Combining stationary wavelet transform and self-organizing maps for brain MR image segmentation

This study presents an image segmentation system that automatically segments and labels T1-weighted brain magnetic resonance (MR) images. The method is based on a combination of unsupervised learning algorithm of the self-organizing maps (SOM) and supervised learning vector quantization (LVQ) methods. Stationary wavelet transform (SWT) is applied to the images to obtain multiresolution information for distinguishing different tissues. Statistical information of the different tissues is extracted by applying spatial filtering to the coefficients of SWT. A multidimensional feature vector is formed by combining SWT coefficients and their statistical features. This feature vector is used as input to the SOM. SOM is used to segment images in a competitive unsupervised approach and an LVQ system is used for fine-tuning. Results are evaluated using Tanimoto similarity index and are compared with manually segmented images. Quantitative comparisons of our system with the other methods on real brain MR images using Tanimoto similarity index demonstrate that our system shows better segmentation performance for the gray matter while it gives average results for white matter.     

Stereo vision correspondence using a multichannel graph matching technique

A multichannel feature-based stereo vision technique is described in this paper where curve segments are used as the feature primitives in the matching process. Curve segments are extracted by tracking the zero-crossings of the left and right images. The generalized Hough transform of each curve and the curve length are used as a local feature vector in representing the distinctive characteristics of the curve segment. The feature vector of each curve segment in the left image is used as a constraint to find an instance of the same curve segment in the right image. The epipolar constraint on the centroids of the curve segment is used to limit the searching space in the right image.

A relational graph is formed from the left image by treating the centroids of the curve segments as the nodes of the graph. The local features of the curve segments are used to represent the local properties of the nodes, and the relationship between the nodes represents the structural properties of the objects in the scene. A similar graph is also formed from the right image curve segments. A graph isomorphism is then formed between the two graphs by using the epipolar constraint on the centroids, the local properties of the nodes, node assignment and the structural relationship (compatibility) between the nodes.     

An object-based image retrieval system for digital libraries

A novel approach to clustering for image segmentation and a new object-based image retrieval method are proposed. The clustering is achieved using the Fisher discriminant as an objective function. The objective function is improved by adding a spatial constraint that encourages neighboring pixels to take on the same class label. A six-dimensional feature vector is used for clustering by way of the combination of color and busyness features for each pixel. After clustering, the dominant segments in each class are chosen based on area and used to extract features for image retrieval. The color content is represented using a histogram, and Haar wavelets are used to represent the texture feature of each segment. The image retrieval is segment-based; the user can select a query segment to perform the retrieval and assign weights to the image features. The distance between two images is calculated using the distance between features of the constituent segments. Each image is ranked based on this distance with respect to the query image segment. The algorithm is applied to a pilot database of natural images and is shown to improve upon the conventional classification and retrieval methods. The proposed segmentation leads to a higher number of relevant images retrieved, 83.5% on average compared to 72.8 and 68.7% for the k-means clustering and the global retrieval methods, respectively.     

采用K-means的脑肿瘤磁共振图像分割与特征提取

大脑肿瘤分割对于医师判断肿瘤恶化程度非常重要。然而,由于肿瘤的不规则形状、与周围组织的低对比度以及出现位置的不固定,给脑肿瘤的精确分割带来很大的困难。传统的K-means分割方法仅仅利用图像的灰度特征,很难准确分割肿瘤边界。利用灰度共生矩阵提取出的纹理特征,并结合图像几何不变矩特征对分割出的脑肿瘤图像进行特征提取。灰度共生矩阵定义为像素对的联合概率分布,是一个对称矩阵,它不仅反映图像灰度在相邻的方向、相邻间隔、变化幅度的综合信息,也反映了相同的灰度级像素之间的位置分布特征,是计算纹理特征的基础;几何矩（不变矩）具有旋转、平移、尺度等特性,能将图像分解为有限特征值,并且通过对比所提取出的同一病人的肿瘤图像的不变矩参数,可以获得该肿瘤几何形状变化程度。实验结果表明,该方法可以同时从纹理和几何特征对图像特征进行描述,与分别采用灰度共生矩阵和不变矩方法进行特征提取相比较,降低了算法计算量,同时提升了算法的抗噪性。     

结合多特征和SVM的SAR图像分割

为实现灰度共生矩阵(GLCM)多尺度、多方向的纹理特征提取, 提出了一种结合非下采样轮廓变换(NSCT)和GLCM的纹理特征提取方法。先用NSCT对合成孔径雷达(SAR)图像进行多尺度、多方向分解; 再对得到的子带图像使用GLCM提取灰度共生量; 然后对提取的灰度共生量进行相关性分析, 去除冗余特征量, 并将其与灰度特征构成多特征矢量; 最后, 充分利用支持向量机(SVM)在小样本数据库和泛化能力方面的优势, 由SVM完成多特征矢量的划分, 实现SAR图像分割。实验结果表明, 基于NSCT域的GLCM纹理提取方法和多特征融合用于SAR图像分割, 可以提高分割准确率, 获得较好的边缘保持效果。     

