动态场景下基于视觉同时定位与地图构建技术的多层次语义地图构建方法

为提高视觉同时定位与地图构建(SLAM)技术的环境适应性和语义信息理解能力,该文提出一种可以在动态场景下实现多层次语义地图构建的视觉SLAM方案。首先利用被迫移动物体与动态目标间的空间位置关系,并结合目标检测网络和光流约束判断真正的动态目标,从而剔除动态特征点;其次提出一种基于超体素的快速点云分割方案,将基于静态区域构建的3维地图进行优化,构建了物体级的点云语义地图;同时构建的语义地图可以提供更高精度的训练数据样本,进一步用来提升目标检测网络性能。在TUM和ICL-NUIM数据集上的实验结果表明,该方法在定位精度上远优于目前主流的动态场景下的视觉SLAM方案,证明了该方法在高动态场景中具有较好的稳定性和鲁棒性;在建图精度和质量上,经过将重建的不同种类地图与各个现有方法进行比较,验证了提出的多层次语义地图构建的方法在静态和高动态场景中的有效性与适用性。     

基于激光点云NDT特征的两步回环检测

针对传统激光雷达配准算法进行大规模同时定位与建图(SLAM)时,存在较大累计误差的问题,本文提出一种基于正态分布变换(NDT)的两步回环检测方法,充分利用NDT配准中点云均值与方差特征,并将所提方法加入SLAM完整框架。点云匹配中采用重叠网格,首先根据各网格特征值,构建点云外观描述,进行粗回环检测。符合粗回环条件后,计算点云网格均值到坐标原点距离的方法使点云具有旋转不变性,进行精确回环检测。本文提出算法在“小旋风”智能车平台进行验证,实验表明,所提算法可以有效减小大规模建图中的累计误差,系统的鲁棒性更强,跟踪性能更好。     

基于深度特征提取和图神经网络匹配的图像复制粘贴篡改检测

针对图像中特征提取不均匀、单尺度超像素划分对伪造定位结果影响较大的问题,提出一种基于深度特征提取和图神经网络(graph neural network,GNN) 匹配的图像复制粘贴篡改检测(cope-move forgery detection,CMFD) 算法。首先将图像进行多尺度超像素分割并提取深度特征,为保证特征点数目充足,以超像素为单位计算特征点分布的均匀度,自适应降低特征提取阈值;随后引入新的基于注意力机制的GNN特征匹配器,进行超像素间的迭代匹配,且用随机采样一致性(random sample consensus,RANSAC) 算法消除误匹配;最后将多尺度匹配结果进行融合,精确定位篡改区域。实验表明,所提算法具有良好的性能,也证明了GNN在图像篡改检测领域的可用性。     

基于点模式匹配的芯片缺陷检测算法研究

针对高速与高精度场景下芯片缺陷检测实时性和准确性难以同时保证的问题,提出一种基于方位环境特征的点模式匹配定位算法(Azimuth Environment Feature Vector-Based Point Pattern Matching Localization Algorithm,AEF-PPMLA),提升检测的实时性、准确性和易用性。该算法含两个部分:方位环境特征向量计算方法(Azimuthal Environment Feature Vector Calculator Method,AEF-VCM)和相似度计算方法 (Matching Degree Calculator Method,MDCM)。AEF-VCM对芯片的方位环境特征进行向量描述,减少点匹配的计算量,提高检测的实时性; MDCM采用卡方检验算法来度量特征向量的相似度,提高检测的准确度。实验部分验证算法的定位精度、耗时以及缺陷识别的准确率,结果表明所提出的AEF-PPMLA能够快速准确定位芯片并识别引脚缺陷,满足高质量生产需要。     

