Resource Type

Journal Article 361

Year

2023 55

2022 60

2021 52

2020 32

2019 36

2018 28

2017 28

2016 18

2015 3

2014 1

2013 1

2012 2

2011 5

2010 2

2009 2

2008 4

2007 8

2006 7

2005 3

2004 3

open ︾

Keywords

Machine learning 42

Deep learning 34

Artificial intelligence 15

Reinforcement learning 14

Active learning 4

Adaptive control 4

Additive manufacturing 3

Bayesian optimization 3

Big data 3

Adaptive dynamic programming 2

Attention 2

Autonomous driving 2

Autonomous learning 2

COVID-19 2

Data-driven 2

Feature learning 2

Materials design 2

Natural language processing 2

Neural network 2

open ︾

Search scope:

排序： Display mode:

Unsupervised object detection with scene-adaptive concept learning Research Articles

Shiliang Pu, Wei Zhao, Weijie Chen, Shicai Yang, Di Xie, Yunhe Pan,xiedi@hikvision.com

Frontiers of Information Technology & Electronic Engineering 2021, Volume 22, Issue 5,   Pages 615-766 doi: 10.1631/FITEE.2000567

Abstract: Object detection is one of the hottest research directions in computer vision, has already made impressive progress in academia, and has many valuable applications in the industry. However, the mainstream detection methods still have two shortcomings: (1) even a model that is well trained using large amounts of data still cannot generally be used across different kinds of scenes; (2) once a model is deployed, it cannot autonomously evolve along with the accumulated unlabeled scene data. To address these problems, and inspired by theory, we propose a novel scene-adaptive evolution algorithm that can decrease the impact of scene changes through the concept of object groups. We first extract a large number of object proposals from unlabeled data through a pre-trained detection model. Second, we build the dictionary of object concepts by clustering the proposals, in which each cluster center represents an object prototype. Third, we look into the relations between different clusters and the object information of different groups, and propose a graph-based group information propagation strategy to determine the category of an object concept, which can effectively distinguish positive and negative proposals. With these pseudo labels, we can easily fine-tune the pre-trained model. The effectiveness of the proposed method is verified by performing different experiments, and the significant improvements are achieved.

Keywords： 视觉知识；无监督视频目标检测；场景自适应学习    

HTML PDF Collect

Automatic image enhancement by learning adaptive patch selection None

Na LI, Jian ZHAN

Frontiers of Information Technology & Electronic Engineering 2019, Volume 20, Issue 2,   Pages 206-221 doi: 10.1631/FITEE.1700125

Abstract:

Today, digital cameras are widely used in taking photos. However, some photos lack detail and need enhancement. Many existing image enhancement algorithms are patch based and the patch size is always fixed throughout the image. Users must tune the patch size to obtain the appropriate enhancement. In this study, we propose an automatic image enhancement method based on adaptive patch selection using both dark and bright channels. The double channels enhance images with various exposure problems. The patch size used for channel extraction is selected automatically by thresholding a contrast feature, which is learned systematically from a set of natural images crawled from the web. Our proposed method can automatically enhance foggy or under-exposed/backlit images without any user interaction. Experimental results demonstrate that our method can provide a significant improvement in existing patch-based image enhancement algorithms.

Keywords： Image enhancement     Contrast enhancement     Dark channel     Bright channel     Adaptive patch based processing    

HTML PDF Collect

Dynamic parameterized learning for unsupervised domain adaptation Research Article

Runhua JIANG, Yahong HAN

Frontiers of Information Technology & Electronic Engineering 2023, Volume 24, Issue 11,   Pages 1616-1632 doi: 10.1631/FITEE.2200631

Abstract: enables neural networks to transfer from a labeled source domain to an unlabeled target domain by learning domain-invariant representations. Recent approaches achieve this by directly matching the marginal distributions of these two domains. Most of them, however, ignore exploration of the dynamic trade-off between and learning, thus rendering them susceptible to the problems of negative transfer and outlier samples. To address these issues, we introduce the dynamic parameterized learning framework. First, by exploring domain-level semantic knowledge, the dynamic alignment parameter is proposed, to adaptively adjust the of and learning. Besides, for obtaining semantic-discriminative and domain-invariant representations, we propose to align training trajectories on both source and target domains. Comprehensive experiments are conducted to validate the effectiveness of the proposed methods, and extensive comparisons are conducted on seven datasets of three visual tasks to demonstrate their practicability.

