| 视觉核感知器 |
| |
| 引用本文： | 邓赵红,王士同.视觉核感知器[J].江南大学学报(自然科学版),2004,3(2):127-130. |
| |
| 作者姓名： | 邓赵红 王士同 |
| |
| 作者单位： | 江南大学,信息工程学院,江苏,无锡,214036 |
| |
| 基金项目： | 江苏省自然科学基金项目(BK2003017)资助课题，国家模式识别重点实验室开放基金项目资助课题，南京大学软件新技术重点实验室开放基金项目资助课题. |
| |
| 摘 要： | 将Mercer核思想和视觉理论相结合，提出了视觉核感知器算法(VKP)。相对于单纯的核感知器，视觉核感知器能更有效地解决了学习的收敛性问题。同时也有效地提高了核感知器学习的精度。大量的实验结果表明了本算法的有效性和某些情况下的优越性。
|
| 关 键 词： | 视觉理论 核函数 感知器 分类 |
| 文章编号： | 1671-7147(2004)02-0127-04 |
Visual Kernel Perceptron |
| |
| DENG Zhao-hong,WANG Shi-tong.Visual Kernel Perceptron[J].Journal of Southern Yangtze University:Natural Science Edition,2004,3(2):127-130. |
| |
| Authors: | DENG Zhao-hong WANG Shi-tong |
| |
| Abstract: | In this paper the novel classification algorithm based on visual kernel perceptron is presented. The proposed method first maps the input data into a high-dimensional feature space using some Mercer kernel function. Then, the perceptron learning is implemented and the learning speed is adjusted to adapt to visual coefficient. Compared with the pure perceptron, the new algorithm not only solves the learning convergence efficiently, but also improves the classification accuracy effectively. The experimental results demonstrate the effectiveness and advantage of this new classification algorithm based on the visual kernel perceptron. |
| |
| Keywords: | visual theory kernel function perceptron classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
