Bridging semantic gap between high-level and low-level features in content-based video retrieval using multi-stage ESN–SVM classifier

Content-based video retrieval system aims at assisting a user to retrieve targeted video sequence in a large database. Most of the search engines use textual annotations to retrieve videos. These types of engines offer a low-level abstraction while the user seeks high-level semantics. Bridging this type of semantic gap in video retrieval remains an important challenge. In this paper, colour, texture and shapes are considered to be low-level features and motion is a high-level feature. Colour histograms convert the RGB colour space into YcbCr and extract hue and saturation values from frames. After colour extraction, filter mask is applied and gradient value is computed. Gradient and threshold values are compared to draw the edge map. Edges are smoothed for sharpening to remove the unnecessary connected components. These diverse shapes are then extracted and stored in shape feature vectors. Finally, an SVM classifier is used for classification of low-level features. For high-level features, depth images are extracted for motion feature identification and classification is done via echo state neural networks (ESN). ESN are a supervised learning technique and follow the principle of recurrent neural networks. ESN are well known for time series classification and also proved their effective performance in gesture detection. By combining the existing algorithms, a high-performance multimedia event detection system is constructed. The effectiveness and efficiency of proposed event detection mechanism is validated using MSR 3D action pair dataset. Experimental results show that the detection accuracy of proposed combination is better than those of other algorithms     

Air Pollution Prediction Via Graph Attention Network and Gated Recurrent Unit

PM2.5 concentration prediction is of great significance to environmental protection and human health. Achieving accurate prediction of PM2.5 concentration has become an important research task. However, PM2.5 pollutants can spread in the earth’s atmosphere, causing mutual influence between different cities. To effectively capture the air pollution relationship between cities, this paper proposes a novel spatiotemporal model combining graph attention neural network (GAT) and gated recurrent unit (GRU), named GAT-GRU for PM2.5 concentration prediction. Specifically, GAT is used to learn the spatial dependence of PM2.5 concentration data in different cities, and GRU is to extract the temporal dependence of the long-term data series. The proposed model integrates the learned spatio-temporal dependencies to capture long-term complex spatio-temporal features. Considering that air pollution is related to the meteorological conditions of the city, the knowledge acquired from meteorological data is used in the model to enhance PM2.5 prediction performance. The input of the GAT-GRU model consists of PM2.5 concentration data and meteorological data. In order to verify the effectiveness of the proposed GAT-GRU prediction model, this paper designs experiments on real-world datasets compared with other baselines. Experimental results prove that our model achieves excellent performance in PM2.5 concentration prediction.     

Research on PM_2.5 Concentration Prediction Algorithm Based on Temporal and Spatial Features

PM_2.5 has a non-negligible impact on visibility and air quality as an important component of haze and can affect cloud formation and rainfall and thus change the climate, and it is an evaluation indicator of air pollution level. Achieving PM_2.5 concentration prediction based on relevant historical data mining can effectively improve air pollution forecasting ability and guide air pollution prevention and control. The past methods neglected the impact caused by PM_2.5 flow between cities when analyzing the impact of inter-city PM_2.5 concentrations, making it difficult to further improve the prediction accuracy. However, factors including geographical information such as altitude and distance and meteorological information such as wind speed and wind direction affect the flow of PM_2.5 between cities, leading to the change of PM_2.5 concentration in cities. So a PM_2.5 directed flow graph is constructed in this paper. Geographic and meteorological data is introduced into the graph structure to simulate the spatial PM_2.5 flow transmission relationship between cities. The introduction of meteorological factors like wind direction depicts the unequal flow relationship of PM_2.5 between cities. Based on this, a PM_2.5 concentration prediction method integrating spatial-temporal factors is proposed in this paper. A spatial feature extraction method based on weight aggregation graph attention network (WGAT) is proposed to extract the spatial correlation features of PM_2.5 in the flow graph, and a multi-step PM_2.5 prediction method based on attention gate control loop unit (AGRU) is proposed. The PM_2.5 concentration prediction model WGAT-AGRU with fused spatiotemporal features is constructed by combining the two methods to achieve multi-step PM_2.5 concentration prediction. Finally, accuracy and validity experiments are conducted on the KnowAir dataset, and the results show that the WGAT-AGRU model proposed in the paper has good performance in terms of prediction accuracy and validates the effectiveness of the model.     

