基于谱直方图表示和支持向量机的纹理分类

在纹理分类中采用谱直方图表示(SHR),每个图像窗表示一个包含滤波后图像直方图的特征向量,而直方图是图像谱表示的连接桥梁.在滤波器选择算法之前,结合每个图像分块和滤波器的独立谱表示和直方图,可以获得更加低层的局部特征.最后,时所有独立滤波器采用滤波器选择算法来得到所需的少量滤波器.为了保证分类的可靠性,选择高斯径向基函数(RBF)进行谱直方图表示,采用支持向量机(SVMs)作为分类函数.对本文方法和其它两种方法:Gabor滤波和独立成分分析(ICA)进行了纹理分类和脸部识别的比较实验.实验结果表明,本文方法具有更高的分类准确性,也证明了SVMs优秀的泛化能力.     

Classification of brain hemorrhage computed tomography images using OzNet hybrid algorithm

Classification of brain hemorrhage computed tomography (CT) images provides a better diagnostic implementation for emergency patients. Attentively, each brain CT image must be examined by doctors. This situation is time-consuming, exhausting, and sometimes leads to making errors. Hence, we aim to find the best algorithm owing to a requirement for automatic classification of CT images to detect brain hemorrhage. In this study, we developed OzNet hybrid algorithm, which is a novel convolution neural networks (CNN) algorithm. Although OzNet achieves high classification performance, we combine it with Neighborhood Component Analysis (NCA) and many classifiers: Artificial neural networks (ANN), Adaboost, Bagging, Decision Tree, K-Nearest Neighbor (K-NN), Linear Discriminant Analysis (LDA), Naïve Bayes and Support Vector Machines (SVM). In addition, Oznet is utilized for feature extraction, where 4096 features are extracted from the fully connected layer. These features are reduced to have significant and informative features with minimum loss by NCA. Eventually, we use these classifiers to classify these significant features. Finally, experimental results display that OzNet-NCA-ANN excellent classifier model and achieves 100% accuracy with created Dataset 2 from Brain Hemorrhage CT images.     

使用支持向量机的纹理识别方法

将支持向量机应用到纹理识别领域,提出了一种基于支持向量机和小波变换的新型纹理识别方法。该方法用小波变换各子带图像共生矩阵参数、分析窗口大小、像素均值和像素标准差等参数作为纹理特征,解决了描述不同尺度纹理的难题。以多类支持向量机作为分类器,用输出纠错码把二分类器扩展到多类,提高了分类器的泛化能力。在包含25类单色自然纹理的图像库上进行识别试验,结果表明,该方法识别错误率小于10%,识别正确率比传统的贝叶斯等方法提高了2%左右,获得了更高的识别正确率,且推广性更好。     

Performance analysis of EM,SVD, and SVM classifiers in classification of carcinogenic regions of medical images

In this article, the performance analysis of Expectation Maximization (EM), Singular Value Decomposition (SVD), and Support Vector Machines (SVM) classifiers for classification of carcinogenic regions from various medical images is carried out. Cancer detection is one of the critical issues where excessive care needs to be taken for better diagnosis. Any classifier needs to detect the cancer with respect to the efficiency in time of detection and performance. Due to these, three classifiers are selected: Expectation Maximization (EM), Singular Value Decomposition (SVD), and Support Vector Machines (SVM). EM classifier performs as the optimizer and SVD classifier performs as the dual class classifier. SVM classifier is used as both optimizer and classifier for multiclass classification procedure and for wide stage cancer detection procedures. The performance analysis of all the three classifiers are analyzed for a group of 100 cancer patients based on the benchmark parameter such as Performance Measures and Quality Metrics. From the experimental results it is evident, that the SVM classifier significantly outperforms other classifiers in the classification of carcinogenic regions of medical images.     

