结合SIFT算法的视频场景突变检测

视频场景变化检测对于视频的标注以及语义检索具有非常重要的作用。本文提出了一种结合SIFT(Scale Invariant Feature Transformation)特征点提取的场景变化检测算法。首先利用SIFT算法分别提取出视频前后帧的特征点并分别统计其数量,然后对视频前后帧进行图像匹配,统计匹配上的特征点数量,最后将该帧的匹配特征点数量与该帧前一帧的特征点数量做比值,从而通过该比值判断场景变化情况。实验结果表明,视频场景突变检测率平均可以达到95.79%。本算法可以在视频帧进行图像匹配的过程中对场景的变化情况进行判断,因此该算法不仅应用范围较广,还可以保证场景变化检测的精度,仿真结果证明了算法的有效性。     

改进SIFT变换与客观评价结合的图像配准算法

针对SIFT(Scale Invariant Feature Transform)算子在大幅复杂图像中提取的过多不稳定特征点及在只有少量重合区域下图像配准过程中出现的过多误匹配,导致图像配准精度下降。提出一种改进的SIFT算法,在对目标图像提取SIFT特征后,利用双向BBF(Best-Bin-First)匹配算法对提取的特征点进行匹配,采用SIFT描述子的尺度以及梯度方向信息建立最小邻域匹配剔除误匹配点,通过随机抽取一致性算法(RANSAC)进一步筛选匹配点,并利用最小二乘法结合多项式近似拟合出变换模型,利用局部均方根有效值(RMS)评价映射矩阵与实际图像的误差,找出并删除引起误差的误匹配点,迭代至配准图像符合评价标准后,计算出精确变换模型。实验结果表明,该算法提高了大幅复杂图像在少量重合区域时的配准精度。     

A feature-based robust digital image watermarking scheme using bandelet transform

Desynchronization attacks are among the most difficult attacks to resist, for it can desynchronize the location of the watermark and hence cause incorrect watermark detection. The design of an image watermarking scheme that is robust against desynchronization attacks is challenging. Based on a multi-scale SIFT (scale invariant feature transform) detector and Bandelet transform theory, we propose a new content based image watermarking algorithm with good visual quality and reasonable resistance toward desynchronization attacks. Firstly, the stable image feature points are extracted from the original host by using the multi-scale SIFT detector, and the local feature regions (LFRs) are constructed adaptively according to the feature scale theory. The Bandelet transform is then performed on the LFRs. Finally, the digital watermark is embedded into the LFRs by modifying the significant Bandelet coefficients. By binding the watermark with the geometrically invariant image features, the watermark detection can be done without synchronization error. Experimental results show that the proposed image watermarking is not only invisible and robust against common signal processing such as sharpening, noise adding, JPEG compression, etc., but also robust against the desynchronization attacks such as rotation, translation, scaling, row or column removal, cropping, etc.     

基于多尺度协同的人头检测方法

针对HOG特征本身不具有尺度不变性,在实际应用中仅能检测出与样本图片大小相差不大的目标对象这一弊端,提出多尺度窗口融合的头部检测的方法;利用线性支持向量机在分类决策方面的优势,与提取的HOG特征结合作分类器的离线训练;在实时的目标检测阶段,采用高斯金字塔式缩放对输入的视频序列作多尺度处理,得到对应的不同分辨率下的待检测帧,在不同的尺度空间作人头的扫描检测并存储结果;之后融合各尺度的检测结果并在相应位置决策标定;实验对某监控视频作检测分析,结果表明,该方法在检出率、召回率、准确度等方面均有较大提升。     

一种基于特征不变描述的图像无缝拼接算法

针对尺度变化不明显场景的拼接,提出一种基于特征不变描述的图像无缝拼接算法.利用Harris特征检测算子进行特征点提取,在此基础上对提取出的特征点采用SIFT描述子进行处理,使其具备旋转不变性,然后利用k-d树算法进行遍历搜索定位,并用RANSAC算法进行图像间单应性变换矩阵计算,最后利用加权平均的融合方法进行图像无缝平滑,得到无缝拼接图像.实验结果表明:该算法在图像任意角度旋转的情况下,能有效地实现图像无缝拼接,并具有较强的鲁棒性.     

Region-based SIFT approach to iris recognition

Traditional iris recognition systems transfer iris images to polar (or log-polar) coordinates and have performed very well on data that tends to have a centered gaze. The patterns of an iris are part of a 3-D structure that is captured as a two-dimensional (2-D) image and cooperative iris recognition systems are capable of correctly matching these 2-D representations of iris features. However, when the gaze of an eye changes with respect to the camera lens, many times the size, shape, and detail of iris patterns will change as well and cannot be matched to enrolled images using traditional methods. Additionally, the transformation of off-angle eyes to polar coordinates becomes much more challenging and noncooperative iris algorithms will require a different approach. The direct application of the scale-invariant feature transform (SIFT) method would not work well for iris recognition because it does not take advantage of the characteristics of iris patterns. We propose the region-based SIFT approach to iris recognition. This new method does not require polar transformation, affine transformation or highly accurate segmentation to perform iris recognition and is scale invariant. This method was tested on the iris challenge evaluation (ICE), WVU and IUPUI noncooperative databases and results show that the method is capable of cooperative and noncooperative iris recognition.     

