本论文着重对整数变换与量化、去方块滤波做了研究。
In this dissertation, the integer transform, quantization, and deblocking filter are expatiated.
264标准的帧内预测和DCT整数变换技术进行了较深入地探讨。
Then we made some more deep research on the intra-prediction and integer DCT transform technologies of the H. 264 standard.
整数到整数变换的基础是提升矩阵，所以fft、DCT整数变换的重点就是构造提升矩阵。
Lifting matrixes is the base of integers transforms, so structuring lifting matrixes is the emphasis in FFT and DCT integers transform.
水印信息经随机置换与LDPC编码，对经过二次整数变换之后的直流分量进行修改实现水印的嵌入。
The watermarking information that is stochastically permuted and coded by LDPC is embedded in the modified DC components which are attained by integer transform twice.
针对低码率应用，分析各编码工具的性能并选取合理编码框架，同时利用单指令多数据(SIMD)技术并行实现快速搜索及整数变换。
For bit rate video application, a rational profile is selected according to analysis of main encoder tools. Fast motion search method and integer transform are implemented by using SIMD technique.
文中分析了整数倍、分数倍和NCO控制的正实数倍的采样率变换方法。
Sampling rate conversion methods of integral ratio, fractional ratio, and NCO controlled positive real ratio are investigated.
通过模糊集合运算与算子变换把模糊数学模型最终转化为一个0-1整数规划问题。
Then, the fuzzy set-covering model is transformed to a 0-1 integer planning problem through fuzzy set operation and operator transformation.
基于提升方法构造的整数小波变换，采用行索引稀疏存储方案，可以对海量影像数据进行高保真的压缩。
The method of sparsely row-indexed storage has been used to compress large volumes of image data, on the base of integer wavelet transform constructed by lift scheme.
介绍基于双正交整数叠式变换（LBT）的图像压缩及编码方法的原理，以及该系统的硬件设计。
The paper introduced the method to realize the image compression based on Integer Lapped Biorthogonal Transform(LBT) and the theory of coding as well as its hardware design.
首先探讨了基于提升方案的整数小波变换，结合线性预测技术，提出了一种机载多光谱遥感图像的无损压缩方法。
A new method combining integer wavelet transform and linear prediction technique for lossless compression of airborne multispectral imagery is proposed in this paper.
提出了一种改进的快速付立叶变换(MFFT)的计算方法，在采样的持续时间不是信号周期整数倍的条件下，能够准确地提供此信号中各周期分量的特性。
Anew approach, named Modified Fast Fourier Transform (MFFT), is proposed to analyze the components of a periodic signal when sampling duration is not equal to an integer multiple of the fundamentals.
对感兴趣区域(ROI)进行无损压缩编码过程中，采用整数小波变换然后进行算术编码的方案降低运算复杂度、保证图像质量。
In the process of compressing the ROI, to decrease the complexity of computation and ensure the quality of the image, integer wavelets transform and arithmetic algorithm are used.
为了对图像进行无损隐秘传输，给出了一种新的基于整数线性变换处理的无损隐秘传输方法。
In order to realize lossless covert communication in image, a novel lossless covert communication method based on integer linear transform was proposed.
快速傅里叶变换在非同步采样和非整数周期截断的情况下存在较大误差，无法获得较精确的间谐波参数值。
The fast Fourier transform (FFT) will bring on higher error under nonsynchronous sampling and truncated non-integral period, thus more accurate interharmonic parameter values cannot be obtained.
针对这一情况，提出了一种基于整数小波变换和多级树集合分割编码压缩新算法。
This algorithm based on integer wavelet transform and set partitioning in hierarchical trees for controlling data's compression ratio and reconstruction difference expediently.
整数提升小波变换结构简单，不需要额外的存储空间，易于实现，在图像压缩中具有很好的应用价值。
Integer to integer wavelet transform has a simple structure, needs no more storage space and is very easy to be implemented, which is suitable for image compression.
本文基于三维整数拟仿射变换和四维混沌系统对图像加密。
In this paper, the symmetric image encryption scheme based on 3-d quasi-affine transformation over limited integer grids and 4-d chaotic map.
对经典DCT变换和整数DCT变换进行研究，并把它们和有限状态矢量量化编码算法相结合，提高了对图像的压缩性能。
Combined with FSVQ algorithm, the traditional DCT transform and integer DCT transform are used to improve the performance of image compression.
研究了基于整数到整数形状自适应离散小波变换对任意形状感兴趣区域的有损和无损编码技术。
Arbitrary shape region of interest lossily and losslessly coding technology based on integer-to-integer shape adaptive discrete wavelet transform has also been studied.
建立了基于浇次计划的炉次计划模型，经过变换，将混合整数规划问题转化为线性规划迭代求解问题。
A model of furnace plan based on casting plan was established, which was converted from mixed integral program problem to linear program problem.
根据整数DCT变换和量化的特性，通过提前判定全零系数块来终止没有意义的运动搜索过程。
According to the characteristics of integer discrete cosine transformation(DCT) and quantization, the coefficient-all-zero block was determined in advance to terminate meaningless search.
为了用较低算法复杂度使数字水印具有较好的鲁棒性与掩蔽性，提出了一种基于整数DCT变换的新的鲁棒水印算法。
In order to design a complicated scheme with both preferable robustness and invisibility with low complexity, in this paper, a new watermarking based on Integer DCT scheme is proposed.
“原子”解码算法只通过一次傅氏变换即完成解码；而整数傅氏变换避免了傅氏变换中最耗时的浮点运算。
The "atomic" algorithm performs decoding process using Fourier transformation only once and the integral Fourier transformation works without floating operation.
当原始图像点乘以变换矩阵，得到新的图像位置点时，其位置可能不是整数值，该位置的灰度值需要通过灰度插值获得。
When the original image point multiply the transformation matrix, a new image position, the position may not integer, this position gray value by gray interpolation gain needed.
给出了一种线性变换的整数SERM分解的选取方法。
A selection method of SERM factorizations for linear transforms is presented.
本文给出利用矩阵初等变换求一组整数的最大公因数，以及把它表示成这组数的组合的一个方法，此法常比一般“初等数论”教材中所给方法简单。
This paper first gets definition and theorems of integers matrices, and then to discusses a new method to find the greatest common factor of integers and solves linear indeterminate equation.
本文给出利用矩阵初等变换求一组整数的最大公因数，以及把它表示成这组数的组合的一个方法，此法常比一般“初等数论”教材中所给方法简单。
This paper first gets definition and theorems of integers matrices, and then to discusses a new method to find the greatest common factor of integers and solves linear indeterminate equation.
应用推荐