基于生物视觉特征和视觉心理学的视频显著性检测算法

提出了一种空域和时域相结合的视频显著性检测算法.对单帧图像,受视觉皮层层次化感知特性和Gestalt视觉心理学的启发,提出了一种层次化的静态显著图检测方法.在底层,通过符合生物视觉特性的特征图像（双对立颜色特征及亮度特征图像）的非线性简化模型来合成特征图像,形成多个候选显著区域;在中层,根据矩阵的最小Frobenius-范数（F-范数）性质选取竞争力最强的候选显著区域作为局部显著区域;在高层,利用Gestalt视觉心理学的核心理论,对在中层得到的局部显著区域进行整合,得到具有整体感知的空域显著图.对序列帧图像,基于运动目标在位置、运动幅度和运动方向一致性的假设,对Lucas-Kanade算法检测出的光流点进行二分类,排除噪声点的干扰,并利用光流点的运动幅度来衡量运动目标运动显著性.最后,基于人类视觉对动态信息与静态信息敏感度的差异提出了一种空域和时域显著图融合的通用模型.实验结果表明,该方法能够抑制视频背景中的噪声并且解决了运动目标稀疏等问题,能够较好地从复杂场景中检测出视频中的显著区域.     

A medical image fusion method based on energy classification of BEMD components

A medical image fusion method based on bi-dimensional empirical mode decomposition (BEMD) and dual-channel PCNN is proposed in this paper. The multi-modality medical images are decomposed into intrinsic mode function (IMF) components and a residue component. IMF components are divided into high-frequency and low-frequency components based on the component energy. Fusion coefficients are achieved by the following fusion rule: high frequency components and the residue component are superimposed to get more textures; low frequency components contain more details of the source image which are input into dual-channel PCNN to select fusion coefficients, the fused medical image is achieved by inverse transformation of BEMD. BEMD is a self-adaptive tool for analyzing nonlinear and non-stationary data; it doesn’t need to predefine filter or basis function. Dual-channel PCNN reduces the computational complexity and has a good ability in selecting fusion coefficients. A combined application of BEMD and dual-channel PCNN can extract the details of the image information more effectively. The experimental result shows the proposed algorithm gets better fusion result and has more advantages comparing with traditional fusion algorithms.     

基于音视频信息融合的目标检测与跟踪算法

针对单一视觉跟踪算法易受遮挡影响的缺陷,提出一种基于音视频信息融合的目标检测与跟踪算法。整个算法框架包括视频检测与跟踪、声源定位、音视频信息融合跟踪3个模块。视频检测与跟踪模块采用YOLOv5m算法作为视觉检测的框架,使用无迹卡尔曼滤波和匈牙利算法实现多目标的跟踪与匹配;声源定位模块采用十字型麦克风阵列获取音频信息,结合各麦克风接收信号的时延计算声源方位;音视频信息融合跟踪模块构建音视频似然函数和音视频重要性采样函数,采用重要性粒子滤波作为音视频融合跟踪的算法,实现对目标的跟踪。在室内复杂环境下对算法性能进行测试,结果表明该算法跟踪准确率达到90.68%,相较于单一模态算法具有更好的性能。     

News Video Summarization Combining SURF and Color Histogram Features

Because the data volume of news videos is increasing exponentially, a way to quickly browse a sketch of the video is important in various applications, such as news media, archives and publicity. This paper proposes a news video summarization method based on SURF features and an improved clustering algorithm, to overcome the defects in existing algorithms that fail to account for changes in shot complexity. Firstly, we extracted SURF features from the video sequences and matched the features between adjacent frames, and then detected the abrupt and gradual boundaries of the shot by calculating similarity scores between adjacent frames with the help of double thresholds. Secondly, we used an improved clustering algorithm to cluster the color histogram of the video frames within the shot, which merged the smaller clusters and then selected the frame closest to the cluster center as the key frame. The experimental results on both the public and self-built datasets show the superiority of our method over the alternatives in terms of accuracy and speed. Additionally, the extracted key frames demonstrate low redundancy and can credibly represent a sketch of news videos.     