Analysis and classification of malignancy in pancreatic magnetic resonance images using neural network techniques

Computer-aided diagnosis (CAD) is a computerized way of detecting tumors in MR images. Magnetic resonance imaging (MRI) has been generally used in the diagnosis and detection of pancreatic tumors. In a medical imaging system, soft tissue contrast and noninvasiveness are clear preferences of MRI. Inaccurate detection of tumor and long time consumption are the disadvantages of MRI. Computerized classifiers can greatly renew the diagnosis activity, in terms of both accuracy and time necessity by normal and abnormal images, automatically. This article presents an intelligent, automatic, accurate, and robust method to classify human pancreas MRI images as normal or abnormal in terms of pancreatic tumor. It represents the response of artificial neural network (ANN) and support vector machine (SVM) techniques for pancreatic tumor classification. For this, we extract features from MR images of pancreas using the GLCM method and select the best features using JAFER algorithm. These features are analyzed by five classification techniques: ANN BP, ANN RBF, SVM Linear, SVM Poly, and SVM RBF. We compare the results with benchmark data set of MR brain images. The analytical outcome presents that the two best features used to classify the MR images using ANN BP technique have 98% classification accuracy.     

基于PCA-GA-SVM的露天矿爆破振动速度预测模型研究

露天矿爆破振动速度受很多因素的影响,传统的经验公式和单一的神经网络模型无法满足现代爆破安全的要求。为提高预测爆破振动速度的精度,利用主成分分析(PCA)提取4个影响爆破振动速度的主成分作为模型的输入变量;结合遗传算法(GA)寻优获得支持向量机(SVM)惩罚系数c和核函数参数g为1.899 1和1.971 2,建立了基于PCA-GA-SVM的露天矿爆破振动速度预测模型,并成功应用于现场爆破工程。结果表明:该模型的平均相对误差为14.60%,建模时间为3.12 s,均方误差为0.131 5,与BP神经网络、传统SVM和GA-SVM对比,此模型具有更快的收敛速度和更高的准确率,为多因素影响下爆破振动速度预测提供了一种新的方法。     

基于遗传算法和最小二乘支持向量机的织物剪切性能预测

提出了一种基于最小二乘支持向量机的织物剪切性能预测模型,并且采用遗传算法进行最小二乘支持向量机的参数优化,将获得的样本进行归一化处理后,将其输入预测模型以得到预测结果.仿真结果表明,基于最小二乘支持向量机的预测模型比BP神经网络和线性回归方法具有更高的精度和范化能力.     

基于PSO改进深度置信网络的滚动轴承故障诊断

针对深度置信网络(Deep Belief Network,DBN)用于轴承故障诊断时,网络层结构调试比较费时等问题,提出一种基于粒子群优化(Particle Swarm Optimization,PSO)的DBN算法,以及基于该算法的轴承故障诊断模型。该模型利用PSO算法优选DBN网络结构,并通过自适应时刻估计法微调模型参数,随后运用具有最优结构的DBN模型直接从原始振动信号中提取低维故障特征,并将其输入到Soft-max分类器中识别轴承的故障模式。该算法与支持向量机、BP神经网络、DBN、堆叠降噪自编码等方法进行对比分析,实验结果表明,PSO改进的DBN算法具有更高的准确率以及更好的鲁棒性。     

