Segmentation of ultrasound images by using a hybrid neural network

A hybrid neural network is presented for the segmentation of ultrasound images.

Feature vectors are formed by the discrete cosine transform of pixel intensities in region of interest (ROI). The elements and the dimension of the feature vectors are determined by considering only two parameters: The amount of ignored coefficients, and the dimension of the ROI.

First-layer-nodes of the proposed hybrid network represent hyperspheres (HSs) in the feature space. Feature space is partitioned by intersecting these HSs to represent the distribution of classes. The locations and radii of the HSs are found by the genetic algorithms.

Restricted Coulomb energy (RCE) network, modified RCE network, multi-layer perceptron and the proposed hybrid neural network are examined comparatively for the segmentation of ultrasound images.     

Medical image segmentation with transform and moment based features and incremental supervised neural network

In this study, a novel incremental supervised neural network (ISNN) is proposed for the segmentation of medical images. Performance of the ISNN is investigated for tissue segmentation in medical images obtained from various imaging modalities. Two feature extraction methods based on transform and moments are comparatively investigated to segment the tissues in medical images. Two-dimensional (2D) continuous wavelet transform (CWT) and the moments of the gray-level histogram (MGH) are computed in order to form the feature vectors of ultrasound (US) bladder and phantom images, X-ray computerized tomography (CT) and magnetic resonance (MR) head images. In the 2D-CWT method, feature vectors are formed by the intensity of one pixel of each wavelet-plane of different energy bands. The MGH represents the tissues within the sub-windows by using the spatial variation of image intensities. In this study, the ISNN and Grow and Learn (GAL) network are employed for the segmentation task. It is observed that the ISNN has significantly eliminated the disadvantages of the GAL network in the segmentation of the medical images.     

Counter-Propagation Neural Networks for Molecular Sequence Classification: Supervised LVQ and Dynamic Node Allocation

A modified counter-propagation (CP) algorithm with supervised learning vector quantizer (LVQ) and dynamic node allocation has been developed for rapid classification of molecular sequences. The molecular sequences were encoded into neural input vectors using an n–gram hashing method for word extraction and a singular value decomposition (SVD) method for vector compression. The neural networks used were three-layered, forward-only CP networks that performed nearest neighbor classification. Several factors affecting the CP performance were evaluated, including weight initialization, Kohonen layer dimensioning, winner selection and weight update mechanisms. The performance of the modified CP network was compared with the back-propagation (BP) neural network and the k–nearest neighbor method. The major advantages of the CP network are its training and classification speed and its capability to extract statistical properties of the input data. The combined BP and CP networks can classify nucleic acid or protein sequences with a close to 100% accuracy at a rate of about one order of magnitude faster than other currently available methods.     

Image-based quality monitoring system of limestone ore grades

In this study, an image analysis-based ore quality monitoring system was developed. The study was conducted at a limestone mine located in India. The samples were collected based on a stratified random sampling method, and images of these samples were taken in a simulated environment in a laboratory. The image preprocessing and segmentation were performed using different segmentation methods to extract morphological, colour and textural features. A total of 189 features was extracted during this study. Principal components analysis was conducted to reduce the feature vector for modeling purposes. Five principal components, which were extracted from the feature vectors, captured 95% of the total feature variance. A neural network model was used as a mapping function for ore grade prediction. The five principal components were used as input, and four grade attributes of limestone (CaO, Al₂O₃, Fe₂O₃ and SiO₂) were used as output. The developed model was then used for day to day quality monitoring at 3 different face locations of the mine. Results revealed that this technique can be successfully used for ore grade monitoring at the mine level in a controlled environment.     

Robust and adaptive techniques in self-organizing neural networks

Robust and adaptive training algorithms aiming at enhancing the capabilities of self-organizing and Radial Basis Function (RBF) neural networks are reviewed in this paper. The following robust variants of Learning Vector Quantizer (LVQ) are described: the order statistics LVQ, the L ₂ LVQ and the split-merge LVQ. Successful application of the marginal median LVQ that belongs to the class of order statistics LVQs in the self-organized selection of the centers in RBF neural networks is reported. Moreover, the use of the median absolute deviation in the estimation of the covariance matrix of the observations assigned to each hidden unit in RBF neural networks is proposed. Applications that prove the superiority of the proposed variants of LVQ and RBF neural networks in noisy color image segmentation, color-based image recognition, segmentation of ultrasonic images, motion-field smoothing and moving object segmentation are outlined.     