Machine Learning and Synthetic Minority Oversampling Techniques for Imbalanced Data: Improving Machine Failure Prediction

Prediction of machine failure is challenging as the dataset is often imbalanced with a low failure rate. The common approach to handle classification involving imbalanced data is to balance the data using a sampling approach such as random undersampling, random oversampling, or Synthetic Minority Oversampling Technique (SMOTE) algorithms. This paper compared the classification performance of three popular classifiers (Logistic Regression, Gaussian Naïve Bayes, and Support Vector Machine) in predicting machine failure in the Oil and Gas industry. The original machine failure dataset consists of 20,473 hourly data and is imbalanced with 19945 (97%) ‘non-failure’ and 528 (3%) ‘failure data’. The three independent variables to predict machine failure were pressure indicator, flow indicator, and level indicator. The accuracy of the classifiers is very high and close to 100%, but the sensitivity of all classifiers using the original dataset was close to zero. The performance of the three classifiers was then evaluated for data with different imbalance rates (10% to 50%) generated from the original data using SMOTE, SMOTE-Support Vector Machine (SMOTE-SVM) and SMOTE-Edited Nearest Neighbour (SMOTE-ENN). The classifiers were evaluated based on improvement in sensitivity and F-measure. Results showed that the sensitivity of all classifiers increases as the imbalance rate increases. SVM with radial basis function (RBF) kernel has the highest sensitivity when data is balanced (50:50) using SMOTE (Sensitivity_test =0.5686, F_test= 0.6927) compared to Naïve Bayes (Sensitivity_test =0.4033, F_test= 0.6218) and Logistic Regression (Sensitivity_test =0.4194, F_test= 0.621). Overall, the Gaussian Naïve Bayes model consistently improves sensitivity and F-measure as the imbalance ratio increases, but the sensitivity is below 50%. The classifiers performed better when data was balanced using SMOTE-SVM compared to SMOTE and SMOTE-ENN.     

Timing and Classification of Patellofemoral Osteoarthritis Patients Using Fast Large Margin Classifier

Surface electromyogram (sEMG) processing and classification can assist neurophysiological standardization and evaluation and provide habitational detection. The timing of muscle activation is critical in determining various medical conditions when looking at sEMG signals. Understanding muscle activation timing allows identification of muscle locations and feature validation for precise modeling. This work aims to develop a predictive model to investigate and interpret Patellofemoral (PF) osteoarthritis based on features extracted from the sEMG signal using pattern classification. To this end, sEMG signals were acquired from five core muscles over about 200 reads from healthy adult patients while they were going upstairs. Onset, offset, and time duration for the Transversus Abdominus (TrA), Vastus Medialis Obliquus (VMO), Gluteus Medius (GM), Vastus Lateralis (VL), and Multifidus Muscles (ML) were acquired to construct a classification model. The proposed classification model investigates function mapping from real-time space to a PF osteoarthritis discriminative feature space. The activation feature space of muscle timing is used to train several large margin classifiers to modulate muscle activations and account for such activation measurements. The fast large margin classifier achieved higher performance and faster convergence than support vector machines (SVMs) and other state-of-the-art classifiers. The proposed sEMG classification framework achieved an average accuracy of 98.8% after 7 s training time, improving other classification techniques in previous literature.     

基于多超平面支持向量机的图像语义分类算法

由于图像的低层可视特征与高层语义内容之间存在巨大的语义鸿沟,而基于内容的图像分类和检索准确性极大依赖低层可视特征的描述,本文提出了一种基于多超平面支持向量机的图像语义分类方法.多超平面分类器从优化问题的复杂度和运行泛化能力两方面进行研究,是最优分离超平面分类器一种显而易见的扩展.实验结果表明,本文提出的方法在图像语义分类的准确性方面要优于诸如采用色彩特征和纹理特征的支持向量机分类器的其它方法.     

Real and Altered Fingerprint Classification Based on Various Features andClassifiers

Biometric recognition refers to the identification of individuals through their unique behavioral features (e.g., fingerprint, face, and iris). We need distinguishing characteristics to identify people, such as fingerprints, which are world-renowned as the most reliable method to identify people. The recognition of fingerprints has become a standard procedure in forensics, and different techniques are available for this purpose. Most current techniques lack interest in image enhancement and rely on high-dimensional features to generate classification models. Therefore, we proposed an effective fingerprint classification method for classifying the fingerprint image as authentic or altered since criminals and hackers routinely change their fingerprints to generate fake ones. In order to improve fingerprint classification accuracy, our proposed method used the most effective texture features and classifiers. Discriminant Analysis (DCA) and Gaussian Discriminant Analysis (GDA) are employed as classifiers, along with Histogram of Oriented Gradient (HOG) and Segmentation-based Feature Texture Analysis (SFTA) feature vectors as inputs. The performance of the classifiers is determined by assessing a range of feature sets, and the most accurate results are obtained. The proposed method is tested using a Sokoto Coventry Fingerprint Dataset (SOCOFing). The SOCOFing project includes 6,000 fingerprint images collected from 600 African people whose fingerprints were taken ten times. Three distinct degrees of obliteration, central rotation, and z-cut have been performed to obtain synthetically altered replicas of the genuine fingerprints. The proposal achieved massive success with a classification accuracy reaching 99%. The experimental results indicate that the proposed method for fingerprint classification is feasible and effective. The experiments also showed that the proposed SFTA-based GDA method outperformed state-of-art approaches in feature dimension and classification accuracy.     

