensing data, filter tensor analysis (FTA) has achieved great success which outperforms other target detection algorithms for single-temporal data. Yet FTA is designed only for single target detection, which means it can not meet the need for practical applications in circumstances where it has to detect more than one target simultaneously. So, in this paper, a modified algorithm for multi-target detection in multi-temporal data has been proposed based on the target constraints in multiple target constrained energy minimization and the tensor filter in FTA. Both the experiment results on simulation data and real remote sensing data from Landsat 8 prove that the algorithm proposed in this paper can effectively detect several targets in multi-temporal data.