胸部CT图像中孤立性肺结节良恶性快速分类

目的：为突破医学影像诊断学依据医学征象进行定性诊断准确度不高的瓶颈,针对胸部CT图像中孤立性肺结节（SPN）定性诊断问题,寻求能够有效表示SPN病理特性的图像特征,实现快速准确地SPN良恶性计算机辅助诊断系统。方法：首先,采取交互式分割方法从胸部CT图像中提取出SPN;其次,直接计算SPN图像的多分辨率直方图得到768维空间信息特征样本集;然后,充分利用具有处理高维数据集优势的支持向量机（SVM）构造SPN良恶性分类器;最后,通过测试样本集对经训练后的SVM分类器进行测试以评价分类性能。结果：经214例病例实验结果表明：240个SPN图像的768维特征计算所用时间为4.83秒,SVM分类器训练测试所用时间为2.24秒,敏感性71.33%,特异性70%,准确度71.67%,接受者操作特性曲线(ROC)下面积(AUC) 0.7864。结论：该系统提取的高维图像空间信息特征能够有效表示SPN特性;没有考虑医学征象进行SPN定性诊断的准确度就达到了71.67%,同时分类速度比传统纹理算法提高了近50倍,为医学影像学解决SPN定性诊断问题提供了便捷、客观的辅助手段。     

基于图像特征统计分析的PCB焊点检测方法

提出了一种基于图像特征统计分析的炉后焊点检测方法,以提高在线自动光学检测系统的检测性能和可操作性.提出双阈值的AdaBoost算法用于设计分类器,在训练的同时进行最优特征选择和分类器的增强,实现了焊点图像特征的自动提取和检测参数的自动设定.采用分类和回归树方法将焊点缺陷决策方法优化为一棵二叉决策树,提高了检测速度.实验结果表明,该方法训练速度较快,可以满足实际生产需要.与目前已经实用化的图像对比算法和图像分析算法相比,在保持现有检测速度基本不变的情况下,该方法的检测精度更高.     

近似稀疏正则化的红外图像超分辨率重建

针对红外图像分辨率低、受噪声影响严重等问题,引入近似稀疏正则化和K-奇异值分解（K-SVD）法,提出了基于近似稀疏表示模型的红外图像超分辨率重建方法。考虑到红外图像受到噪声污染,首先建立了稳健近似稀疏表示模型。针对已有字典训练方法时间消耗巨大问题,在假定低分辨率图像空间和高分辨率图像空间具有相似流形的前提下,联合近似稀疏表示模型和K-SVD方法,提出近似稀疏约束的基于K-SVD的高低分辨率字典对学习算法。最后,通过高分辨字典和对应的红外图像群稀疏表示系数重建得到高分辨率的红外图像。为了验证算法的性能,对提出的算法与稀疏性正则化的图像超分辨模型（SRSR）和Zeyde算法进行了实验比较。结果表明,本文方法能够较好地减少红外图像中的噪声,同时获得更好的超分辨率重建效果。     

基于小波矩和模糊核聚类的示功图诊断方法研究

提出了基于小波矩特征和模糊核聚类算法的示功图故障诊断方法.通过边缘检测和形态学细化的方法完成示功图的图像分割,采用极坐标下小波不变矩算法提取示功图的形状特征,通过参数选择确定12个小波矩特征量,将特征量输入到模糊核聚类分类器中进行故障类型的分类识别,得到了良好的实验效果,验证了该算法对于示功图故障诊断的有效性.     

