Implementation of a polychromatic Hamming net for color image classification

A color exemplar-based Hamming net as applied to a color image classifier is presented. The color decomposition and composition techniques for constructing a polychromatic Hamming layer and winner-take-all (WTA) layer interconnection weight matrix (IWM) are discussed. We have simplified the WTA algorithm to reduce the number of iteration cycles in the WTA layer. To confirm the operation of the color Hamming net we provide simulation and experimental demonstrations.     

基于卷积神经网络图像分类的轴承故障模式识别

针对传统的基于数据驱动的机械故障模式识别方法中需要人工构造算法提取特征以及人工构造特征提取算法繁琐的问题,结合卷积神经网络(CNN)在图像特征自动提取与图像分类识别中的广泛应用,提出了一种基于CNN图像分类的轴承故障模式识别方法。首先,利用集合经验模态分解(EEMD)方法对轴承振动信号进行自适应分解并用相关系数对得到的本征模函数分量进行筛选。其次,对筛选得到的本征模函数分量进行伪魏格纳-威利时频分析(PWVD)计算得到信号的时频分布图,并对时频图进行预处理。最后,将轴承15种不同工况预处理后的时频图利用CNN进行特征提取与分类识别。将该方法与同类方法进行了对比,分类正确率提高了4.26%。     

基于灰度-梯度共生矩阵和模糊核聚类的振动图形识别方法

以往复机械振动参数图形为对象,提出了基于灰度-梯度共生矩阵和模糊核聚类的振动图形识别方法。利用灰度-梯度共生矩阵直接提取振动参数图形中的特征信息,将得到的纹理特征参量作为样本输入空间,通过Mercer核把输入样本映射到高斯特征空间后,在高维特征空间中进行聚类,从而实现往复机械故障智能诊断。实验结果表明,该方法可以获得较高的诊断精度,具有一定的可行性和有效性。     

彩色目标高维检测与空间分集检测技术

提出一种彩色目标高维检测和空间分集检测技术,它利用 RGB 颜色的均值趋同性校正色偏,减少光照强度和光源颜色对目标检测的影响;采用检测门限自适应变化的彩色目标高维检测算法检测出疑似目标,进一步避免了传统方法受环境光照影响大的缺点;使用基于区域关联性的彩色信号自适应空间分集检测技术,提高目标检测效果。试验证明,检测并降噪以后,97.5%以上的图像的提取结果可以达到 30db 以上的信噪比。     

Block truncation coding with color clumps: A novel feature extraction technique for content based image classification

The paper has explored principle of block truncation coding (BTC) as a means to perform feature extraction for content based image classification. A variation of block truncation coding, named BTC with color clumps has been implemented in this work to generate feature vectors. Classification performance with the proposed technique of feature extraction has been compared to existing techniques. Two widely used public dataset named Wang dataset and Caltech dataset have been used for analyses and comparisons of classification performances based on four different metrics. The study has established BTC with color clumps as an effective alternative for feature extraction compared to existing methods. The experiments were carried out in RGB color space. Two different categories of classifiers viz. K Nearest Neighbor (KNN) Classifier and RIDOR Classifier were used to measure the classification performances. A paired t test was conducted to establish the statistical significance of the findings. Evaluation of classifier algorithms were done in receiver operating characteristic (ROC) space.     

Optimal wavelength band clustering for multispectral iris recognition

This work explores the possibility of clustering spectral wavelengths based on the maximum dissimilarity of iris textures. The eventual goal is to determine how many bands of spectral wavelengths will be enough for iris multispectral fusion and to find these bands that will provide higher performance of iris multispectral recognition. A multispectral acquisition system was first designed for imaging the iris at narrow spectral bands in the range of 420 to 940 nm. Next, a set of 60 human iris images that correspond to the right and left eyes of 30 different subjects were acquired for an analysis. Finally, we determined that 3 clusters were enough to represent the 10 feature bands of spectral wavelengths using the agglomerative clustering based on two-dimensional principal component analysis. The experimental results suggest (1) the number, center, and composition of clusters of spectral wavelengths and (2) the higher performance of iris multispectral recognition based on a three wavelengths-bands fusion.     

