一种具有边界保持特性的图像修补方法

将边界优先级的图像修补方法引入GMRF修补模型中,通过对待修补区域边界像素点优先级的计算,确定像素的修补顺序,并和传统GMRF修补算法的修补结果做出比较。实验结果表明,基于边界优先级的图像修补算法对待修补区域包含边缘的图像有较好的修补效果,能够保持图像缺损区域的结构信息。     

基于图象方向性信息测度的图象象素分类

针对以往象象图处理方法缺乏通用性的问题,提出了一种新的基于图象方向性信息测度的图象素数分方法,即利用不同类型的图象象素的方向性信息测度,以及方向性信息测试随观观察尺度变化时的改变不同,来将图象象素分成边缘点,平滑点和纹理点以及用于图象分析,该方法不仅可用于任意图象的前期分析,而且具有良好的抗噪能力,实验证明,该方法具有令人满意的效果。     

一种粗糙不确定的图像分割方法

图像分割是计算机视觉领域的一个基础问题,涉及图像检索、物体检测、物体识别、行人跟踪等众多后续任务。目前已有大量研究成果,有基于阈值、聚类、区域生长的传统方法,也有基于神经网络的流行算法。由于图像区域边界的不确定性问题,现有算法并没有很好地解决图像部分区域渐变导致的边界模糊问题。粒计算是解决复杂问题的有效工具之一,在不确定的、模糊的问题上取得了良好的效果。针对现有图像分割算法在不确定性问题上的局限性,基于粒计算思想,提出了一种粗糙不确定性的图像分割方法。该算法在K均值算法的基础上,结合邻域粗糙集模型,先对类别边界区域的像素点进行粒化,运用邻域关系矩阵,得到各类别对各粒化像素点的包含度,从而对边界区域类别模糊的像素点进行重新划分,优化了图像分割的结果。在Matlab2019编程环境中,实验选取了BSDS500数据集中的一张马术训练图片和一张建筑物图片来测试算法性能。实验先对彩色图像进行灰度处理,用K均值算法对图像进行初步分割,再设置邻域因子值,依据边界像素点邻域信息重新划分边界点。对比K均值算法的分割结果可知,所提算法取得了更佳的效果。实验结果表明,该方法在粗糙度这一评价标准上优于K均值算法,可以有效降低图像区域边界的模糊性,实现灰度边界模糊的图像渐变区域的分割。     

改进的图像纹理检索方法在矿石识别中的应用

为了将图像纹理检索应用到某种矿石成分的识别中，在使用灰度共现矩阵和灰度行程统计矩阵进行特征提取的基础上，考虑到将纹理识别时的计算时间限制在一定范围内。提出了仅使用一个综合特征进行目标纹理检索的有效方法。实验结果表明，只要对按照欧氏距离定义的相似度进行匹配时，选取适当的相似度阈值，则该方法便能够获得满意的纹理识别结果。     

A fuzzy target recognition system with homomorphic invariant feature extraction

A target recognition system based on the concept of fuzzy set theory and homomorphic system is proposed. It involves automatic threshold selection, feature extraction, and classification. An optimal threshold is selected by the fuzzy risk criterion, i.e. to separate a given image into meaningful grey level classes under the assumption that the object and pixel grey level values are normally distributed. An edge measure for evaluating thresholding methods is also presented. When an image is segmented, a set of invariant features called mean, variance, skewness, and kurtosis, which are derived from the spectrum histogram of the target image, is calculated. The classification is then accomplished using the membership function of the feature space of an image and stored patterns. By simulation results, we find that the fuzzy risk thresholding achieves significant performance according to the uniformity and edge measures, and the fuzzy filter with the invariant features has good performance even in low signal-to-noise ratio conditions.     

A comparison of contextual classification methods using Landsat TM

The performance of contextual classification methods is evaluated using Landsat TM data. Classes of pixels adjacent to the pixel to be classified are assumed to be conditionally independent given the class of the pixel to be classified. An assumption of autocorrelated spectral reflectance is made in three of the methods. Methods that utilize information from one image and images from two different occasions are compared.Our results indicate that an autocorrelation method utilizing images from two different occasions performs optimally.     

