| 基于改进的局部敏感哈希算法实现图像型垃圾邮件过滤 |
| |
| 引用本文： | 曹玉东.基于改进的局部敏感哈希算法实现图像型垃圾邮件过滤[J].计算机应用研究,2016,33(6). |
| |
| 作者姓名： | 曹玉东 |
| |
| 作者单位： | 辽宁工业大学 |
| |
| 基金项目： | 国家自然科学基金资助项目 |
| |
| 摘 要： | 提出一种快速的图像型垃圾邮件过滤方案，结合半监督机器学习技术改进局部敏感哈希(LSH)算法，基于改进的LSH算法构建垃圾图像特征库索引，提高图像的查找速度。构造了60000个垃圾图像样本，实验结果表明利用改进的LSH算法能有效地提高垃圾图像的过滤速度。
|
| 关 键 词： | 垃圾图像过滤 局部敏感哈希 图像特征提取 高维数据索引 |
| 收稿时间： | 2015/2/12 0:00:00 |
| 修稿时间： | 2016/4/28 0:00:00 |
Image Spam Filtering System with improved LSH Algorithm |
| |
| CAO Yudong.Image Spam Filtering System with improved LSH Algorithm[J].Application Research of Computers,2016,33(6). |
| |
| Authors: | CAO Yudong |
| |
| Affiliation: | liaoning university of technology |
| |
| Abstract: | A fast image-based spam filtering scheme is presented in this paper. LSH algorithm is improved based on semi-supervised technology, which is used to building index of feature collection of 60000 spam images. The experimental results show that the improved LSH can speed up the image spam filtering because the space complexity of improved LSH algorithm is lower than that of original LSH. |
| |
| Keywords: | image spam classifier LSH feature extraction high dimension data index |
|
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