Ant colony recognition systems for part clustering problems

Cellular manufacturing requires an effective part clustering method to start up the manufacturing cell design. This paper presents a new part clustering algorithm that uses the concept of the recognition system of artificial ants. The proposed algorithm mimics the random meetings of real ants to build up the ability of object recognition and then to form many initial part clusters with high similarities. These initial part clusters are further merged into larger and larger clusters in an agglomerative way until the designated number of part families is reached. The characteristics of artificial ants, such as randomization and collective behaviour, allow the algorithm to re-cluster wrongly grouped parts into the proper clusters. As a result this can eliminate the chaining effects resulting from the interference of abnormal parts during the clustering process. This algorithm has been developed into a software system called the ant colony recognition system (ACRS). A number of problems selected from the literature have been solved by ACRS, and the evaluation results indicate that ACRS is able to solve the cell formation problems effectively.     

Improving representation-based classification for robust face recognition

The sparse representation classification (SRC) method proposed by Wright et al. is considered as the breakthrough of face recognition because of its good performance. Nevertheless it still cannot perfectly address the face recognition problem. The main reason for this is that variation of poses, facial expressions, and illuminations of the facial image can be rather severe and the number of available facial images are fewer than the dimensions of the facial image, so a certain linear combination of all the training samples is not able to fully represent the test sample. In this study, we proposed a novel framework to improve the representation-based classification (RBC). The framework first ran the sparse representation algorithm and determined the unavoidable deviation between the test sample and optimal linear combination of all the training samples in order to represent it. It then exploited the deviation and all the training samples to resolve the linear combination coefficients. Finally, the classification rule, the training samples, and the renewed linear combination coefficients were used to classify the test sample. Generally, the proposed framework can work for most RBC methods. From the viewpoint of regression analysis, the proposed framework has a solid theoretical soundness. Because it can, to an extent, identify the bias effect of the RBC method, it enables RBC to obtain more robust face recognition results. The experimental results on a variety of face databases demonstrated that the proposed framework can improve the collaborative representation classification, SRC, and improve the nearest neighbor classifier.     

Neuro-fuzzy network for flavor recognition and classification

This paper presents the neuro-fuzzy Takagi-Sugeno-Kang (TSK) network for the recognition and classification of flavor. The important role in this process fulfills the self-organizing process used for the creation of the inference rules. The self-organizing neurons perform the role of clustering data into fuzzy groups with different membership values (the preprocessing stage). Applying the automatic control of clusters, we have the optimal size of the TSK network. The developed measuring system has been applied for the recognition of flavor of different brands of beer. The fuzzy neural network is used for processing signals obtained from the semiconductor sensor array. The results of numerical experiments have confirmed the excellent performance of such solutions.     

Comparison of clustering approaches with application to dual colour protein data

Cells communicate with their environment via proteins, located at the plasma membrane separating the interior of a cell from its surroundings. The spatial distribution of these proteins in the plasma membrane under different physiological conditions is of importance, since this may influence their signal transmission properties. In this study, the authors compare different methods such as hierarchical clustering, extensible Markov models and the gammics method for analysing such a spatial distribution. The methods are examined in a simulation study to determine their optimal use. Afterwards, they analyse experimental imaging data and extend these methods to simulate dual colour data.Inspec keywords: proteins, molecular biophysics, cellular biophysics, biomembranes, Markov processes, bioinformatics, statistical analysisOther keywords: clustering approaches, dual colour protein data, cell communication, plasma membrane, spatial protein distribution, signal transmission, hierarchical clustering, extensible Markov models, gammics method, experimental imaging data     

用于彩色图像分割的改进遗传FCM算法

本文提出了一种适用于彩色图像分割的遗传模糊C均值聚类(GAFCM)算法.该算法使用Ohta等人提出的彩色特征集中的第一个分量作为图像像素的一维特征向量,并利用由像素空间到特征空间的映射来改进目标函数,从而大大降低了运算量;使用对特征空间结构没有特殊要求的特征距离代替欧氏距离,从而克服了特征空间结构对聚类结果的影响;使用引入FCM优化的遗传算法来搜索最优解,从而提高了搜索速度.实验表明,该算法不但能很好地分割彩色图像,而且具有运算量小、收敛速度快的优点.     