基于深度学习的视觉SLAM闭环检测方法

针对在摄像机视角、光照、气候、地貌等条件的大幅度变化或者存在快速移动物体的复杂场景下,视觉即时定位与地图构建(simultaneous localization and mapping,SLAM)的精确性和鲁棒性较低等问题,闭环检测作为解决SLAM位姿漂移的重要环节,提出了一种基于神经网络的闭环检测方法。该方法通过传感器获取视觉图像的数据,不同于传统方法的特征提取,采用改进三重约束损失函数训练Darknet提取图像特征,构造对应特征向量矩阵。由于Darknet借鉴了残差网络(resnet)的思想,在具有较深网络层数的同时,仍保持较高的准确率,减少了特征提取误差。经过自编码器方法对数据进行降维处理,通过余弦相似度计算,设定合理阈值,能够更快的得到闭环检测结果。最后通过在两个公开视觉SLAM闭环检测数据集,New College数据集和光照及角度变化更明显的City Centre数据集上进行实验,结果表明复杂环境下本文提出的方法比现有闭环检测方法,能够得到更高准确率和速率,更好满足了视觉SLAM系统对消除累计误差和实时性的要求。     

基于Wi-Fi即时定位与映射像素模板匹配的室内运动地图构建与定位

传统指纹定位方法由于建库人力时间开销大、系统通用性不强约束着指纹定位系统的推广,为了解决该问题同时结合即时定位与映射(SLAM)技术的优势,该文提出一种新的Wi-Fi/微机电系统(MEMS)融合室内运动地图构建与定位方法。首先利用行人航迹推算(PDR)、最小描述长度(MDL)原则和基于密度的空间聚类算法(DBSCAN)对众包运动轨迹进行预处理,提出基于轨迹主路径的运动地图构建方法。之后提出基于像素模板的地图匹配方法获取地图的绝对位置,并采用抗差扩展卡尔曼滤波(EKF)对目标位置进行最优估计。实验结果表明,所提聚类方法可以准确构建各区域运动地图,在少量的人力时间开销下实现较高的定位精度。     

一种基于匹配的Android系统漏洞检测方法

目前Android系统占领了智能移动操作系统的绝大多数份额,然而Android系统的安全问题却令人忧虑,近年来频频爆出高危漏洞,给广大Android用户带来了很大的安全隐患。对Android系统漏洞检测进行了研究,提出了一种基于匹配的快速检测Android系统漏洞的方法,同时设计和实现了一个基于脚本的漏洞扫描工具。实验表明,该工具可有效检测Android系统漏洞,运行效果良好。     

Illumination robust image transformations for feature-based SLAM using photometric and feature matches loss

Simultaneous localization and mapping (SLAM) technology becomes more and more important in robot localization. The purpose of this paper is to improve the robustness of visual features to lighting changes and increase the recall rate of map re-localization under different lighting environments by optimizing the image transformation model. An image transformation method based on matches and photometric error (name the method as MPT) is proposed in this paper, and it is seamlessly integrated into the pre-processing stage of the feature-based visual SLAM framework. The results of the experiment show that the MPT method has a better matching effect on different visual features. In addition, the image transformation module encapsulated by a robot operating system (ROS) can be used with multiple visual SLAM systems and improve its re-localization effect under different lighting environments.     

基于纹元森林和显著性先验的弱监督图像语义分割方法

弱监督语义分割任务常利用训练集中全体图像的超像素及其相似度建立图模型,使用图像级别标记的监督关系进行约束求解。全局建模缺少单幅图像结构信息,同时此类参数方法受到复杂度限制,无法使用大规模的弱监督训练数据。针对以上问题,该文提出一种基于纹元森林和显著性先验的弱监督图像语义分割方法。算法使用弱监督数据和图像显著性训练随机森林分类器用于语义纹元森林特征(Semantic Texton Forest, STF)的提取。测试时,先将图像进行过分割,然后提取超像素语义纹元特征,利用朴素贝叶斯法进行超像素标记的概率估计,最后在条件随机场(CRF)框架下结合图像显著性信息定义了新的能量函数表达式,将图像的标注(labeling)问题转换为能量最小化问题求解。在MSRC-21类数据库上进行了验证,完成了语义分割任务。结果表明,在并未对整个训练集建立图模型的情况下,仅利用单幅图像的显著性信息也可以得到较好的分割结果,同时非参模型有利于规模数据分析。     