An experimental comparison of modelling techniques for speaker recognition under limited data condition

Most of the existing modelling techniques for the speaker recognition task make an implicit assumption of sufficient data for speaker modelling and hence may lead to poor modelling under limited data condition. The present work gives an experimental evaluation of the modelling techniques like Crisp Vector Quantization (CVQ), Fuzzy Vector Quantization (FVQ), Self-Organizing Map (SOM), Learning Vector Quantization (LVQ), and Gaussian Mixture Model (GMM) classifiers. An experimental evaluation of the most widely used Gaussian Mixture Model-Universal Background Model (GMM-UBM) is also made. The experimental knowledge is then used to select a subset of classifiers for obtaining the combined classifiers. It is proposed that the combined LVQ and GMM-UBM classifier provides relatively better performance compared to all the individual as well as combined classifiers.     

Effective Classification of Synovial Sarcoma Cancer Using Structure Features and Support Vectors

In this research work, we proposed a medical image analysis framework with two separate releases whether or not Synovial Sarcoma (SS) is the cell structure for cancer. Within this framework the histopathology images are decomposed into a third-level sub-band using a two-dimensional Discrete Wavelet Transform. Subsequently, the structure features (SFs) such as Principal Components Analysis (PCA), Independent Components Analysis (ICA) and Linear Discriminant Analysis (LDA) were extracted from this sub-band image representation with the distribution of wavelet coefficients. These SFs are used as inputs of the Support Vector Machine (SVM) classifier. Also, classification of PCA + SVM, ICA + SVM, and LDA + SVM with Radial Basis Function (RBF) kernel the efficiency of the process is differentiated and compared with the best classification results. Furthermore, data collected on the internet from various histopathological centres via the Internet of Things (IoT) are stored and shared on blockchain technology across a wide range of image distribution across secure data IoT devices. Due to this, the minimum and maximum values of the kernel parameter are adjusted and updated periodically for the purpose of industrial application in device calibration. Consequently, these resolutions are presented with an excellent example of a technique for training and testing the cancer cell structure prognosis methods in spindle shaped cell (SSC) histopathological imaging databases. The performance characteristics of cross-validation are evaluated with the help of the receiver operating characteristics (ROC) curve, and significant differences in classification performance between the techniques are analyzed. The combination of LDA + SVM technique has been proven to be essential for intelligent SS cancer detection in the future, and it offers excellent classification accuracy, sensitivity, specificity.     

Medicinal and illegal drugs among Danish car drivers

The objective of this study was to get an insight into the prevalence of medicinal and illegal drugs among car drivers in a Danish rural area. The police randomly stopped about 1000 car drivers and asked them to deliver a saliva sample and gave them a questionnaire to fill in at home. Laboratory analyses by specific methods of samples, which a screening found positive, confirmed that 2% were positive for benzodiazepines or illegal drugs (amphetamine, cannabis, cocaine or opiates): 1.3% were positive for illegal drugs and 0.7% for benzodiazepines. Questionnaire statements from some of the drivers confirm that occasionally some of these drive despite a suspicion to be under the influence of an illegal drug (2.8%), an illegal drug including alcohol (4%), a hazardous medicinal drug including alcohol (8.5%), or alcohol alone above the legal limit (24.5%). These results are considered reliable for the survey area and may not reflect national conditions. The overall results indicate that in this study driving under the influence of illegal drugs or alcohol seems to be associated to especially men, aged 22–44 years. Driving under the influence of hazardous medicinal drugs seems to be associated to middle-aged/elderly drivers, both men and women.     

