检索结果-内蒙古大学图书馆

Dynamics of Masked image Modeling in Hyperspectral image Classification

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Ma, Chen Li, Huayi Jiang, Junjun Aybar, Cesar Yao, Jiaqi Camps-Valls, Gustau Harbin Institute of Technology Research Center of Satellite Technology State Key Laboratory of Micro-Spacecraft Rapid Design and Intelligent Cluster School of Astronautics Harbin150001 China Harbin Institute of Technology School of Computer Science and Technology Harbin150001 China Image Signal Processing Laboratory University of València València46980 Spain Tianjin Normal University Academy of Eco-civilization Development for Jing-Jin-Ji Megalopolis Tianjin300387 China

Masked image modeling (MIM), a common selfsupervised learning (SSL) technique, has been extensively studied for remote sensing image processing. Nevertheless, its effectiveness for hyperspectral imagery (HSI) remains underexplored due to the distinct data structures and high dimensionality. This paper aims to provide a detailed understanding of MIM from different perspectives of representation learning and statistical analysis for HSI classification tasks. Our study reveals that the MIM paradigm injects inductive bias in the attention mechanism of the transformer model, which is advantageous for capturing the local discrepancies between the spectra. We also show that MIM can increase the diversity of the attention heads in every layer, which is beneficial for the model in extracting more discriminative features from different spectral bands. The similarity of representations from various layers further proves this. Furthermore, our investigation highlights how MIM introduces a dynamic perspective to spectral representations, enabling the model to learn more robust and discriminative features. The final numerical experiments indicate that a moderate mask ratio can enhance the performance of downstream tasks. This suggests that designing a more targeted masking strategy might be necessary to achieve higher and more stable gains in downstream task performance. Without bells and whistles, the vanilla MIM improves the overall classification accuracy by an average of 2.69% over its SL counterpart. We hope that our findings can advance the understanding of MIM in HSI and inspire the design of a more stable SSL paradigm for HSI processing. © 1980-2012 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Asymptotically Optimal Sequence Sets With Low/Zero Ambiguity Zone Properties

引用

IEEE Transactions on Information Theory 2025年

作者： Tian, Liying Song, Xiaoshi Liu, Zilong Li, Yubo Northeastern University School of Computer Science and Engineering Shenyang110819 China Northeastern University Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Shenyang110189 China University of Essex School of Computer Science and Electronics Engineering ColchesterCO4 3SQ United Kingdom Yanshan University School of Information Science and Engineering Qinhuangdao066004 China Hebei Key Laboratory of Information Transmission and Signal Processing Qinhuangdao066004 China

Sequences with low/zero ambiguity zone (LAZ/ZAZ) properties are useful in modern communication and radar systems operating over mobile environments. This paper first presents a new family of ZAZ sequence sets motivated by the "modulating" zero correlation zone (ZCZ) sequences which were first proposed by Popovic and Mauritz. We then introduce a second family of ZAZ sequence sets with comb-like spectrum, whereby the local Doppler resilience is guaranteed by their inherent spectral nulls in the frequency domain. Finally, LAZ sequence sets are obtained by exploiting their connection with a novel class of mapping functions. These proposed unimodular ZAZ and LAZ sequence sets are cyclically distinct and asymptotically optimal with respect to the existing theoretical bounds on ambiguity functions. © 1963-2012 IEEE.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

A NEW ALGORITHM FOR ADAPTIVE ARRAYS

引用

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND signal processing 1987年第9期35卷 1314-1319页

作者： GARDNER, WA BROWN, WA Signal and Image Processing Laboratory Department of Electrical and Computer Engineering University of California슠Davis Davis CA USA Signal and Image Processing Laboratory Department of Electrical and Computer Engineering University of California Davis Davis CA USA

A new deflected stochastic-gradient algorithm for adaptive sensor arrays is introduced. It is shown to have the capability of converging substantially faster than the nondeflected stochastic gradient (LMS) algorithm but slower than the recursive least squares (RLS) algorithm. Because of its amenability to reduction of computational complexity by data quantization, it has the potential for processing digital signals with bandwidths one to two orders of magnitude larger than the bandwidths manageable with the RLS algorithm for sensor arrays containing a number of adjustable weights in the approximate range of 10-50.