Image segmentation scheme based on SOM–PCNN in frequency domain

A hybrid scheme for the image segmentation of high-resolution images is proposed in this study. Our methodology is based on combining both supervised and unsupervised segmentation. The entire process is performed in the frequency domain, rather than the spatial domain, using the Shift Invariant Shearlet Transform (SIST). Initially, the input image is filtered using an anisotropic filter to enhance the texture features. Then, it is separated into low and high sub-band frequencies using SIST. Subsequently, we built a feature vector from coarser coefficients complemented with texture information extracted from high-frequency coefficients of the input image. SOM is used for the preliminary classification of the input image coefficients, and the network training process is performed using the previously built feature vector. Lastly, the modified PCNN is used to augment the SOM results to reduce the over-segmentation artefacts. We used the Berkeley Segmentation Database (BSR) and Quick-Bird Satellite images to validate the results. It was found that the proposed scheme is superior to the Fuzzy-C-Means-based, SOM-based, and PCNN-based segmentation algorithms in terms of quantitative criteria and visual interpretation.     

支持向量机在核磁共振脑组织图像分割中的应用

在医学图像分析中,脑组织图像分割有着重要的研究与应用价值。采用支持向量机方法对核磁共振脑图像进行研究。传统的支持向量机方法在图像分割中一般选用方形的区域,用该区域的像素灰度和纹理特征作为支持向量的训练样本,对图像进行提取和分析,得到分类结果。提出了一种新型的研究区域,在该区域上提取训练样本,对核磁共振脑图像进行分类。分类结果显示用新型区域做的图像分割提高了正确率。     

AS-OCT图像下的自动皮质性白内障分类框架

白内障是一种主要导致视觉损伤的眼病. 早期干预和白内障手术是改善患者视力和生活质量的主要手段. 眼前节光学相干断层成像图像 (anterior segment optical coherence tomography, AS-OCT) 是一种新型眼科图像, 其具有非接触、高分辨率、检查快速等特点. 在临床上, 眼科医生已经逐渐采用AS-OCT图像进行眼科疾病如青光眼的诊断, 然而尚未有研究工作利用它进行皮质性白内障 (cortical cataract, CC) 自动分类. 为此, 提出了一个基于AS-OCT图像的自动皮质性白内障分类框架, 由图像预处理、特征提取、特征筛选和分类等4部分组成. 首先, 利用反光区域去除和对比度增强方法进行图像预处理; 紧接着使用灰度共生矩阵 (grey level co-occurrence matrix, GLCM)、灰度区域大小矩阵 (grey level size zone matrix, GLSZM) 和邻域灰度差矩阵 (neighborhood grey tone difference matrix, NGTDM) 方法从皮质区域提取了22个特征; 然后, 采用斯皮尔曼相关系数方法对提取的特征进行特征重要性分析并筛除冗余特征; 最后利用线性支持向量机方法进行分类. 在一个临床AS-OCT图像数据集上的实验结果表明, 所提出的皮质性白内障分类框架准确率、召回率、精确率和F1分别达到86.04%, 86.18%, 88.27%和86.35%, 取得与先进的深度学习算法接近的性能, 表明其具有作为辅助眼科医生进行皮质性白内障临床诊断工具的潜力.     

融合SIFT特征的熵图估计医学图像非刚性配准

配准准确性是医学图像配准算法的一项重要指标,像素灰度是目前图像配准中广泛使用的特征,但是灰度特征来源单一,而且忽略空间信息,在一些情况下容易产生误配。针对这个问题,本文提出一种融合SIFT特征的熵图估计医学图像非刚性配准算法。该算法首先使用基于互信息的刚性配准算法对两幅待配准图像进行粗配;然后,在采样点上提取像素灰度和SIFT高维特征,并在此基础上构造k-最邻近图(kNNG);最后,使用k-最邻近图来估计α互信息(αMI)。实验结果表明:和传统的基于互信息和像素灰度的刚性配准算法,基于熵图估计和单一像素灰度特征的非刚性配准算法相比,本文提出的算法具有更高的配准准确性。     