采用目标背景建模的毫米波弱小目标检测

基于被动毫米波成像特性,提出了改进的稀疏表示——圆周中心差(ISR-CSCD)算法来解决被动毫米波图像中弱小目标与背景区分度较弱,目标可提取特征较少的问题。该算法通过改进稀疏表示方法完成背景抑制与目标增强。依据目标与周围背景特征先验,提出了圆周中心差背景抑制算法对检测图像进行背景抑制。然后,融合改进稀疏表示方法和圆周中心差背景抑制算法的结果得到抑制了背景的目标增强图像。最后,基于恒虚警率的检测方法完成了弱小目标的检测。对不同场景下的毫米波图像进行了实验检测,结果表明,与主流算法图像稀疏表示(SR)法、鲁棒规则核回归牛顿算法(NRRKR),空时联合分类稀疏表示算法(STCSR)和累积中心与周边差异测量算法(ACSDM)相比,ISR-CSCD算法具有更低的虚警率、更高的检测精度、更强的鲁棒性。对各种虚警率、信噪比之下的毫米波弱小目标检测结果显示,ISR-CSCD检测率相对于其它算法平均提高了约15%。     

一种自适应的分布式肺结节检测方法

提出了一种自适应的分布式肺结节检测方法。该方法由肺结节检测代理和中心管理器两部分组成。中心管理器负责数据收集以及调节肺结节检测代理。肺结节检测代理完成肺实质分割、多尺度增强提取感兴趣区域(ROI)、ROI的特征提取以及基于Mahalanobis距离的ROI分类4部分。在理想情况下,推导了控制中心管理器的自适应调节函数;在现实的分布式系统中,设计了基于遗传算法的自适应分布式肺结节检测算法。实验结果表明,所提出的自适应分布式肺结节检测方法能随着检测数据的增加不断调节分类算法中的阈值,提高分类检测的性能;同时,使用该方法检测肺结节具有较低的误诊率与漏诊率,能够辅助医生进行诊断。     

基于ROI的海洋卫星遥感图像压缩算法

针对海洋监视卫星遥感图像的特点,提出了一种基于感兴趣区域(ROI: Regions Of Interest)的自适应海洋遥感图像近无损压缩算法。对图像进行了形态Harr小波最大提升后,在一个分辨率低的高频子带中利用阈值和八邻域连通分析方法检测出目标,使用外接矩形与环面的交集来描述ROI,其他分辨率级的高频子带的ROI通过Mosaic放大得到。高频子带中的ROI采用Rice无损熵编码方法,非ROI进行比特平面编码。低频子带采用DPCM和Rice结合的无损编码方法。实验结果表明,该算法能有效地划分ROI,在中低比特率情况下,获得了比JPEG2000更好的重构图像质量,且没有明显的ROI分割痕迹。本文算法的计算复杂度较低,易于硬件实现,而且还具有自适应性和数据包独立的优点,适合于海洋监视卫星遥感图像近无损压缩应用。     

基于张量表示的高光谱图像目标检测算法

高光谱图像目标检测是当前一个研究热点。其在军事和民用方面的应用越来越广泛。为了能同时利用高光谱图像数据的空谱信息,本文提出一种新的基于张量表示的高光谱图像目标检测算法。算法使用CP张量分解技术和张量块分解(BTD)分别对高光谱数据进行盲源分析,提取了有效的局部图像块空谱特征,建立了一个基于稀疏表示和协作表示的检测模型,针对多种类型背景复杂的场景数据进行实验,并与当前流行的目标检测算法进行比较。从可视化检测结果来看,本文算法在复杂背景和强噪声环境下,有效提取了空谱特征,对背景具有较好的抑制能力,检测的目标显著。此外,本文从接收机操作曲线(ROC)和ROC曲线下面积(AUC)等定量指标分析算法性能。以较为流行的Sandiego图像为例,在10%的虚警率下,本文算法取得90%的检测精度,AUC大于0.95。本文算法相较几种流行算法而言具有较高的检测精度,更强的鲁棒性。     

ROI的海洋监视卫星遥感图像压缩算法

针对海洋监视卫星遥感图像的特点,提出了一种基于感兴趣区域(ROI)的自适应海洋遥感图像近无损压缩算法.对图像进行了形态Harr小波最大提升后,在一个分辨率低的高频子带中利用阈值和八邻域连通分析方法检测出目标,使用外接矩形与环面的交集来描述ROI,其他分辨率高频子带的ROI通过Mosaic放大得到.高频子带中的ROI采用Rice无损熵编码方法,非ROI进行比特平面编码.低频子带采用DPCM和Rice结合的无损编码方法.实验结果表明,该算法能有效地划分ROI,且没有明显的ROI分割痕迹.在中低比特率时,本文算法的峰值信噪比(PSNR)比JPEG2000的高2～5 dB,在高比特率时,两者的PSNR均在45 dB以上.本文算法计算复杂度低,易于硬件实现,且具有自适应性和数据包独立的优点,适用于海洋监视卫星遥感图像的近无损压缩.     