关键词： Adaptive arrays Sensor arrays Least squares approximation Resonance light scattering Bandwidth Stochastic processes Least squares methods Computational complexity Quantization signal processing

来源：评论

学校读者我要写书评

暂无评论

RECURSIVE BLOCK CODING - A NEW APPROACH TO TRANSFORM CODING

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 1986年第2期34卷 161-179页

作者： FARRELLE, PM JAIN, AK Signal and Image Processing Laboratory Department of Electrical and Computer Engineering University of California Davis CA USA

The concept of fast KL transform coding introduced earlier [7], [8] for first-order Markov processes and certain random fields has been extended to higher order autoregressive (AR) sequences and practical images yielding what we call recursive block coding (RBC) algorithms. In general, the rate-distortion performance for these algorithms is significantly superior to that of the conventional block KL transform algorithm. Moreover, these algorithms permit the use of small size transforms, thereby removing the need for fast transforms and making the hardware implementation of such coders more appealing. This improved performance has been verified for practical image data and results in suppression of the block-boundary effect commonly observed in traditional transform coding techniques. This is illustrated by comparing RBC with cosine transform coding using both one- and twodimensional algorithms. Examples of RBC encoded images at various rates are given.

关键词： Transform coding Karhunen-Loeve transforms Block codes Data compression image coding signal processing image processing Bit rate Markov processes Rate-distortion

来源：评论

学校读者我要写书评

暂无评论

SPECTRAL CORRELATION OF MODULATED signalS - .1. ANALOG MODULATION

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 1987年第6期35卷 584-594页

作者： GARDNER, WA Signal and Image Processing Laboratory Department of Electrical and Computer Engineering University of California Davis CA USA

The importance of the concept of cyclostationarity in design and analysis of signal detectors, synchronizers, and extractors in communication systems is briefly discussed, and the central role of spectral correlation, in the characterization of random processes that are cyclostationary in the wide sense, is explained. A spectral correlation function that is a generalization of the power spectral density function is described, and a corresponding generalization of the Wiener-Khinchine relation and several other fundamental spectral correlation relations also are described. Explicit formulas for the spectral correlation function for various types of analog-modulated signals are derived. This includes pulse and carrier amplitude modulation, quadrature amplitude carrier modulation, and phase and frequency carrier modulation. To illustrate the differing spectral correlation characteristics of different modulation types, the magnitudes of the spectral correlation functions are graphed or described in graphical terms as the heights of surfaces above a bifrequency plane.

关键词： Pulse modulation Amplitude modulation Phase modulation Quadrature amplitude modulation signal design signal analysis signal processing signal detection Detectors Frequency synchronization

来源：评论

学校读者我要写书评

暂无评论

SPECTRAL CORRELATION OF MODULATED signalS .2. DIGITAL MODULATION

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 1987年第6期35卷 595-601页

作者： GARDNER, WA BROWN, WA CHEN, CK Signal and Image Processing Laboratory Department of Electrical and Computer Engineering University of California Davis CA USA

As a continuation of Part I, the spectral correlation function is presented for a variety of types of digitally modulated signals. These include digital pulse-amplitude, pulse-width, and pulse-position modulation, and various types of phase-shift keying and frequency-shift keying. The magnitudes of the spectral correlation functions are graphed as the heights of surfaces above a bifrequency plane, and these graphs are used as visual aids for comparison and contrast of the spectral correlation properties of different modulation types.

