- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
PV shading fault detection and classification based on I-V curve using principal component analysis: Application to isolated PV system
摘要： Health monitoring and diagnosis of photovoltaic (PV) systems is becoming crucial to maximise the power production, increase the reliability and life service of PV power plants. Operating under faulty conditions, in particular under shading, PV plants have remarkable shape of current-voltage (I-V) characteristics in comparison to reference condition (healthy operation). Based on real electrical measurements (I-V), the present work aims to provide a very simple, robust and low cost Fault Detection and Classi?cation (FDC) method for PV shading faults. At ?rst, we extract the features for di?erent experimental tests under healthy and shading conditions to build the database. The features are then analysed using Principal Component Analysis (PCA). The accuracy of the data classi?cation into the PCA space is evaluated using the confusion matrix as a metric of class separability. The results using experimental data of a 250 Wp PV module are very promising with a successful classi?cation rate higher than 97% with four di?erent con?gurations. The method is also cost e?ective as it uses only electrical measurements that are already available. No additional sensors are required.
关键词： Fault classi?cation,Principal component analysis,Fault detection,I-V curves,PV shading faults
更新于2025-09-23 15:23:52
-
[IEEE 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC) - Chongqing (2018.6.27-2018.6.29)] 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC) - Instance Selection in the Projected High Dimensional Feature Space for SVM
摘要： Data classi?cation is a supervised learning task where a training set with previously known information is used to construct a classi?er. The classi?er is then used to predict the class of unforeseen test instances. It is often bene?cial to use a subset of the training set to construct the classi?er, in particular when the size of the data set is large. For example, support vector machine (SVM), one of the most effective classi?ers, only needs the support vectors to make the prediction. Therefore, all non-support vectors can be eliminated without affecting the classi?cation performance. However, it is usually unknown which instances in the training set are support vectors before the training is completed. Researchers have developed different methods to delete the potential non-support vectors while retaining the likely support vectors before the training starts. This preprocessing to the training data set is often known as instance selection. Many of the instance selection methods are based on the geometry of the training samples. Measures in the original feature space are usually used. We propose to use measures in the projected high dimensional feature space for SVM since this is where the separating hyperplanes are determined. We compare the performance with some existing methods on a few benchmark data sets. The experiments show that using measures in the projected feature space may improve the classi?cation accuracy, sometimes substantially.
关键词： data classi?cation,SVM,instance selection,support vector machine
更新于2025-09-23 15:23:52
-
An Automated Galaxy Spectra Recognition Method Basing on Spectral Lines Information
摘要： For the vast amounts of spectra produced by LAMOST, the pipeline basing on PCAZ method is limited by the bad ?ux calibration and low S/N data. This work focuses on the study of the e?cient recognition methods of galaxy spectra of LAMOST basing on spectral lines information. The new method searches spectral lines and extracts the information of spectral lines (position, height, and width et al.) automatically. Using the spectral lines information which are less in?uenced by the quality of ?ux calibration and the S/N ratio, galaxy spectra are recognized with the redshift measured through spectral lines matching method. The experiment veri?ed it is feasible for the LAMOST galaxy spectra: the correct recognition rate > 80% for the data with SN R g > 5, and > 90% for the data with SN R r > 5. Compared with the redshift of SDSS, the systematic error of our method is 0, and the standard deviation of the error is 0.0002.
关键词： galaxies: fundamental parameters (classi?cation,methods: data analysis,telescopes:LAMOST,luminosities,colors,radii,techniques: spectroscopic,etc.),masses
更新于2025-09-23 15:22:29
-
Extended attribute profiles on GPU applied to hyperspectral image classification
摘要： Extended pro?les are an important technique for modelling the spatial information of hyperspectral images at different levels of detail. They are used extensively as a pre-processing stage, especially in classi?cation schemes. In particular, attribute pro?les, based on the application of morphological attribute ?lters to the connected components of the image, have been shown to provide very good results. In this paper we present a parallel implementation of the attribute pro?les in CUDA for multispectral and hyperspectral imagery considering the attributes area and standard deviation. The pro?le computation is based on the max-tree approach but without building the tree itself. Instead, a matrix-based data structure is used along with a recursive ?ooding (component merging) and ?lter process. Additionally, a previous feature extraction stage based on wavelets is applied to the hyperspectral image in order to extract the most valuable spectral information, reducing the size of the resulting pro?le. This scheme ef?ciently exploits the thousands of available threads on the GPU, obtaining a considerable reduction in execution time as compared to the OpenMP CPU implementation.