一种改进的均值移位红外目标跟踪算法

为了增强复杂背景条件下红外目标跟踪的稳健性,提出了一种改进的均值移位目标跟踪算法。该算法融合了均值移位的梯度匹配搜索策略的优势与基于特征分类跟踪算法强鲁棒性的优点,建立了灰度似然比加权的核直方图目标表征模型。模型中加入了目标与局部背景灰度特征的似然比作为原始核直方图的权值,应用该模型的均值移位算法能够进一步提高目标像素灰度的移位权重,有效抑制背景干扰,进而提高低对比度目标跟踪的稳健性。同时,基于跟踪复杂度估计提出了目标遮挡情况下的模型更新判别准则,提高了算法的自适应性能。实测红外目标跟踪实验表明了该算法简单、有效。     

Microscopic skin laceration segmentation and classification: A framework of statistical normal distribution and optimal feature selection

Among precision medical techniques, medical image processing is rapidly growing as a successful tool for cancer detection. Skin cancer is one of the crucial cancer types. It is identified through computer vision (CV) techniques using dermoscopic images. The early diagnosis skin cancer from dermoscopic images can be decrease the mortality rate. We propose an automated system for skin lesion detection and classification based on statistical normal distribution and optimal feature selection. Local contrast is controlled using a brighter channel enhancement technique, and segmentation is performed through a statistical normal distribution approach. The multiplication law of probability is implemented for the fusion of segmented images. In the feature extraction phase, optimized histogram, optimized color, and gray level co‐occurrences matrices features are extracted and covariance‐based fusion is performed. Subsequently, optimal features are selected through a binary grasshopper optimization algorithm. The selected optimal features are finally fed to a classifier and evaluated on the ISBI 2016 and ISBI 2017 data sets. Classification accuracy is computed using different Support Vector Machine (SVM) kernel functions, and the best accuracy is obtained for the cubic function. The average accuracies of the proposed segmentation on the PH2 and ISBI 2016 data sets are 93.79 and 96.04%, respectively, for an image size 512 × 512. The accuracies of the proposed classification on the ISBI 2016 and ISBI 2017 data sets are 93.80 and 93.70%, respectively. The proposed system outperforms existing methods on selected data sets.     

Adaptive segmentation of digital mammograms through reinforcement learning

1Introduction Computer aideddiagnosisofmammo grams[12]usestheoutputofthecomputerizeda nalysisofmammogramsasa"secondopinion"to assistaradiologistindetectinglesionsandin makingdiagnosisdecisions.Automatedextrac tionofthebreastregionfrommammogramsis animportantpreprocessingstepforthecomput er aideddiagnosis.Byexcludingthebackground region,furthersearchoperationforabnormali tiesarelimitedtothebreastregion.Withoutun dueinfluencefromthebackgroundregion,the reliabilityandefficiencyoffurtheranalysisca…     

发动机异常检测多目标优化方法

针对柴油发动机异常检测中的特征选择和分类器参数与检测精度之间的耦合关系,提出了一种基于非支配排序粒子群优化的柴油发动机异常检测封装式多目标同步优化方法.利用双树复小波包的分解与重构,对发动机振动信号进行时域、频域和时频域多角度特征提取,构建了较完备的特征参数集,分析了故障诊断中特征选择与分类器参数优化对检测精度的影响,运用非支配排序粒子群优化算法对多个优化目标进行协调和折衷处理,同时追求特征参数子集维数最小化和分类正确率最大化.实验数据分析表明,该方法能够寻找出最优的特征子集和分类器参数,提高柴油发动机异常检测的精度和效率.     