基于区域霍夫变换的机弹分离初始信息提取研究

为提供机弹分离数值仿真的精准初始物理信息，提出了基于区域霍夫变换的导弹物理信息提取方案，利用高速摄像机提取视频帧中导弹弹身与弹尾处的标识圆圆心位置，并通过坐标变换计算出导弹的物理信息。在帧处理上，利用帧前后时间间隔短、速度矢量变化可以忽略的特点，提出基于区域霍夫变换的圆搜索算法。该算法针对实际中运动矢量在搜索窗口具有偏置而非均匀分布的特性，对搜索框进行固定移位，预测区域与实际区域的中心误差不超过5%，且耗时较传统算法减少41.6%以上，计算结果与实际结果平均误差不超过2.39%，同时还具有复杂背景的适应能力，满足方案时效性与精度需求。     

基于可能性信息质量合成的双模态红外图像融合算法选取

针对双模态红外图像在融合时异类差异特征两两合成出现信息冗余导致所选择的融合算法相互冲突,造成融合效果差甚至失效的问题,提出了一种基于可能性信息质量合成的双模态红外图像融合算法选取方法。首先计算双模态红外图像多融合算法下不同差异特征的融合有效度,利用可能性框架得到对应的可能性分布向量子集;其次计算向量子集的信息量和可信度,并对多个向量子集进行加权合成;然后构建基于信息质量的排序函数,得到每种融合算法下的非支配子集;最后构建多融合算法得分函数的联合分布对多种融合算法优化选择。实验结果表明,将基于质量来整合多个差异特征的方法运用于双模态红外图像融合算法选取中,所选出的融合算法在加权综合指标上高于其他算法均值55%以上,证明了本文方法的有效性和合理性;由多组实验算得本文方法平均耗时10.083 s,在时间效率上也符合实时图像融合应用的工程需求。     

Multi-sensor image fusion using discrete wavelet frame transform

An algorithm is presented for multi-sensor image fusion using discrete wavelet frame transform (DWFT).The source images to be fused are firstly decomposed by DWFT. The fusion process is the combining of the source coefficients. Before the image fusion process, image segmentation is performed on each source image in order to obtain the region representation of each source image. For each source image, the salience of each region in its region representation is calculated. By overlapping all these region representations of all the source images, we produce a shared region representation to label all the input images. The fusion process is guided by these region representations. Region match measure of the source images is calculated for each region in the shared region representation. When fusing the similar regions, weighted averaging mode is performed; otherwise selection mode is performed. Experimental results using real data show that the proposed algorithm outperforms the traditional pyramid transform based or discrete wavelet transform (DWT) based algorithms in multi-sensor image fusion.     

潜在低秩表示框架下基于卷积神经网络结合引导滤波的红外与可见光图像融合

为提高融合图像的可视性,解决传统红外与可见光图像融合算法中存在的边缘特征缺失、细节模糊的问题,提出了一种潜在低秩表示框架下基于卷积神经网络结合引导滤波的红外与可见光图像融合算法。该算法首先利用潜在低秩表示对源图像进行分解,得到源图像的低秩分量和显著分量。其次,利用卷积神经网络根据源图像的特征信息,得到权值图。再次,通过引导滤波算法对权值图进行边缘锐化,然后再将优化后的权值图分别与源图像的低秩分量和显著分量融合,得到融合图像的低秩分量和显著分量。最后,将融合图像的低秩分量和显著分量叠加,得到最终的融合图像。实验结果表明,该算法在主观评价和客观指标上均优于传统的红外与可见光图像融合算法。     

Moving object detection based on improved VIBE and graph cut optimization

Extracting foreground moving objects from video sequences is an important task and also a hot topic in computer vision and image processing. Segmentation results can be used in many object-based video applications such as object-based video coding, content-based video retrieval, intelligent video surveillance and video-based human–computer interaction. In this paper, we present a novel moving object detection method based on improved VIBE and graph cut method from monocular video sequences. Firstly, perform moving object detection for the current frame based on improved VIBE method to extract the background and foreground information; then obtain the clusters of foreground and background respectively using mean shift clustering on the background and foreground information; Third, initialize the S/T Network with corresponding image pixels as nodes (except S/T node); calculate the data and smoothness term of graph; finally, use max flow/minimum cut to segmentation S/T network to extract the motion objects. Experimental results on indoor and outdoor videos demonstrate the efficiency of our proposed method.     