支持向量机集成和特征选择联合算法

提出了两种基于支持向量机集成和特征选择联合算法。联合算法的核心思想是在构建基础分类器的同时选择有效特征。通过对实测舰船数据和公共数据的识别实验,证明了两种算法都可以用于舰船目标识别。算法一更适用于冗余特征较多的情况。算法二在对舰船目标识别时,选择的特征数目降低为原来特征数目的30%,正确分类率比单个支持向量机高近10%。     

基于佳点集粒子群算法的SVM参数优化方法

针对传统的粒子群算法（PSO）初始种群随机生成而导致的算法稳定性差和易出现早熟等问题,提出了基于佳点集改进的粒子群算法（GSPSO）,并将其优化支持向量机（SVM）,构建一种高效的预测评估模型（GSPSO-SVM）。首先采用佳点集方法使PSO中初始粒子均匀分布,然后利用GSPSO优化SVM的惩罚因子C和径向基核函数参数g以获取最佳参数值,提高SVM分类性和稳定性,最后将模型应用于旱情数据的评估预测。仿真实验结果表明：本模型在平均准确率和方差方面的准确都取得了很好的效果;对比分别用PSO和遗传算法（GA）优化的SVM模型,本模型的性能更好。     

基于GWO的SVM在红外甲烷传感器测量误差分析中的应用

针对传统支持向量机回归模型应用在红外甲烷传感器测量数据处理时出现预测精度低的问题,提出了一种基于灰狼优化算法的支持向量机回归模型。该模型在传统支持向量机的基础上,利用灰狼优化算法自适应搜索特征空间来选择最佳特征组合,经过循环比较,能快速、准确地搜索到最优的惩罚因子C与gamma参数。用实验室研制的红外甲烷传感器对0～5.05%浓度范围的标准甲烷气体进行测量后,建立了3种SVM回归模型,并进行对比。结果表明,采用灰狼优化算法建立的支持向量机回归模型其绝对误差和相对误差小,精度高。     

基于SVM的高通量dPCR基因芯片荧光图像分类研究

目的 为了实现高通量dPCR基因芯片荧光图像的亮点分类与计数,提出一种基于支持向量机（SVM）的荧光图像分类与计数方法。方法 首先对荧光图像进行去噪、对比度增强等图像预处理,对预处理后荧光图像进行亮点区域提取标注,去除背景与暗点的冗余信息,利用方向梯度直方图（Histogram of Oriented Gradient, HOG）提取鉴别特征,计算合并所有样本的亮点特征得到HOG特征向量,根据已得到的HOG特征向量创建一个线性SVM分类器,利用训练好的SVM分类器对荧光图像亮点进行分类与计数。结果 对比传统算法,文中算法具有较高的分类识别精度,平均准确率高达98%以上,可以很好地实现荧光图像亮点分类与计数。结论 在有限的小样本标注数据下,文中算法具有良好的分类性能,能够有效识别荧光图像中的亮点,对其他荧光图像分类研究也具有一定参考价值。     

路堑开挖爆破中邻近民房风险评价的网格支持向量机模型

应用支持向量机理论并结合路堑开挖爆破特点,提出路堑开挖爆破中临近民房安全性评价的支持向量机回归模型。考虑爆破参数、地质条件和民房结构状况因素,选取最小抵抗线、孔距、排距、炸药单耗和民房的自振周期等16个影响较大的因素作为该模型的输入参数,房屋安全等级系数作为模型输出,利用网格搜索寻优方法对支持向量机模型的参数进行了优化。以19组路堑开挖爆破实测数据作为学习样本进行训练,对另外3组待判样本进行判别,并与多元回归、BP神经网络回归和实测结果进行对比。研究结果表明:建立的支持向量机回归模型对路堑开挖爆破中临近民房安全性评价效果良好,具有较高的预测精度。     