基于高阶统计特征实值阴性克隆选择算法的轴承故障检测

为解决轴承故障检测领域中异常样本数据不易收集的现实应用问题,提出一种基于实值阴性克隆选择算法(Real-valued negative clone selection,RNCS)的一类轴承故障检测模型。该模型只需要正常样本数据进行训练,利用改进的RNCS生成故障检测器集合以此实现轴承故障检测。该算法通过引入自适应变异算子和克隆成熟度判定算子,能够提高原有算法抗体的检测能力并加快算法收敛速度。为解决因高阶统计特征(Higher order statistics,HOS)信息繁多而无法有效实现智能检测的不足,模型利用HOS特征矩阵分解的奇异值谱为特征进行检测,该方法不仅有效地减少了数据维度及训练时间,同时还降低了噪声影响提高了检测性能。试验中对不同参数选择及不同正常训练样本个数情况下的检测器性能进行了分析,不同检测器个数之间的性能比较也在试验中给出。将建议的方法同原有算法进行比较,试验结果验证了设计思想的正确性和算法的高效检测性能。     

人脸特征选择中的SVM泛化误差估计

根据统计学习理论,特征选择可以通过有效的特征搜索策略最小化某个预测泛化误差及其它相关性能来实现。本文研究通过递归特征排除法（Recursive Feature Elimination,RFE）最小化SVM VC留一法(Leave-One-Out, LOO)误差或支持向量span误差估计选择优化特征子集问题,并将最小化VC LOO误差或支持向量span误差估计作为Wrapper特征选择模型的选择判据。人脸识别实质是稀疏超高维空间、典型的小样本模式识别问题。解决这类问题的关键在于如何获得对分类有意义的特征。将特征选择与分类器设计结合,理论上优于传统的特征提取或特征选择方法。为此,本论文将WT和KPCA作为过滤模型（Filter）,最小化SVM泛化误差估计作为封装模型（Wrapper）,结合这两种模型的优势提出人脸特征选择及识别的新框架。并在UMIST人脸数据库上进行了相应的实验,结果显示提出的特征选择方法和特征搜索策略及人脸特征选择构架有效可行。     

A new hybrid pattern recognition scheme for automatic discrimination of power quality disturbances

In this paper an intelligent method for automatic detection of Power Quality Disturbances (PQDs) is presented. The proposed automatic scheme is noticed to retaining informative features or eliminating redundant features simultaneously. This paper presents an effective method, for extracting features, so-called “integrated approach”, using integration of Discrete Wavelet Transform and Hyperbolic S Transform. Moreover, a new efficient feature selection method namely Orthogonal Forward Selection by incorporating Gram Schmidt procedure and forward selection is applied for selection of the best subset features. Some different classifiers are empirically compared in order to determine the best classifier. In this automatic scheme, the variable parameters of classifiers are optimized using a powerful method namely Particle Swarm Optimization. The sensitivity of the proposed method under noisy conditions has been investigated. The average rate of correct classification using the proposed hybrid scheme for automatic discrimination of PQDs in various noise conditions 99.55% is obtained.     

Microscopic abnormality classification of cardiac murmurs using ANFIS and HMM

Auscultation of heart dispenses identification of the cardiac valves. An electronic stethoscope is used for the acquisition of heart murmurs that is further classified into normal or abnormal murmurs. The process of heart sound segmentation involves discrete wavelet transform to obtain individual components of the heart signal and its separation into systole and diastole intervals. This research presents a novel scheme to develop a semi‐automatic cardiac valve disorder diagnosis system. Accordingly, features are extracted using wavelet transform and spectral analysis of input signals. The proposed classification scheme is the fusion of adaptive‐neuro fuzzy inference system (ANFIS) and HMM. Both classifiers are trained using the extracted features to correctly identify normal and abnormal heart murmurs. Experimental results thus achieved exhibit that proposed system furnishes promising classification accuracy with excellent specificity and sensitivity. However, the proposed system has fewer classification errors, fewer computations, and lower dimensional feature set to build an intelligent system for detection and classification of heart murmurs.     