Keywords： Unsupervised domain adaptation     Optimization steps     Domain alignment     Semantic discrimination    

HTML PDF Collect

Layer-wise domain correction for unsupervised domain adaptation Article

Shuang LI, Shi-ji SONG, Cheng WU

Frontiers of Information Technology & Electronic Engineering 2018, Volume 19, Issue 1,   Pages 91-103 doi: 10.1631/FITEE.1700774

Abstract: Deep neural networks have been successfully applied to numerous machine learning tasks because of their impressive feature abstraction capabilities. However, conventional deep networks assume that the training and test data are sampled from the same distribution, and this assumption is often violated in real-world scenarios. To address the domain shift or data bias problems, we introduce layer-wise domain correction (LDC), a new unsupervised domain adaptation algorithm which adapts an existing deep network through additive correction layers spaced throughout the network. Through the additive layers, the representations of source and target domains can be perfectly aligned. The corrections that are trained via maximum mean discrepancy, adapt to the target domain while increasing the representational capacity of the network. LDC requires no target labels, achieves state-of-the-art performance across several adaptation benchmarks, and requires significantly less training time than existing adaptation methods.

Keywords： Unsupervised domain adaptation     Maximum mean discrepancy     Residual network     Deep learning    

HTML PDF Collect

Deep learning based wavefront sensor for complex wavefront detection in adaptive optical microscopes Research Articles

Shuwen Hu, Lejia Hu, Wei Gong, Zhenghan Li, Ke Si,weigong@zju.edu.cn,kesi@zju.edu.cn

Frontiers of Information Technology & Electronic Engineering 2021, Volume 22, Issue 10,   Pages 1277-1288 doi: 10.1631/FITEE.2000422

Abstract: The Shack-Hartmann wavefront sensor (SHWS) is an essential tool for wavefront sensing in adaptive optical microscopes. However, the distorted spots induced by the complex wavefront challenge its detection performance. Here, we propose a based method which combines point spread function image based Zernike coefficient estimation and wavefront stitching. Rather than using the centroid displacements of each micro-lens, this method first estimates the of local wavefront distribution over each micro-lens and then stitches the local wavefronts for reconstruction. The proposed method can offer low root mean square wavefront errors and high accuracy for complex , and has potential to be applied in adaptive optical microscopes.

Keywords： 自适应光学；波前探测；深度学习；泽尼克系数；显微成像    

HTML PDF Collect

Self-supervised graph learning with target-adaptive masking for session-based recommendation Research Article

Yitong WANG, Fei CAI, Zhiqiang PAN, Chengyu SONG,wangyitong20@nudt.edu.cn,caifei08@nudt.edu.cn,panzhiqiang@nudt.edu.cn,songchengyu@nudt.edu.cn

Frontiers of Information Technology & Electronic Engineering 2023, Volume 24, Issue 1,   Pages 73-87 doi: 10.1631/FITEE.2200137

Abstract: aims to predict the next item based on a user's limited interactions within a short period. Existing approaches use mainly recurrent neural networks (RNNs) or (GNNs) to model the sequential patterns or the transition relationships between items. However, such models either ignore the over-smoothing issue of GNNs, or directly use cross-entropy loss with a softmax layer for model optimization, which easily results in the over-fitting problem. To tackle the above issues, we propose a self-supervised graph learning with (SGL-TM) method. Specifically, we first construct a global graph based on all involved sessions and subsequently capture the self-supervised signals from the global connections between items, which helps supervise the model in generating accurate representations of items in the ongoing session. After that, we calculate the main supervised loss by comparing the ground truth with the predicted scores of items adjusted by our designed module. Finally, we combine the main supervised component with the auxiliary self-supervision module to obtain the final loss for optimizing the model parameters. Extensive experimental results from two benchmark datasets, Gowalla and Diginetica, indicate that SGL-TM can outperform state-of-the-art baselines in terms of Recall@20 and MRR@20, especially in short sessions.

