Study of improved Otsu algorithm and its ration evaluation analysis for PCB photoelectric image segmentation

The algorithm of maximum variance between clusters (traditional Otsu algorithm) is discussed, and its advantage is given also. In order to segment the PCB photoelectric image better, on the basis of the traditional Otsu algorithm, considering the different influence of image segmentation about the factors of the distance between target and background as well as each kind of cohesion, an improved Otsu algorithm is proposed, and its basic principle and segmentation advantages are analyzed in detail. In order to evaluate these segmentation results impersonally by using different algorithms, the quantitative criteria of gray-level contrast and district interior uniformity are adopted to evaluate these segmentation results impersonally. Finally, the different segmentation experiment contrasts of PCB photoelectric image between our algorithm and other algorithms is executed, the results of experiment indicate that our algorithm has relatively better segmentation quality.     

基于水平集的闪光照相图像分割算法

针对Chan-Vese(CV)模型局部控制能力差的缺点,将基于区域的CV模型和分割曲线的局部信息结合起来,提出了一种新的水平集图像分割算法。该算法以CV法的分割曲线为初始曲线,以获得全局收敛性,在后继分割中引入分割曲线的局部信息,以提高模型对图像中微弱信号的分割能力。对闪光照相图像的数值实验表明,该算法噪声抵抗能力强,对初始轮廓位置不敏感,能实现对含细长拓扑结构和微小孔洞的弱边界闪光图像的自动分割。     

Multiple-phase segmentation approach for blood vessel extraction on cervical MRA image sequence

The segmentation of regions is an important first step for a variety of image-related applications such as image analysis, computer vision and visualization tasks. Specifically, the computer algorithm for the delineation of anatomical structures and other regions of interest is an essential component in assisting and automating specific radiological tasks. In this article, we propose a multiple-phase segmentation algorithm for carotid artery (CA) extraction. The seed position was automatically selected on the initially thresholded image by using a priori knowledge of the CA anatomic structure. In consideration of the preserved connectivity between consecutive slice images, the selected seed was maintained within the CA area throughout the entire segmentation process. The average intensity value was adaptively adjusted as a homogeneity criterion for each slice image. In addition, the stack feature should be used to automatically locate the branch, and the duplicated stack was used to save branch detection time for subsequent segmentation processes. This algorithm provided fine segmentation results compared with well-known single-phase segmentation approaches and other combined segmentation methods. This multiple-phase segmentation approach could be applicable to segment tree-like organ structures such as renal artery, coronary artery and airway tree from medical imaging modalities.     

一种快速提取连续CT脑肿瘤病灶区的算法

脑肿瘤图像提取就是将肿瘤病灶区域(水肿、坏死、癌变)从正常的脑部组织(灰质、白质、脑脊液)分开,精确的脑肿瘤分割对脑瘤的诊断、研究和治疗有重要的临床意义。针对传统脑部CT肿瘤病灶提取的缺点,即需要耗费大量时间并且分割精度不高的问题,提出一种综合了形态学重建、分水岭分割和改进的区域生长算法。先用形态学重建进行去噪,再用结合多尺度梯度分水岭分割提取整个图像的边界,然后在肿瘤病灶区域内选取种子点进行区域生长,提取肿瘤区域轮廓,滤除其他封闭区域,得到的图像作为改进的区域生长法的初始分割区域,使用改进的区域生长法,滤除过分割区域。实验结果显示该算法分割出的结果有效区域大,分割精度高。结论：该算法提高了分割精度,由于不用匹配结构参数,加快了分割速度,具有一定的临床价值。     

Automatic leukocyte nucleus segmentation by intuitionistic fuzzy divergence based thresholding

The paper proposes a robust approach to automatic segmentation of leukocyte's nucleus from microscopic blood smear images under normal as well as noisy environment by employing a new exponential intuitionistic fuzzy divergence based thresholding technique. The algorithm minimizes the divergence between the actual image and the ideally thresholded image to search for the final threshold. A new divergence formula based on exponential intuitionistic fuzzy entropy has been proposed. Further, to increase its noise handling capacity, a neighborhood-based membership function for the image pixels has been designed. The proposed scheme has been applied on 110 normal and 54 leukemia (chronic myelogenous leukemia) affected blood samples. The nucleus segmentation results have been validated by three expert hematologists. The algorithm achieves an average segmentation accuracy of 98.52% in noise-free environment. It beats the competitor algorithms in terms of several other metrics. The proposed scheme with neighborhood based membership function outperforms the competitor algorithms in terms of segmentation accuracy under noisy environment. It achieves 93.90% and 94.93% accuracies for Speckle and Gaussian noises, respectively. The average area under the ROC curves comes out to be 0.9514 in noisy conditions, which proves the robustness of the proposed algorithm.     