路堑开挖爆破中邻近民房风险评价的网格支持向量机模型

应用支持向量机理论并结合路堑开挖爆破特点,提出路堑开挖爆破中临近民房安全性评价的支持向量机回归模型。考虑爆破参数、地质条件和民房结构状况因素,选取最小抵抗线、孔距、排距、炸药单耗和民房的自振周期等16个影响较大的因素作为该模型的输入参数,房屋安全等级系数作为模型输出,利用网格搜索寻优方法对支持向量机模型的参数进行了优化。以19组路堑开挖爆破实测数据作为学习样本进行训练,对另外3组待判样本进行判别,并与多元回归、BP神经网络回归和实测结果进行对比。研究结果表明:建立的支持向量机回归模型对路堑开挖爆破中临近民房安全性评价效果良好,具有较高的预测精度。     

DeepESC网络的环境声分类方法研究

为进一步提升环境声分类的识别率，提出了一种仿深度隐藏身份特征（Deep Hidden Identity Feature，DeepID）网络连接方式的卷积神经网络——深度环境声分类网络（Deep Environment Sound Classification，DeepESC）。DeepESC网络共有六层——三层卷积层、两层全连层以及一层聚合层，为使网络在自动抽取高层次特征的同时能有效地兼顾低层次特征，网络将三层卷积层的输出聚合为一层，该层充分包含不同层次的特征，提升了卷积神经网络的特征表达能力。ESC-10和ESC-50数据集上的仿真结果表明：在相同的识别框架下，与随机森林分类器相比，本文网络识别率分别平均提升了7.6%和22.4%，与传统的卷积神经网络相比，识别率分别平均提升4%和2%，仿真实验验证了本文分类器的有效性。     

An Improved Deep Fusion CNN for Image Recognition

With the development of Deep Convolutional Neural Networks (DCNNs), the extracted features for image recognition tasks have shifted from low-level features to the high-level semantic features of DCNNs. Previous studies have shown that the deeper the network is, the more abstract the features are. However, the recognition ability of deep features would be limited by insufficient training samples. To address this problem, this paper derives an improved Deep Fusion Convolutional Neural Network (DF-Net) which can make full use of the differences and complementarities during network learning and enhance feature expression under the condition of limited datasets. Specifically, DF-Net organizes two identical subnets to extract features from the input image in parallel, and then a well-designed fusion module is introduced to the deep layer of DF-Net to fuse the subnet’s features in multi-scale. Thus, the more complex mappings are created and the more abundant and accurate fusion features can be extracted to improve recognition accuracy. Furthermore, a corresponding training strategy is also proposed to speed up the convergence and reduce the computation overhead of network training. Finally, DF-Nets based on the well-known ResNet, DenseNet and MobileNetV2 are evaluated on CIFAR100, Stanford Dogs, and UECFOOD-100. Theoretical analysis and experimental results strongly demonstrate that DF-Net enhances the performance of DCNNs and increases the accuracy of image recognition.     

一种改进的雁群扩展粒子群算法研究

针对粒子群算法易陷入局部最优的问题,结合雁群启示粒子群算法和扩展粒子群算法提出了基于雁群启示的扩展粒子群(GeEPSO)算法。该算法在利用雁群飞行方向的多样性同时融合了所有粒子的个体极值信息,提高了种群多样性。为进一步提高改进算法的收敛速度,引入简化粒子群提出了 GeESPSO算法。基准函数的仿真表明:改进算法GeESPSO较好地平衡了收敛速度和局部最优两个矛盾,总体较优。为进一步验证算法在实际应用中的有效性,又分别用两种改进算法优化BP神经网络,并用相关气象数据对PM2.5的值进行预测。     

基于粒子群算法优化支持向量机的铝热连轧机轧制力预报

为了提高热轧带材的轧制力预报精度,提出了粒子群算法和支持向量机结合的方法来预报轧制力。根据轧制原理用支持向量机建立轧制力预报的模型,通过粒子群算法优化支持向量机参数来提高预报精度。为了进一步提高轧制力预报精度,还提出了支持向量机网络与数学模型相结合的方法,对某“1+4”铝热连轧厂现场采集的5052铝合金轧制数据进行离线仿真,仿真结果可以看出支持向量机网络与数学模型结合的方法预报轧制力,提高了轧制力预报速度并使其轧制力预报精度控制在7%以内。     

