Three stage cervical cancer classifier based on hybrid ensemble learning with modified binary PSO using pretrained neural networks

ABSTRACT

Cervical cancer is one of the major challenges in developing nations like India.In recent years, a lot of research has been done todetect cervical cancer at an early stage through the pap-smear test, human papillomavirus test (HPV), etc. In this study, we have proposed athree-stage cervical cancer classifier to classify cervical cells among normal and abnormal cells using a hybrid ensemble classifier based onfeatures extracted using pre-trained neural networks. Furthermore, this work extends to classify the cells among different levels of dysplastic mainly mild, moderate and severe. The accuracy achieved for 2-class classification among normal and abnormal cells is up to 100% while for 4-class classification among normal, mild, moderate and severe dysplastic cells is up to 98.91% and 99.12% for new and old Herlev university hospital datasets respectively.     

Constructing a new fuzzy classifier based on hierarchical fuzzy entropy

Abstract

In an earlier work, Lee et al. (Lee et al., 2001) presented a simple and fast fuzzy classifier that employed fuzzy entropy to evaluate pattern distribution information in a pattern space. In this paper, we extend his work to propose a new fuzzy classifier based on hierarchical fuzzy entropy (FC‐HFE). We retained the main parts of the original structure and modified some methods (e.g., methods for deciding the number of intervals in each dimension and for assigning class labels). In addition, the hierarchical fuzzy entropy is proposed for partitioning the decision region. The proposed FC‐HFE improves classification accuracy and overcomes some of the drawbacks in the Lee et al method (Lee et al., 2001). The simulation results show that the classification rate of the proposed FC‐HFE is better than earlier methods.     

Structure optimization of rotor cage blades for turbo air classifier based on entropy production analysis

Entropy production theory based on the second law of thermodynamics was introduced for evaluating the flow field inside the turbo air classifier. The three new types of rotor cage with the wedge blades, the inverted wedge blades and the spindle blades were designed, and the flow field and the classification performance of the classifiers were investigated. The results show that, compared to the rectangular blades, the productions of total entropy, turbulent entropy and wall entropy of the wedge blades are reduced by 17.3%, 25.86% and 3.34%, respectively. The corresponding effective airflow area increases by 7.5%, and the residence time of 5 μm particle is shorten by 16%. The classifier with the wedge blades has smaller cut size and higher classifying sharpness. The results validate that the turbulent entropy generation can be an indicator for monitoring the overall flow field and the classifiers’ performance.     

An ensemble model considering health index based classification for remaining useful life prediction

Accurate prediction of remaining useful life (RUL) plays an important role in the formulation of maintenance strategies. However, due to the diversity of the failure mode of equipment, there are significant differences between the degradation data, which greatly affects the accuracy of RUL prediction. In this case, an ensemble prediction model considering health index-based (HI-based) classification is proposed in this paper. Firstly, the stacked autoencoder (SAE) is employed to construct the HI. Then, the time window is used to sequentially process the HI sequence, so that many data segments with the same length can be achieved. To differentiate the data with the similar degradation process, K-means and Xgboost are selected to construct offline and online data classification models respectively. Finally, according to the results of the data classification, the ensemble model that integrates multiple machine learning methods are separately trained and then adaptively used for RUL prediction. In addition, integrating multiple methods helps to improve the generalization ability of the model. The NASA C-MAPSS dataset is applied to verify the effectiveness of the proposed method, and the results show that the proposed method achieves a higher prediction accuracy and shorter computational time than other existing prediction models.     

An efficient gait recognition based on a selective neural network ensemble

The neural network ensemble is a learning paradigm where a collection of neural networks is trained for the same task. Generally, the ensemble shows better generalization performance than a single neural network. In this article, a selective neural network ensemble is applied to gait recognition. The proposed method selects some neural network based on the minimization of generalization error. Since the selection rule is directly incorporated into the cost function, we can obtain adequate component networks to constitute an ensemble. Experiments are performed with the NLPR database to show the performance of the proposed algorithm. © 2008 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 18, 237–241, 2008; Published online in Wiley InterScience (www.interscience.wiley.com).     

