Deep networks have been used for semantic segmentation tasks on scenes of outdoor environments with increasing popularity. However, the majority existing work centers daytime favorable illumination and weather conditions, relies supervision pixel-level annotations. This paper seeks to address problem rainy, night-time without using We introduce a near scene approach that uses images as bridge transferring knowledge from pre-trained models rainy night images. Specifically, we first present...

To improve the robustness and accuracy of corner-detection algorithm, this paper proposes a camera-calibration method based on EDLines algorithm for automatic detection chessboard corners. The is initially used to perform straight-line calibration image. features broken straight lines at corners are then filter remove background outside chessboard. pixels in rectangular area around filtered line sorted by gray gradient. After using results fit line, coordinates intersection taken as initial...

Subsampling or subdata selection is a useful approach in large-scale statistical learning. Most existing studies focus on model-based subsampling methods which significantly depend the model assumption. In this article, we consider model-free strategy for generating from original full data. order to measure goodness of representation with respect data, propose criterion, generalized empirical <inline-formula><tex-math notation="LaTeX">$F$</tex-math></inline-formula> -discrepancy (GEFD), and...

Due to recent advances in learning-based semantic segmentation, road scene parsing can usually achieve satisfactory results under normal illumination conditions. However, training a robust model for nighttime scenes is still very challenging, especially when labels of samples are absent. In this paper, we propose convolutional neural network (CNN)-based method an unsupervised manner. The proposed system includes appearance transferring module and segmentation module, which coupled together...

Multiple vision sensors (multi-VSs) are difficult to calibrate in large fields of view measurement owing the lack overlapping areas. Thus, this study proposes a novel method for external parameter calibration multi-VSs by using an encoded 1D target. The quick response code is used uniquely identify feature circle. Hough transform and ellipse center solving jointly applied optimize points further improve collinearity points. As cross-ratio involves projection invariance, rotation translation...

Light Detection and Ranging (LiDAR) is widely used in the perception of physical environment to complete object detection tracking tasks. The current methods datasets are mainly developed for autonomous vehicles, which could not be directly roadside perception. This paper presents a 3D point cloud stitching method with wide horizontal field view (FoV) using LiDAR. Firstly, base model trained by KITTI dataset has achieved accuracy 88.94. Then, new range 180° can inferred break limitation...

A novel series of conformationally constrained tricyclic tropane analogues, (Z)-9-(substituted arylmethylene)-7-azatricyclo[4.3.1.03,7]decanes, were prepared, and their abilities to inhibit high-affinity uptake dopamine (DA), serotonin (5-HT), norepinephrine (NE) into rat brain nerve endings (synaptosomes) evaluated. First, a systematic screening variety different substituents on the phenyl ring indicated that substitution pattern plays an important role in monoamine transporter activity....

The vasculature inside breast cancers is one important component of the tumour microenvironment. investigation its spatial morphology, distribution and interactions with cancer cells, including stem essential for elucidating mechanisms development treatment response. Using confocal microscopy fluorescent markers, we have acquired three-dimensional images within mammary tumours normal gland mouse models. However, it difficult to segment reconstruct complex accurately from in vivo owing...

Interactions between cancer cells and factors within the tumor microenvironment (mE) are essential for understanding development. The spatial relationships blood vessel cells, e.g. initiating (TICs), an important parameter. Accurate segmentation of is necessary quantization their relationships. However, this remains open problem due to uneven intensity low signal noise ratio (SNR). To overcome these challenges, we propose a novel approach that integrates oriented hidden Markov random field...

With the development of augmented reality (AR) technologies, more and approaches are proposed for medical applications. help AR technology, doctor can highly improve spatial perception obtain information beneath displayed image. Generally, 3D reconstruction, registration, tracking depth visualization all important an system, which directly determine accuracy system. However, none recent systems have achieved a perfect vision doctors. It is particularly challenging system to perform well...

Range image segmentation is one of the most common problem in field computer vision. In view defect about traditional range methods, this paper introduces a new method based on snake active contour model (SCAM). First, expresses form parameters, and illustrates line data points inside outside movement pseudo code description motion point algorithm; then constructs energy function, pushing Euler equation discretization; finally gets results by Cholesky decomposition. Numerical experiment show...

In this paper, we establish the matching relation between implication operator and aggregation operator, which provides a new solution for design construction of multi-rule fuzzy inference system. Firstly, according to definition monotonicity classification method is proposed, then process using different operators classified. Then, dynamic maximum minimum are proposed. Based on compositional rule (CRI) method, basis systems given illustrated with examples. Finally, applicability proposed in...

A problem of multi-sensor robust information fusion for uncertain systems with colored noises is investigated. new optimal steady-state filter weighted by matrix and scalars parameters given using theory based on linear least variance criterion. Compared existing filter, the proposed considers perturbation factor in solution to weighting matrices, parameter filter. Simulation example shows effectiveness method.

Fish is a common and widely distributed creature. Its skin has unique physiological structure plays an important role in many fields. also potential value for bionics research. This study aims to provide method reliable data the of bionics. A measuring mechanical properties fish samples using binocular stereo digital image correlation (DIC) system combined with synchronous tensile testing machine was proposed. The (e.g., elastic modulus E strain) grass (GFSA) were tested hydrophilic dry...

&lt;p&gt;Fraud and error are two underlying sources of misstated financial statements. Modern machine learning techniques provide a potential direction to distinguish the factors in such In this paper, thorough evaluation is conducted on how off-the-shelf tools perform for fraud/error classification. particular, task treated as standard binary classification problem; i.e., mapping from an input vector indices class label which either or fraud. With real dataset restatements, study...

Natural language question answering over knowledge graphs is an important and interesting task as it enables common users to gain accurate answers in easy intuitive manner. However, remains a challenge bridge the gap between unstructured questions structured graphs. To address problem, natural discipline building query represent input question. Searching graph can produce Distinct from existing methods that are based on semantic parsing or templates, we propose effective approach qaSQP...

It is becoming study hotspot of academia that using visible 3D expression face curved surface to recognize face. These issues involved in founding 3D-face recognition system were analyzed such as basic process, function, technology line, architecture and database programming. Then some key technologies explored which development management platform, analysis video frequency flow, automatic demarcation character points on the face, calculation shooting angle so on.

Given a natural language phrase, relation linking aims to find (predicate or property) from the underlying knowledge graph match phrase. It is very useful in many applications, such as question answering, personalized recommendation and text summarization. However, previous algorithms usually produce single for input phrase pay little attention more general challenging problem, i.e., combinational that extracts subgraph pattern compound (e.g. mother-in-law). In this paper, we focus on task...

For 2D image data target detection, this paper proposes an improved detection algorithm for YoloV5s based on the attention mechanism. By replacing Focus module in with a CBL module, it facilitates model export. The four modules, SELayer, EcaLayer, CBAM and CoordAtt are added to convolutional network of YoLoV5s. Before P4 layer, before P5 SPPF layer last original backbone network, respectively. experimenting algorithms that include YoloV5s, It was concluded among five tested evaluation....

In this paper, we introduced the design and implementation of a general-purpose assembler that can be written modified to adaptive for any given assembly language generate target machine code carry out execution. The system consists parsing module an execution module. former generates from inputted code, latter simulates hardware operation computer executes instructions. Due its versatility flexibility, allows easy modification reconfigure itself implementing instruction set architecture. As...