Gaussian Support Vector Machine Algorithm Based Air Pollution Prediction

Air pollution is one of the major concerns considering detriments to human health. This type of pollution leads to several health problems for humans, such as asthma, heart issues, skin diseases, bronchitis, lung cancer, and throat and eye infections. Air pollution also poses serious issues to the planet. Pollution from the vehicle industry is the cause of greenhouse effect and CO2 emissions. Thus, real-time monitoring of air pollution in these areas will help local authorities to analyze the current situation of the city and take necessary actions. The monitoring process has become efficient and dynamic with the advancement of the Internet of things and wireless sensor networks. Localization is the main issue in WSNs; if the sensor node location is unknown, then coverage and power and routing are not optimal. This study concentrates on localization-based air pollution prediction systems for real-time monitoring of smart cities. These systems comprise two phases considering the prediction as heavy or light traffic area using the Gaussian support vector machine algorithm based on the air pollutants, such as PM2.5 particulate matter, PM10, nitrogen dioxide (NO2), carbon monoxide (CO), ozone (O3), and sulfur dioxide (SO2). The sensor nodes are localized on the basis of the predicted area using the meta-heuristic algorithms called fast correlation-based elephant herding optimization. The dataset is divided into training and testing parts based on 10 cross-validations. The evaluation on predicting the air pollutant for localization is performed with the training dataset. Mean error prediction in localizing nodes is 9.83 which is lesser than existing solutions and accuracy is 95%.     

Diabetes Prediction Algorithm Using Recursive Ridge Regression L2

At present, the prevalence of diabetes is increasing because the human body cannot metabolize the glucose level. Accurate prediction of diabetes patients is an important research area. Many researchers have proposed techniques to predict this disease through data mining and machine learning methods. In prediction, feature selection is a key concept in preprocessing. Thus, the features that are relevant to the disease are used for prediction. This condition improves the prediction accuracy. Selecting the right features in the whole feature set is a complicated process, and many researchers are concentrating on it to produce a predictive model with high accuracy. In this work, a wrapper-based feature selection method called recursive feature elimination is combined with ridge regression (L2) to form a hybrid L2 regulated feature selection algorithm for overcoming the overfitting problem of data set. Overfitting is a major problem in feature selection, where the new data are unfit to the model because the training data are small. Ridge regression is mainly used to overcome the overfitting problem. The features are selected by using the proposed feature selection method, and random forest classifier is used to classify the data on the basis of the selected features. This work uses the Pima Indians Diabetes data set, and the evaluated results are compared with the existing algorithms to prove the accuracy of the proposed algorithm. The accuracy of the proposed algorithm in predicting diabetes is 100%, and its area under the curve is 97%. The proposed algorithm outperforms existing algorithms.     

Kernel-like impurity detection according to colour band spectral image using GA/SVM

Kernel-like impurities (KLIs) have the similar colour, shape, texture and specific gravity with sound kernels. The amount of the KLIs is an important parameter for evaluating the quality of wheat. However, it is difficult to classify KLIs from sound kernels with normal methods because of these similar features. In this study, a machine vision system with a linear colour charged coupled device used to acquire images of kernels and a software package developed to extract various features from the images were used to classify 1169 sound kernels and 896 KLIs. Three methods—genetic algorithm (GA)/support vector machine (SVM), principal components analysis/SVM and linear discriminant analysis—were applied for the classification. The performance of GA/SVM for detecting KLIs was very outstanding, and the accuracy of testing sets could reach 99.34%. GA/SVM has the potential to improve the KLI classification accuracy in machine vision system. It is feasible to extract a small quantity of useful features without any extra image or data processing for online KLI classification.