关键词： Remote sensing,Attribute pro?les,GPU,Real-time,Hyperspectral,Supervised classi?cation
更新于2025-09-23 15:22:29
-
[IEEE 2018 IEEE International Conference on Multimedia and Expo (ICME) - San Diego, CA (2018.7.23-2018.7.27)] 2018 IEEE International Conference on Multimedia and Expo (ICME) - Feature Reinforcement Network for Image Classification
摘要： Deep Learning has attracted much attention these years as it produces fabulous performance in various applications. Most researchers have mainly focused on improving and optimizing the network structure, e.g., deeper and deeper networks are constructed to extract high-level features from raw data. In this paper, we propose a two-wing deep convolutional network, called Feature Reinforcement Networks (FRN). One wing acts as the traditional operation in VGG, ResNet and DenseNet, while the other wing called feature reinforcement block (FRB) also conducts layer-wise convolution operations which share the convolution parameters of the former layer. Then, Relu function is employed in FRB to rectify the feature maps except the output layer. The outputs of these wings are integrated as the input of the next convolution layer. It is con?rmed that the proposed FRN is more sensitive to the informative features. Our experiments on a few multimedia datasets prove FRN outperforms the original deep neural networks.
关键词： Deep learning,Two-wing deep convolutional network,Feature reinforcement block,Image classi?cation
更新于2025-09-23 15:21:01
-
Hyperspectral Image Classification Based on Improved Rotation Forest Algorithm
摘要： Hyperspectral image classi?cation is a hot issue in the ?eld of remote sensing. It is possible to achieve high accuracy and strong generalization through a good classi?cation method that is used to process image data. In this paper, an ef?cient hyperspectral image classi?cation method based on improved Rotation Forest (ROF) is proposed. It is named ROF-KELM. Firstly, Non-negative matrix factorization( NMF) is used to do feature segmentation in order to get more effective data. Secondly, kernel extreme learning machine (KELM) is chosen as base classi?er to improve the classi?cation ef?ciency. The proposed method inherits the advantages of KELM and has an analytic solution to directly implement the multiclass classi?cation. Then, Q-statistic is used to select base classi?ers. Finally, the results are obtained by using the voting method. Three simulation examples, classi?cation of AVIRIS image, ROSIS image and the UCI public data sets respectively, are conducted to demonstrate the effectiveness of the proposed method.
关键词： extreme learning machine,rotation forest,hyperspectral image classi?cation,Q-statistic
更新于2025-09-23 15:21:01
-
Cloud Masking Technique for High-Resolution Satellite Data: An Artificial Neural Network Classifier Using Spectral & Textural Context
摘要： Cloud masking is a very important application in remote sensing and an essential pre-processing step for any information derivation applications. It helps in estimation of usable portion of the images. Many popular spectral classi?cation techniques rely upon the presence of a short-wave infrared band or bands of even higher wavelength to differentiate between clouds and other land covers. However, these methods are limited to sensors equipped with higher wavelength bands. In this paper, a generic and ef?cient technique is attempted using the Cartosat-2 series (C2S) satellite which is having high-resolution multispectral sensor in the visible and near-infrared bands. The methodology is based on textural features from the available spectral context, and using a feedforward neural network for the classi?cation is proposed. The method was shown to have an overall accuracy of 97.98% for a large manually pre-classi?ed validation dataset with more than 2 million data points. Experimental results and cloud masks generated for various scenes show that the method may be viable as a reasonable cloud masking algorithm for C2S data.
