Energy-efficient Hardware Accelerators for the SA-DCT and Its Inverse

The explosive growth of the mobile multimedia industry has accentuated the need for efficient VLSI implementations of the associated computationally demanding signal processing algorithms. In particular, the short battery life caused by excessive power consumption of mobile devices has become the biggest obstacle facing truly mobile multimedia. We propose novel hardware accelerator architectures for two of the most computationally demanding algorithms of the MPEG-4 video compression standard––the forward and inverse shape adaptive discrete cosine transforms (SA-DCT/IDCT). These accelerators have been designed using general low-energy design philosophies at the algorithmic/architectural abstraction levels. The themes of these philosophies are avoiding waste and trading area/performance for power and energy gains. Each core has been synthesised targeting TSMC 0.09 μm TCBN90LP technology, and the experimental results presented in this paper show that the proposed cores improve upon the prior art.

Adaptive Algorithms of Nonlinear Approximation with Finite Terms

This paper deals with realizable adaptive algorithms of the nonlinear approximation with finite terms based on wavelets. We present a concrete algorithm by which we may find the required index set Am for the greedy algorithm Gm^P（., Ψ）. This makes the greedy algorithm realize the near best approximation in practice. Moreover, we study the efficiency of the finite-term approximation of another Mgorithm introduced by Birge and Massart.     

频域图像数字水印技术

数字水印技术作为一种新型的数据隐藏技术,它通过在宿主数字媒体中嵌入一定的数字信息,实现数字产品的版权保护、安全传输等作用。本文在介绍水印技术的基本要求之后,重点讨论几种典型的数字水印嵌入和检测算法。     

Economic Cycles of Carnot Type

Originally, the Carnot cycle was a theoretical thermodynamic cycle that provided an upper limit on the efficiency that any classical thermodynamic engine can achieve during the conversion of heat into work, or conversely, the efficiency of a refrigeration system in creating a temperature difference by the application of work to the system. The first aim of this paper is to introduce and study the economic Carnot cycles concerning Roegenian economics, using our thermodynamic–economic dictionary. These cycles are described in both a Q−P diagram and a E−I diagram. An economic Carnot cycle has a maximum efficiency for a reversible economic “engine”. Three problems together with their solutions clarify the meaning of the economic Carnot cycle, in our context. Then we transform the ideal gas theory into the ideal income theory. The second aim is to analyze the economic Van der Waals equation, showing that the diffeomorphic-invariant information about the Van der Waals surface can be obtained by examining a cuspidal potential.     

结构光测量中快速相位解包裹算法的讨论

对Schofield等提出的快速傅里叶变换（采用4次傅里叶变换和4次逆傅里叶）解包裹算法和Volkov等提出的相位重建算法（仅采用两次傅里叶变换和一次逆傅里叶变换）进行了讨论,提出基于离散余弦变换的快速相位解包裹算法.指出前两种算法在处理一般带噪音的模拟相位图时具有明显的优势,其处理结果非常接近理想值,而在处理相位变化剧烈或不连续区域的实验相位图（采用四步相移法测量雕刻佛像的三维面形）时出现较大误差,甚至无法处理,而基于离散余弦变换算法能很好的解决这个问题,并且拥有比前两种算法更快的运算速度.在实际应用中,针对不同的相位图把Schofield所提出的算法与基于离散余弦变换的算法结合起来,可以解决大部分相位解包裹问题.     

基于MATLAB的DCT变换在JPEG图像压缩中的应用

介绍了JPEG图像压缩算法,并在MATLAB数学分析工具环境下从实验角度出发,较为直观地探讨了DCT在JPEG图像压缩中的应用。仿真实验表明,用MATLAB来实现离散余弦变换的图像压缩,具有方法简单、速度快、误差小的优点,大大提高了图像压缩的效率和精度。     

采用冗余字典稀疏表达的红外与水汽云图融合

卫星云图作为典型的多光谱遥感图像,因各个遥感器成像波段的差异,致使云图间既有一定的相关性,又存在一定的差异,故可认为云图包含2种特征:共性特征和个性特征.一种稀疏表示的云图融合方法,能够把多幅云图在一个过完备字典上进行稀疏表示,使用稀疏系数作为云图的特征,然后对不同图像的个性特征根据稀疏系数向量的1范数决定权重因子,融合云图可以由共性特征和融合后的个性特征联合表示.实验表明,该方法的融合云图无论在客观指标还是视觉效果上都优于传统方法,蕴藏了更为丰富的天气信息.     

ISAR成像的相参多普勒质心跟踪相位补偿方法

目标强散射中心的转动相位分量对传统多普勒质心跟踪（DCT）相位补偿方法性能具有较大的影响。本文通过分析得出STRETCH处理得到的目标回波初相不具备相参性,而中频直接采样回波信号脉间具备相参的特点,并提出相参多普勒质心跟踪相位补偿（CDCT）方法。该方法在DCT方法的基础上,利用中频直接采样得到的目标回波的脉间相参性提高平动相位分量估计的精度,从而提高相位补偿效果,改善ISAR成像的质量。相对于MDCT法,本文能够达到更好的相位补偿效果。实测数据验证了本文方法的有效性。      

面向有限带宽信道的字典学习图像超分辨率重建

固定邻域回归(ANR)算法采用K层奇异值分解(K-SVD)算法进行字典训练, 在字典学习过程中存在稀疏表示系数不准确的问题, 导致重建的结果不理想. 因此, 引入一种改进的K-SVD算法对字典进行训练, 该算法对字典训练改变了传统K-SVD算法更新稀疏表示系数的方式, 使得稀疏表示系数更加准确, 而且加快了字典的收敛速度, 使得训练得到的字典具有更好的稀疏表达能力. 同时, 针对ANR算法的不足, 提出一种面向有限带宽信道基于字典学习的图像超分辨率方法, 该方法采用改进的K-SVD算法训练字典对 , 并将其应用到ANR算法中, 实现图像的超分辨率重建. 实验结果表明, 本文提出的方法不仅能够保持ANR算法快速重建的优势, 而且提高了图像的重建质量, 具有更高的峰值信噪比和结构相似度.     

基于数据驱动字典和稀疏表示的语音增强

本文提出了一种基于数据驱动字典和过完备稀疏表示的自适应语音增强方法。首先在训练阶段采用干净语音基于K奇异值分解(K singular value decomposition, K SVD)算法训练过完备字典,然后在测试阶段根据含噪语音的噪声方差自适应选择最优的阈值,采用正交匹配追踪算法对含噪语音信号在过完备字典上进行稀疏分解,最后利用系数稀疏表示重构语音信号,从而达到语音增强的目。该方法不像传统语音增强方法那样减少或消去噪声,而是从字典中选取适当的原子表示纯净信号,从而把纯净信号从含噪信号中分离出来。对白噪声和有色噪声环境下重构语音进行了主客观评价。仿真结果显示:该方法能有效去除加性噪声,并且改善了语音质量。