一种提高人耳特征点识别度的目标区域提取方法

对人耳进行特征识别多采用SURF算法,但该算法应用时极易受到图像中非目标区域的干扰,进而影响人耳特征点的检测和匹配准确度.基于目标区域的人耳特征识别算法可以突出目标区,而尽可能地抑制背景区域的影响.针对此问题,提出一种复合图像分割算法—KRM法作为人耳识别的预处理方法,将图像中人耳所在目标区域提取出来.该KRM法分为3步:首先利用k-means聚类算法将图像初步分割为前景目标区域和背景两类;再通过区域生长算法对过度分割的区域进行合并;最后应用形态学腐蚀的方法进行滤波得到人耳所在的目标区域.将KRM目标区域提取和SURF方法联用(简称KRM-SURF算法)应用于50组人耳图像,进行人耳特征点的检测与匹配,实验结果表明,特征点识别度(RD)均值达到0.924,KRM法的使用能极大地提高基于SURF算法的人耳特征识别的准确性.     

基于纹理的运动阴影检测方法

针对在视频对象分割时,运动阴影常被误分为视频对象,给出一种新的运动阴影检测方法.首先,进行基于自适应高斯混合模型的背景提取,获得包含运动阴影的前景分割,然后根据灰度图像中阴影区域和已获取背景相应位置的纹理相似性,进行阴影检测.纹理采用局部二元图(Local Binary Patterns,LBP)统一模式的直方图表征.实验表明,方法检测效果好,速度快,可应用于运动目标检测分割及跟踪等领域.     

基于纹理相似度的颜色迁移

提出一种以纹理相似度为依据的颜色迁移算法。通过提取图像的多维纹理特征进行主成分分析和线性判别分析,构建纹理特征空间,以度量像素点邻域的纹理相似度,并以纹理相似度为依据,对图像进行分割,在分割后的局部区域,建立纹理相似度与色度信息的映射关系,实现颜色迁移。实验结果表明,基于纹理相似度的颜色迁移,可解决颜色在边界处的误扩散问题,颜色迁移效果较好。     

基于双参数颗粒分析的纹理分割

形态学颗粒分析是一种有效的颗粒纹理图像分割方法,但由于单一的颗粒度参数,使得其难以分割如颗粒度相同而空间排列不同的纹理图像,为此提出了一种基于双参数颗粒分析的纹理图像分割方法,该方法将传统的颗粒分析从单一参数扩展到以颗粒度和空间位置为参数的二维空间,使得扩展后的颗粒分析通过分布函数不仅能够提取纹理的颗粒度特征,而且能够获取纹理的空间排列特征,克服了传统颗粒分析难以区分颗粒度相同而空间排列不同纹理区域的问题.仿真实验表明,该方法在运算复杂度增加不大的情况下,纹理分割效果优于颗粒分析法,区域像素错分率低于颗粒分析法和Gabor滤波器法.     

基于超像素分类的海天线检测方法

海天线检测是海空背景图像中海面目标提取的一项重要研究内容。根据海空背景图像的特点,提出了一种基于超像素分类的海天线检测方法。该方法首先利用简单线性迭代聚类算法(SLIC)将图像预分割成具有相似特性的图像块,然后提取每个图像块的特征组成该区域的特征向量,利用k-均值聚类算法将图像分成海水和天空两个区域,最后根据海天区域的连接点,采用RANSAC算法确定出海天线直线参数。实验结果表明,该方法可以有效地检测出复杂海天背景下的海天线。     

OLED显示屏表面缺陷自动检测方法

目的 为了解决OLED显示屏表面周期性纹理背景和缺陷边界模糊、对比度低的特征导致其表面缺陷检测困难的问题,开展OLED显示屏表面缺陷自动检测方法研究.方法 对OLED显示屏图像进行奇异值分解,选择前2个较大的奇异值重构图像纹理背景,对原图像和重构图像进行差分运算,获得残差图像.将残差图像像素随机赋予初始隶属度值,采用模糊C均值聚类法获得像素最终隶属度值.根据隶属度大小,将残差图像像素聚成2类,并从残差图像中准确地分割缺陷.结果 选取较大的2个奇异值可以有效地重构OLED显示屏的周期性纹理背景;模糊C均值聚类法分割缺陷获得的区域灰度一致性(U)平均值为0.9846.结论 基于奇异值分解的背景重构方法可以有效地检测OLED显示屏表面缺陷;与分水岭法和Otsu方法相比,模糊C均值聚类可以准确地分割模糊边界的缺陷区域.     

纸塑复合袋表面缺陷图像分割算法的设计与实现

目的 纸塑复合袋表面缺陷图像受到噪声、光照不均以及自身缺陷等因素的影响,在对图像缺陷区域进行分割时会造成过分割或欠分割.针对此现象提出一种将边缘检测和自适应区域生长法相结合的纸塑复合袋表面缺陷图像的分割算法.方法 首先利用Sobel算子和形态学运算对双边滤波后的缺陷图像进行第1次分割;然后对缺陷区域进行最小外接矩形标记并计算其形状特征,通过判定形状特征大小来决定是否继续分割;最后将符合继续分割的图像缺陷区域质心作为初始种子点,在原始图像上进行自适应区域生长,形成第2次分割结果,完成缺陷图像分割.结果 与其他算法相比,该算法对各类常见缺陷均能取得较好的分割效果,Dice系数均在0.93以上.结论 该算法分割精度较高,有较强的鲁棒性,可以满足工业上的生产需求.     