融合颜色词袋特征的视觉词汇树图像检索

针对由图像灰度空间产生的传统词袋模型SIFT特征无法体现图像的颜色信息的问题,提出了一种融合颜色特征的视觉词汇树来对图像进行描述。提取SIFT特征并建立词汇树,获取图像的SIFT表示向量。利用K-means方法对图像库中的所有图像的HSV值进行聚类,获得基于HSV空间的颜色词袋表示向量,避免了传统颜色直方图方法所带来的量化误差。将SIFT特征与颜色词袋特征进行融合,完成了图像的全局特征和局部特征的融合。然后,计算融合特征的相似度,将相似度从高到低排序,完成图像检索。为了验证本方法的有效性,选择Corel图像库对算法性能进行实验分析,从主观评价和客观评价标准分别进行评价,并与传统方法进行了对比。结果表明,特征融合的检索性能与单一特征方法相比有较大提高。特征融合方法的平均检索查准率和查全率-查准率等评价指标,对比传统方法均有不同程度提高。     

基于可变模板和支持向量机的人体检测

随着图像处理技术和模式识别技术的发展，人体检测在监控系统、驾驶员辅助系统、图像索引等领域已得到广泛应用。针对静态图像中站姿人体检测问题，提出了一种新的特征选取方法，并应用可变模板和支持向量机相结实现对图像中的人体检测和定位。通过对图像进行轮廓提取和网格划分，选择水平方向和垂直方向上的特征组成图像的特征向量，然后使用搜集到的样本特征向量对模型进行训练，用可变模板搜索待检测图像，根据检测区域的特征和训练好的模型对图像进行分类。实验结果表明，该方法可以快速准确地检测出多种背景图像中的站姿人体,正确分类率达92%以上。     

基于支持向量机的连续超声图像集分割算法

针对传统的支持向量机（SVM）模型对连续超声图像集进行分割时需要为图像集中每张图片提取样本点来建立分割模型的问题,提出了一个对整个连续超声图像集的统一的SVM分割模型。首先,从图像的灰度直方图中提取灰度特征作为表征图像集中图像连续性的特征;其次,从图像集中选取部分图像作为样本,并从中提取像素点的灰度特征;最后,将各像素点的灰度特征与各像素点所在图像中表征图像集连续性的特征相结合,用SVM的方法训练出分割模型对整个图像集进行分割。实验结果表明,与传统SVM分割方法相比,新模型在面对大量的有连续变化的图像集的分割问题上,大幅地减少了人工选取样本点的工作量,并且在分割的准确率上也有保证。     

Computer-assisted categorization of brain computerized tomography pixels into cerebrospinal fluid, white matter, and gray matter

A computer-assisted method was employed to estimate the amounts of cerebrospinal fluid (CSF), white matter, and gray matter in individual computerized tomography (CT) scans of brains. By means of an image processing procedure (DMORPH), the means +/- SD CT numbers of "pure" CSF, white matter, and gray matter were determined in each scan and stored. A CATSEG program used these means to define ranges for CT numbers for each of the three tissues on each scan, and to assign each pixel in a scan to one of the three categories. Summing over seven serial scans provided volumetric estimates of CSF, white matter, and gray matter in a brain segment. For 10 subjects aged 21 to 43 years, CSF volume equaled 1.4 to 4.7% of the total segment volume, white matter equaled 37.5 to 48.2%, and gray matter equaled 50.2 to 58.9%. Image processing hardware and software which allow standardized sampling from CT images for the evaluation of surface areas and CT numbers are described. These procedures, as applied to CT scans of the human brain, can be used to estimate the volumes of CSF, white matter, and gray matter in a selected intracranial segment.     

局部特征与多示例学习结合的超声图像分类方法

利用全局特征对超声图像进行描述具有一定的局限性,而且对图像进行手工标注的成本过高, 为解决上述问题,本文提出了一种利用局部特征描述超声图像,并结合多示例学习对超声图像进行分类的新方法. 粗略定位图像中的感兴趣区域 (Region of interest, ROI),并提取局部特征,将感兴趣区域看作由局部特征构成的示例包, 采用自组织映射(Self-organizing map, SOM)的方法对示例特征进行矢量量化,采用Bag of words方法将示例特征映射到示例包空间,进而采用传统的支持向量机对示例包进行分类.本文提出的方法在临床超声图像上进行了实验,实验结果表明,该方法具有良好的泛化能力和较高的准确性.     

A computer-aided diagnosis system for malignant melanomas

The aim of this study is to provide an efficient way to segment the malignant melanoma images. This method first eliminates extra hair and scales using edge detection; afterward, it deduces a color image into an intensity image and approximately segments the image by intensity thresholding. Some morphological operations are used to focus on an image area where a melanoma boundary potentially exists and then used to localize the boundary in that area. The distributions of texture and a new feature known as AIBQ features in the next step provide a good discrimination of skin lesions to feature extraction. Finally, we rely on quantitative image analysis to measure a series of candidate attributes hoped to contain enough information to differentiate malignant from benign melanomas. The selected features are applied to a support vector machine to classify the melanomas as malignant or benign. By our approach, we obtained 95 % correct classification of malignant or benign melanoma on real melanoma images.     