An efficient ensemble of radial basis functions method based on quadratic programming

Radial basis function (RBF) surrogate models have been widely applied in engineering design optimization problems to approximate computationally expensive simulations. Ensemble of radial basis functions (ERBF) using the weighted sum of stand-alone RBFs improves the approximation performance. To achieve a good trade-off between the accuracy and efficiency of the modelling process, this article presents a novel efficient ERBF method to determine the weights through solving a quadratic programming subproblem, denoted ERBF-QP. Several numerical benchmark functions are utilized to test the performance of the proposed ERBF-QP method. The results show that ERBF-QP can significantly improve the modelling efficiency compared with several existing ERBF methods. Moreover, ERBF-QP also provides satisfactory performance in terms of approximation accuracy. Finally, the ERBF-QP method is applied to a satellite multidisciplinary design optimization problem to illustrate its practicality and effectiveness for real-world engineering applications.     

Lymph node detection method based on multisource transfer learning and convolutional neural network

Recently years, convolutional neural networks (CNNs) have proven to be powerful tools for a broad range of computer vision tasks. However, training a CNN from scratch is difficult because it requires a large amount of labeled training data, which remains a challenge in medical imaging domain. To this end, deep transfer learning (TL) technique is widely used for many medical image tasks. In this paper, we propose a novel multisource transfer learning CNN model for lymph node detection. The mechanism behind it is straightforward. Point-wise (1 × 1) convolution is used to fuse multisource transfer learning knowledge. Concretely, we view the transferred features as priori domain knowledge and 1 × 1 convolutional operation is implemented after pre-trained convolution layers to adaptively combine the transfer information for target task. In order to learn non-linear transferred features and prevent over-fitting, we present an encode process for the pre-trained convolution kernels. At last, based on convolutional factorization technique, we train the proposed CNN model and the encoder process jointly, which improves the feasibility of our approach. The effectiveness of the proposed method is verified on lymph node (LN) dataset: 388 mediastinal LNs labeled by radiologists in 90 patient CT scans, and 595 abdominal LNs in 86 patient CT scans for LN detection. Our method demonstrates sensitivities of about 85%/71% at 3 FP/vol. and 92%/85% at 6 FP/vol. for mediastinum and abdomen respectively, which compares favorably to previous methods.     

Experimental and CFD study on a cyclonic classifier with new flow pattern

Physical principle of conventional top-inlet classifier (CTC) with reverse-flow pattern leads to the heavily fine particles entrainment in coarse fraction. Present work concentrates on the flow-field design for less downward airflow at near-wall region of the classifier. A new middle-inlet classifier (NMC) is proposed and analyzed using computational fluid dynamics (CFD) method and powder classification experiments. The results showed that new flow pattern characterized by a pair of vortexes was created in the new classifier. The upper vortex with 80% of the total air volume moves upward and forms the washing effect at near-wall region, which effectively reduces the fine particles entrainment in coarse fraction. The downer vortex with reverse-flow pattern discharges the coarse particles timely. The radial centrifugal sedimentation combined with the axial counter-current washing effect dominates the particle classification in the NMC. Compared to the CTC, classification accuracy index of the NMC with double-vortex averagely increases by 27% with a pressure drop reduction of more than 38%. This work offers a new principle for high-efficiency particle classification and new strategy for improving the classification performance of turbo air classifiers and hydrocyclones.     

机器学习分布式网络传输异常数据智能检测方法

对于链路状态数据库的网络传输异常数据检测存在检测数据不完整、较为敏感、检测效率差的问题,提出基于机器学习的分布式网络传输异常数据智能检测方法,通过K最近邻分簇算法对分布式网络节点实施分簇,利用贝叶斯分类算法检测簇头是否出现异常;确定异常簇后,选取小波阈值降噪方法对异常簇内数据进行降噪处理,在此基础上,采用遗传算法检测降...     

多维数据雷达图和模糊推理的分类器研究

提出了一种新颖的基于多维数据雷达图表示原理结合模糊推理规则的自动分类器设计方法.该方法首先采用多元分析中的雷达图表示多维数据,建立已知类模板,然后应用模糊推理方法识别未知类的雷达图形,应用模板匹配法确定其归属,从而完成自动分类.基于混合油品132个测试数据的实验表明,此分类器可以将全部合理混合油品做出正确分类,且具有良好的分类效果和分类精度.     

