- 标题
- 摘要
- 关键词
- 实验方案
- 产品
过滤筛选
- 2018
- Conditional Random Fields (CRF)
- Convolutional Neural Network (CNN)
- Fine Classification
- Airborne hyperspectral
- Optoelectronic Information Science and Engineering
- Wuhan University
- Central South University
- Hubei University
-
Hyperspectral inversion of heavy metal content in reclaimed soil from a mining wasteland based on different spectral transformation and modeling methods
摘要： Conventional methods for investigating heavy metal contamination in soil are time consuming and expensive. We explored reflectance spectroscopy as an alternative method for assessing heavy metals. Four spectral transformation methods, first-order differential (FDR), second-order differential (SDR), continuum removal (CR) and continuous wavelet transform (CWT), are used for the original spectral data. Spectral preprocessing effectively eliminated the noise and baseline drifting and also highlighted the locations of the spectral feature bands. Partial least squares regression (PLSR) and radial basis function neural network (RBF) were used to study the hyperspectral inversion of four heavy metals (Cr, As, Ni, Cd). The inversion models of four heavy metals were established in the bands with the highest correlation coefficient. The inversion effects were evaluated by the coefficient of determination (R2), root mean square error (RMSE) and residual predictive deviation (RPD) indexes. The R values of the correlation coefficient were significantly improved after smoothing and spectral transformation compared to the original waveband. The method combining continuous wavelet transform (CWT) with radial basis function neural network (RBF) had the best inversion effect on the four heavy metals. When compared to partial least squares regression (PLSR), the RMSE values were reduced by approximately 2. The CWT-RBF method can be used as a means of inversion of heavy metals in mining wasteland reclaimed land.
关键词： Continuous Wavelet Transform,Heavy metal,Spectral analysis,Radial Basis Function Neural Network,Reclamation soil
更新于2025-09-04 15:30:14
-
[Lecture Notes in Computer Science] Pattern Recognition and Computer Vision Volume 11257 (First Chinese Conference, PRCV 2018, Guangzhou, China, November 23-26, 2018, Proceedings, Part II) || Deep Supervised Auto-encoder Hashing for Image Retrieval
摘要： Image hashing approaches map high dimensional images to compact binary codes that preserve similarities among images. Although the image label is important information for supervised image hashing methods to generate hashing codes, the retrieval performance will be limited according to the performance of the classi?er. Therefore, an e?ective supervised auto-encoder hashing method (SAEH) is proposed to generate low dimensional binary codes in a point-wise manner through deep convolutional neural network. The auto-encoder structure in SAEH is designed to simultaneously learn image features and generate hashing codes. Moreover, some extra relaxations for generating binary hash codes are added to the objective function. The extensive experiments on several large scale image datasets validate that the auto-encoder structure can indeed increase the performance for supervised hashing and SAEH can achieve the best image retrieval results among other prominent supervised hashing methods.
关键词： Image hashing,Image retrieval,Supervised learning,Deep neural network,Convolutional auto-encoder
更新于2025-09-04 15:30:14
-
[Lecture Notes in Computer Science] Neural Information Processing Volume 11301 (25th International Conference, ICONIP 2018, Siem Reap, Cambodia, December 13-16, 2018, Proceedings, Part I) || Proposal of Complex-Valued Convolutional Neural Networks for Similar Land-Shape Discovery in Interferometric Synthetic Aperture Radar
摘要： We propose a complex-valued convolutional neural network to extract the areas having land shapes similar to samples in interferometric synthetic aperture radar (InSAR). InSAR extends its application to various earth observations such as volcano monitoring and earthquake damage estimation. Since the amount of data is increasing drastically in these years, it is necessary to structurize them in a big data framework. In this paper, experiments demonstrate that similar small volcanoes are grouped into a single class. We ?nd that the neural network is capable of discovering unidenti?ed lands similar to prepared samples successfully.
关键词： Complex-valued neural network (CVNN),Interferometric synthetic aperture radar (InSAR),Feature discovery
更新于2025-09-04 15:30:14
-
Rapid tracking of extrinsic projector parameters in fringe projection using machine learning
摘要： In this work, we propose to enable the angular re-orientation of a projector within a fringe projection system in real-time without the need for re-calibrating the system. The estimation of the extrinsic orientation parameters of the projector is performed using a convolutional neural network and images acquired from the camera in the setup. The convolutional neural network was trained to classify the azimuth and elevation angles of the projector approximated by a point source through shadow images of the measured object. The images used to train the neural network were generated through the use of CAD rendering, by simulating the illumination of the object model from di?erent directions and then rendering an image of its shadow. The accuracy to which the azimuth and elevation angles are estimated is within 1 classi?cation bin, where 1 bin is designated as a ± 10° patch of the illumination dome. To evaluate use of the proposed system in fringe projection, a pyramidal additively manufactured object was measured. The point clouds generated using the proposed method were compared to those obtained by an established fringe projection calibration method. The maximum dimensional error in the point cloud generated when using the convolutional network as compared to the established calibration method for the object measured was found to be 1.05 mm on average.
关键词： real-time tracking,convolutional neural network,fringe projection,machine learning,projector calibration
更新于2025-09-04 15:30:14
-
[IEEE 2018 26th International Conference on Geoinformatics - Kunming, China (2018.6.28-2018.6.30)] 2018 26th International Conference on Geoinformatics - An Improved Bag-of-Visual-Word Based Classification Method for High-Resolution Remote Sensing Scene
摘要： Remote sensing (RS) scene classification is important for RS imagery semantic interpretation. Yet complex scenes make the task difficult. The Bag-of-Visual-Words (BoVW) method is an effective method for RS scene classification while most BoVW methods only consider local features and ignore the import global features of the scene. This paper aims to improve the traditional scale-invariant feature transform (SIFT) based Bag-of-Visual-Words (BoVW) method which only captures local information by fusing a global feature extracted from deep convolutional neural network (DCNN) for high-resolution remote sensing (HRRS) scene classification. The proposed method enhances representation ability for HRRS scenes by considering local and global features simultaneously and outperforms the sate-of-the-arts for obtaining accuracies of 95% on the widely used UC Merced dataset and SIRI-WHU dataset.
