共查询到20条相似文献,搜索用时 980 毫秒
1.
2.
分层垃圾邮件过滤器的设计与实现 总被引:1,自引:0,他引:1
由于垃圾邮件变化多端,日益泛滥,传统的、单一的反垃圾邮件方法对新型的垃圾邮件已经无能为力,为了攻克这一难题,我们设计了一个包含贝叶斯算法、反向DNS查询以及系统级和用户级过滤等的分层反垃圾邮件技术的邮件服务器.该服务器除了能收发邮件、过滤邮件,还可以通过自动和手动两种方式进行学习,以适应垃圾邮件的变化,有效的屏蔽垃圾邮件.本文着重介绍该服务器的垃圾过滤器所采用的各种技术和主要结构. 相似文献
3.
基于免疫原理的个性化Spam过滤算法 总被引:2,自引:0,他引:2
受生物免疫系统工作机制的启发,本文提出一种基于免疫原理的个性化Spam过滤算法.其主要思想是根据用户兴趣和邮件特征定义垃圾邮件社区,将各垃圾邮件归类于不同的垃圾邮件社区,抽取各个垃圾邮件社区的特征并用一组特征检测器来表示,检测时通过判断待检测邮件是否归属于某垃圾邮件社区来进行过滤.该算法是一个增量学习算法,能连续过滤垃圾邮件.算法中免疫学习与免疫记忆机制的采用不仅能提高垃圾邮件过滤的检出率与正确率还能加快邮件过滤的速度.文中通过测试实验和分析表明,本文算法的垃圾邮件过滤性能优于AISEC与Naive Bayesian算法. 相似文献
4.
5.
从图片垃圾邮件的现状着手,通过对图片垃圾邮件的分析,将图片垃圾邮件与文本垃圾邮件之间的不同点进行了对比,并对图片垃圾邮件的特征进行了总结.与此同时,对图片垃圾邮件过滤中常用的一些过滤方法,例如OCR(最优字符识别)以及指纹技术进行了介绍,分析了其优缺点,并结合它们自身的缺点提出了一些建设性看法.最后对最新的反垃圾邮件研究成果作了简略描述,并对垃圾邮件的发展作出了展望. 相似文献
6.
基于多任务学习的邮件过滤系统的研究 总被引:1,自引:0,他引:1
随着电子邮件的广泛使用,如何有效地避免和防范垃圾邮件的侵扰已成为一个亟待解决的问题.受机器学习在邮件过滤中研究和应用的启发,利用多任务学习(multitask learning)的特性,将判断一个用户的邮件是否为垃圾邮件看作一个任务(task),利用多任务学习中任务相关性假设,提出一种基于多任务学习的邮件过滤系统.实验表明,该系统对中英文邮件语料都是可靠和有效的,尤其对于同一邮件列表(mail list)中的用户的邮件. 相似文献
7.
8.
垃圾邮件的内容因人而异,现有的垃圾邮件过滤系统大多采用统一的过滤标准对用户的邮件进行过滤,因而忽略了垃圾邮件的这种个性化特征.针对这一情况提出一种个性化垃圾邮件过滤的计算模型,它事先不需要对模型进行针对性的训练,从对用户日常处理不同类型邮件的行为中分析和挖掘垃圾邮件的个性化特征,然后利用这种个性化特征在对垃圾邮件进行识别的同时不断强化这种个性化特征,以实现逐步提升对垃圾邮件识别率的目的.据此实现了相应的原型系统,通过对此系统的实验验证,该方法在现实环境下对垃圾邮件具有很好的过滤效果. 相似文献
9.
垃圾邮件制造者采用图像来隐藏垃圾信息,把文字隐藏在图像中并带一定的倾斜。在对含有倾斜文字的图像邮件深入分析后,针对图片中含有倾斜文字的垃圾邮件,文章给出了一种在进行OCR识别之前对图像进行hough变换,并结合支持向量机识别垃圾邮件的算法,实验结果表明,该方法具有比较满意的过滤效果。 相似文献
10.
11.
We propose an active image information system based upon the concept of smart images. A smart image is an image with an associated knowledge structure, consisting of protocols, hot spots, active indexes and attributes. This active image information system enables us to accomplish the objectives of timely delivery and easy accessibility by handling long-duration operations and supporting unitary image information usage. The experimental prototype of the smart image system is described. 相似文献
12.
针对图像经过信息隐藏与经过噪声干扰具有不同的特性,从图像隐藏信息和加噪的数学原理入手,引入图像直方图和差分直方图的模型,在图像小波分解高频系数差分直方图的数学模型基础上,利用高频系数差分数值的方差对图像是否隐藏信息或被噪声干扰进行分辨,给出了具体的计算判断方法。最后,在标准图像集上进行实验。结果表明,即使在图像直观统计参数基本相同的情况下,该方法也能明显分辨出两种不同的干扰现象,取得良好结果。 相似文献
13.
