With the rapid prevalence of smart mobile devices and dramatic proliferation location-based social networks (LBSNs), recommendation has become an important means to help people discover attractive interesting points interest (POIs). However, extreme sparsity user-POI matrix cold-start issue create severe challenges, causing CF-based methods degrade significantly in their performance. Moreover, requires spatiotemporal context awareness dynamic tracking user's latest preferences a real-time manner.

Based on evolutionary computation, a novel real-time route planner for unmanned air vehicles is presented. In the planner, individual candidates are evaluated with respect to workspace so that computation of configuration space not required. The incorporates domain-specific knowledge, can handle unforeseeable changes environment, and take into account different kinds mission constraints such as minimum leg length flying altitude, maximum turning angle, fixed approach vector goal position....

Aircraft detection in satellite images is generally difficult due to the variations of aircraft type, pose, size and complex background. In this paper, we propose a new framework based on objectiveness techniques (e.g., BING) Convolutional Neural Networks (CNN). The advantages are two folds. On one hand, first introduce CNN for detection, as can learn rich features from raw data automatically has yielded state-of-the-art performance many object tasks. other use candidate regions proposed by...

Supervised metric learning plays a substantial role in statistical classification. Conventional algorithms have limited utility when the training data and testing are drawn from related but different domains (i.e., source domain target domain). Although this issue has got some progress feature-based transfer learning, most of work area suffers non-trivial optimization pays little attention to preserving discriminating information. In paper, we propose novel algorithm knowledge an...

Based on mathematic simulations, the impact of spectral filtering pulse breaking up and noise-like generation in all-normal-dispersion fiber lasers are investigated. Three types spectrum filters employed which have a Gaussian-shaped profile, super-Gaussian-shaped sinusoidal-shaped respectively. With filter, breaking-up process is discussed. The filter two different formation mechanisms for pulses revealed. In addition, with dissipative solitons central wavelengths achieved.

This paper addresses the problem of water extraction from high resolution remote sensing images (including R, G, B, and NIR channels), which draws considerable attention in recent years. Previous work on mainly faced two difficulties. 1) It is difficult to obtain accurate position boundary because using low images. 2) Like all other image based object classification problems, phenomena "different objects same image" or object" affects extraction. Shadow elevated (e.g. buildings, bridges,...

With the rapid technological development of various different satellite sensors, a huge volume high-resolution image data sets can now be obtained and widely used in military civilian fields. Detecting typical targets images is challenging task due to complicated background. Traditional manually engineered features (i.e. HOG, Gabor feature Hough transform, etc.) do not work well for massive remote sensing data. Thus, we are expected find an efficient way automatically learn presentations...

Marine radar is employed across multiple fields, notably in navigation, meteorology, defense, and security. images are highly sensitive to sea clutter, highlighting the crucial importance of clutter suppression image processing. However, existing algorithms for often struggle effectively generalize complex marine environments. In this paper, we introduce RadarDiff, a novel approach that leverages diffusion models enhance plan-position indicator (PPI) images. We treat as an image-to-image...

Knowledge discovery in texts (KDT) has been widely applied for business data analysis, but it only reveals a common pattern based on large amounts of data. Since 2000, chance (CD) as an extension KDT proposed to detect rare significant events or situations regarded candidates human decision making. Key Graph is useful and important algorithm well tool CD mining visualizing these chances. However, scenario graph visualized by machine-oriented, causing bottleneck cognition. Traditional...

Spectrum prediction in the cognitive radio system attracts more and attention. It can predict future spectrum holes to save energy of sensing improve efficiency access. The current research on is use model such as back propagation (BP) neural network predict. However, performance conventional not satisfied meet real for its using inaccurate states defects BP network. Therefore, we propose a based improved-BP networks. In proposed model, channel power values information instead are used...

Extracting opinion words and targets from online reviews is an important task for fine-grained mining. Usually, traditional extraction methods under the pipeline-based framework have higher precision but lower recall, while in propagation-based possess greater recall poorer precision. To achieve better performance both this paper proposes a unified mining, combining propagation with refinement dynamic iterative process. In process, syntactic patterns are chosen as relations to extract new...

Learning an appropriate feature representation across source and target domains is one of the most effective solutions to domain adaptation problems. Conventional cross-domain learning methods rely on Reproducing Kernel Hilbert Space (RKHS) induced by a single kernel. Recently, Multiple (MKL), which bases classifiers combinations kernels, has shown improved performance in tasks without distribution difference between domains. In this paper, we generalize framework MKL for propose novel...

With the growing demand for geospatial data, aerial imagery with high spatial, spectral, and temporal resolution achieves great development. It is imperative to evaluate whether acquired images are qualified enough, since further image mosaic asks strict time consistency a re-flight involves considerable resources. In this paper, we address problem of quick series quality assessment. An analysis system proposed, which includes single assessment, assessment based on matching, offering visual...

We consider multi-agent decision making problems in which agents need to communicate with other make socially optimal decisions but, at the same time, have some private information that they do not want share. Abstract argumentation has been widely used both single-agent and problems, because of its ability for reasoning incomplete conflicting information. In this work, we propose an abstract argumentation-based knowledge representation communication protocol, such can find strategies by...

In recent years, optical systems near the diffraction limit have been widely used in high-end applications. Evidently, an analytical solution of point spread function (PSF) will help to enhance both understanding and dealing with imaging process. This paper analyzes Fresnel diffraction-limited defocused conditions. For this work, PSF was obtained using series expansion confluent hypergeometric functions. The expression transfer is also presented herein for comparison PSF. Additionally, some...

In order to obtain complete and correct information for road line extraction from remote sensing image, the complementary of two kinds feature is discussed a fusion method proposed. The implemented by cross validating features LSD SRM according their spatial relationship. results experiments demonstrated these features, performance with fused can be improved obviously than any single kind feature.

In the cognitive radio system, spectrum prediction attracts more and attention, which predicts future holes to save energy of sensing improve efficiency access. The current research on is similar hard decision in communication system. However, loses amount channel information during process obtaining statuses, decreases predictive accuracy prediction. Therefore, we propose a "soft decision" model for based back-propagation neural networks. proposed model, power values frequency sampling...

The land use and cover classification is an important hot research topic in remote sensing image processing. How to information effectively data categorize different scenes needs urgent attention. In this paper, we analysis the Bag-of-Word model based feature extracting method systematically, propose Saliency Map Cooperated (SMC) coding strategy according characters of images. proposed SMC takes into account both primary objects partial large scale images, with little effecting on texture...