基于非参数变换的尿沉渣细胞图像识别方法

首先提出一种改进的局部秩变换(D-LRT)方法,通过计算图像像素的局部秩进行图像变换,根据阈值选取实现低对比度和局部模糊的尿沉渣细胞图像分割。然后提出一种局部直方图统计(LHS)方法提取细胞图像平移、旋转、光照不变特征。该方法通过高斯模糊获取不同尺度下的细胞高斯模糊图像,并利用RIUP-LBP算子获取细胞高斯模糊图像的RIUP-LBP特征图谱,基于距离变换采取由内而外的方式,分层统计细胞高斯模糊图像及其RIUP-LBP特征图谱的直方图,将所有直方图串联作为细胞图像的LHS局部特征。还同时将LHS特征结合图像的几何特征、Harris角点及灰度共生矩阵特征,从局部和全局角度构造尿沉渣细胞图像的特征向量。最后采用支持向量机(SVM)对7类典型尿沉渣细胞图像进行分类。实验结果表明:本文提出的D-LRT方法在低对比度和局部模糊的细胞图像分割中完整度明显提高,提出的LHS方法可以有效地提取7类尿沉渣细胞图像特征,7类细胞图像识别的平均准确率可达到93.0%,平均召回率可达93.2%。     

A novel fuzzy c-means method for classifying heartbeat cases from ECG signals

This study proposes a simple and reliable method termed the fuzzy c-means method for classifying the heartbeat cases from electrocardiogram (ECG) signals. The proposed method has the advantages of good detection results, no complex mathematic computations, low memory space and low time complexity. The FCMM can accurately classify and distinguish the difference between normal heartbeats and abnormal heartbeats. Classifying the heartbeat cases from ECG signals consists of four main procedures: (i) Procedure-DOM for detecting QRS waveform using the Difference Operation Method; (ii) qualitative features stage (Procedure-ROM) for qualitative feature selection using the Range-Overlaps Method on ECG signals; (iii) Procedure-CCC is used to compute the cluster center for each class; and (iv) Procedure-HCD is used to determine the heartbeat case for the patient. The experiments show that the sensitivities were 98.28%, 90.35%, 86.97%, 92.19%, and 94.86% for NORM, LBBB, RBBB, VPC and APC, respectively. The total classification accuracy was approximately 93.57%.     

Control chart pattern recognition using RBF neural network with new training algorithm and practical features

The control chart patterns are the most commonly used statistical process control (SPC) tools to monitor process changes. When a control chart produces an out-of-control signal, this means that the process has been changed. In this study, a new method based on optimized radial basis function neural network (RBFNN) is proposed for control chart patterns (CCPs) recognition. The proposed method consists of four main modules: feature extraction, feature selection, classification and learning algorithm. In the feature extraction module, shape and statistical features are used. Recently, various shape and statistical features have been presented for the CCPs recognition. In the feature selection module, the association rules (AR) method has been employed to select the best set of the shape and statistical features. In the classifier section, RBFNN is used and finally, in RBFNN, learning algorithm has a high impact on the network performance. Therefore, a new learning algorithm based on the bees algorithm has been used in the learning module. Most studies have considered only six patterns: Normal, Cyclic, Increasing Trend, Decreasing Trend, Upward Shift and Downward Shift. Since three patterns namely Normal, Stratification, and Systematic are very similar to each other and distinguishing them is very difficult, in most studies Stratification and Systematic have not been considered. Regarding to the continuous monitoring and control over the production process and the exact type detection of the problem encountered during the production process, eight patterns have been investigated in this study. The proposed method is tested on a dataset containing 1600 samples (200 samples from each pattern) and the results showed that the proposed method has a very good performance.