一种新的红外成像末制导目标跟踪方法

为了稳定跟踪导弹末制导阶段的红外目标,提出了一种基于尺度不变特征变换的红外目标跟踪算法.尺度不变性特征变换所提取的图像纹理特征具有尺度和旋转不变性,跟踪算法分别提取目标模板和待跟踪图像的尺度不变特征变换特征.根据最小欧氏距离准则提取目标模板与待跟踪图像间相匹配的尺度不变特征变换特征点对,利用该特征点对拟合反映两图像间映射关系的仿射模型,并据此估计目标中心位置及调整目标模板尺寸.仿真结果表明,跟踪算法能够较好地实现在导弹末制导阶段对红外地面杂波背景下目标的稳定跟踪,其跟踪准确度和稳定度优于传统方法.
关键词:末制导跟踪|尺度不变性特征变换|特征匹配|仿射模型     

基于SIFT特征的彩色图像拼接方法研究

针对传统特征提取拼接算法在复杂图像中配准过程中出现的过多误匹配,导致拼接后图像出现鬼影、模糊等问题,从而影响拼接图像的质量,提出一种改进的SIFT配准算法。在对目标图像提取SIFT特征后,利用SIFT描述子的尺度以及梯度方向信息建立最小邻域匹配剔除误匹配点,之后利用局部均方根误差(RMSE)评价映射矩阵与RANSAC算法相结合,迭代出精确变换模型。在对图像进行几何矫正后,提出一种自适应的混合线性算法对重合区域图像变换至HIS颜色空间进行图像拼接,最后得到平滑无缝的完整彩色全景拼接图像。实验结果证明,该算法在拼接复杂场景并且重合区域不多时仍有较好的准确性及稳定性。     

Geometrically robust image watermarking using scale-invariant feature transform and Zernike moments

In order to resist geometric attacks, a robust image watermarking algorithm is proposed using scaleinvariant feature transform (SIFT) and Zernike moments. As SIFT features are invariant to rotation and scaling, we employ SIFT to extract feature points. Then circular patches are generated using the most robust points. An invariant watermark is generated from each circular patch based on Zernike moments.The watermark is embedded into multiple patches for resisting locally cropping attacks. Experimental results show that the proposed scheme is robust to both geometric attacks and signal processing attacks.     

结合Harris与SIFT算子的图像快速配准算法

本文提出了一种结合Harris与SIFT算子的快速图像配准方法。首先,对Harris算法进行两方面的改进:一是构建高斯尺度空间,提取具有尺度不变性的角点特征;二是采用Forsnter算子对提取的角点精定位,提高配准精度。然后,利用SIFT算子的特征描述方法描述提取到的特征点,通过随机kd树算法对两幅影像的特征点进行匹配。最后采用RANSAC算法对匹配点对进行提纯,并通过最小二乘法估计两幅影像间的空间变换单应矩阵,完成图像配准。实验结果表明:本文方法在基本保持配准精度的同时,在配准过程的时间消耗上比标准SIFT算法减少了64%。     

特征增强SSD算法及其在遥感目标检测中的应用

为了提高对复杂场景下多尺度遥感目标的检测精度,提出了基于多尺度单发射击检测(SSD)的特征增强目标检测算法.首先对SSD的金字塔特征层中的浅层网络设计浅层特征增强模块,以提高浅层网络对小目标物体的特征提取能力;然后设计深层特征融合模块,替换SSD金字塔特征层中的深层网络,提高深层网络的特征提取能力;最后将提取的图像特征与不同纵横比的候选框进行匹配以执行不同尺度遥感图像目标检测与定位.在光学遥感图像数据集上的实验结果表明,该算法能够适应不同背景下的遥感目标检测,有效地提高了复杂场景下的遥感目标的检测精度.此外,在拓展实验中,文中算法对图像中的模糊目标的检测效果也优于SSD.     

Adaptive optics retinal image registration from scale-invariant feature transform

With the use of adaptive optics(AO), high-resolution microscopic imaging of the living human retina in the single cell level has been achieved. In an AO retinal imaging system, with a small field size (about 1°, 300 μm) the motion of the eye severely affects the stabilization of the real-time video images results in significant distortions of the retina images. Scale-invariant feature transform (SIFT) algorithm is applied to automatically abstract corner points with subpixel resolution and match the points in two frames. With the matched corner points, we estimate and remove the motions of 20 frames of photoreceptor cells and capillary blood vessels, respectively. The maximal translational motion is about 30 and 44 pixels in the 20 frames whose size is 416 × 416 pixels. More general motions can be considered by the SIFT algorithm, but only simple translational motion can be considered by cross-correlation algorithm.     