基于梯度的动态分块迭代阈值图像二值化方法

在传统的Ostu和Bernsen算法基础上,针对亮度不均的大幅面图像二值化问题,提出了一种基于图像梯度的动态分块迭代求取阈值的二值化方法.实验验证了该方法对于由线阵CCD扫描获得的低对比度、亮度不均的大幅面图像二值化的有效性.     

一种基于灰度变换的红外图像增强算法

针对红外图像，采用双门限分割法进行图像分割，然后采用分段灰度变换法进行图像增强。根据具体图像，通过人机交互的方式确定2个阈值，将图像分割为目标区、过渡区和背景区3部分；按照每一部分的特点，设计不同的灰度变换，对图像进行分段线性增强，得到感兴趣目标区的最佳视觉效果。通过对16位红外图像进行实验，得到了满意的结果。实验表明,该算法灵活便捷，在增加对比度和去除噪声的同时，还抑制了背景，达到了预期的效果。     

基于伪魏格纳分布分解的合成孔径雷达图像目标与阴影分割

在二维魏格纳分布的框架内,针对魏格纳变换的交叉项问题和计算量大的问题,提出了合成孔径雷达图像局部伪魏格纳变换的目标和目标阴影的分割方法.首先,将合成孔径雷达图像进行二维伪魏格纳变换,得到各像素点的二维能量谱图|然后提取各像素点的二维能量谱图对应位置值形成多个不同频段的与原图像同大小的能量谱图|最后,对不同频段的能量谱图采用不同的处理方法后,将各能量谱图相加处理后形成区域标识图像,最终得到原图像的目标和目标阴影分割图像.本文利用该方法对MSTAR切片图像进行了分割试验,并对分割图像与频谱最大值距离或方位分割算法和基于双参量CFAR与隐马尔科夫联合分割算法进行了分割图像对比度对比.实验结果表明,采用本文算法的合成孔径雷达分割图像,对比度明显提高,且保留了目标图像细节.     

Stable Atlas-based Mapped Prior (STAMP) machine-learning segmentation for multicenter large-scale MRI data

Machine learning (ML)-based segmentation methods are a common technique in the medical image processing field. In spite of numerous research groups that have investigated ML-based segmentation frameworks, there remains unanswered aspects of performance variability for the choice of two key components: ML algorithm and intensity normalization. This investigation reveals that the choice of those elements plays a major part in determining segmentation accuracy and generalizability. The approach we have used in this study aims to evaluate relative benefits of the two elements within a subcortical MRI segmentation framework. Experiments were conducted to contrast eight machine-learning algorithm configurations and 11 normalization strategies for our brain MR segmentation framework. For the intensity normalization, a Stable Atlas-based Mapped Prior (STAMP) was utilized to take better account of contrast along boundaries of structures. Comparing eight machine learning algorithms on down-sampled segmentation MR data, it was obvious that a significant improvement was obtained using ensemble-based ML algorithms (i.e., random forest) or ANN algorithms. Further investigation between these two algorithms also revealed that the random forest results provided exceptionally good agreement with manual delineations by experts. Additional experiments showed that the effect of STAMP-based intensity normalization also improved the robustness of segmentation for multicenter data sets. The constructed framework obtained good multicenter reliability and was successfully applied on a large multicenter MR data set (n > 3000). Less than 10% of automated segmentations were recommended for minimal expert intervention. These results demonstrate the feasibility of using the ML-based segmentation tools for processing large amount of multicenter MR images. We demonstrated dramatically different result profiles in segmentation accuracy according to the choice of ML algorithm and intensity normalization chosen.     

石油岩芯CT图像裂缝分割算法研究

针对裂缝区域分割的需求和石油岩芯CT图像的特点, 改进了现有的水平集分割算法。首先对图像中值滤波去噪后运用C-V模型对图像进行初分割, 把背景区域和岩芯区域准确分开, 得到岩芯区域的轮廓;然后调整轮廓外区域的灰度值, 使之等于岩芯区域平均灰度值, 增强目标区域;最后再进行RSF模型细分割, 得到最终分割结果。对于高斯噪声污染严重的岩芯图像, 先采用了邻域加窗的非局部均值去噪方法, 再用改进水平集算法分割, 实验结果表明该分割方法是有效的。     