一种基于同质映射的视频图像中文本检测方法

提出了一种基于同质映射域的纹理特征的文本检测方法,并通过实验验证了该方法的性能.该方法与传统的文本检测方法的不同之处在于,首先将图像映射到同质性空间域中,在此空间域中计算纹理特征,然后通过支持向量机(SVM)分类器确定文本区域.与直接在图像空间域中提取纹理特征的方法相比,该方法对复杂背景下的文本检测更为有效,能有效地解决场景纹理特征与文本区域相近似造成的文本检测错误.     

采用多源图像分形特征的多目标检测方法

针对多目标的检测,本文提出一种采用多源图像分形特征的特征级融合检测方法.首先对多目标检测的特点进行了分析,对分形理论进行了介绍,然后详细介绍了该融合检测算法的思路和原理.该算法首先由红外图像阈值分割出部分目标;然后利用分维数图的统计特征可以增强分形维数的奇异性,在可见光图像的分维数图中搜索与已检测出的目标区域具有相近分形统计特征的区域,进行标记;再根据"距离相似度准则"进行目标的聚类识别,排除背景干扰,最终检测出全部目标.实验结果表明该融合检测算法能有效地进行多目标的检测与识别.     

时空联合的红外运动目标提取算法

针对红外图像对比度差、边缘模糊的特点,提出了一种基于时空联合的红外序列图像目标提取的新方法.算法充分利用了红外目标的亮度特征、背景信息以及运动信息.时域分割中通过建立帧差图像背景的高斯分布模型,采用变化检测模板来确定红外目标约束区域.然后,构造图像像素与区域之间的空间关系隶属度矩阵并约束到传统的模糊聚类算法中,空域分割则利用该模糊聚类来对目标约束区域进行有效分割.最后将时空分割结果融合便能实现最终的红外目标提取.实验结果表明,该方法简单有效,能准确提取动态场景中的红外目标.     

融合色彩分割和加权非参数变换的立体匹配

针对传统非参数变换的局限性,提出了一种立体匹配算法。利用参考图像的色彩分割信息获得基于任意形状和大小支持区域的匹配代价,并在相似色彩区域内计算加权非参数变换匹配代价,再将这两种匹配代价融合构成联合匹配代价,通过局部优化方法获得稠密视差图。实验结果表明算法提高了低纹理区域、遮挡区域和不连续区域的匹配精度,并对幅度失真具有鲁棒性。     

Fast unsupervised texture segmentation using Texel similarity map

Unsupervised texture segmentation is a challenging topic in computer vision. It is difficult to obtain boundaries of texture regions automatically in real-time, especially for cluttered images. This paper presents a new fast unsupervised texture segmentation method. First, the Texel similarity map (TSM) is used to compare the changes of intensity and gray level of neighboring pixels to determine whether they are identical. Then, a scheme called multiple directions integral images (MDII) is proposed to quickly evaluate the TSM. With the aid of MDII, one pixel’s similarity value can be computed in constant time. Finally, the proposed segmentation method is tested on both artificial texture and natural images. Experimental results show that the proposed method performs well on a wide range of images, and outperforms state-of-the-art method on segmentation speed.     

An Attribute Fusion System Based on the Choquet Integral to Evaluate the Quality of Composite Parts

This paper presents an information fusion system based on the Choquet integral for the quality evaluation of composite material parts. The application deals with the detection of typical regions inside parts with the help of images. For this purpose, several attributes related to texture homogeneity and intensity gradient orientation are extracted from the X-ray images of composite material parts. These attributes are transformed into region detection degrees expressed under the form of similarity maps. To improve the detection of the typical regions, a fusion system based on the 2-additive Choquet integral is applied on the similarity maps to take the interactions between attributes into account. An unsupervised learning algorithm is used to estimate the Choquet integral parameters from the reference regions pointed out by experts. The results are compared to those obtained by a previous fusion approach based on the ldquoDecision Templatesrdquo.     