高频子带特征图像的人工智能配准仿真

在很多的应用领域上,都要求图像的配准精度达到亚像素级别。传统的SIFT算法是图像配准中用来描写局部特征较为精准、可拓展性较强的一种方法,但对于图像关键点特征向量描述有着冗杂,配准精度低等缺陷。为了进一步提高图像的配准精度,提出一种高频子带特征图像人工智能配准方法。提取高频子带图像特征点,处理高频子带图像滤波,将掺杂噪声与不含噪声的高频子带图像,采用图像像素点的灰度值、空间元素以及像素加权灰度密度这三个特征进行划分,完成图像去噪处理。根据去噪后的图像像素点提取出高频子带图像特征点,将提取的高频子带图像特征点进行粗配准。根据仿真结果表明,所提方法与传统的特征提取方法相比,有效的提高了配准精度,同时配准后的高频子带图像噪声滤除效果更佳。     

An improved SOM algorithm and its application to color feature extraction

Reducing the redundancy of dominant color features in an image and meanwhile preserving the diversity and quality of extracted colors is of importance in many applications such as image analysis and compression. This paper presents an improved self-organization map (SOM) algorithm namely MFD-SOM and its application to color feature extraction from images. Different from the winner-take-all competitive principle held by conventional SOM algorithms, MFD-SOM prevents, to a certain degree, features of non-principal components in the training data from being weakened or lost in the learning process, which is conductive to preserving the diversity of extracted features. Besides, MFD-SOM adopts a new way to update weight vectors of neurons, which helps to reduce the redundancy in features extracted from the principal components. In addition, we apply a linear neighborhood function in the proposed algorithm aiming to improve its performance on color feature extraction. Experimental results of feature extraction on artificial datasets and benchmark image datasets demonstrate the characteristics of the MFD-SOM algorithm.     

基于Multi-Path RefineNet的多特征高分辨率SAR图像道路提取算法

为解决现有高分辨率SAR图像道路提取算法自动化较差、普适性不高的问题,提出了一种基于多路径优化网络的多特征提取算法。首先,对SAR图像进行Gabor变换及灰度梯度共生矩阵变换,获取丰富的道路特征信息,联结级联优化网络和残差网络形成多路径优化网络;然后,对SAR原图、获取的低级特征图和标签图进行训练,充分利用每层网络提取的道路特征获取初始分割的道路结果;最后,利用数学形态学运算连接初始道路断裂处并去除虚警。利用所提算法对不同分辨率的SAR图像进行道路提取,实验结果表明,该算法在提取SAR图像道路方面适用范围广且道路提取效果佳。     

基于倒金字塔深度学习网络的三维医学图像分割

基于深度学习的医学图像分割对医学研究和临床疾病诊断具有重要意义。然而,现有三维脑图像分割网络仅依赖单一模态信息,且最后一层网络的特征表达不准确,导致分割精度降低。引入注意力机制,提出一种基于深度学习的多模态交叉重构的倒金字塔网络MCRAIP-Net。以多模态磁共振图像作为输入,通过三个独立的编码器结构提取各模态的特征信息,并将提取的特征信息在同一分辨率级进行初步融合。利用双通道交叉重构注意力模块实现多模态特征的细化与融合。在此基础上,采用倒金字塔解码器对解码器各阶段不同分辨率的特征进行整合,完成脑组织的分割任务。在MRBrainS13和IBSR18数据集上的实验结果表明,相比3D U-Net、MMAN、SW-3D-Unet等网络,MCRAIP-Net能够充分利用多模态图像的互补信息,获取更准确丰富的细节特征且具有较优的分割精度,白质、灰质、脑脊液的Dice系数分别达到91.67%、88.95%、84.79%。     

基于人工蜂群算法的W SN 覆盖优化研究

As an emerging network structure, the capsule network uses vector output instead of scalar output, which can capture the spatial relationship between image features and improve the limitations of convolutional neural network. This paper firstly trains the capsule network to achieve image classification, obtains the predictive label of the image, determines the category of the query im- age, and then uses the feature parameters in the digital capsule layer of the network as the feature vector of the image. The feature vector is used to find images similar to the query image in the category set of the query image. In this paper, experiments are carried out on the FASHION-MNIST and CIFAR10 datasets respectively. The experimental results show that the proposed method can bet- ter extract the features of the images and obtain good image retrieval results.