Optical dual-scale architecture for neural image recognition

A novel neural-network architecture that combines image data reduction with focus of attention to achieve reduced training cost, improved noise tolerance, and better generalization performance than comparable conventional networks for image-recognition tasks is presented. The dual-scale architecture is amenable to optical implementation, and an example optical system is demonstrated. For one example problem, the best-case improvements of the dual-scale network over its conventional counterpart were found through simulation to be a factor of 6.7 in training cost, 67.3% in noise tolerance, and 61.6% in generalization to distortions. The dual-scale network is also applied to one instance of a human face recognition problem.     

Efficient image segmentation based on deep learning for mineral image classification

Mineral image segmentation plays a vital role in the realization of machine vision based intelligent ore sorting equipment. However, the existing image segmentation methods still cannot effectively solve the problem of adhesion and overlap between mineral particles, and the segmentation performance of small and irregular particles still needs to be improved. To overcome these bottlenecks, we propose a deep learning based image segmentation method to segment the key areas in mineral images using morphological transformation to process mineral image masks. This investigation explores four aspects of the deep learning-based mineral image segmentation model, including backbone selection, module configuration, loss function construction, and its application in mineral image classification. Specifically, referring to the designs of U-Net, FCN, Seg Net, PSP Net, and DeepLab Net, this experiment uses different backbones as Encoder to building ten mineral image segmentation models with different layers, structures, and sampling methods. Simultaneously, we propose a new loss function suitable for mineral image segmentation and compare CNNs-based segmentation models' training performance under different loss functions. The experiment results show that the proposed mineral image segmentation has excellent segmentation performance, effectively solves adhesion and overlap between adjacent particles without affecting the classification accuracy. By using the Mobile Net as backbone, the PSP Net and DeepLab can achieve a high segmentation performance in mineral image segmentation tasks, and the 15 × 15 is the most suitable size for erosion element structure to process the mask images of the segmentation models.     

语音识别方法在水下目标识别中的应用

水下目标识别是潜艇在海战中,先敌发现并有效进行水声对抗的关键技术。然而,如何根据声纳接收到的舰船辐射噪声对三类目标进行分类识别是长期困扰人们的问题。研究了四种语音识别中常用的方法——线性预测系数（LPC）,线性预测倒谱系数（LPCC）,美尔倒谱系数（MFCC）和最小均方无失真响应（MVDR）,在水下目标识别中的应用效果,并比较了这四种方法在无噪声情况下的识别概率,以及在不同信噪比下的识别概率,并通过比较找到在无噪声和有噪声情况下的最佳方法。实验表明,在无噪声的情况下,MFCC方法总体识别率最高,第一类目标MFCC方法的识别率最高,第二类目标MFCC和MVDR方法识别率相似,好于其他两者,第三类目标MVDR方法识别率最高。在加入噪声的情况下,MVDR方法对三类目标的识别和抗噪声性能明显好于其余三者。     

Unsupervised tissue classification of brain MR images for voxel‐based morphometry analysis