融合目标检测与距离阈值模型的露天矿行车障碍预警

针对当前行车预警方法无法适应露天矿非结构化道路问题,本文提出一种融合目标检测和障碍距离阈值的预警方法。首先根据露天矿障碍特点改进原有的Mask R-CNN检测框架,在骨架网络中引入扩张卷积,在不缩小特征图的情况下扩大感受野范围保证较大目标的检测精度。然后,根据目标检测结果构建线性距离因子,表征障碍物在输入图像中的深度信息,并建立SVM预警模型。最后为了保证预警模型的泛化能力采用迁移学习的方法,在COCO数据集中对网络进行预训练,在文中实地采集的数据集中训练C5阶段和检测层。实验结果表明,本文方法在实地数据检测中精确率达到98.47%,召回率为97.56%,人工设计的线性距离因子对SVM预警模型有良好的适应性。     

基于多级关联信号树的高效可重构网包分类方法研究

针对高速网络中包分类严重影响路由系统性能提升的问题,进行了深入的实验性研究.针对传统包分类算法通过扩展规则搜索空间实现匹配,占用内存空间大,功耗高,吞吐率低的问题,研究了基于多级关联信号树的高效可重构网包分类方法.通过分析网包分类规则集合特点,提出了一种基于多级关联信号树的逻辑匹配结构,从中抽取出三类可重构的粗粒度网包分类基本计算单元——固定型匹配器、前缀型匹配器和范围型匹配器,用这三类匹配器构成了一个可重构网包分类阵列,通过配置匹配器的重构功能单元(RFU)层和匹配器之间的互联结构——重构互联网络(RIN)层实现了高速分类计算.该方法能够有效节省内存空间,降低功耗,大幅提升匹配速度.为了验证算法性能,在Xilinx公司的Virtex-6(model:XC6 VSX475T)芯片上进行仿真实验,实验结果表明该算法吞吐率可以达到100Gbp以上.     

基于阈值判断的自适应中值滤波算法

针对标准的中值滤波算法在去除噪声与保留图像细节方面难以取舍的缺陷,在自适应中值滤波算法的基础上提出了一种改进的基于噪声点检测的自适应中值滤波算法.该算法在进行噪声点检测时采用了一种阈值判断法,充分利用了当前像素点与邻域像素点的灰度值之间的关系.结果表明,在噪声浓度较高时仍然可以区分噪声点与边缘点,滤波的同时有效地保护了图像的细节.     

基于深度学习检测器的多角度人脸关键点检测

针对人脸关键点检测(人脸对齐)在应用场景下的速度和精度需求,首先在SSD基础之上融合更多分布均匀的特征层,对人脸框坐标进行级联预测,形成对于多尺度人脸信息均具有更加鲁棒响应的深度学习检测器MR-SSD。其次在局部二值特征LBF的级联形状回归方法基础上,提出了基于面部像素差值的多角度初始化算法。采用端正人脸正负90°倾斜范围内的五组特征点形状进行初始化,求取每组回归后形状的眼部特征点像素均方差值并以最大者对应方案作为最终回归形状,从而实现对多角度倾斜人脸优异的拟合效果。本文所提出的最优架构可以实时获得极具鲁棒性的人脸框坐标并且可实现对于多角度倾斜人脸的关键点检测。     

基于随机卷积特征和集成超限学习机的快速SAR目标识别

深度卷积神经网络在目标检测与识别等方面表现出了优异性能,但将其用于SAR目标识别时,较少的训练样本和深度模型的优化设计是必须解决的两个问题。本文设计了一种结合二维随机卷积特征和集成超限学习机的SAR目标识别算法。首先,随机生成具有不同宽度的二维卷积核,对输入图像进行卷积与池化操作,提取随机卷积特征向量。其次,为提高分类器的泛化能力,并降低训练时间,基于集成学习思想对提取的卷积特征进行随机采样,然后采用超限学习机训练基分类器。最后,通过投票表决法对基分类器的分类结果进行集成。采用MSTAR数据集进行了SAR目标识别实验,实验结果表明,由于采用的超限学习机具有快速训练能力,训练时间降低了数十倍,在无需进行数据增强的情况下,分类精度与采用数据增强和多层卷积神经网络的深度学习算法相当。提出的算法具有实现简单、需要调整参数少等优点,采用集成学习思想提高了分类器的泛化能力。