An energy minimization method for MS lesion segmentation from T1-w and FLAIR images

In this paper, we extend the multiplicative intrinsic component optimization (MICO) algorithm to multichannel MR image segmentation, with focus on segmentation of multiple sclerosis (MS) lesions. The MICO algorithm was originally proposed by Li et al. in Ref. [1] for normal brain tissue segmentation and intensity inhomogeneity correction of a single channel MR image, which exhibits desirable advantages over other methods for MR image segmentation and intensity inhomogeneity correction in terms of segmentation accuracy and robustness. In this paper, we extend the MICO algorithm to multi-channel MR image segmentation and enable the segmentation of MS lesions. We assign different weights for different channels to control the impact of each channel. The weighted channels allow the enhancement of the impact of the FLAIR image on the segmentation of MS lesions by assigning a larger weight to the FLAIR image channel than the other channels. With the inherent mechanism of estimation of the bias field, our method is able to deal with the intensity inhomogeneity in the input multi-channel MR images. In the application of our method, we only use T1-w and FLAIR images as the input two channel MR images. Experimental results show promising result of our method.     

基于OB-HMAD算法和光谱特征的高分辨率遥感影像变化检测

高空间分辨率遥感影像蕴涵丰富的地物细节信息,针对高分辨率多时相遥感影像的变化检测可以更清楚认识到地理单元的变化情况,传统的遥感变化检测算法面对高分辨率遥感影像时,会出现明显的"椒盐现象"。本文借鉴面向对象图像分析的思想,以高分辨率遥感影像对象的光谱特征为分析对象,在多变量变化检测算法(multivariate alternative detection, MAD)的基础上,提出一种半自动阈值选取的OB-HMAD(object based-hybrid MAD)算法,并利用该算法进行变化检测实验对比分析。首先对高分辨率多时相遥感影像进行多尺度分割,形成多通道的影像对象;其次利用MAD变换,形成差异影像对象,并对其进行MNF变换,提高影像对象的信噪比;然后采用直方图曲率分析(histogram curvature analysis, HCA)进行半自动阈值选取,提取变化区域;最后结合实地样本数据对变化检测结果进行混淆矩阵的精度验证。结合2012年和2013年北京地区Worldview-2影像的实验可知,OB-HMAD算法融合多通道的光谱信息,可以有效的实现多时相高分影像的变化检测,基本消除了基于像元变化检测中"椒盐"现象的干扰,并在一定程度上降低建筑物阴影和几何配准误差的影响,总体精度和kappa系数也较优于其他变化检测算法,但存在较大的漏检误差。MNF变换可以有效的提高影像的信噪比,使差异信息更集中,直方图曲率分析的阈值分割算法相对其他阈值算法,自动化程度更高。     

基于C-V模型的改进快速水平集图像分割法

针对水平集方法计算复杂度高,无法满足实时系统要求的缺陷,提出一种改进的快速水平集算法。该算法对快速水平集算法进行简化,采用单链表表示轮廓曲线。利用C-V模型的二值拟合项来设计曲线演化的速度函数,保留了C-V模型的全局优化特性。还给出了一个基于单链表中轮廓点个数变化的水平集演化终止准则。该算法不仅明显提高了分割速度,且对噪声图像也能实现高效的分割。     

基于OTSU分割和HOG特征的行人检测与跟踪方法

传统的HOG算法针对整幅图像进行行人特征提取,大量的非人窗口计算必然降低检测的准确率和效率。为此,提出一种基于OTSU分割和HOG特征的行人检测与跟踪方法。利用OTSU算法以最佳阈值分割图像,在分割区域的基础上进行Canny边缘检测,通过边缘的对称性计算确定行人候选区,继而采用经PCA方法降维后的HOG特征和隐马尔可夫模型对行人候选区进行检测验证。最后,以确定的行人区域为跟踪窗口,利用CamShift算法跟踪行人。多组实验结果证明,本文方法的行人检测效率和精度均有所提高,跟踪性能稳定、可靠。     