Keywords： Session-based recommendation     Self-supervised learning     Graph neural networks     Target-adaptive masking    

HTML PDF Collect

Self-adaptive Creating of Truss Ground Structure Based on ANFIS

Li Ying,Hu Yunchang,Cao Hongduo

Strategic Study of CAE 2004, Volume 6, Issue 10,   Pages 24-27

Abstract:

The ground structure's intelligent and automatical creating is researched. The ground structures of truss structures are intelligently and automatically created by using collateral ANFIS (Adaptive-Network-Based Fuzzy Inference Systems). The essentials of this method is to form a mapping between configurations of the trusses existed and those to be found. The ground structure is inferred by the mechanism contained in sample trusses. To produce appropriate data format of ANFIS, the topology configuration of trusses is described as a series of decimal fraction. The simulation shows the efficiency of this system.

Keywords： truss     ANFIS     intelligent     self-adaptive    

HTML PDF Collect

An Adaptive Demodulation Method for BPSK Signals

Li Yanxin,Hu Aiqun,Song Yubo

Strategic Study of CAE 2006, Volume 8, Issue 5,   Pages 49-51

Abstract:

The paper presents a novel method for demodulating the binary phase shift keying (BPSK) signals basing on adaptive filtering. The commonly used least mean square (LMS) error adaptive filtering algorithm is employed for studying the demodulating procedure and the performance of the novel adaptive BPSK demodulation. The novel adaptive BPSK demodulation does not need the adaptive filter completing convergence. The performance of the method in theory is compared with computer simulating results. It shows that the error rates in simulation agree well with that in theory. Also, it is indicated that the demodulation method has many advantages over conventional ones, such as the powerful anti-noise ability, the small transfer delay, and the convenient implementation with DSP technology, and has lower error rates than correlation modulation at the same sample rate.

Keywords： digital communications     adaptive signal processing     demodulation     BPSK    

HTML PDF Collect

An ANFIS-based Approach for Predicting MiningInduced Surface Subsidence

Ding Dexin,Zhang Zhijun,Bi Zhongwei

Strategic Study of CAE 2007, Volume 9, Issue 1,   Pages 33-39

Abstract:

Current approaches for predicting mining induced surface subsidence have a drawback in common that they predict the subsidence only on the basis of a physical or mechanical approach irrespective of the practical examples in engineering practice in mining induced surface subsidence.However,these experiences created in engineering practice are of great value and full use should be made of them to establish an approach for predicting mining induced surface subsidence.Therefore,this paper accumulated a lot of practical examples of mining induced surface subsidence,integrated these examples by using adaptive neuro-fuzzy inference system (ANFIS)and established an ANFIS-based approach for predicting mining induced surface subsidence.The approach was further tested by using practical examples of mining induced surface subsidence.The results show that the approach can converge quickly,fit the data in very good agreement and make generalization prediction with high accuracy.

Keywords： underground mining     mining induced surface subsidence     adaptive neuro唱fuzzy inferencesystem    

HTML PDF Collect

Pricing Based Adaptive Call Admission Control Algorithm for Wireless Networks

Zhang Xue

Strategic Study of CAE 2006, Volume 8, Issue 4,   Pages 32-38

Abstract:

In order to efficiently and effectively control the use of wireless network resources, in this paper, according to the characteristic of adaptive multimedia applications in which bandwidths can be adjusted dynamically, and the influence of pricing on the users' behavior, an adaptive admission control algorithm integrated with pricing is proposed. The algorithm, in with the price is adjusted dynamically based on the current network conditions, is fit for the multi-priorilies services. Attempt is tried to make best balance between the efficiency and simplicity for the pricing scheme. Comparison of the performance of the proposed approach is made with the corresponding results of conventional systems where pricing is not taken into consideration in CAC process. The performance results verify the considerable improvement achieved by the integration of pricing with CAC in wireless networks.