关键词： Bag-of-Visual-Words (BoVW),scene classification,scale-invariant feature transform (SIFT),deep convolutional neural network (DCNN),high-resolution remote sensing (HRRS) scene
更新于2025-09-04 15:30:14
-
[ACM Press the 24th ACM Symposium - Tokyo, Japan (2018.11.28-2018.12.01)] Proceedings of the 24th ACM Symposium on Virtual Reality Software and Technology - VRST '18 - Real-time human motion forecasting using a RGB camera
摘要： We propose a real-time human motion forecasting system which visualize the future pose in virtual reality using a RGB camera. Our system consists of three parts: 2D pose estimation from RGB frames using a residual neural network, 2D pose forecasting using a recurrent neural network, and 3D recovery from the predicted 2D pose using a residual linear network. To improve the prediction learning quantity of temporal feature, we propose a special method using lattice optical flow for the joints movement estimation. After fitting the skeleton, a predicted 3d model of target human will be built 0.5s in advance in a 30-fps video.
关键词： Deep neural network,Real-time pose prediction,Motion forecasting
更新于2025-09-04 15:30:14
-
[IEEE 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China (2018.8.20-2018.8.24)] 2018 24th International Conference on Pattern Recognition (ICPR) - Multi-layer CNN Features Aggregation for Real-time Visual Tracking
摘要： In this paper, we propose a novel convolutional neural network (CNN) based tracking framework, which aggregates multiple CNN features from different layers into a robust representation and realizes real-time tracking. We found that some feature maps have interference for effectively representing objects. Instead of using original features, we build an end-to-end feature aggregation network (FAN) which suppresses the noisy feature maps of CNN layers. The feature significantly benefits to represent objects with both coarse semantic information and fine details. The FAN, as a light-weight network, can run at real-time. The highlighted region of feature maps obtained from the FAN is the tracking result. Our method performs at a real-time speed of 24 fps while maintaining a promising accuracy compared with state-of-the-art methods on existing tracking benchmarks.
关键词： real-time tracking,convolutional neural network,feature aggregation,visual tracking
更新于2025-09-04 15:30:14
-
Robust fusion algorithm based on RBF neural network with TS fuzzy model and its application to infrared flame detection problem
摘要： A robust fusion algorithm based on Radial Basis Function (RBF) neural network with Takagi-Sugeno (TS) fuzzy model is proposed in view of the data loss, data distortion or signal saturation which is usually occurred in the process of infrared flame detecting with multiple sensors. To initialize the model, the traditional K-means clustering algorithm is used to obtain the number of the fuzzy rules and the center of the membership function. Compared with the traditional RBF neural network with TS fuzzy model, the output of the node in the proposed model is constructed taking into account the membership degree of the feature components in each item of the output polynomial of the hidden layer nodes in consequent fuzzy network. A new weighted activation degree (WAD) is defined to calculate the firing strength (i.e., fuzzy rule applicability) of the fuzzy node instead of the commonly used Mahalanobis distance. The feature representation coefficients used in the above WAD fully consider the variant representation degree of different features in different fuzzy clusters, thus the developed method can deal with the abnormal outputs of the fuzzy rules caused by the variation of the feature components of the raw data obtained from the complex industrial environments. The robustness of the proposed approach is validated with experimental data obtained from a developed triple-channel infrared flame detector and the experiment results show that the convergence rate, accuracy and generalization ability of the proposed method are improved compared with the traditional RBF neural network with TS fuzzy model in [1] and the GA-BP (Genetic Algorithm-Back Propagation) model in [2]. In particular, the required number of the hidden layer nodes in the proposed approach is the least among the aforementioned methods.
关键词： robustness,feature representation coefficient,Infrared flame detector,TS fuzzy model,RBF neural network
更新于2025-09-04 15:30:14
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Diversifying Deep Multiple Choices for Remote Sensing Scene Classification
摘要： Recently, deep models have shown powerful ability for remote sensing scene representation. However, the training process of these deep methods requires large amount of labelled samples while usual remote sensing image datasets cannot provide enough training samples. Therefore, the learned model is usually suboptimal. To solve the problem, this work focuses on obtaining multiple choices by training multiple models simultaneously, and then the human oracle can choose a proper one from these choices. However, training several models separately usually makes the obtained results similar. This paper tries to diversify the obtained choices by encouraging the obtained choices to repulse from each other. Experiments are conducted on Ucmerced Land Use dataset to validate the effectiveness of the proposed method to provide multiple diversified choices.
关键词： Remote Sensing Image,Cross Entropy,Diversity,Convolutional Neural Network,Classification
更新于2025-09-04 15:30:14
-
Neural network prediction of K and L-shell X-ray production cross sections
摘要： The ionization and X-ray production cross section are fundamental parameters in elemental analysis by PIXE technique. Unfortunately no exact general analytical expression exists, from which the interest of this work. In this paper, we apply the neural network technique in the evaluation of the X-ray production cross sections. The calculations are based on Mukoyama’s PWBA data. Our results are compared with experimental data for protons and alpha particles for energies ranging from hundreds KeV to tens MeV.
关键词： PWBA,cross section,PIXE,neural network
更新于2025-09-04 15:30:14