Search - vc get

Search - vc get - List

[Graph Recognize] SVM-and--Face-Recognition DL : 0: 支持向量机及其在人脸识别中的应用研究上海交通大学博士论文，在知网上面付费下载得到的。本文从应用的角度出发，较为全面地对一些相关问题进行探讨，并使用Visual C++实现了一个基于支持向量机的人脸识别软件—idTeller。论文的主要工作和创新点包括： ·提出了两种基于VC边界的支持向量机参数选择算法—固定C算法和VC-CV算法。VC边界是两类支持向量机参数选择的一个理想准则，但它的一些固有缺点使其应用变得困难。本文通过将VC边界转化为VC指标，最终把问题归结为对最小包围体的求解，从理论上和计算上为VC边界的使用铺平了道路。在此基础之上，本文提出了两种基于VC边界的参数选择算法—固定C算法和VC-CV算法。在数个基准数据集上的实验表明，相比交叉验证算法，VC-CV算法不仅能获得性能更好的分类器，而且具有较低的计算复杂度。 ·使用序贯最小优化算法解决了最小包围体求解问题。最小包围体求解是计算VC指标的一个关键步骤，本文使用序贯最小优化算法对其求解，并对算法初始化、参数选择及更新等若干实现问题进行了深入地研究。在多个基准数据集上的实验表明，序贯最小优化算法能够快速而准确地解决最小包围体求解问题。- Support vector machine and its application to face recognition Shanghai Jiaotong University doctoral thesis, in HowNet above pay to get the download. From the application point of view, to more fully explore some related issues, and using Visual C-idTeller a support vector machine-based face recognition software. The main work and innovation of the paper include: two kinds of parameters of support vector machine based on the VC boundary selection algorithm- fixed-C algorithm and the VC-CV algorithm. VC boundaries are two types of support vector machine parameters to select the ideal criteria, but some of its inherent shortcomings make it difficult. This article by VC boundary for the VC index, and ultimately the problem is reduced to the solution of the minimum bounding volume, and paved the way for the use of the VC boundary from the theory and calculations. On this basis, we propose two parameter selection algorithm based on the VC boundary- fixed-C algorithm and the VC-CV algorit
Date : 2025-12-31 Size : 9.9mb User : Jessicaying
[Graph Recognize] moving DL : 0: 运动目标检测系统，VC++编程系统实现，根据背景图像，可以得到分步。-Moving target detection system, VC++ programming system implementation, according to the background image, you can get step by step.
Date : 2025-12-31 Size : 5.16mb User : bai
[Graph Recognize] fileinfo DL : 0: VC获取任意文件属性信息，比如修改时间，是否隐藏，文件创建时间，访问时间，文件名称及路径等信息。-VC get any file attribute information, such as the modification time, if hidden, file creation time, access time, file name and path information.
Date : 2025-12-31 Size : 12kb User : jaiko
[Graph Recognize] CarNumber DL : 0: 本算法进行车牌识别，首先提取车牌的位置，然后对车牌的字符进行分割，最后进行识别，将识别结果输出。本算法能得到较好的识别结果。(The algorithm for license plate recognition, the first extraction of the location of the license plate, and then the license plate character segmentation, and finally identify the recognition results will be output. The algorithm can get a better recognition result.)
Date : 2025-12-31 Size : 15.27mb User : 王老头