基于分区动态规划和航迹关联的弱小目标检测

序列图像中低信噪比运动弱小目标的实时检测算法,是精确制导系统中的关键算法之一。提出一种图像序列弱小目标实时检测新算法,采用分方向区间的动态规划算法和二值图像航迹关联检测。通过在动态规划能量累加过程中引入方向限制,减小了噪声轨迹能量积累和目标轨迹能量扩散,提高了算法的目标检测能力。对算法的检测性能进行了仿真实验,结果表明该算法能有效检测深空背景下信噪比大于1.8、运动方向任意、速度不大于1像素/帧的多个运动弱小目标。     

Robust tracking based on local structural cell graph

Structure information has been increasingly incorporated into computer vision, however most trackers have ignored the inner spatial structure of the object. In this paper, we develop a simple yet robust tracking algorithm based on local structural cell graph (LSCG). This approach exploits both partial and spatial information of the target via representing the object with local structural cells (LSCs) and constructing a graph to model the spatial structure between the inner parts of the object. The tracking is formulated as matching LSCG, whose nodes are target parts and edges are the interaction between two parts. Within the Bayesian framework, we achieve object tracking by matching graphs between the reference and candidates. Eventually, the candidate with the highest similarity is the target. In addition, an updating strategy is adopted to help our tracker adapt to the fast time-varying object appearance. Experimental results demonstrate that the proposed method outperforms several state-of-the-art trackers.     

基于颜色空间和模板匹配的交通标志检测方法

交通标志检测技术是交通标志识别系统的重要前提和基础。由于背景的复杂性,在进行颜色分割时部分区域可能会受到干扰,为交通标志的检测带来困难。文中利用HSV颜色空间和RGB颜色空间对不同颜色的交通标志进行粗检测,标记不同的值实现ROI分割。然后利用模板匹配的方法对交通标志进行处理,使用模板在ROI区域上滑动,得到模板相似度的最大值,以此来实现检测过程。实验结果表明该方法能获得较好的检测结果。     

一种基于正交匹配追踪的压缩感知信号检测算法

针对当前压缩感知信号检测算法没有充分利用稀疏系数幅值信息的不足,提出了一种新的检测算法。从正交匹配追踪算法切入,通过深入分析归一化残差的变化信息,提出归一化余差概念,建立了一种基于归一化残差和归一化余差二维判决的信号检测算法。仿真结果表明,算法的有效检测阈值区间随着信噪比的降低而不断减小,且在信噪比为-8 dB、压缩比为0.25时,该算法的检测概率仍能满足要求,具备较好的适应性。     

Temporal matching prior network for vehicle license plate detection and recognition in videos

In real‐world intelligent transportation systems, accuracy in vehicle license plate detection and recognition is considered quite critical. Many algorithms have been proposed for still images, but their accuracy on actual videos is not satisfactory. This stems from several problematic conditions in videos, such as vehicle motion blur, variety in viewpoints, outliers, and the lack of publicly available video datasets. In this study, we focus on these challenges and propose a license plate detection and recognition scheme for videos based on a temporal matching prior network. Specifically, to improve the robustness of detection and recognition accuracy in the presence of motion blur and outliers, forward and bidirectional matching priors between consecutive frames are properly combined with layer structures specifically designed for plate detection. We also built our own video dataset for the deep training of the proposed network. During network training, we perform data augmentation based on image rotation to increase robustness regarding the various viewpoints in videos.     