基于支持向量机的油封缺陷图像检测方法

提出一种基于支持向量机分类的油封缺陷图像检测方法,把油封外观中的有无缺陷看作两种不同的类别模式,应用支持向量机对两类不同的样本采样学习,然后进行分类判断。采集油封各部位图像并进行预处理,利用算法切割出各个检测区域图像,根据油封主要部位的各类缺陷特点,选取不同特征参数。应用径向基核函数建立支持向量机识别模型,并通过实验实现核函数参数寻优。实验结果表明,该方法具有检测成本低、可靠性高、泛化能力强、容易在线实施等特点,具有实用推广价值。     

基于多核Boosting方法的通用目标分类和识别

针对传统基于SVM分类器的多核学习方法优化参数多、优化过程复杂、计算量大的缺点,本文提出基于Real Adaboost的多核学习方法解决通用目标分类与识别问题.该方法根据核函数能将高维特征映射到低维空间的特性,采用核函数空间上的线性平面分割构建弱分类器,并用Real Adaboost学习框架对弱分类器进行学习.先用分层...     

The culpability of drivers killed in New Zealand road crashes and their use of alcohol and other drugs

Over a period of five years, blood samples were taken from 1046 drivers killed as a result of a motor vehicle crash on New Zealand roads. These were analysed for the presence of alcohol and a range of both illicit drugs and psychoactive medicinal drugs. Driver culpability was determined for all crashes. The control group of drug- and alcohol-free drivers comprised 52.2% of the study population. Drivers positive for psychoactive drugs were more likely to be culpable (odds ratio (OR) 3.5, confidence interval (CI) 95% 2.4–5.2) than the control group. Driver culpability exhibited the expected positive association with alcohol use (OR 13.7, 95% CI 4.3–44) and with combined alcohol and cannabis use (OR 6.9, 95% CI 3.0–16). There was only a weak positive association between cannabis use (with no other drug) and culpability (OR 1.3, CI 95% 0.8–2.3). Furthermore, the OR for drivers with blood tetrahydrocannabinol (THC) concentrations greater than 5 ng/mL was lower (OR 1.0, CI 95% 0.4–2.4) than drivers with blood THC concentrations less than 2 ng/mL (OR 3.1, CI 95% 0.9–10). This is inconsistent with results reported by other studies where a significant increase in crash risk was found with blood THC levels greater than 5 ng/mL. In this study, there were very few drivers who had used a single drug, other than cannabis or alcohol. Therefore, from this study, it is not possible to comment on any relationship between opioid, stimulant or sedative drug use and an increased risk of being killed in a crash for the drivers using these drugs. The results from a multivariate analysis indicate that driver gender, age group and licence status, (P = 0.022, P = 0.016, P = 0.026, respectively), the type of vehicle being driven (P = 0.013), the number of vehicles in the crash (P < 0.001), the blood alcohol concentration of the driver (P < 0.001) and the use of any drug other than alcohol and cannabis (P = 0.044), are all independently associated with culpability.     

Machine Learning Based Psychotic Behaviors Prediction from Facebook Status Updates

With the advent of technological advancements and the widespread Internet connectivity during the last couple of decades, social media platforms (such as Facebook, Twitter, and Instagram) have consumed a large proportion of time in our daily lives. People tend to stay alive on their social media with recent updates, as it has become the primary source of interaction within social circles. Although social media platforms offer several remarkable features but are simultaneously prone to various critical vulnerabilities. Recent studies have revealed a strong correlation between the usage of social media and associated mental health issues consequently leading to depression, anxiety, suicide commitment, and mental disorder, particularly in the young adults who have excessively spent time on social media which necessitates a thorough psychological analysis of all these platforms. This study aims to exploit machine learning techniques for the classification of psychotic issues based on Facebook status updates. In this paper, we start with depression detection in the first instance and then expand on analyzing six other psychotic issues (e.g., depression, anxiety, psychopathic deviate, hypochondria, unrealistic, and hypomania) commonly found in adults due to extreme use of social media networks. To classify the psychotic issues with the user's mental state, we have employed different Machine Learning (ML) classifiers i.e., Random Forest (RF), Support Vector Machine (SVM), Naïve Bayes (NB), and K-Nearest Neighbor (KNN). The used ML models are trained and tested by using different combinations of features selection techniques. To observe the most suitable classifiers for psychotic issue classification, a cost-benefit function (sometimes termed as ‘Suitability’) has been used which combines the accuracy of the model with its execution time. The experimental evidence argues that RF outperforms its competitor classifiers with the unigram feature set.     

