基于图像的对称性识别算法研究

在视觉特征描述中,对称性是一个重要特征,对称性检测是图像特征检测的重要内容,其广泛应用于图像处理、计算机视觉、信息传输等领域,且拥有广阔的应用前景。文中给出了对称性分类及几种相关典型的算法,重点介绍了基于相位信息和基于SIFT算子的检测图像对称性的方法。并采用这两种方法通过选取不同类别的图像进行实验,对实验结果图进行对比分析,提出了一种基于封闭轮廓角点信息的对称性检测方法,并对其进行了初步探索。     

基于多特征融合技术的商品图像分类

针对网上商品图像的特点,提出了一种多特征融合的分类方法。本文针对颜色和商品图案风格两方面对图像进行分类。首先对商品图像进行分割,再提取特征,颜色特征选择提取颜色直方图特征和颜色矩特征;提取PHOG和SIFT特征来描述图案风格。然后采用基于决策的加权融合方法将两种特征结合起来进行分类,最后在数据集上进行实验,与仅用单一特征分类和使用普通多特征拼接方法作比较,使用本文融合特征的方法进行分类准确率较高,并且其准确率有8%~10%的提升。实验结果表明本文提出的方法是一种有效的商品图像分类方法。     

基于多特征扩展pLSA模型的场景图像分类

场景图像分类近年来受到人们的广泛关注,而基于统计模型的方法更是场景分类中的研究热点。我们提出了一种新的基于多特征融合和扩展pLSA模型的场景图像分类框架。对每幅图像首先用多尺度规则分割确定局部基元,然后提取每个局部基元的多分辨率直方图矩特征和SIFT特征,最后用扩展的概率生成模型对图像集进行建模,测试。我们的方法不仅能够很好的表示图像的语义特性而且在模型的训练阶段是无监督的。我们针对目前常用的3个数据库,做了三组对比实验,均取得了比以前的方法更好的识别结果。      

基于SIFT/ORB几何约束的红外与可见光图像特征点匹配

红外图像与可见光图像记录着地物的不同属性信息,两者融合能够优势互补,弥补单一数据源信息的不足。然而由于两者成像原理不同,热红外传感器与可见光传感器对同一场景获取的图像灰度差异较大,二者图像误匹配多,融合难度大。本文在分析红外与可见光图像共有特征的基础上,提出了一种基于SIFT与ORB特征检测的匹配方法,利用SIFT算子与ORB算子同时进行特征点检测,先基于RANSAC对SIFT匹配得到的同名点进行筛选,同时结合最近邻比次近邻算法获取ORB匹配点,再利用SIFT匹配点对ORB匹配点进行距离和角度的几何约束进一步剔除误匹配,最终得到特征点分布均匀、可靠度更高的匹配结果,解决因灰度差异较大产生的匹配效果不佳的问题。利用4组红外与可见光图像进行实验,结果表明,本文算法特征点正确匹配数量相较于SIFT分别提高了约3.7倍、3.2倍、3.6倍、3倍,大幅地提高了红外与可见光图像的匹配数量,为两者间的匹配提供了一种有效的方法。     

$n$ -SIFT: $n$ -Dimensional Scale Invariant Feature Transform

We propose the $n$ -dimensional scale invariant feature transform ( $n$-SIFT) method for extracting and matching salient features from scalar images of arbitrary dimensionality, and compare this method's performance to other related features. The proposed features extend the concepts used for 2-D scalar images in the computer vision SIFT technique for extracting and matching distinctive scale invariant features. We apply the features to images of arbitrary dimensionality through the use of hyperspherical coordinates for gradients and multidimensional histograms to create the feature vectors. We analyze the performance of a fully automated multimodal medical image matching technique based on these features, and successfully apply the technique to determine accurate feature point correspondence between pairs of 3-D MRI images and dynamic $3{rm D} + {rm time}$ CT data.      

基于SIFT算法的异源遥感影像自动匹配研究

由于不同传感器、多时相、多分辨率、多波段的遥感图像的光谱特征、空间特征、纹理特征等存在较大差异,为遥感图像的匹配带来了困难。主要利用图像特征点提取方法,使用具有尺度不变特性的SIFT(Scale Invariance Feature Transform)方法,对异源遥感图像进行配准和图像进行拼接操作,并进一步对SIFT算法进行优化,采用双向匹配策略。实验证明该算法具有稳定、可靠、快速等特点,适用于存在光谱特征、空间特征、纹理特征差异的异源遥感图像的精确配准,同时实验验证了双向匹配算法用于SIFT特征点匹配中的优越性,证明其为一种好的匹配测度。     

一种基于多级空间视觉词典集体的图像分类方法

针对单一特征时存在提取的信息量不足,对图像内容描述比较片面,提出将传统的SIFT特征与KDES-G特征进行串行融合,生成一个联合向量作为新的特征向量.针对传统的视觉词典构造方法缺乏考虑视觉词汇在空间的分布特点,本文引入图像空间信息,提出了一种空间视觉词典的构造方法,先对图像进行空间金字塔划分,再把空间各子区域内的特征分别聚类,构建属于对应子空间区域的空间视觉词典.在图像表示阶段,图像各子区域内的特征基于其对应的空间视觉词典进行LLC稀疏编码,根据各子区域对图像贡献程度的不同,把编码后各子区域的特征向量赋予不同的权重加权处理,再连接形成最终的图像描述.最后,利用线性SVM进行图像分类,实验结果表明了本文方法的有效性和鲁棒性.     