关键词： Digital modulation Space vector pulse width modulation signal processing Pulse modulation Phase shift keying Frequency shift keying Pulse width modulation Filters Matrices Parameter estimation

来源：评论

学校读者我要写书评

暂无评论

INTERFERENCE-TOLERANT TIME-DIFFERENCE-OF-ARRIVAL ESTIMATION FOR MODULATED signalS

引用

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND signal processing 1988年第9期36卷 1385-1395页

作者： GARDNER, WA CHEN, CK Signal and Image Processing Laboratory Department of Electrical Engineering and Computer Science University of California슠Davis Davis CA USA Signal and Image Processing Laboratory Department of Electrical Engineering and Computer Science University of California Davis Davis CA USA

A method for estimation of the difference in the times of arrival of wavefronts of two separate sensors is introduced. The method, called SPECCORR, exploits the spectral correlation property, called spectral coherence, that essentially all modulated signals exhibit to obtain estimates that are highly tolerant to severely corruptive noise and interference. This tolerance is explained theoretically and demonstrated with simulations.< >

关键词： Interference Time difference of arrival Noise measurement Narrowband Autocorrelation Direction of arrival estimation Antenna measurements Design methodology signal design

来源：评论

学校读者我要写书评

暂无评论

signal INTERCEPTION - A UNIFYING THEORETICAL FRAMEWORK FOR FEATURE DETECTION

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 1988年第8期36卷 897-906页

作者： GARDNER, WA Signal and Image Processing Laboratory Department of Electrical Engineering and Computer Science University of California Davis CA USA

The unifying framework of the spectral-correlation theory of cyclostationary signals is used to present a broad treatment of weak, random signal detection for interception purposes. The relationships among a variety of previously proposed ad hoc detectors, optimum detectors, and newly proposed detectors are established. The spectral-correlation-plane approach to the interception problem is put forth as especially promising for detection, classification, and estimation in particularly difficult environments involving unknown and changing noise levels and interference activity. A fundamental drawback of the popular radiometric methods in such environments is explained.

关键词： Cyclostationary signal Environments Frameworks Unknown interceptions RANDOM signalS feature detections correlation theory

来源：评论

学校读者我要写书评

暂无评论

NONSTATIONARY LEARNING CHARACTERISTICS OF THE LMS ALGORITHM

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS 1987年第10期34卷 1199-1207页

作者： GARDNER, WA Signal and Image Processing Laboratory Department of Electrical and Computer Engineering University of California Davis USA

Upper and lower bounding first-order linear recursions for the mean-squared error realized with the LMS algorithm subjected to a sequence of independent nonstationary training vectors are derived. These bounds coincide to give the exact evolution of mean-squared error for the problem of identification of a nonrecursive time-varying system with white-noise excitation. This leads to an exact formula for time-averaged mean-squared error that is used to study optimization of the step-size parameter for minimum time-average misadjustment. New results on dependence of the minimal step size and the minimum misadjustment on the degree of nonstationarity are obtained.

关键词： Least squares approximation System identification Adaptive filters Algorithm design and analysis Vectors Time varying systems Gaussian distribution signal processing image processing

来源：评论

学校读者我要写书评

暂无评论

CONTROL OF NONLINEAR-SYSTEMS WITH MARKOVIAN PARAMETER CHANGES

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 1991年第2期36卷 233-238页

作者： MARITON, M Signal and Image Processing Laboratory Matra Sep Imagerie et Informatique Saint Quentin-en-Yvelines France

The control of hybrid systems is considered where a totally observed state process is generated by a nonlinear stochastic differential equation whose parameters are functions of a random jump process, called the plant regime. Regime transitions typically represent target maneuvers in a tracking system. The control task is split between a jump detection and identification (JDI) filter and a discrete adaptation, or reconfiguration. A design is presented that takes into account the errors of the JDI device and the plant nonlinearities to produce a near optimal passively adaptive control law with an easily implemented affine structure.

关键词： Nonlinear control systems Control systems Nonlinear systems Hybrid power systems Stochastic systems Differential equations Target tracking Filters Error correction Adaptive control

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：