结合稀疏编码和空间约束的红外图像聚类分割研究

提出了结合稀疏编码和空间约束的红外图像聚类分割新算法, 在稀疏编码的基础上融合聚类算法, 扩展了传统的基于K-means聚类的图像分割方法. 结合稀疏编码的聚类分割算法能有效融合图像的局部信息, 便于利用像素之间的内在相关性, 但是对于分割会出现过分割和像素难以归类的问题.为此, 在字典的学习过程中, 将原子的聚类算法引入其中, 有助于缩减字典中原子所属类别的数目, 防止出现过分割; 考虑到像素及其邻域像素具有类别属性一致性的特点, 引入了空间类别属性约束信息, 并给出了一种交替优化算法. 联合学习字典、稀疏系数、聚类中心和隶属度, 将稀疏编码系数同原子对聚类中心的隶属程度相结合, 构造像素归属度来判断像素所属的类别. 实验结果表明, 该方法能够有效提高红外图像重要区域的分割效果, 具有较好的鲁棒性. 关键词： 图像分割 稀疏编码 聚类 空间约束     

一种边缘提取的图像分割方法

图像分析技术从广义上来讲 ,是指从图像中提取有用的数据或其它信息。图像分割是图像分析技术的重要手段 ,在前人的理论基础上经过实验提出了一种边缘提取的图像分割方法———行扫描空间带通滤波法。该方法在电视图像的自动跟踪识别中取得了很好的效果 ,是一种较为可取的图像分割方法     

Automatized segmentation of photovoltaic modules in IR-images with extreme noise

Local electric defects may result in considerable performance losses in solar cells. Infrared thermography is an essential tool to detect these defects on photovoltaic modules. Accordingly, IR-thermography is frequently used in R&D labs of PV manufactures and, furthermore, outdoors in order to identify faulty modules in PV-power plants. Massive amount of data is acquired which needs to be analyzed. An automatized method for detecting solar modules in IR-images would enable a faster and automatized analysis of the data.However, IR-images tend to suffer from rather large noise, which makes an automatized segmentation challenging. The aim of this study was to establish a reliable segmentation algorithm for R&D labs. We propose an algorithm, which detects a solar cell or module within an IR-image with large noise. We tested the algorithm on images of 10 PV-samples characterized by highly sensitive dark lock-in thermography (DLIT). The algorithm proved to be very reliable in detecting correctly the solar module. In our study, we focused on thin film solar cells, however, a transfer of the algorithm to other cell types is straight forward.     

Infrared super-resolution imaging based on compressed sensing

The theoretical basis of traditional infrared super-resolution imaging method is Nyquist sampling theorem. The reconstruction premise is that the relative positions of the infrared objects in the low-resolution image sequences should keep fixed and the image restoration means is the inverse operation of ill-posed issues without fixed rules. The super-resolution reconstruction ability of the infrared image, algorithm’s application area and stability of reconstruction algorithm are limited. To this end, we proposed super-resolution reconstruction method based on compressed sensing in this paper. In the method, we selected Toeplitz matrix as the measurement matrix and realized it by phase mask method. We researched complementary matching pursuit algorithm and selected it as the recovery algorithm. In order to adapt to the moving target and decrease imaging time, we take use of area infrared focal plane array to acquire multiple measurements at one time. Theoretically, the method breaks though Nyquist sampling theorem and can greatly improve the spatial resolution of the infrared image. The last image contrast and experiment data indicate that our method is effective in improving resolution of infrared images and is superior than some traditional super-resolution imaging method. The compressed sensing super-resolution method is expected to have a wide application prospect.     

基于图像分割的去雾算法

针对雾霾条件下,所得的单幅图像出现降质现象,因而对视觉效果要求高的图像进行复原具有必要性。基于图像分割的去雾算法以暗通道先验模型为基础对大气光矢量A值的求取和透射率t(x,y)的处理方法实现改进。首先对单幅图像进行阈值分割找到天空区域,在所获取的天空区域部分结合skyline算法,可以找到精确的大气光矢量A值;进而对初始透射率t(x,y)采用改进的约束最小二乘方滤波进行优化,得到优化透射率t1(x,y),最后将所得的大气光矢量A值和优化透射率t1(x,y)利用大气光传输物理模型复原。改进算法的去雾结果具备保留细致的边缘细节,同时具有高效的去除图像噪声能力。实验结果表明,与he方法相对比,改进去雾算法的处理时间大程度缩短的同时,图像效果得到了提升。     

基于混合自适应小波基的织物疵点检测算法

为了改进具有复杂纹理织物的疵点检测效果,提出了一种基于混合自适应小波基的织物疵点检测算法,采用各自优化的自适应小波基实现对不同层织物图像的分解变换。首先对正常图像和其经一层小波分解后的低频子图像优化得到混合自适应小波基,然后用该小波基将织物疵点图像进行二层小波分解,最后采用阈值法对径向子图像进行分割得到检测结果。实验结果证明,本文提出的算法能有效实现疵点检测,具有较好的疵点分割和定位结果。