在当今社会，随着信息技术的高速发展，信息安全越来越引起人们的重视。生物识别技术利用人体本身固有的生理特征（如指纹、掌纹等）或行为特征（如步态、签名等）确定人的身份，具有很高的安全性与可靠性。如今很多生物识别技术已经被广泛地应用到不同的领域当中。 掌纹识别与其它生物技术相比包含丰富的信息，如：几何特征，掌纹主线、皱纹、三角点、细节点特征等。通过结合手掌上的多种特征可建立起高精度的生物识别系统，因此，在近年来，人们越来越关注这一技术，有很多学者都在进行着掌纹识别技术的研究。 本文在着重针对高分辨率掌纹图像的分割，掌纹图像线特征的提取及后处理算法等进行了深入的研究。主要研究工作有以下几点： 1）对高分辨率掌纹图像的分割方法进行了研究。利用均值MEAN、方差VAR作为标准实现由纹线组成的掌纹图像的前景分割。针对于手指部分的干扰，文章中将其分为孤立的手指部分和与手掌相连的手指部分分别进行处理。利用区域生长法提取最大连通域达到去除孤立手指部分；利用轮廓跟踪比较CD、PD距离，定位与手掌相连的手指部分的位置，从而实现将其去除。通过实验，该方法很好的实现了高分辨率掌纹图像的分割，为后续的特征提取奠定了基础。 2）在对掌纹线特征进行提取时，根据线特征能够从较低分辨率的图像中提取出来的特点，将原图像分辨率降低。然后使用基于改进的有限Radon变换的线特征提取方法提取掌纹的线特征。 3）由于使用MFRAT方法提取出来的线特征仍然存在不少的干扰，所以需要对其进行进一步的后处理。本文中使用去除小的连通域、图像细化、删除短线和毛刺、断点的连接四步进行后处理，得到单个象素宽，连通性较好的，去除了短线毛刺等噪声的线特征图。 实验结果证明，本文方法很好的从高分辨率的掌纹图像中提取出了线特征，这些特征能够为以后掌纹图像的匹配与识别及后续的处理提供有用的信息。

Nowadays, people attach more importance to technology security with the rapid growth of information technology. Biometrics, which exploits physical features in human body such as fingerprint and palmprint or behavioral features such as gait and signature to identify a person, is deemed to be secure and convenient. Actually many biometrics technologies have been used in various domains. Compared to other biometric technologies, palmprint contains more information, such as palm geometry, principal lines and wrinklesridge,and ridge and valley features,etc. A highly accurate biometrics system can be built by combining all features of palms. Therefore palmprint recognition has recently attracted an increasing amount of attention from researchers. In this paper, the high-resolution palmprint segmentation, the extraction of palm lines and the post treatment algorithm were lucubrated. The main achievements in this paper are as follows: 1)Make a research of high-resolution palmprint. The mean and variance of an image make up a criterion for palmprint segmentation. Towards the influnce of finger region, Isolated finger regions and connected finger regions are distinguished and solved respectively. The Isolated finger regions are wiped off with region growing algorithm to find out the biggest connected component. The two distances, CD and PD, which are caculated and compared by counter tracing algorithm, locate the connected finger regions, in order to wipe off them. According to the experiments, the method segments high-resolution palmprint image successfully, and lay a good foundation for the subsequent palmprint feature extraction. 2)When extract the line feature from palmprint, because of line features can be detected from low-resolution images, the resolution of original palmprints could be reduced. Then a modified finite Radon transform (MFRAT) can be used to extract the line features from palmprint. 3)Due to some hindrance exists in the line feature extracted by MFRAT, a post treatment is necessary. In this paper, the 4 steps: wipe off small connected component, image thinning, delete short line and burr, and combine the breakpoints, consist of the post treatment. Afer that, a line feature image with sigal pixel width, good connectivity and noises like short lines and burr deleted is obtained. The experement proves that, the proposed method can extract line features form high-resolution palmprint commendably. The line features can provide usefull information for the palmprint matching and identification, amd the furture manipulation.