随着多媒体技术的飞速发展，视频编解码技术也日益凸显出它的重要性，新的视频编解码技术不断涌现。MPEG-4是国际标准化组织 ISO/IEC 的运动图像专家组制定的新一代视频编解码标准，这一编码标准可获得很高的编码效率，适合低宽带、高质量的网络视频应用的需要。其传输速率要求较低，可以利用很窄的带宽，通过帧重建技术，压缩和传输数据，以求以最少的数据获得最佳的图像质量。 运动估计算法作为视频编码领域中的关键技术，它直接影响到视频数据压缩编码的效率和编码质量。运动估计的越准确，编码的效率就越高，解码出来的视频质量就越好。所以研究设计高效、快速的运动估计算法已经成为目前视频压缩技术中的重要方向。本文主要工作是对运动估计算法进行研究并对运动估计模块进行逻辑设计及验证。 本文按照基于图像灰度信息和基于图像特征的分类，详细讨论了现有常见运动估计算法的原理和实现方法，分析了它们各自的特点。着重介绍了块匹配运动估计的基本原理和研究现状，在充分了解块匹配运动估计主要技术，并且考虑运动矢量的统计和分布特性的基础上，选择用大菱形模板确定搜索方向，用小菱形模板进行精细定位搜索，找出最佳运动矢量，这样不仅增加了搜索的精度，并且降低了陷入局部最优的可能性。在此基础上使用一种新的取数方法，按模块取数，每次取进9个像素点的值，使用完后即抛弃，这样就能够节省搜索时间和数据存储空间。在进行残差值计算时采用pe阵列，一共8个pe模块一起进行运算，提高了运算速度，节省了运算时间。 在完成逻辑设计的基础上，用Ncverilog对其进行仿真验证，得到的结果与预期目标相符。并将所设计的模块运用到了视频芯片中，此视频芯片已经流片成功。

With the rapid development of multimedia technology, the importance of video decoding technology increasingly highlights, and some new video decoding technologies constantly emerges. MPEG-4 is an international standard organization, and the ISO/IEC moving picture is a new generation of a new generation of video codec standards which was worked out by the expert group. And this coding standard can achieve very high coding efficiency and satisfied the low broadband and high quality network video's application needs. Furthermore, this standard has a low demand for transmission rate; can use a very narrow broadband, through the frame reconstruction technique, compression and transmit data, so that with the least amount of data to get the best image quality. The motion estimation algorithm as the key technique in the filed of video coding which can directly affect the efficiency and quality of data compression. The more accurate of motion estimation, the higher coding efficiency and better video quality we can get. Therefore, studying and designing an efficient and fast motion estimation algorithm has become an important direction of video compression technology. This article has launched a study of the motion estimation algorithm and carried out the logic design and testing for the motion estimation module. In this paper, according to the image information and image feature based classification, discusses the existing common motion estimation algorithm principle and realization method, and their feartures are analyzed. Emphatically introduces the block matching motion estimation principle and research status quo, in full knowledge of block matching motion estimation techniques, and consider the motion vector statistic and distribution characteristics of foundation, choose to use large diamond template to determine the search direction, with a small diamond pattern for fine positioning search, to find the optimal motion vector, which not noly increases the search accuracy, and reduce the possibility of trapping in local optimum. Use of a new access method, according to the module check number, each taking into the 9 pixel value, after the use is thrown out, it will be able to save the searching time and data storage space. The residual value is calculated using PE array, a total of 8 PE modules together for operation, improves the operation speed, saving operation time. After the completion of logic design, Ncverilog software was employed to carry out, the results obtained with the expected goal is the same. The design used in the video chip has been taped out.