多光谱遥感卫星图像的精确配准方法研究

针对一组多光谱遥感图像中,各谱段图像之间配准不一致的问题,本文提出了一种基于特征点的快速自动配准方法。在图像信息熵的基础上,利用环形移动窗口,自动快速寻找感兴趣区域, 并利用尺度不变特征转换（SIFT）算法提取特征。为提高精度,文中对特征初匹配方法作了改进,并用余弦定理和空间距离约束条件剔除误匹配点,之后提取最稳定的特征点对计算变换参数,完成配准。最后根据配准前后图像的互信息和特征点的均方根误差（RMSE）来衡量配准的程度。通过对大量中巴地球资源卫星拍摄的多光谱图像进行实验,该方法能达到亚像素级配准精度,并能快速对各谱段图像进行配准。      

改进图像配准算法在无人机遥感图像拼接中的应用

针对传统图像拼接方法的不足, 提出一种基于改进SIFT算法的图像拼接方法, 并将其应用于无人机遥感图像拼接算法中。首先, 采用Harris算子角点检测遥感图像的特征点, 然后用改进的SIFT算法进行特征点的描述, 通过对高维数据进行降维处理, 减小运算量; 匹配完成后, 采用随机抽样一致性(RANSAC)算法消除误匹配; 最后采用渐入渐出加权平均融合法进行图像融合。实验结果表明: 采用所提出算法能有效剔除遥感图像之间的误匹配, 减小时间复杂度, 更好地消除拼接缝隙。     

基于SIFT的多焦距图像特征点提取算法

将尺度不变特征变换（SIFT）算法应用到图像的特征点提取与匹配中,SIFT算法可在尺度空间寻找极值点,提取对图像焦距变化具有稳定性的特征点及其特征描述符。在采用SIFT算法提取图像的特征点及其特征描述符后,提出了一种特征点精匹配算法进行特征点的匹配,并通过仿真证明该算法具有很好的效果。     

改进图像配准算法在无人机遥感图像拼接中的应用

针对传统图像拼接方法的不足,提出一种基于改进SIFT算法的图像拼接方法,并将其应用于无人机遥感图像拼接算法中。首先,采用Harris算子角点检测遥感图像的特征点,然后用改进的SIFT算法进行特征点的描述,通过对高维数据进行降维处理,减小运算量;匹配完成后,采用随机抽样一致性(RANSAC)算法消除误匹配;最后采用渐入渐出加权平均融合法进行图像融合。实验结果表明:采用所提出算法能有效剔除遥感图像之间的误匹配,减小时间复杂度,更好地消除拼接缝隙。     

Fusion of block and keypoints based approaches for effective copy-move image forgery detection

Keypoint-based and block-based methods are two main categories of techniques for detecting copy-move forged images, one of the most common digital image forgery schemes. In general, block-based methods suffer from high computational cost due to the large number of image blocks used and fail to handle geometric transformations. On the contrary, keypoint-based approaches can overcome these two drawbacks yet are found difficult to deal with smooth regions. As a result, fusion of these two approaches is proposed for effective copy-move forgery detection. First, our scheme adaptively determines an appropriate initial size of regions to segment the image into non-overlapped regions. Feature points are extracted as keypoints using the scale invariant feature transform (SIFT) from the image. The ratio between the number of keypoints and the total number of pixels in that region is used to classify the region into smooth or non-smooth (keypoints) regions. Accordingly, block based approach using Zernike moments and keypoint based approach using SIFT along with filtering and post-processing are respectively applied to these two kinds of regions for effective forgery detection. Experimental results show that the proposed fusion scheme outperforms the keypoint-based method in reliability of detection and the block-based method in efficiency.     

Analysis of BICM-ID Receivers Exploiting Transformations of Extrinsic Information

One of the key problems in the field of Computer Vision is recovering the geometry from multiple views of the same scene. Once the homography of two images is known, the motion of a stereo camera system can be determined, images can be rectified or image registration can be performed. A feature-based approach to determine the homography between two images bases on the extraction and matching of SIFT features (SIFT, Scale-Invariant Feature Transform). By extracting image features from varying images of one scene and finding corresponding image features in both images, the homography of the scene can be determined. The extraction of image features, which provide sufficient quality for computation of the homography of a scene, leads to an algorithm complexity, that prevents real-time applications on conventional CPUs. Therefore, we present and discuss an application-specific instruction-set extensions for a Tensilica Xtensa LX5 ASIP to accelerate a SIFT feature extraction (ASIP, Application-Specific Instruction-set Processor). In total, the complete SIFT feature extraction, executed on an extended processor is accelerated by a factor of x125 compared to the baseline processor. At the same time, the accuracy of the SIFT features is preserved. In addition, the proposed processor extensions maintain the full flexibility of an ASIP for a fast integration of further feature extractors.     