基于小波分析和多项式细分定位的超分辨率图像重建算法

图像超分辨率重建是图像增强和图像复原研究中的一项重要课题,广泛应用于高清晰电视、医学成像和遥感成像等领域。在小波分析边缘检测的基础上,通过多项式细分算法定位亚像素边缘,将图像分为平滑区域、边缘区域和微细边缘区域。根据不同的区域特性,采用不同的插值方式进行超分辨率图像重建。仿真结果显示所提算法重建的高分辨率图像边界部分清晰自然,其主观判断和客观评价结果明显好于传统重建算法,从而验证了本算法的可行性和有效性。     

局部自适应的灰度图像彩色化

目的 现有的灰度图像彩色化方法为了保证彩色化结果在颜色空间上的一致性,往往采用全局优化的算法,使得图像边界区域易产生过渡平滑现象。为此提出一种局部自适应的灰度图像彩色化方法,在迁移过程中考虑局部邻域像素信息,同时自动调节邻域像素权重,在颜色正确迁移的同时保证清晰的边界信息。方法 首先结合SVM（support vector machine）和ISLIC（improved simple linear iterative clustering）算法获取彩色图像和灰度图像分类结果图;然后在分类基础上,确定灰度图像高置信度像素点,并根据图像纹理特征,在彩色图像中寻找灰度图像的像素匹配点;最后利用自适应权重均值滤波实现高置信度匹配像素点的颜色迁移,并利用迁移结果对低置信度像素点进行颜色扩散,以完成灰度图像彩色化。结果 实验结果显示,本文方法获得的彩色化迁移结果评分均高于3.5分,特别是局部放大区域评价结果均接近或高于4.0分,高于其他现有彩色化方法评价分数。表明本文方法不仅能够保证颜色迁移的准确性和颜色空间的一致性,同时也能获取颜色区分度高的边界细节信息。与现有的典型灰度图像彩色化方法相比,彩色化结果图在颜色迁移的正确性和抑制边界区域颜色的过渡平滑上都有更优的表现。结论 本文算法为灰度图像彩色化过程中抑制颜色越界问题提供了新的指导方法,能有效地应用于遥感、黑白图像/视频处理、医学图像着色等领域。     

Vectorial boundary-based sub-pixel mapping method for remote-sensing imagery

This article presents a vectorial boundary-based sub-pixel mapping (VBSPM) method to obtain the land-cover distribution with finer spatial resolution in mixed pixels. With inheritance from the geometric SPM (GSPM), VBSPM first geometrically partitions a mixed pixel using polygons, and then utilizes a vectorial boundary extraction model (VBEM), rather than the rasterization method in GSPM, to determine the location and length of each edge in the polygon, while these edges are located at the boundary of and within the interior of the mixed pixel. Furthermore, VBSPM uses a decay function to manage the mixed pixels along the image boundary region due to the missing parts of their neighbours. Finally, a ray-crossing algorithm is employed to determine the land-cover class of each sub-pixel in terms of vectorial boundaries. The experiments with artificial and remotely sensed images have demonstrated that VBSPM can reduce the inconsistency between the boundaries of different land-cover classes, approximately calculating errors with an odd zoom factor, and achieve more accurate sub-pixel mapping results than the hard classification methods and GSPM.     

基于空间聚类和边缘梯度的图像分割算法

提出一种结合空间聚类和边缘梯度信息的图像自动分割算法。在判断超像素颜色及纹理相似性的同时,进一步给出更加精确的分段边缘梯度计算方法,并采用测地距离来刻画超像素之间的相似性,使得分割结果更好地融合边缘不连续性与区域相似性。大量图像分割实验结果表明,该方法能更准确地找出分割边界,提高图像分割的准确性。     