Keywords： wireless networks     adaptive call admission control     microeconomic theory     pricing     connection level QoS    

HTML PDF Collect

Adaptive Wavelet Thresholding Denoising Used in Gravitational Signal Processing

Zhao Liye,Zhou Bailing,Li Kunyu

Strategic Study of CAE 2006, Volume 8, Issue 3,   Pages 49-52

Abstract:

The theory of wavelet thresholding denoising is analyzed and applied to process the data measured by gravimeter in order to effectively alleviate the effect of different noise in high precise gravitational system. The signal to noise ration (SNR) is used as the index for evaluating the performance of the data processing methods. Theoretical analysis and emulation experiments indicate that wavelet thresholding denoising, adaptive wavelet thresholding denoising and adaptive Kalman filtering are all effective in alleviating the effects of different noise, but the performance of adaptive wavelet thresholding denoising is best.

Keywords： gravimeter     signal processing     wavelet transform     adaptive wavelet threshold     adaptive Kalman     filering    

HTML PDF Collect

The Adaptive Robust Controller of the Centrifuge

Li Guo,Zhang Peichang,Hu Jianfei,Yu Dafei

Strategic Study of CAE 2006, Volume 8, Issue 9,   Pages 30-34

Abstract:

This paper investigates the use of the adaptive robust controller for improving control performance and stability of the centrifuge. Based on its structural merit that the electric motor is connected to the centrifuge, the implementation of a control system is expected to achieve satisfactory control performance. An adaptive robust control algorithm of the centrifuge is presented in the paper, and the adaptive robust controller is designed according to the centrifuge model. The effectiveness of the algorithm is verified by the experimental results. It is clarified that the control performance and stability of the centrifuge is improved and the control system still maintains satisfactory control performance despite the change of environment conditions.

Keywords： centrifuge     adaptive control     robust control    

HTML PDF Collect

High Precision Adaptive Predictive Control for Cruise Missile

Sun Mingwei,Chen Zengqiang,Yuan Zhuzhi,Ren Qiang,Yang Ming

Strategic Study of CAE 2005, Volume 7, Issue 10,   Pages 23-27

Abstract:

Cruise missile achieves good flight performance by means of stabilization and regulation of its attitudes. Based on analysis of the perturbation model of missile*s dynamic characteristics, series control structures are constructed for attitude control loop, and their discrete models are served as controlled plant for recursive least square (RLS) based adaptive predictive control, thus the mass center control with slow response transforms into trajectory angle control with fast response and high precision. On the basis of missile’s characteristics, generalized predictive control (GPC) is used in inner attitude loop, and an integral form of predictive control is adopted in outter trajectory loop. Effective transformation from mass center command to trajectory reference has achieved to realize high precision tracking. This method realizes the integration of attitude reference signal with guidance command, and that of attitude control with mass center control. It can reduce precision requirements on aerodynamic data and the control parameters can be easily selected. The numerical simulations demonstrate its effectiveness. Finally, some further academic directions are presented.

Keywords： cruise missile     adaptive control     model based predictive control     robustness    

HTML PDF Collect

Interactive image segmentation with a regression based ensemble learning paradigm Article

Jin ZHANG, Zhao-hui TANG, Wei-hua GUI, Qing CHEN, Jin-ping LIU

Frontiers of Information Technology & Electronic Engineering 2017, Volume 18, Issue 7,   Pages 1002-1020 doi: 10.1631/FITEE.1601401

Abstract: To achieve fine segmentation of complex natural images, people often resort to an interactive segmentation paradigm, since fully automatic methods often fail to obtain a result consistent with the ground truth. However, when the foreground and background share some similar areas in color, the fine segmentation result of conventional interactive methods usually relies on the increase of manual labels. This paper presents a novel interactive image segmentation method via a regression-based ensemble model with semi-supervised learning. The task is formulated as a non-linear problem integrating two complementary spline regressors and strengthening the robustness of each regressor via semi-supervised learning. First, two spline regressors with a complementary nature are constructed based on multivariate adaptive regression splines (MARS) and smooth thin plate spline regression (TPSR). Then, a regressor boosting method based on a clustering hypothesis and semi-supervised learning is proposed to assist the training of MARS and TPSR by using the region segmentation information contained in unlabeled pixels. Next, a support vector regression (SVR) based decision fusion model is adopted to integrate the results of MARS and TPSR. Finally, the GraphCut is introduced and combined with the SVR ensemble results to achieve image segmentation. Extensive experimental results on benchmark datasets of BSDS500 and Pascal VOC have demonstrated the effectiveness of our method, and the com-parison with experiment results has validated that the proposed method is comparable with the state-of-the-art methods for in-teractive natural image segmentation.