集成SIFT算法与检测模型优化的UAV影像匹配方法

低空无人机(UAV)测量凭借着低成本、高效率、高精度的数据采集模式,可快速获取高空间分辨率的影像数据,已经成为遥感领域的一种重要技术手段.其中,影像匹配技术是UAV影像数据处理的重要步骤,图像间的匹配直接影响后期三维场景的精度及视觉效果.针对高原山地的高差起伏变化大地形复杂,植被覆被率高及地物分布不规则等问题存在,致使...     

An infrared thermal image processing framework based on superpixel algorithm to detect cracks on metal surface

Infrared thermography has been used increasingly as an effective non-destructive technique to detect cracks on metal surface. Due to many factors, infrared thermal image has low definition compared to visible image. The contrasts between cracks and sound areas in different thermal image frames of a specimen vary greatly with the recorded time. An accurate detection can only be obtained by glancing over the whole thermal video, which is a laborious work. Moreover, experience of the operator has a great important influence on the accuracy of detection result. In this paper, an infrared thermal image processing framework based on superpixel algorithm is proposed to accomplish crack detection automatically. Two popular superpixel algorithms are compared and one of them is selected to generate superpixels in this application. Combined features of superpixels were selected from both the raw gray level image and the high-pass filtered image. Fuzzy c-means clustering is used to cluster superpixels in order to segment infrared thermal image. Experimental results show that the proposed framework can recognize cracks on metal surface through infrared thermal image automatically.     

A non-rigid cardiac image registration method based on an optical flow model

According to non-rigid medical image registration, new method of classification registration is proposed. First, Feature points are extracted based on SIFT (Scale Invariant Feature Transform) from reference images and floating images to match feature points. And the coarse registration is performed using the least square method. Then the precise registration is achieved using the optical flow model algorithm. SIFT algorithm is based on local image features that are with good scale, rotation and illumination invariance. Optical flow algorithm does not extract features and use the image gray information directly, and its registration speed is faster. The both algorithms are complementary. SIFT algorithm is used for improving the convergence speed of optical flow algorithm, and optical flow algorithm makes the registration result more accurate. The experimental results prove that the algorithm can improve the accuracy of the non-rigid medical image registration and enhance the convergence speed. Therefore, the algorithm has some advantages in the image registration.     

Good match exploration for thermal infrared face recognition based on YWF-SIFT with multi-scale fusion

Stable local feature detection is a critical prerequisite in the problem of infrared (IR) face recognition. Recently, Scale Invariant Feature Transform (SIFT) is introduced for feature detection in an infrared face frame, which is achieved by applying a simple and effective averaging window with SIFT termed as Y-styled Window Filter (YWF). However, the thermal IR face frame has an intrinsic characteristic such as lack of feature points (keypoints); therefore, the performance of the YWF-SIFT method will be inevitably influenced when it was used for IR face recognition. In this paper, we propose a novel method combining multi-scale fusion with YWF-SIFT to explore more good feature matches. The multi-scale fusion is performed on a thermal IR frame and a corresponding auxiliary visual frame generated from an off-the-shelf low-cost visual camera. The fused image is more informative, and typically contains much more stable features. Besides, the use of YWF-SIFT method enables us to establish feature correspondences more accurately. Quantitative experimental results demonstrate that our algorithm is able to significantly improve the quantity of feature points by approximately 38%. As a result, the performance of YWF-SIFT with multi-scale fusion is enhanced about 12% in infrared human face recognition.     

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.     

Novel averaging window filter for SIFT in infrared face recognition

The extraction of stable local features directly affects the performance of infrared face recognition algorithms.Recent studies on the application of scale invariant feature transform(SIFT) to infrared face recognition show that star-styled window filter(SWF) can filter out errors incorrectly introduced by SIFT.The current letter proposes an improved filter pattern called Y-styled window filter(YWF) to further eliminate the wrong matches.Compared with SWF,YWF patterns are sparser and do not maintain rotation invariance;thus,they are more suitable to infrared face recognition.Our experimental results demonstrate that a YWF-based averaging window outperforms an SWF-based one in reducing wrong matches,therefore improving the reliability of infrared face recognition systems.     

Imaging a periodic moving/state-changed object with Hadamard-based computational ghost imaging

We propose a method for imaging a periodic moving/state-changed object based on computational ghost imaging with Hadamard speckle patterns and a slow bucket detector, named as PO-HCGI. In the scheme, speckle patterns are produced from a part of each row of a Hadamard matrix. Then, in each cycle, multiple speckle patterns are projected onto the periodic moving/state-changed object, and a bucket detector with a slow sampling rate records the total intensities reflected from the object as one measurement. With a series of measurements, the frames of the moving/state-changed object can be obtained directly by the second-order correlation function based on the Hadamard matrix and the corresponding bucket detector measurement results. The experimental and simulation results demonstrate the validity of the PO-HCGI. To the best of our knowledge, PO-HCGI is the first scheme that can image a fast periodic moving/state-changed object by computational ghost imaging with a slow bucket detector.