保留边缘信息的静态数字图象超分辨率重建

插补方法是图象超分辨率 (super- resolution)技术的一个重要方面 .插补处理后 ,数字图象的视觉质量主要依赖于图象轮廓边缘的两个因素 :(1)跨越边缘方向的灰度值变化比较尖锐 ;(2 )沿边缘方向的灰度值变化比较平滑 .着重考虑以上两个因素 ,提出了一种新的插补方法 ,以获得清晰的图象边缘 ,从而获得较好的数字图象放大效果     

一种新的图像椒盐噪声的非线性滤波算法

为了在滤除椒盐噪声的同时能很好地保持图像的边缘细节,提出了一种新颖的图像椒盐噪声非线性滤波算法。利用局部统计信息,先将图像像素点分为信号点和可能的噪声点两类。然后将可能的噪声点进一步细分为边缘点、噪声点和信号点：利用方向信息、均方差来判断是否为边缘点,利用自适应阈值的方法来判断是否为噪声点,并且对边缘点和噪声点采取不同的方法进行滤波。经过仿真实验并与其它滤波算法进行比较表明,文中的算法具有更好的效果。     

基于TOF深度图像修复的输送带煤流检测方法

传统的带式输送机煤流检测装置中,核子胶带秤存在一定安全和环保隐患,电子胶带秤检测精度易受输送带张力、刚度等因素的影响;而基于超声波、线激光条纹、双目视觉等技术的非接触式检测方法存在实时性差、测量误差较大等问题。提出了一种基于飞行时间(TOF)深度图像修复的输送带煤流检测方法。通过TOF相机获取输送带运煤图像;对TOF图像进行均衡化处理,采用帧差法和边界跟随算法去除背景噪声,获得感兴趣的煤料区域;针对TOF深度图像因边缘处存在飞行像素噪声与多径误差噪声而导致的边缘信息不准确问题,提出强度图像引导的深度图像修复算法,通过Canny边缘检测算法寻找深度图像和强度图像的相似边缘,基于强度图像的有效边缘信息对深度图像边缘处的不可靠数据进行校正,并进一步基于Navier-Stokes方程和中值滤波器得到高精度深度图像;对煤料区域进行像素级分割,并建立煤料体积计算模型,结合输送带速度得出输送带煤流。实验结果表明,该方法的检测误差不超过3.78%,标准差不超过0.491,平均处理时间为83 ms,满足实际生产要求。     

区域的深度计算及其在图象渐变处理中的应用

为了解决局部图象处理容易产生边界效应的问题 ,可以区域中任意一点的深度值为基础 ,来得到权函数的数学表达式 ,然后根据权函数对图象上给定区域范围内的象素进行处理 ,使区域中离区域边界较近的象素的灰度值改正量小一些 ,而使区域中离区域边界较远的象素的灰度值改正量大一些 ,从而实现区域图象的渐变处理 .另外 ,为了克服迭代算法速度慢的缺点 ,还提出了一种快速标记多边形区域内部点的深度值算法 .实验结果表明 ,所提出的渐变处理方法 ,在克服区域图象增强处理时容易产生的边界效应方面是十分有效的 ,可以获得满意的效果 .     