     

基于深度学习的合作目标靶球检测

为了解决复杂场景下激光跟踪仪对合作目标靶球的精确识别难题,提出了基于深度学习的合作目标靶球高效检测方法。首先分析了合作目标靶球的图像特征,然后采用改进的YOLOv2模型,针对合作目标靶球多尺度与小目标占比多的特点,提出了一种基于注意力机制的改进方法,同时为提高网络模型对复杂背景的抗干扰能力,提出了一种数据增强方法。测试结果表明,所提出的基于注意力机制与数据增强的改进YOLOv2模型对复杂背景的抗干扰能力较强,且对合作目标靶球的检测精度有显著提高,在合作目标靶球测试集上的检测准确率达到92.25%,能够有效满足激光跟踪仪在大型装置精密装配过程中的目标检测精度需求。     

基于YOLOv4-tiny的溜筒卸料煤尘检测方法

煤炭港在使用装船机的溜筒卸载煤的过程中会产生扬尘,港口为了除尘,需要先对粉尘进行检测。为解决粉尘检测问题,本文提出一种基于深度学习(YOLOv4-tiny)的溜筒卸料煤粉尘的检测方法。利用改进的YOLOv4-tiny算法对溜筒卸料粉尘数据集进行训练和测试,由于检测算法无法获知粉尘浓度,本文将粉尘分为四类分别进行检测,最后统计四类粉尘的检测框总面积,通过对这些数据做加权和计算近似判断粉尘浓度大小。实验结果表明,四类粉尘的检测精度(AP)分别为93.98%、93.57%、80.03%和57.43%,平均检测精度(mAP)为81.27%,接近YOLOv4的83.38%,而检测速度(FPS)为25.1,高于YOLOv4的13.4。该算法较好地平衡了粉尘检测的速率和精度,可用于实时的粉尘检测以提高抑制溜筒卸料产生的煤粉尘的效率。     

综合边缘检测和区域生长的红外图像分割方法

针对红外图像的特点,提出了一种综合应用边缘检测和区域生长方法的图像分割方法。其思路为:先对图像进行边缘提取,得到边缘像素点集;然后利用该点集的平均灰度和目标区域的连通性作为生长判决条件,采用区域生长法实现图像分割。仿真结果表明,该方法能快速准确有效地实现红外图像分割,避免了单独使用边缘提取或区域生长法进行图像分割时的典型分割错误。     

Shi-Tomasi角点区域的拷贝图像检测

针对由于裁剪、翻转和旋转等产生的图像拷贝问题,提出一种Shi-Tomasi角点的拷贝检测算法.先使用Shi-Tomasi角点检测算法提取图像的局部角点;然后在以Shi-Tomasi角点为中心的圆环区域内计算特征向量的协方差描述子(多特征融合);最后通过协方差描述子的相似性度量来检测圆环区域的相似性,并以此判断检测图像是否为原图像的拷贝.实验结果证明,该检测算法对图像的裁剪、旋转等攻击具有较好的鲁棒性.     

一种基于线谱特征函数提取LOFAR图线谱的方法

水声信号被动检测中广泛使用LOFAR图对接收信号进行处理和分析。针对LOFAR图中线谱信号检测问题,根据线谱信号特征设计特征函数,提出频域滑动窗线谱特征累积检测法。该方法在频率轴移动观察窗,用多步决策算法计算每个观察窗的最优解,得到最优路径,如果最优路径特征值大于阈值,则累积LOFAR图像素点被该最优路径经过的次数,次数越多对应点为线谱点的概率越大。仿真研究表明,该方法对频率时变、低信噪比的线谱信号具有良好的检测能力,可实现多根线谱的增强与检测。海试数据处理结果证明了该方法的可行性和稳健性。该算法对于辐射线谱信号的水下目标远距离探测识别有较高的参考价值。