Video Summarization Based on Mutual Information and Entropy Sliding Window Method

This paper proposes a video summarization algorithm called the Mutual Information and Entropy based adaptive Sliding Window (MIESW) method, which is specifically for the static summary of gesture videos. Considering that gesture videos usually have uncertain transition postures and unclear movement boundaries or inexplicable frames, we propose a three-step method where the first step involves browsing a video, the second step applies the MIESW method to select candidate key frames, and the third step removes most redundant key frames. In detail, the first step is to convert the video into a sequence of frames and adjust the size of the frames. In the second step, a key frame extraction algorithm named MIESW is executed. The inter-frame mutual information value is used as a metric to adaptively adjust the size of the sliding window to group similar content of the video. Then, based on the entropy value of the frame and the average mutual information value of the frame group, the threshold method is applied to optimize the grouping, and the key frames are extracted. In the third step, speeded up robust features (SURF) analysis is performed to eliminate redundant frames in these candidate key frames. The calculation of Precision, Recall, and Fmeasure are optimized from the perspective of practicality and feasibility. Experiments demonstrate that key frames extracted using our method provide high-quality video summaries and basically cover the main content of the gesture video.     

利用反向投影的flash场景自适应视频编码算法

针对闪光造成的光照变化会导致视频帧之间巨大的强度差异问题,提出利用反向投影的flash场景自适应视频编码算法。根据直方图差异提取闪光和非闪光帧,相应地为每个帧分配适当的编码类型,并在加权预测(WP)参数集确定中采用运动向量导数,通过反向投影保证flash场景的全局一致性。实验结果表明,相比其他几种加权预测算法,本文算法在峰值信噪比(PSNR)及计算复杂度方面获得了更加优越的编码性能。     

Filtering of multichannel video sequences distorted by noise, based on the fuzzy-set theory

A new approach based on the fuzzy-set theory is proposed, which allows detection of object motion from frame to frame of video sequences and filtering of images distorted by additive noise. New fuzzy rules and membership functions introduced in the study make it possible to perform efficient filtering of video sequences using the correlation between channels and in video sequence neighboring frames. Simulation of the proposed and known algorithms of video sequence processing shows the advantage of the new filtering scheme in terms of conventional PSNR, MAE, NCD, SSIM criteria and visual comparison of filtered video sequences.     

基于大数据分析的移动对象轨迹预测方法

对移动对象的轨迹预测将在移动目标跟踪识别中具有较好的应用价值。移动对象轨迹预测的基础是移动目标运动参量的采集和估计,移动目标的运动参量信息特征规模较大,传统的单分量时间序列分析方法难以实现准确的参量估计和轨迹预测。提出一种基于大数据多传感信息融合跟踪的移动对象轨迹预测算法。首先进行移动目标对象进行轨迹跟踪的控制对象描述和约束参量分析,对轨迹预测的大规模运动参量信息进行信息融合和自正整定性控制,通过大数据分析方法实现对移动对象运动参量的准确估计和检测,由此指导移动对象轨迹的准确预测,提高预测精度。仿真结果表明,采用该算法进行移动对象的运动参量估计和轨迹预测的精度较高,自适应性能较强,稳健性较好,相关的指标性能优于传统方法。     

Removal of rain in video based on motion and shape characteristics of raindrops

Rain degrades the performance of the outdoor computer vision system, which adversely impacts tracking, object detection and recognition. To solve this problem, a novel rain detection and removal method is proposed in this paper. Firstly, raindrops are considered as the motion component and extracted by FastICA algorithm. Secondly, raindrops are identified based on its properties of area, direction and width. Finally, a two-step removal method is proposed to replace pixels covered by rain with estimated background pixels. Experimental results show that the algorithm can effectively remove the raindrops in various videos. Compared with traditional rain removal methods, it does not require any parameters about rain, camera and scenes, and only needs three successive frames to recover a raindrops free video.     

一种基于特征跟踪的视频序列稳像算法

提出一种基于特征跟踪的视频序列稳像算法.该算法从视频序列的参考帧中提取出一组角点特征,然后在后续帧中基于模糊Kalman滤波进行特征窗跟踪,通过比较各帧图像中特征窗间的对应关系计算出补偿摄像机运动所必需的参数,使用这些参数将后续帧向参考帧对准,从而得到稳定的视频序列.实验结果表明该算法稳像效果好,运算复杂度低,且具有较强的鲁棒性.     