基于多目标优化的SIFT特征匹配算法

SIFT算子在实际应用中,由于地面图像本身特征不明显且提取出的特征点多、乱以及灰度变化不明显等特点的影响,从而导致特征点误匹配。为此提出一种改进的SIFT图像特征匹配算法。该算法是在SIFT特征匹配的基础上,利用多目标优化算法,建立相关匹配模板,利用给定同一场景的两幅图像,寻找同一场景点投影到图像中的模板之间的相关性建立数学模型即目标函数,根据同一幅图像中模板间的距离建立边界约束条件,从而剔除一些误匹配点。实验表明,该算法可以有效地提高图像匹配精度。     

Iterative colour correction of multicamera systems using corresponding feature points

Multiview images captured by multicamera systems are generally not uniform in colour domain. In this paper, we propose a novel colour correction method of multicamera systems that can (i) be applied to not only dense multicamera system, but also sparse multicamera configuration and (ii) obtain an average colour pattern among all cameras. Our proposed colour correction method starts from any camera on the array sequentially, following a certain path, for pairs of cameras, until it reaches the starting point and triggers several iterations. The iteration stops when the correction applied to the images becomes small enough. We propose to calculate the colour correction transformation based on energy minimisation using a dynamic programming of a nonlinearly weighted Gaussian-based kernel density function of geometrically corresponding feature points, obtained by the modified scale invariant feature transformation (SIFT) method, from several time instances and their Gaussian-filtered images. This approach guarantees the convergence of the iteration procedure without any visible colour distortion. The colour correction is done for each colour channel independently. The process is entirely automatic, after estimation of the parameters through the algorithm. Experimental results show that the proposed iteration-based algorithm can colour-correct the dense/sparse multicamera system. The correction is always converged with average colour intensity among viewpoint, and out-performs the conventional method.     

基于粗糙集近似约简和SIFT特征的图像匹配算法

SIFT(Scale Invariant Feature Transform)描述符由于具有尺度、旋转和光照不变等特性在图像匹配领域获得了广泛的应用。但是,SIFT特征点采用128维特征向量表示,当图像特征点较多时,匹配算法所需的存储空间大、匹配时间长,且匹配精度不理想。针对以上问题,本文给出了一种基于Rough-SIFT描述符的图像匹配算法。首先,利用排序法求出图像的稳健特征点,然后为提高后续匹配处理运算效率,将粗糙集约简理论引入到基于SIFT特征的匹配算法中,通过构建一种新的近似约简算法来对稳健特征点的128维特征向量进行降维处理,最后利用约简后的特征点对图像进行匹配。仿真实验表明, 本文方法使得约简后的SIFT特征点更加精确、稳定、可靠,有效减小了匹配算法的存储空间,提高了匹配算法的效率和准确率。      

一种基于改进ORB特征匹配的无人机视觉导航方法

为了解决在全球导航卫星系统(Global Navigation Satellite System)拒止情况下无人机导航能力缺失等问题,提出了一种基于改进快速提取旋转描述子(Oriented FAST and Rotated Brief, ORB)图像特征匹配的无人机视觉导航方法。首先,为了实现无人机的绝对定位,提出了一种特征图像基准数据库构建方法;其次,为提取图像数据集的特征点,采用了一种结合尺度不变特征变换(Scale Invariant Feature Transform, SIFT)的尺度空间优化ORB特征提取算法;最后,为了将图像特征与图像基准数据库快速匹配并提高其匹配精度,提出了一种改进ORB特征匹配算法——ORB+GMS+PROSAC算法。通过在ArcGIS中分割图像构建基准数据库并进行实验分析,结果表明,基于ORB+GMS+PROSAC特征匹配算法性能显著提升,其中匹配准确率上升5.05%,匹配时间减少41.61%,明显优于其他传统特征匹配算法。     

Singular value decomposition based fusion for super-resolution image reconstruction

In this paper, we address a super-resolution problem of generating a high-resolution image from low-resolution images. The proposed super-resolution method consists of three steps: image registration, singular value decomposition (SVD)-based image fusion and interpolation. The contribution of this work is two-fold. First we customize an image registration approach using Scale Invariant Feature Transform (SIFT), Belief Propagation and Random Sampling Consensus (RANSAC) for super-resolution. Second, we propose SVD-based fusion to integrate the important features from the low-resolution images. The proposed image registration and fusion steps effectively maintain the important features and greatly improve the super-resolution results. Results, for a variety of image examples, show that the proposed method successfully generates high-resolution images from low-resolution images.