SNRNet: A Deep Learning-Based Network for Banknote Serial Number Recognition

The banknote serial number recognition (SNR) plays an important role in the banking business and attracts much attention recently. However, most of the existing SNR methods take character segmentation and character classification as two separate steps, so that the accuracy of SNR heavily relies on the character segmentation, which is a challenging problem due to complicated background and uneven illumination. In this paper, the SNR is cast into a sequence prediction problem, which integrates such two steps into a unified network, and we propose a deep learning-based serial number recognition network, which can be trained end-to-end to avoid the preliminary character-segmentation with three steps as follow. First, the improved convolutional neural networks are employed to extract the feature sequence of the input image. Second, the feature sequence is used as an input to the bidirectional recurrent neural networks (BRNNs), where the character segmentation is not required. Finally, the label recognition is implemented using the connectionist temporal classification to decode the BRNNs’ output. The experimental results demonstrate that the proposed method outperforms the state-of-the-art methods in both accuracy and efficiency: it achieves character and serial number recognition of the renminbi (RMB) with accuracies 99.96% and 99.56%, respectively.

     

Tool wear monitoring using artificial neural network based on extended Kalman filter weight updation with transformed input patterns

The condition of the tool in a turning operation is monitored by using artificial neural network (ANN). The recursive Kalman filter algorithm is used for weight updation of the ANN. To monitor the status of the tool, tool wear patterns are collected. The patterns are transformed from n-dimensional feature space to a lower dimensional space (two dimensions). This is done by using two discriminant vectors j₁ {\varphi_{1 }} and j₂{\varphi_{2}}. These discriminant vectors are found by optimal discriminant plane method. Thirty patterns are used for training the ANN. A comparison between the classification performances of the ANN trained without reducing the dimensions of the input patterns and with reduced dimensions of the input patterns is done. The ANN trained with transformed tool wear patterns gives better results in terms of improved classification performance in less iteration, when compared with the results of the ANN trained without transforming the dimensions of the input patterns to a lower dimension.     

基于神经网络的颗粒图像边缘混合检测方法

针对颗粒图像的特点,提出一种基于神经网络的边缘混合检测方法,该方法包含边界候选象素提取和神经网络边缘检测两部分,神经网络由用于图像信息压缩与图像信息编码的自组织竞争子神经网络（ＡＳＣＳＮＮ）和用于获取图像边缘矢量信息的基于径向函数子神经网络（ＲＢＦＳＮＮ）组成,实验结果表明,该方法分割颗粒图像得到的边缘图像封闭性好,边界描述真实,适用于堆积颗粒物料图像的边缘检测。     

Neural Networks for Full-Scale Protein Sequence Classification: Sequence Encoding with Singular Value Decomposition

A neural network classification method has been developed as an alternative approach to the search/organization problem of protein sequence databases. The neural networks used are three-layered, feed-forward, back-propagation networks. The protein sequences are encoded into neural input vectors by a hashing method that counts occurrences ofn-gram words. A new SVD (singular value decomposition) method, which compresses the long and sparsen-gram input vectors and captures semantics ofn-gram words, has improved the generalization capability of the network. A full-scale protein classification system has been implemented on a Cray supercomputer to classify unknown sequences into 3311 PIR (Protein Identification Resource) superfamilies/families at a speed of less than 0.05 CPU second per sequence. The sensitivity is close to 90% overall, and approaches 100% for large superfamilies. The system could be used to reduce the database search time and is being used to help organize the PIR protein sequence database.     

