图象

In this paper, based on MPEG 2 Test Model 5 (TM5) bit rate control algorithm, there are some improvements proposed:① an intelligent integral action is proposed for improving the control precision of bit rate;② after using DCT AC coefficients to define the measure of macroblock and picture coding difficulty, an improvement on target bits allocation is proposed for the decoded picture quality and conformance including scene change, which can adjust the target bits allocation adaptively by their measure.

针对MPEG－2测试模型5（TM5）比特速率控制算法进行了以下改进：①引入一种智能积分作用，有利于提高比特速率的控制精度；②采用DCT交流系数定义宏块和图象的编码难度测度，然后提出了改进的目标编码量分配策略，即根据编码难度测度自适应调整图象和宏块的目标编码量，有利于提高包括场景变化在内的图象解码质量及其一致性。

After a brief review about the problems of bit rate control for MPEG video coding, two optimizations are proposed based on MPEG-2 Test Model 5 (TM5) algorithm:①An intelligent integral action is introduced for improving the control precision of bit rate;②After using DCT AC coefficients to define the measure of macroblock and picture coding difficulty, an improvement on target bits allocation is proposed for the decoded picture quality and conformance including scene change, which can ajust the target bits allocation adaptively by their measure.

文中首先对MPEG视频编码比特速率的控制问题作了简要回顾，然后针对MPEG-2测试模型5(TM5)算法进行了以下优化：①引入一种智能积分作用，有利于提高比特速率的控制精度；②采用DCT交流系数定义宏块和图象的编码难度测度，然后提出了优化的目标编码量分配策略，即根据编码难度测度自适应调整图象和宏块的目标编码量，这样有利于提高包括场景变化在内的图象解码质量及其一致性。

The image registration firstly assumes that there is some transformation between frame and frame of images, such as translation, affine and planar surface etc. and then registers images.

图象配准算法首先假设图象帧间存在某些变换，例如平移变换、仿射变换和平面变换等，然后对序列图象进行运动配准。

This paper, firstly, after a deep study of image mathematical models, sets up two kinds of image models namely Laplace partial differential equation model and biharmonic equation model and then obtains the eigenvectors of image by the methods of eigenfunction expansion and boundary recurrence.

本文首先较深入地研究了图象的数学描述，提出图象的Laplace偏微分方程和双调和方程两种模型，继而采用特征函数展开法和边界递推法分别求得图象的特征矢量。

The dissertation aims at automatic target recognition in images. For that purpose, a variety of algorithms proposed by the dissertation have been used in different stages of image recognition, which deal with image segmentation, boundary feature point extraction, curve feature extraction, object invariant representation and recognition.

本论文的研究目标是实现图象中目标的自动识别，为此，论文围绕图象识别的各阶段，研究了图象分割、轮廓特征点提取、曲线特征提取、物体的不变性表征及识别等。

Our experiments show that DSNR of different images in the same scene changes congruously with PSNR, thus can be used to evaluate the quality of decompressed images and image sequences.

实验证明，在相同场景中DSNR能较好地反映PSNR的变化，因而可以用于对同一图象的不同解压缩图象以及图象序列的评价。

But these methods limited by the expense of computing fundamental matrices and trifocal tensors or of the nonlinear optimization involved in computing epipolar tangencies. Through the studies of the invariants of the single axis motions, computational theories have been developed in this thesis to provide practical solutions for the problem of structure and motion from fitting the corresponding points in the whole sequence to its conic locus or conics for short.

在最近的研究中，一些研究人员利用视觉几何的方法从非定标的图象中对单轴旋转运动进行分析，提出了利用基本矩阵和三焦距张量求解单轴旋转运动中的关键参数——旋转角度的算法，但是该算法需要计算一系列相邻2图象的基本矩阵和一系列相邻3图象的三焦距张量，由于每个基本矩阵的自由度为7，每个三焦距张量的自由度为18，导致算法参数数量非常之大。

Connected component detection is an elementary and time-consuming operation in digital image processing. It is widely applied in pattern recognition, characteristic extraction, image understanding and so on.

连通检测是计算机数字图象处理中一个重要的基本操作，它在模式识别、图象理解、图象特征提取等方面有着广泛的用途。

The two stage method by-passes the correspondence problem and ill-posed problem encountered in the general motion analysis, and avoids the local minimum problem of the spatial-constrain-based iteration method.

本文推广了全景图象方法和外极面图象方法，从而使之能适用于具有抖动的图象序列分析，并避免了一般运动视觉方法的不适定问题和特征对应问题、基于空域约束的迭代方法的局部最小化等问题。

First scanning the known handwriting materials then number them ,In pretreatment, we convert the valid part of the image into a standard size, as images carries out duotone and go throw off chirp handling in order to achieve better effect.draw features from the known handwriting materials with the Co-occurence,especially,we divided a copy of handwriting into 25 little pieces with the size of 128*128 ,drawing features from every little piecese with four directions(0 degree, 45 degrees, 90 degrees as well as 135 degrees) and calculate the four major feature values( veins and the statistical quantity of veins contrast and the statistical quantity of veins consistency Shang the statistical quantity of statistical quantity as well as the veins correlation of gray scale ), preservation all the feature value that drawn from all known ma terials to the handwriting characteristic database,then input the unknown handwriting materials, also using the method of the Co-occurence to draw those features, recycling the minimum European Distance law match the unknown writing material feature value with the handwriting characteristic feature database, export the label of the known hand writing materials which is most similar to the unknown material with minimum European Distance, and then we can confirm who is the author of the unknown material.

首先将笔迹材料作为图象扫描输入，并对其进行编号。预处理部分可将笔迹图象的有效部分规范化到一个统一尺寸，接着对其进行二值化和去除噪声的处理，以便于更好的提取图像的特征。在此我们采用了灰度共生矩阵法提取手写笔迹材料的纹理特征，与以往有所不同的是，我们将一份手写材料分割成64块大小为80*80象素的子图象，每个小块都从四个方向（0度、45度、90度以及135度）来更全面的提取特征，并计算出四个最主要的特征值（纹理一致性的统计量、纹理反差的统计量、纹理熵的统计量以及纹理灰度相关性的统计量），将从所有已知材料提取的特征值保存到纹理特征库中，对于待检手写材料，同样采用灰度共生矩阵的方法提取其纹理特征，再利用最小欧氏距离分类法将从待检手写材料中提取的特征值与纹理特征库中的特征值进行比对，与欧氏距离比对值最小的相匹配，输出匹配成功的原材料的标号，进而识别出待检材料书写者的身份。

They weren't aggressive, but I yelled and threw a rock in their direction to get them off the trail and away from me, just in case.

他们没有侵略性，但我大喊，并在他们的方向扔石头让他们过的线索，远离我，以防万一。

In slot 2 in your bag put wrapping paper, quantity does not matter in this case.

在你的书包里槽2把包装纸、数量无关紧要。