基于Haar小波的地形匹配技术

在传统相关地形匹配的基础上，提出了一种基于Haar小波的多尺度地形匹配方法。这种方法用Haar小波对地形作多尺度分解，然后，利用分解后的HL和LH分量分别进行相关匹配，对匹配后的结果用Dempster-Shafer证据理论进行融合，得到正确匹配结果。由于Haar小波分解后减少了数据量，保留了原始地形的主要几何特征，所以，它的计算速度和匹配精度都优于传统的匹配方法。     

基于序贯相似性检测的NCC目标跟踪快速匹配方法

基于特征点的图像匹配方法的关键是准确快速地将可靠的特征点提取出来。经典归一化互相关匹配法属于基于特征点匹配方法中的一种,归一化互相关匹配法具有操作简单,匹配精度高等优点,但其计算量庞大,难以满足实时跟踪的要求。提出了一种基于序贯相似性检测的归一化互相关快速匹配方法,并对提出的改进方法进行实验验证。实验表明,该方法可以准确快速的进行特征点匹配,减少了算法的计算时间,有效地减少了发生误匹配的概率。     

Interference coordination method based on graph theory in two-tier cellular networks

This paper studies an interference coordination method by means of spectrum allocation in Long-Term Evolution (LTE) multi-cell scenario that comprises of macrocells and femtocells. The purpose is to maximize the total throughput of femtocells while ensuring the Signal-to-Interference plus Noise Ratio (SINR) of the edge macro mobile stations (mMSs) and the edge femtocell Mobile Stations (fMSs). A new spectrum allocation algorithm based on graph theory is proposed to reduce the interference. Firstly, the ratio of Resource Blocks (RBs) that mMSs occupy is obtained by genetic algorithm. Then, after considering the impact of the macro Base Stations (mBSs) and small scale fading to the fMS on different RBs, multi-interference graphs are established and the spectrum is allocated dynamically. The simulation results show that the proposed algorithm can meet the Quality of Service (QoS) requirements of the mMSs. It can strike a balance between the edge fMSs’ throughput and the whole fMSs’ throughput.     

FilterFA: a multiple string matching algorithm based on specification of character set

Multiple string matching is one of the core techniques of intrusion detection system, where Aho-Corasick al-gorithm is widely used. To solve the problem that huge storage overhead of AC would influence performance deeply, an improved algorithm ——FilterFA, based on specification of character set was proposed. This algorithm compressed large character by the character set mapping function, and constructed a new automata based on the mapped character set,then space complexity decreased to O(|P||Σ′|). Experiments on synthetic datasets and real-world datasets (such as ClamAV) show that the storage overhead of FilterFA is only about 3% of that of AC, while the size of the character set is 8, and the false recognition rate is less than 2%.     

RLP-AGMC: Robust label propagation for saliency detection based on an adaptive graph with multiview connections

There have been remarkable improvements in the salient object detection in the recent years. During the past few years, graph-based saliency detection algorithms have been proposed and made advances. Nevertheless, most of the state-of-the-art graph-based approaches are usually designed with low-level features, misleading assumption, fixed predefined graph structure and weak affinity matrix, which determine that they are not robust enough to handle images with complex or cluttered background.In this paper, we propose a robust label propagation-based mechanism for salient object detection throughout an adaptive graph to tackle above issues. Low-level features as well as deep features are integrated into the proposed framework to measure the similarity between different nodes. In addition, a robust mechanism is presented to calculate seeds based on the distribution of salient regions, which can achieve desirable results even if the object is in contact with the image boundary and the image scene is complex. Then, an adaptive graph with multiview connections is constructed based on different cues to learn the graph affinity matrix, which can better capture the characteristics between spatially adjacent and distant regions. Finally, a novel RLP-AGMC model, i.e. robust label propagation throughout an adaptive graph with multiview connections, is put forward to calculate saliency maps in combination with the obtained seed vectors. Comprehensive experiments on six public datasets demonstrate the proposed method outperforms fourteen existing state-of-the-art methods in terms of various evaluation metrics.     

基于初始尺度变换的SIFT匹配算法

直接使用检测到的SIFT(Scale-Invariant Feature Transformation)特征点进行特征点匹配,匹配性能仍然有待提升.提出了改进的SIFT匹配算法,利用匹配特征点的尺度比直方图,估计出近似的图像尺度比k,然后将空间分辨率较高的图像初始尺度增大到k倍,再次提取特征点进行匹配.实验结果表明,相比于其它用尺度约束条件提升性能的匹配算法,基于初始尺度变化的SIFT匹配算法在处理结构型图像时性能得到了很大的提升.