Representation of an alpine treeline ecotone in SPOT 5 HRG data

An ecotone is a zone of vegetation transition between two communities, often resulting from a natural or anthropogenic environmental gradient. In remotely sensed imagery, an ecotone may appear as an edge, a boundary of mixed pixels or a zone of continuous variation, depending on the spatial scale of the vegetation communities and their transition zone in relation to the spatial resolution of the imagery. Often in image classification, an ecotone is either ignored if it falls within a width of one or two pixels, or part of it may be mapped as a separate vegetation community if it covers an area of several pixel widths. A soft classification method, such as probability mapping, is inherently appealing for mapping vegetation transition. Ideally, the probability of membership each pixel has to each vegetation class corresponds with the proportional composition of vegetation classes per pixel. In this paper we investigate the use of class probability mapping to produce a softened classification of an alpine treeline ecotone in Austria using a SPOT 5 HRG image. Here the transition with altitude is from dense subalpine forest to treeless alpine meadow and herbaceous vegetation. The posterior probabilities from a Maximum Likelihood algorithm are shown to reflect the land-cover composition of mixed pixels in the ecotone. The relationships between the posterior probability of class membership for the two end-member classes of ‘scrub and forest’ and ‘non-forest vegetation’ and the percentage ground cover of these vegetation classes (enumerated in 15 quadrats from 1:1500 aerial photographs) were highly significant: r² = 0.83 and r² = 0.85 respectively (p < 0.001, n = 15). We identify thresholds (alpha-cuts) in the posterior probabilities of class membership of ‘scrub and forest’ and ‘non-forest vegetation’ to map the alpine treeline ecotone as a transition zone of five intermediate vegetation classes between the end-member communities. In addition, we investigate the representation of the ecotone as a ratio between the posterior probabilities of ‘scrub and forest’ and ‘non-forest vegetation’. This displays the vegetation transition without imposing subjective boundaries, and has greater emphasis on the ecotone transition rather than on the end-member communities. We comment on the fitness for purpose of the different ways investigated for representing the alpine treeline ecotone.     

结合深度学习与条件随机场的遥感图像分类

目的 为进一步提高遥感影像的分类精度,将卷积神经网络（CNN）与条件随机场（CRF）两个模型结合,提出一种新的分类方法。方法 首先采用CNN对遥感图像进行预分类,并将其类成员概率定义为CRF模型的一阶势函数;然后利用高斯核函数的线性组合定义CRF模型的二阶势函数,用全连接的邻域结构代替常见的4邻域或8邻域;接着加入区域约束,使用Mean-shift分割方法得到超像素,通过计算超像素的后验概率均值修正各像素的分类结果,鼓励连通区域结果的一致性;最后采用平均场近似算法实现整个模型的推断。结果 选用3组高分辨率遥感图像进行地物分类实验。本文方法不仅能抑制更多的分类噪声,同时还可以改善过平滑现象,保护各类地物的边缘信息。实验采用类精度、总体分类精度OA、平均分类精度AA,以及Kappa系数4个指标进行定量分析,与支持向量机（SVM）、CNN和全连接CRF相比,最终获得的各项精度均得到显著提升,其中,AA提高3.28个百分点,OA提高3.22个百分点,Kappa提高5.07个百分点。结论 将CNN与CRF两种模型融合,不仅可以获得像元本质化的特征,而且同时还考虑了图像的空间上下文信息,使分类更加准确,后加入的约束条件还能进一步保留地物目标的局部信息。本文方法适用于遥感图像分类领域,是一种精确有效的分类方法。     

The use of prior probabilities in maximum likelihood classification of remotely sensed data

The expected distribution of classes in a final classification map can be used to improve classification accuracies. Prior information is incorporated through the use of prior probabilities—that is, probabilities of occurrence of classes which are based on separate, independent knowledge concerning the area to be classified. The use of prior probabilities in a classification system is sufficiently versatile to allow (1) prior weighting of output classes based on their anticipated sizes; (2) the merging of continuously varying measurements (multispectral signatures) with discrete collateral information datasets (e.g., rock type, soil type); and (3) the construction of time-sequential classification systems in which an earlier classification modifies the outcome of a later one. The prior probabilities are incorporated by modifying the maximum likelihood decision rule employed in a Bayesian-type classifier to calculate a posteriori probabilities of class membership which are based not only on the resemblance of a pixel to the class signature, but also on the weight of the class which is estimated for the final output classification. In the merging of discrete collateral information with continuous spectral values into a single classification, a set of prior probabilities (weights) is estimated for each value which the discrete collateral variable may assume (e.g., each rock type or soil type). When maximum likelihood calculations are performed, the prior probabilities appropriate to the particular pixel are used in classification. For time-sequential classification, the prior classification of a pixel indexes a set of appropriate conditional probabilities reflecting either the confidence of the investigator in the prior classification or the extent to which the prior class identified is likely to change during the time period of interest.     