Providing an efficient intelligent transportation system through detection,tracking and recognition of the region of interest in traffic signs by using non-linear SVM classifier in line with histogram oriented gradient and Kalman filter approach

With a focus on new researches in the area of intelligent transportation systems (ITS), an efficient approach has been investigated here. Based on the present view point, analysis of traffic signs are first considered via intelligence based approach, which is carried out through three main stages including detection, tracking and recognition, respectively, in this research. The key role of detection is to identify traffic signs by classification of road sign shapes in accordance with their signatures. This classification consists of four different shapes of circle, semicircle, triangle and square, as well. The linear classification of traffic sign is also carried out via support vector machine (SVM) by using one against all (OAA), since the present SVMs classifiers realized via linear kernel. The next step is to track traffic sign. It should be noted that this technique is now developed to reduce the searching mode in case of the whole area to be optimized its computational processing, consequently. This research work is investigated by realizing Kalman filter approach, where, finally, in recognition step, a feature of the region of interest (ROI) has been extracted for SVM classification. Histogram of oriented gradient (HOG) is realized in organizing the approach, as long as Gaussian kernel is also developed for non-linear SVM classifier.     

Classification of magnetic resonance images from rabbit renal perfusion

The feasibility of using chemometric techniques for the automatic detection of whether a rabbit kidney is pathological or not is studied. Sequential images of the kidney are acquired using Dynamic Contrast-Enhanced Magnetic Resonance Imaging with contrast agent injection. A segmentation approach based upon principal component analysis (PCA) is used to separate out the cortex from the rest of the kidney including the medulla, the renal pelvic, and the background. Two classifiers (Soft Independent Method of Class Analogy, SIMCA; Partial Least Squares Discriminant Analysis, PLS-DA) are tested for various types of data pre-treatment including segmentation, feature extraction, centering, autoscaling, standard normal variate transformation, Savitsky-Golay smoothing, and normalization. It is shown that (i) the renal cortex contains more discriminating information on kidney perfusion changes than the whole kidney, and (ii) the PLS-DA classifiers outperform the SIMCA classifiers. PLS-DA, preceded by an automated PCA-based segmentation of kidney anatomical regions, correctly classified all kidneys and constitutes a classification tool of the renal function that can be useful for the clinical diagnosis of renovascular diseases.     

Investigation on effect of fibre hybridization and orientation on mechanical behaviour of natural fibre epoxy composite

Nowadays bio fibre composites play a vital role by replacing conventional materials used in automotive and aerospace industries owing to their high strength to weight ratio, biodegradability and ease of production. This paper aims to find the effect of fibre hybridization and orientation on mechanical behaviour of composite fabricated with neem, abaca fibres and epoxy resin. Here, three varieties of composites are fabricated namely, composite 1 which consists of abaca fibre and glass fibre, composite 2, which consists of neem fibre and glass fibre, whereas composite 3 consists of abaca, neem fibres and glass fibres. In all the above three varieties, fibres are arranged in three types of orientations namely, horizontal (type I), vertical (type II) and 45\(^{\circ }\) inclination (type III). The result shows that composites made up of abaca and neem fibres with inclined orientation (45\(^{\circ }\)) have better mechanical properties when compared with other types of composites. In addition, morphological analysis is carried out using scanning electron microscope to know the fibre distribution, fibre pull out, fibre breakage and crack propagation on tested composites.     

An efficient lymphocytic leukemia detection based on EfficientNets and ensemble voting classifier

Early detection of leukemia increases the chances of a speedier recovery. If a patient exhibits any symptoms, doctors would often examine a blood sample slide under a microscope to detect hematological malignancies. Manually categorizing leukocytes as normal or abnormal requires examining the many characteristics of the cells, which is time-consuming and error-prone. This research aims to create a transfer learning-based Acute Lymphocytic Leukemia (ALL) detection system that is both efficient and easy. To overcome the critical challenges associated with feature extraction, we used EfficientNet, the most recent and most substantial deep learning model. In this article, eight EfficientNets variations are used to extract features and are compared based on classification accuracy. This work uses an ensemble of three sophisticated classifiers, namely Support Vector Machine (SVM), Random Forest, and Logistic Regression, which achieves a classification accuracy of 98.5%.