深度区域网络方法的细粒度图像分类

目的 在细粒度视觉识别中,难点是对处于相同层级的大类,区分其具有微小差异的子类,为实现准确的分类精度,通常要求具有专业知识,所以细粒度图像分类为计算机视觉的研究提出更高的要求。为了方便普通人在不具备专业知识和专业技能的情况下能够区分物种细粒度类别,进而提出一种基于深度区域网络的卷积神经网络结构。方法 该结构基于深度区域网络,首先,进行深度特征提取任务,使用VGG16层网络和残差101层网络两种结构作为特征提取网络,用于提取深层共享特征,产生特征映射。其次,使用区域建议网络结构,在特征映射上进行卷积,产生目标区域;同时使用兴趣区域（RoI）池化层对特征映射进行最大值池化,实现网络共享。之后将池化后的目标区域输入到区域卷积网络中进行细粒度类别预测和目标边界回归,最终输出网络预测类别及回归边框点坐标。同时还进行了局部遮挡实验,检测局部遮挡部位对于分类正确性的影响,分析局部信息对于鸟类分类的影响情况。结果 该模型针对CUB_200_2011鸟类数据库进行实验,该数据库包含200种细粒度鸟类类别,11 788幅鸟类图片。经过训练及测试,实现VGG16+R-CNN （RPN）和Res101+R-CNN （RPN）两种结构验证正确率分别为90.88%和91.72%,两种结构Top-5验证正确率都超过98%。本文模拟现实环境遮挡情况进行鸟类局部特征遮挡实验,检测分类效果。结论 基于深度区域网络的卷积神经网络模型,提高了细粒度鸟类图像的分类性能,在细粒度鸟类图像的分类上,具有分类精度高、泛化能力好和鲁棒性强的优势,实验发现头部信息对于细粒度鸟类分类识别非常重要。     

An insect classification analysis based on shape features using quality threshold ARTMAP and moment invariant

The main objective of this paper is to investigate the use of Quality Threshold ARTMAP (QTAM) neural network in classifying the feature vectors generated by moment invariant for the insect recognition task. In this work, six different types of moment invariant technique are adopted to extract the shape features of the insect images. These moment techniques are Geometrical Moment Invariant (GMI), United Moment Invariant (UMI), Zernike Moment Invariant (ZMI), Legendre Moment Invariant (LMI), Tchebichef Moment Invariant (TMI) and Krawtchouk Moment Invariant (KMI). All the moment techniques are analyzed using the concept of intraclass and interclass analysis. In intraclass analysis, several computation methods are introduced in order to examine the invariance properties of adopted moment techniques for the same insect object. Meanwhile, the classification accuracy of neural networks is adopted to measure the interclass characteristic and the effectiveness of moment technique in extracting the shape features of insect images. Other types of neural networks are also utilized in this research work. This includes novel enhancement technique based on the Gaussian and Mahalanobis function that design to increase its prediction accuracy. All the other networks used to classify the feature vectors are based on the Fuzzy ARTMAP (FAM) neural network. The experimental results indicated that the Krawtchouk Moment Invariant technique generated the highest classification accuracy for most of the networks used and generated the smallest error for the intraclass analysis. Using different normalization technique, the Quality Threshold ARTMAP and Mahalanobis distance function (QTAM-m) network gave the highest insect recognition results when compared to other networks.     

Neural network based methods for cloud classification on AVHRR images

Artificial neural networks trained on spectral and textural features extracted from Advanced Very High Resolution Radiometer (AVHRR) images have been used to develop an automated cloud classification system. Selection of the optimum combination of features was achieved by using statistical methods presented in earlier work by Gu et al. and by running large numbers of neural network simulations on test datasets. The performance of these methods surpasses that of other approaches such as the use of Gabor filters for texture segmentation and the maximum likelihood classifier. A particular architecture for an operational classification system is presented based on a two-stage multiple network configuration which is shown to segment complex images to a high degree of accuracy and achieves an overall accuracy on an independent, representative test set of 91%.     

残差密集空间金字塔网络的城市遥感图像分割

目的 遥感图像语义分割是根据土地覆盖类型对图像中每个像素进行分类,是遥感图像处理领域的一个重要研究方向。由于遥感图像包含的地物尺度差别大、地物边界复杂等原因,准确提取遥感图像特征具有一定难度,使得精确分割遥感图像比较困难。卷积神经网络因其自主分层提取图像特征的特点逐步成为图像处理领域的主流算法,本文将基于残差密集空间金字塔的卷积神经网络应用于城市地区遥感图像分割,以提升高分辨率城市地区遥感影像语义分割的精度。方法 模型将带孔卷积引入残差网络,代替网络中的下采样操作,在扩大特征图感受野的同时能够保持特征图尺寸不变;模型基于密集连接机制级联空间金字塔结构各分支,每个分支的输出都有更加密集的感受野信息;模型利用跳线连接跨层融合网络特征,结合网络中的高层语义特征和低层纹理特征恢复空间信息。结果 基于ISPRS （International Society for Photogrammetry and Remote Sensing） Vaihingen地区遥感数据集展开充分的实验研究,实验结果表明,本文模型在6种不同的地物分类上的平均交并比和平均F₁值分别达到69.88%和81.39%,性能在数学指标和视觉效果上均优于SegNet、pix2pix、Res-shuffling-Net以及SDFCN （symmetrical dense-shortcut fully convolutional network）算法。结论 将密集连接改进空间金字塔池化网络应用于高分辨率遥感图像语义分割,该模型利用了遥感图像不同尺度下的特征、高层语义信息和低层纹理信息,有效提升了城市地区遥感图像分割精度。     