基于小波域多分辨率MRF的声呐图像目标分割

声呐图像受噪声影响严重,分辨率低,传统算法对其目标分割效果较差,为此,提出了小波域多分辨率MRF模型的声呐图像分割算法。小波域多分辨率分析有利于提取声呐图像弱特征信息;每一分辨率中的观测特征采用高斯混合模型建模,尺度内同标记的观测特征用高斯模型建模,用各向同性的双点多级逻辑(Multi-Level Logistic,MLL)模型建模每一尺度的标记场;最后,用迭代条件模式(Iterated Conditional Mode,ICM)实现多分辨率马尔可夫随机场(Multi-Resolution Markov Random Field,MRA-MRF)中能量函数的最优解,获取标记场,完成声呐图像分割。从视觉效果和定量分析两方面验证。对比实验的结果表明,该文算法能有效地提取声呐图像的弱目标信息,较好地将目标区域和背景区域分割出来,具有较高的分割精度和鲁棒性。     

Glioma brain tumor detection and segmentation using weighting random forest classifier with optimized ant colony features

The uncontrolled growth of cells in brain regions leads to the tumor regions and these abnormal tumor regions are scanned by magnetic resonance imaging (MRI) technique as an image. This paper proposes random forest classifier based Glioma brain tumor detection and segmentation methodology using feature optimization technique. The texture features are derived from brain MRI image and these derived feature set are now optimized by ant colony optimization algorithm. These optimized set of features are trained and classified using random forest classification method. This classifier classifies the brain MRI image into Glioma or non-Glioma image based on the optimized set of features. Furthermore, energy-based segmentation method is applied on the classified Glioma image for segmenting the tumor regions. The proposed methodology for Glioma brain tumor stated in this paper achieves 97.7% of sensitivity, 96.5% of specificity, and 98.01% of accuracy.     

基于区域邻接图的立体视觉边缘匹配算法

针对自然场景轮廓边缘的立体匹配问题,提出了基于区域邻接图的快速匹配算法.首先利用分水岭变换进行图像分割,根据分割区域边界确定图像中场景的轮廓边缘.基于由全局到局部、自上而下的分层匹配思想,匹配过程分为两步:第一步将轮廓边缘按其所属区域进行分组作为匹配基元进行匹配,匹配过程中根据边缘所属区域的位置,尺寸和灰度特征建立区域约束,并在边缘特征角点的引导下,按照区域邻接图采用类似区域生长的匹配策略实现边缘匹配,区域约束大大减少了边缘特征匹配的搜索空间、优化了匹配顺序.第二步则根据边缘匹配结果,以已匹配的边缘特征角点为基准点,在其引导下实现其他边缘点的快速立体匹配.实验结果表明,该算法匹配正确率能达到93%以上,是一种快速有效的立体匹配算法.     

Performance comparison of texture feature analysis methods using PNN classifier for segmentation and classification of brain CT images

A computer software system is designed for the segmentation and classification of benign and malignant tumor slices in brain computed tomography images. In this paper, we present a texture analysis methods to find and select the texture features of the tumor region of each slice to be segmented by support vector machine (SVM). The images considered for this study belongs to 208 benign and malignant tumor slices. The features are extracted and selected using Student's t‐test. The reduced optimal features are used to model and train the probabilistic neural network (PNN) classifier and the classification accuracy is evaluated using k fold cross validation method. The segmentation results are also compared with the experienced radiologist ground truth. Quantitative analysis between ground truth and segmented tumor is presented in terms of quantitative measure of segmentation accuracy and the overlap similarity measure of Jaccard index. The proposed system provides some newly found texture features have important contribution in segmenting and classifying benign and malignant tumor slices efficiently and accurately. The experimental results show that the proposed hybrid texture feature analysis method using Probabilistic Neural Network (PNN) based classifier is able to achieve high segmentation and classification accuracy effectiveness as measured by Jaccard index, sensitivity, and specificity.     

WISECODE: wise image segmentation based on community detection

Image segmentation is one of the fundamental problems in image processing and computer vision, since it is the first step in many image analysis systems. This paper presents a new perspective to image segmentation, namely, segmenting input images by applying efficient community detection algorithms common in social and complex networks. First, a common segmentation algorithm is used to fragment the image into small initial regions. A weighted network is then constructed. Each initial region is mapped to a vertex, and all these vertices are connected to each other. The similarity between two regions is calculated from colour information. This similarity is then used to assign weights to the edges. Afterwards, a community detection algorithm is applied, and communities are extracted such that the highest modularity measure is achieved. Finally, a post-processing algorithm merges very small regions with the greater ones, further enhancing the final result. One of the most striking features of the proposed method, is the ability to segment the input image without the need to specify a predefined number of segments manually. This remarkable feature results from the optimal modularity value, which is utilised by this method. It is also able to segment the input image into a user defined number of segments. Extensive experiments have been performed, and the results show that the proposed scheme can reliably segment the input colour image into good subjective criteria.     

Development of morphological technique for segmentation of anatomical objects in abdominal MRI

Abstract

In this paper, a morphological technique for the segmentation of abdominal organs in magnetic resonance imaging (MRI) images is proposed based on watershed segmentation. New morphological based preprocessing and post-processing techniques are developed to reduce oversegmentation by means of removing and merging spurious segments. The preprocessing aims at removing trivial regions as well as background noise by combining thresholding, morphological smoothing, Gaussian smoothing and morphological edge detection. To obtain a more concise region representation, the watershed segmented image is post-processed, where a region adjacency list is built for the region merging process that produces the final segments. To control the merging process, a similarity function is defined, whence the most similar neighbouring regions are merged. The proposed technique produces effective and significant results in successfully segmenting various anatomical objects in axial MRI images of the abdomen, as it is shown in this paper.