Keywords： Interactive image segmentation     Multivariate adaptive regression splines (MARS)     Ensemble learning     Thin-plate spline regression (TPSR)     Semi-supervised learning     Support vector regression (SVR)    

HTML PDF Collect

Improving entity linking with two adaptive features Research Article

Hongbin ZHANG, Quan CHEN, Weiwen ZHANG

Frontiers of Information Technology & Electronic Engineering 2022, Volume 23, Issue 11,   Pages 1620-1630 doi: 10.1631/FITEE.2100495

Abstract:

(EL) is a fundamental task in natural language processing. Based on neural networks, existing systems pay more attention to the construction of the , but ignore latent semantic information in the and the acquisition of effective information. In this paper, we propose two , in which the first adaptive feature enables the local and s to capture latent information, and the second adaptive feature describes effective information for embeddings. These can work together naturally to handle some uncertain information for EL. Experimental results demonstrate that our EL system achieves the best performance on the AIDA-B and MSNBC datasets, and the best average performance on out-domain datasets. These results indicate that the proposed , which are based on their own diverse contexts, can capture information that is conducive for EL.

Keywords： Entity linking     Local model     Global model     Adaptive features     Entity type    

HTML PDF Collect

Title Author Date Type Operation

Unsupervised object detection with scene-adaptive concept learning

Shiliang Pu, Wei Zhao, Weijie Chen, Shicai Yang, Di Xie, Yunhe Pan,xiedi@hikvision.com

Journal Article

Automatic image enhancement by learning adaptive patch selection

Na LI, Jian ZHAN

Journal Article

Dynamic parameterized learning for unsupervised domain adaptation

Runhua JIANG, Yahong HAN

Journal Article

Layer-wise domain correction for unsupervised domain adaptation

Shuang LI, Shi-ji SONG, Cheng WU

Journal Article

Deep learning based wavefront sensor for complex wavefront detection in adaptive optical microscopes

Shuwen Hu, Lejia Hu, Wei Gong, Zhenghan Li, Ke Si,weigong@zju.edu.cn,kesi@zju.edu.cn

Journal Article

Self-supervised graph learning with target-adaptive masking for session-based recommendation

Yitong WANG, Fei CAI, Zhiqiang PAN, Chengyu SONG,wangyitong20@nudt.edu.cn,caifei08@nudt.edu.cn,panzhiqiang@nudt.edu.cn,songchengyu@nudt.edu.cn

Journal Article

Self-adaptive Creating of Truss Ground Structure Based on ANFIS

Li Ying,Hu Yunchang,Cao Hongduo

Journal Article

An Adaptive Demodulation Method for BPSK Signals

Li Yanxin,Hu Aiqun,Song Yubo

Journal Article

An ANFIS-based Approach for Predicting MiningInduced Surface Subsidence

Ding Dexin,Zhang Zhijun,Bi Zhongwei

Journal Article

Pricing Based Adaptive Call Admission Control Algorithm for Wireless Networks

Zhang Xue

Journal Article

Adaptive Wavelet Thresholding Denoising Used in Gravitational Signal Processing

Zhao Liye,Zhou Bailing,Li Kunyu

Journal Article

The Adaptive Robust Controller of the Centrifuge

Li Guo,Zhang Peichang,Hu Jianfei,Yu Dafei

Journal Article

High Precision Adaptive Predictive Control for Cruise Missile

Sun Mingwei,Chen Zengqiang,Yuan Zhuzhi,Ren Qiang,Yang Ming

Journal Article

Interactive image segmentation with a regression based ensemble learning paradigm

Jin ZHANG, Zhao-hui TANG, Wei-hua GUI, Qing CHEN, Jin-ping LIU

Journal Article

Improving entity linking with two adaptive features

Hongbin ZHANG, Quan CHEN, Weiwen ZHANG

Journal Article