In this article, a fully unsupervised method for brain tissue segmentation of T1‐weighted MRI 3D volumes is proposed. The method uses the Fuzzy C‐Means (FCM) clustering algorithm and a Fully Connected Cascade Neural Network (FCCNN) classifier. Traditional manual segmentation methods require neuro‐radiological expertise and significant time while semiautomatic methods depend on parameter's setup and trial‐and‐error methodologies that may lead to high intraoperator/interoperator variability. The proposed method selects the most useful MRI data according to FCM fuzziness values and trains the FCCNN to learn to classify brain’ tissues into White Matter, Gray Matter, and Cerebro‐Spinal Fluid in an unsupervised way. The method has been tested on the IBSR dataset, on the BrainWeb Phantom, on the BrainWeb SBD dataset, and on the real dataset “University of Palermo Policlinico Hospital” (UPPH), Italy. Sensitivity, Specificity, Dice and F‐Factor scores have been calculated on the IBSR and BrainWeb datasets segmented using the proposed method, the FCM algorithm, and two state‐of‐the‐art brain segmentation software packages (FSL and SPM) to prove the effectiveness of the proposed approach. A qualitative evaluation involving a group of five expert radiologists has been performed segmenting the real dataset using the proposed approach and the comparison algorithms. Finally, a usability analysis on the proposed method and reference methods has been carried out from the same group of expert radiologists. The achieved results show that the segmentations of the proposed method are comparable or better than the reference methods with a better usability and degree of acceptance.     

基于子空间字典偶学习的高光谱图像分类

针对高光谱高分辨率带来巨大数据量和空间分辨率引起混合像元的问题,提出了基于子空间(subspace)的字典偶学习(DPL)算法,简称DPLsub算法。DPL算法是对字典学习的改进,它通过学习得到综合字典和分析字典,在模式识别中体现了高效性,而子空间投影的方法能更好地表征噪声和高度混合的像元。将光谱和空间特征融合的方法用于分类研究试验。实验数据是两幅高光谱影像,比较了子空间字典偶学习(DPLsub)模型和其他三种分类器即最小二乘支持向量机(LS-SVM)、稀疏多分类回归(SMLR)和字典学习(DL-OMP)的分类结果。实验结果显示,DPLsub算法无论在时间上还是精度上都优于其他算法,证明了这种子空间字典偶学习方法对高光谱图像分类的可行性与高效性。     

Cosine histogram analysis for spectral image data classification

Conventional multivariate strategies for making qualitative estimates of sample composition rely chiefly on identifying subtle differences in spectral shape. In some instances, such as in biological tissues, the spectra obtained from a single sample class may consist of many shapes. Likewise, two distinctly different sample classes, such as normal and abnormal tissue, may produce similar variations in spectral shape. In our work, we employ statistical analysis of the set of cosine correlation scores obtained from multispectral visible absorption images of stained cervical Papanicolaou samples. By analyzing the cosine correlation score frequency for spectra obtained from the cell nuclei, abnormal cells can be differentiated from the background of normal cells, which vary considerably in their optical properties and morphology. Our method, called cosine histogram analysis (CHA), returns the percent likelihood of abnormality for each pixel in the field of view and is presented here for the first time.     

Adaptive binary joint transform correlator for image recognition

Our research has shown that the autocorrelation peaks of a binary joint transform correlator are affected by input scenes' backgrounds. An adaptive method is proposed to overcome this problem. The image of interest is first extracted from the background based on the position of the highest correlation peak of the input and reference images. The extracted image is then correlated with the reference to obtain the final correlation peak. Numerical simulations showed that the final autocorrelation peak is the maximum constant for a specified reference image.     

Noise robustness of nonlinear filters for image recognition

We analyze the performance of the Fourier plane nonlinear filters in terms of signal-to-noise ratio (SNR). We obtain a range of nonlinearities for which SNR is robust to the variations in input-noise bandwidth. This is shown both by analytical estimates of the SNR for nonlinear filters and by experimental simulations. Specifically, we analyze the SNR when Fourier plane nonlinearity is applied to the input signal. Using the Karhunen-Loève series expansion of the noise process, we obtain precise analytic expressions of the SNR for Fourier plane nonlinear filters in the presence of various types of additive-noise processes. We find a range of nonlinearities that need to be applied that keep the output SNR of the filter stable relative to changes in the noise bandwidth.