关键词： Cloud masking,Feed forward network,High-resolution satellite data,Image classi?cation,Arti?cial neural network,GLCM texture
更新于2025-09-23 15:21:01
-
Classifications of Forest Change by Using Bitemporal Airborne Laser Scanner Data
摘要： Changes in forest areas have great impact on a range of ecosystem functions, and monitoring forest change across di?erent spatial and temporal resolutions is a central task in forestry. At the spatial scales of municipalities, forest properties and stands, local inventories are carried out periodically to inform forest management, in which airborne laser scanner (ALS) data are often used to estimate forest attributes. As local forest inventories are repeated, the availability of bitemporal ?eld and ALS data is increasing. The aim of this study was to assess the utility of bitemporal ALS data for classi?cation of dominant height change, aboveground biomass change, forest disturbances, and forestry activities. We used data obtained from 558 ?eld plots and four repeated ALS-based forest inventories in southeastern Norway, with temporal resolutions ranging from 11 to 15 years. We applied the k-nearest neighbor method for classi?cation of: (i) increasing versus decreasing dominant height, (ii) increasing versus decreasing aboveground biomass, (iii) undisturbed versus disturbed forest, and (iv) forestry activities, namely untouched, partial harvest, and clearcut. Leave-one-out cross-validation revealed overall accuracies of 96%, 95%, 89%, and 88% across districts for the four change classi?cations, respectively. Thus, our results demonstrate that various changes in forest structure can be classi?ed with high accuracy at plot level using data from repeated ALS-based forest inventories.
关键词： classi?cation,dominant height,forest change,ALS,forestry activity,aboveground biomass,disturbance,forest
更新于2025-09-23 15:19:57
-
A High-Performance Deep Learning Algorithm for the Automated Optical Inspection of Laser Welding
摘要： The battery industry has been growing fast because of strong demand from electric vehicle and power storage applications.Laser welding is a key process in battery manufacturing. To control the production quality, the industry has a great desire for defect inspection of automated laser welding. Recently, Convolutional Neural Networks (CNNs) have been applied with great success for detection, recognition, and classi?cation. In this paper, using transfer learning theory and pre-training approach in Visual Geometry Group (VGG) model, we proposed the optimized VGG model to improve the e?ciency of defect classi?cation. Our model was applied on an industrial computer with images taken from a battery manufacturing production line and achieved a testing accuracy of 99.87%. The main contributions of this study are as follows: (1) Proved that the optimized VGG model, which was trained on a large image database, can be used for the defect classi?cation of laser welding. (2) Demonstrated that the pre-trained VGG model has small model size, lower fault positive rate, shorter training time, and prediction time; so, it is more suitable for quality inspection in an industrial environment. Additionally, we visualized the convolutional layer and max-pooling layer to make it easy to view and optimize the model.
关键词： defect classi?cation,optimized VGG model,laser welding,convolutional neural networks (CNNs),automatic optical inspection
更新于2025-09-23 15:19:57
-
[IEEE 2019 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC) - Munich, Germany (2019.6.23-2019.6.27)] 2019 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC) - Challenges in the Miniaturization of Mid-IR Sensors Fully Integrated on Si
摘要： Accurate classi?cation of biological phenotypes is an essential task for medical decision making. The selection of subjects for classi?er training and validation sets is a crucial step within this task. To evaluate the impact of two approaches for subject selection—randomization and clinical balancing, we applied six classi?cation algorithms to a highly replicated publicly available breast cancer data set. Using six performance metrics, we demonstrate that clinical balancing improves both training and validation performance for all methods on average. We also observed a smaller discrepancy between training and validation performance. Furthermore, a simple analytical argument is presented which suggests that we need only two metrics from the class of metrics based on the entries of the confusion matrix. In light of our results, we recommend: 1) clinical balancing of training and validation data to improve signal-to-noise ratio and 2) the use of multiple classi?cation algorithms and evaluation metrics for a comprehensive evaluation of the decision making process.
关键词： strati?ed sampling.,performance metrics,Classi?cation,confusion matrix,random sampling
更新于2025-09-23 15:19:57