Echinococcosis is a worldwide anthropozoonosis which highly endemic over large animal husbandry areas in northwestern China. The current clinical therapeutic medicine against echinococcosis albendazole, although it caused serious side effects patients. component traditional Chinese herb medicine, Sophora moorcroftiana alkaloids (SA), thought to be potential drug treat echinococcosis. In order explore the effect and mechanism of SA treatment echinococcosis, we established model treated rats...

Inferring gaze target or following is an effective way to understand human actions and intentions, which makes quite a challenge. Some existing studies on estimation cannot accurately locate the in 3D scene by direction alone, while other have failed exploit contexts scene. In this article, we make full use of information obtained RGB-D camera innovatively expand from 2D image space through predicted vector. Specifically, rebuild new gaze-following dataset, Attention contains real-world...

Deep Convolutional Neural Networks (DCNNs) have achieved impressive performance in Single Image Super-Resolution (SISR). To further improve the performance, existing CNN-based methods generally focus on designing deeper architecture of network. However, we argue blindly increasing network's depth is not most sensible way. In this paper, propose a novel end-to-end Residual Neuron Attention (RNAN) for more efficient and effective SISR. Structurally, our RNAN sequential integration...

In this paper, we handle the problem of human action recognition by combining covariance matrices as local spatio-temporal (ST) descriptors and ST features extracted densely from video.Unlike traditional methods that separately utilizing gradient-based feature optical flow-based feature, use matrix to fuse two types feature.Since are Symmetric Positive Definite (SPD) matrices, which form a special type Riemannian manifold.To measure distance SPDs while avoid computing geodesic between them,...

When multi-robot systems execute large-area detection tasks, such as airport runways, taxiways, and parking aprons, it is necessary to consider the execution time of tasks entry exit positions robots. To shorten overall for executing inspection ensure that each robot can be allocated a balanced task during allocation. In this paper, based on sequential single-item (SSI) auction algorithm, allocation method considers total distance area proposed solve problem unbalanced load tasks. It better...

Due to its importance, figure/ground segmentation in video has gained interest recently. The key factor of the is construction spatio-temporal coherence. Previous works usually use motion approximation as a measurement coherence, resulting low accuracy. In this paper, we present novel method measure and an algorithm for target tracking proposed. Each image abstracted by some compact perceptually homogeneous elements, representing elements sparse linear combinations dictionary templates,...

Deep Convolutional Neural Networks (DCNNs) have achieved impressive performance in Single Image Super-Resolution (SISR). To further improve the performance, existing CNN-based methods generally focus on designing deeper architecture of network. However, we argue blindly increasing network's depth is not most sensible way. In this paper, propose a novel end-to-end Residual Neuron Attention (RNAN) for more efficient and effective SISR. Structurally, our RNAN sequential integration...

As characterizing videos simultaneously from spatial and temporal cues have been shown crucial for video processing, with the shortage of information soft assignment, vector locally aggregated descriptor (VLAD) should be considered as a suboptimal framework learning spatio-temporal representation. With development attention mechanisms in natural language this work, we present novel model VLAD following self-attention operations, named weighted (ST-SAWVLAD). In particular, sequential...

The complete coverage detection of airport pavement is one the important tasks to ensure safety civil aviation. Due large area and rigid requirements for ultra-high security, humans robots are required interact when performing specific tasks. Therefore, it a challenge divide environment into suitable working areas. In this paper, we propose partition planning method based on adjustable rectangular decomposition environment. First, use rectangles decompose initial areas without obstacles,...

Human action recognition in videos draws huge research interests computer vision. The Bag-of-Word model is quite commonly used to obtain the video level representations, however, BoW roughly assigns each feature vector its nearest visual word and collection of unordered words ignores interest points' spatial information, inevitably causing nontrivial quantization errors impairing improvements on classification rates. To address these drawbacks, we propose an approach for by encoding...

It is important to simultaneously characterize videos with spatial and temporal information, especially for human action recognition, as cue can model the appearance while dynamic motion need be represented by cue. The vector of locally aggregated descriptor (VLAD) whose assignment shortage regarded a suboptimal solution recognition. In this paper, VLAD soft spatio-temporal assignment, named VLAD-SSTA, proposed further boost performance recognition employing characteristic. Specifically,...

Airport runway safety inspections and timely Foreign Object Debris (FOD) removal can effectively guarantee the safe take-off landing of flights. At present, number flights taking off on airport is large, inspection time window short, so there are problems such as FOD in time, which affect flight takeoff landing. The use multiple robots for shorten operation time. In order to solve problem multi-robot task allocation removal, this paper proposes a "inspection-removal" model an extended...

Most human-aware navigation methods usually gather human features like: speed, gesture, and other characteristics to establish a comfortable space. However, they do not consider the direction of vision, which often results in intrusion gaze space by robots. This will bring strong sense discomfort humans. paper proposes framework based on an asymmetric Gaussian model pedestrian trajectory prediction. Firstly, robot's RGB-D camera laser sensors are used detect pedestrian's direction, personal...

The coverage detection of airport pavement is the premise to ensure civil aviation safety. Due hard requirements ultra-high security itself, application robots unmanned in not accomplished overnight, but a process step-by-step and continuous technical iteration. Therefore, for different stages robot airport, an appropriate method should be given guide maintenance engineers flexibly adjust partition operation mode. In this paper, we design evaluation index named Partition Quality Function...