多尺度奇异值分解的偏振图像融合去雾算法与实验

针对现有偏振去雾算法鲁棒性不强和图像增强效果有限的问题,提出一种基于多尺度奇异值分解的图像融合去雾算法.首先,利用偏振测量信息的冗余特性,采用最小二乘法,提高了传统偏振图像去雾算法中偏振信息的准确度;然后,从传统偏振图像去雾算法的局限性出发,定性分析了偏振图像融合去雾的可行性,并提出了一种基于多尺度奇异值分解的偏振图像...     

Infrared dim and small target detecting and tracking method inspired by Human Visual System

Detecting and tracking dim and small target in infrared images and videos is one of the most important techniques in many computer vision applications, such as video surveillance and infrared imaging precise guidance. Recently, more and more algorithms based on Human Visual System (HVS) have been proposed to detect and track the infrared dim and small target. In general, HVS concerns at least three mechanisms including contrast mechanism, visual attention and eye movement. However, most of the existing algorithms simulate only a single one of the HVS mechanisms, resulting in many drawbacks of these algorithms. A novel method which combines the three mechanisms of HVS is proposed in this paper. First, a group of Difference of Gaussians (DOG) filters which simulate the contrast mechanism are used to filter the input image. Second, a visual attention, which is simulated by a Gaussian window, is added at a point near the target in order to further enhance the dim small target. This point is named as the attention point. Eventually, the Proportional-Integral-Derivative (PID) algorithm is first introduced to predict the attention point of the next frame of an image which simulates the eye movement of human being. Experimental results of infrared images with different types of backgrounds demonstrate the high efficiency and accuracy of the proposed method to detect and track the dim and small targets.     

视频文字信息检查工具的设计与实现

针对某些特定视频中,画面文字信息经常包含较为敏感文字信息,导致信息泄露,设计实现了一种视频画面中的文字的检测识别系统,对视频画面中的文字标语、文字条幅,新闻画面中的文字导语等信息进行识别与比对。采用基于双阈值的视频镜头分割算法,根据颜色直方图信息提取关键帧,采用最大稳定极值区域算法提取图像中稳定区域,通过聚类和级联分类器实现文字区域提取,最后将文字区域分割后进行OCR识别,实验表明,设计的系统针对复杂背景中的文字能够达到较高的检测识别率。     

实时运动结构重建在自主导航系统中的应用

实时运动结构重建是自主车辆、机器人导航、空间探测器自主降落、智能监控等领域中的重要研究课题。目前实时运动结构重建主要存在着特征匹配困难、鲁棒性差、系统无法自动获取初始参数和需要大量人工干预等诸多问题。利用高速CMOS摄像机与惯性传感数据融合提高了运动结构重建算法的精度及其鲁棒性。该算法在扩展卡尔曼滤波框架下是通过融合惯性与视觉传感器的数据来进行运动估计的。对场景中的每一个待估计结构的特征点建立对应的卡尔曼滤波器,以估计其空间三维结构信息。运动估计模块与结构估计模块交替运行,减小了系统运算的复杂度,提高了实时性能。通过对真实场景图像序列的实验验证结果表明,惯性传感器的额外信息能够有效地提高运动结构估计的精度,能够增强算法的鲁棒性。     

基于Shearlet变换的自适应图像融合算法

针对多聚焦图像与多光谱和全色图像的成像特点,结合Shearlet变换具有较好的稀疏表示图像特征的性质,提出了一种新的图像融合规则.并基于此融合规则,提出了基于Shearlet变换的自适应图像融合算法.在多聚焦图像的融合算法中,分别对聚焦不同的图像进行Shearlet变换,并基于本文提出的融合规则,对分解后的高低频系数进行融合处理. 通过与多种算法的比较实验证明了本文提出的算法融合的图像具有更高的清晰度和更加丰富的细节信息.在多光谱和全色图像的融合处理中,提出了一种基于Shearlet变换与HSV变换相结合的图像融合方法.该算法首先对多光谱图像作HSV变换,将得到的V分量与全色图像进行Shearlet分解与融合,在融合过程中对分解系数选用特定的融合准则进行融合,最后将融合生成新的分量与H、S分量进行HSV逆变换产生新的RGB融合图像. 该算法在空间分辨率和光谱特性两方面达到了良好的平衡,融合后的图像在减少光谱失真的同时,有效增强了空间分辨率. 仿真实验证明,本文算法融合的图像与传统的多光谱和全色图像融合算法相比,具有更佳的融合性能和视觉效果.