基于改进RCE和RBF神经网络的静态手势识别

针对手势识别的手区域分割、手势特征提取和手势分类的三个过程,提出了一种新的静态手势识别方法。改进了传统的RCE神经网络用于手区域的分割,具有更高的运行速度和更强的抗噪能力。依Freeman链码方向提取手的边缘到掌心的距离作为手势的特征向量。将上一步得到的手势特征向量作为RBF神经网络的输入,进行网络的训练和分类。实验验证了该方法的有效性和可行性,并用其实现了人和仿人机器人的剪刀石头布的猜拳游戏。     

傅里叶变换通道注意力网络的胆管癌高光谱图像分割

目的 胆管癌高光谱图像的光谱波段丰富但存在冗余,造成基于深度神经网络高光谱图像分割方法的分割精度下降,虽然一些基于通道注意力机制的网络能够关注重要通道,但在处理通道特征时存在信息表示不足问题,因此本文研究构建一种新的通道注意力机制深度网络,以提高分割准确性。方法 提出了傅里叶变换多频率通道注意力机制（frequency selecting channel attention,FSCA）。FSCA对输入特征进行2维傅里叶变换,提取部分频率特征,再通过两层全连接层得到通道权重向量,将通道权重与对应通道特征相乘,获得了融合通道注意力信息的输出。针对患癌区域和无癌区域数据不平衡问题引入了Focal损失,结合Inception模块,构建基于Inception-FSCA的胆管癌高光谱图像分割网络。结果 在采集的胆管癌高光谱数据集上进行实验,Inception-FSCA网络的准确率（accuracy）、精度（precision）、敏感性（sensitivity）、特异性（specificity）、Kappa系数分别为0.978 0、0.965 4、0.958 6、0.985 2、0.945 6,优于另外5种对比方法。与合成的假彩色图像的分割结果相比,高光谱图像上的实验指标分别提高了0.058 4、0.105 8、0.087 5、0.039 0、0.149 3。结论 本文所提出的傅里叶变换多频率通道注意力机制能够更有效地利用通道信息,基于Inception-FSCA的胆管癌高光谱图像分割网络能够提升分割效果,在胆管癌医学辅助诊断方面具有研究和应用价值。     

Computer-aided prediction of the Al2O3 nanoparticles’ effects on tensile strength and percentage of water absorption of concrete specimens

In the present paper, two models based on artificial neural networks and genetic programming for predicting split tensile strength and percentage of water absorption of concretes containing Al₂O₃ nanoparticles have been developed at different ages of curing. For purpose of building these models, training and testing using experimental results for 144 specimens produced with 16 different mixture proportions were conducted. The data used in the multilayer feed-forward neural networks models and input variables of genetic programming models are arranged in a format of eight input parameters that cover the cement content, nanoparticle content, aggregate type, water content, the amount of superplasticizer, the type of curing medium, Age of curing and number of testing try. According to these input parameters, in the neural networks and genetic programming models, the split tensile strength and percentage of water absorption values of concretes containing Al₂O₃ nanoparticles were predicted. The training and testing results in the neural network and genetic programming models have shown that every two models have strong potential for predicting the split tensile strength and percentage of water absorption values of concretes containing Al₂O₃ nanoparticles. It has been found that NN and GEP models will be valid within the ranges of variables. In neural networks model, as the training and testing ended when minimum error norm of network gained, the best results were obtained, and in genetic programming model, when 4 gens was selected to construct the model, the best results were acquired. Although neural network have predicted better results, genetic programming is able to predict reasonable values with a simpler method rather than neural network.     

基于神经网络的汽车车图象自动识别

提出了一种利用前向三层神经网络及计算机视觉技术对汽车车型进行自动识别的方法,包括汽车车型图象的分割,特征提取以及基于前向三层神经网络的自动识别和分类,文中的所有算法均在５８６／１３３微机上实现,识别和分类效果良好。