A probabilistic SVM approach for hyperspectral image classification using spectral and texture features

New hyperspectral sensors can collect a large number of spectral bands, which provide a capability to distinguish various objects and materials on the earth. However, the accurate classification of these images is still a big challenge. Previous studies demonstrate the effectiveness of combination of spectral data and spatial information for better classification of hyperspectral images. In this article, this approach is followed to propose a novel three-step spectral–spatial method for classification of hyperspectral images. In the first step, Gabor filters are applied for texture feature extraction. In the second step, spectral and texture features are separately classified by a probabilistic Support Vector Machine (SVM) pixel-wise classifier to estimate per-pixel probability. Therefore, two probabilities are obtained for each pixel of the image. In the third step, the total probability is calculated by a linear combination of the previous probabilities on which a control parameter determines the efficacy of each one. As a result, one pixel is assigned to one class which has the highest total probability. This method is performed in multivariate analysis framework (MAF) on which one pixel is represented by a d-dimensional vector, d is the number of spectral or texture features, and in functional data analysis (FDA) on which one pixel is considered as a continuous function. The proposed method is evaluated with different training samples on two hyperspectral data. The combination parameter is experimentally obtained for each hyperspectral data set as well as for each training samples. This parameter adjusts the efficacy of the spectral versus texture information in various areas such as forest, agricultural or urban area to get the best classification accuracy. Experimental results show high performance of the proposed method for hyperspectral image classification. In addition, these results confirm that the proposed method achieves better results in FDA than in MAF. Comparison with some state-of-the-art spectral–spatial classification methods demonstrates that the proposed method can significantly improve classification accuracies.     

基于局部能量的边缘检测方法

本文从局部能量的角度提出了一种有效的图像边缘检测方法.在以一个像素点为中心的对称区域中,计算区域内所有像素的灰度值与中心像素的灰度值之间的差值,将差值平方的总和作为中心点所对应的局部能量.该局部能量可以有效地用于检测图像的边缘,因为边缘点的局部能量要比对应光滑区域内的像素点大得多.根据本文所构造的局部能量函数可以有效地找到边缘点.本文使用Baddeley误差度量（BEM）方法来评估本文方法检测结果的准确性.实验结果表明本文方法检测效果比较好.     

自适应的线条画绘制

目的 基于参考图像的线条画生成是非真实感绘制最为常见的应用之一。尽可能模拟艺术家的创作风格生成疏密得当、具有层次感的线条画是这类工作的主要目标和挑战。本文提出一个自适应线画图绘制算法。方法 首先,将场景图像分割成若干个区域,分别计算每个区域亮度的方差以及每个像素到边界的最小距离,将每个区域的方差和面积的比值作为该区域的复杂度。然后,计算能反映其显著视觉特征的边缘切向流场。最后,使用基于流的各向异性高斯差分滤波生成线条画。在构造边缘切向流时,每个位置的切向量由其邻域的切向量加权而得到。文中增加了一个新的系数项,对于邻域的任意一个位置,如果它和参考位置在区域分类中属同一个类别。则该位置的权值更大。基于流的高斯差分自适应滤波过程中,高斯差分滤波的尺度参数和复杂度以及到区域边界距离有关。细节越丰富,离边界越近,尺度参数取值越小,这样得到的边缘比较细,同时可以防止将相邻小细线条连接成粗线条。然后,将高斯差分滤波结果沿着流线方向进行高斯滤波,对于细节丰富的区域,边缘比较多,尺度参数取值比较小,所连接边缘比较短,可以减少错误边缘可能。结果 对生物、树林、建筑、山河等具有代表性的图像,采用本文算法进行自动实时进行线条绘制,实验结果表明,采用本文算法所生成的线条随着区域场景的复杂程度呈现不同粗细和浓淡的变化,具有一定的层次感。因而本文算法能生成视觉特征鲜明、风格化效果突出的线条画,且能处理各种复杂场景的图像。结论 本文自适应参数的线条画生成算法,其算法参数调节以及算法效果优于固定参数的算法,本文算法在处理日常生活中各类主题场景的图像时均能取得良好效果。