Alajlan N Kamel MS Freeman GH 《IEEE transactions on pattern analysis and machine intelligence》2008,30(6):1003-1013
In this paper, a geometry-based image retrieval system is developed for multi-object images. We model both shape and topology of image objects using a structured representation called curvature tree (CT). The hierarchy of the CT reflects the inclusion relationships between the image objects. To facilitate shape-based matching, triangle-area representation (TAR) of each object is stored at the corresponding node in the CT. The similarity between two multi-object images is measured based on the maximum similarity subtree isomorphism (MSSI) between their CTs. For this purpose, we adopt a recursive algorithm to solve the MSSI problem and a very effective dynamic programming algorithm to measure the similarity between the attributed nodes. Our matching scheme agrees with many recent findings in psychology about the human perception of multi-object images. Experiments on a database of 13500 real and synthesized medical images and the MPEG-7 CE-1 database of 1400 shape images have shown the effectiveness of the proposed method. 相似文献
14.
Cho-Wei Shih Hui-Chuan Chu Yuh-Min Chen Chuin-Cheng Wen 《Expert systems with applications》2012,39(17):12897-12904
Image annotation is a process of assigning metadata to digital images in the form of captions or keywords, and has been regarded as image management and one of the most crucial processes of image retrieval. And many automatic methods have been proposed. However, these methods still have some problems respectively. Fractals are fragmented geometries and can be considered separate parts; each part is similar to the contracted overall shape. Fractal features provide geometric information of an image that is irrelevant to the shape and size of an object in the image; therefore, fractal features are more robust than color and texture features. Therefore, this study proposed a fractal-driven image annotation (FIA) schema that extracts fractal features through fractal image coding and integrates color and texture as new visual features to conduct image-based annotation. Experimental results indicate that the effect of thresholds on annotating accuracy is insignificant. This finding supports the application of FIA on complex practical environments, reduces the time for identifying the optimal thresholds, and improves the practicality of using FIA in real environments. 相似文献
15.
提出了一种结合图像融合与小波变换的图像插值方法.首先利用一种常用图像插值方法获得初步放大图像,再将初步放大图像进行小波分解,源图像与小波低频子图进行直方图匹配后作为新的低频子图.将初步放大图像小波分解得到的不同方向的高频系数子图结合新的低频子图作为融合图像的多分辨率金字塔.最后进行小波逆变换得到融合图像,也即放大图像.实验结果表明该方法可提高图像插值的主客观效果,峰值信噪比相对常用方法可提高2dB以上. 相似文献
16.
Chi-Chou Kao 《Multimedia Tools and Applications》2017,76(11):12981-12999
In this paper, we proposed stereoscopic image generation methods of adjusting the depth value of edge pixels and improved hole filling procedures. For the conventional system, the smooth of depth map can reduce the incidence of image holes, but cause geometric distortions of the image depth. To solve the problems, the depth map is first expanded to refine the accuracy of image depth and the quality of images. Next, we derive a hardware-oriented method for 3D warping. Finally, appropriate blocks are searched to enhance the performance of image by improving hole-filling procedures. The experimental results demonstrate the proposed methods have great performance and practicability. 相似文献
17.
目前大多数的图像风格迁移方法属于有监督学习,训练数据需要成对出现,并且在处理图像背景时,现有的方法过于繁琐。针对这些问题,提出了一种基于图像蒙板的无监督图像风格迁移方法。在实验中,采用了基于循环一致性的CycleGAN架构,并使用Inception-ResNet结构设计了一个全新的具有内置图像蒙板的生成式模型,最后通过无监督学习将图像的背景与学习到的抽象特征进行自动重组。实验表明,新方法有效地对图像背景和抽象特征进行自动分离与重组,同时解决了特征学习过程中的区域干扰问题,获得了可观的视觉效果。 相似文献
18.
针对水下图像对比度低及细节模糊的问题,提出一种基于图像融合的自适应水下图像增强方法,实现不同类型水下图像的增强效果。基于颜色校正方法对水下图像进行颜色均衡化预处理;对亮度分量L进行Gamma校正,获得对比度提升的亮度图像;对两个亮度分量进行三层小波分解,提出对分解所得的低频分量及高频分量分别采用线性融合和自适应融合策略进行融合。多尺度融合保证了增强图像细节的丰富性,自适应融合策略体现了融合过程的可控性。实验结果表明,增强的水下图像呈现出高对比度和清晰的细节。 相似文献
19.
We discuss a new deblurring problems in this paper. Focus measurements play a fundamental role in image processing techniques. Most traditional methods neglect spatial information in the frequency domain. Therefore, this study analyzed image data in the frequency domain to determine the value of spatial information. but instead misleading noise reduction results . We found that the local feature is not always a guide for noise reduction. This finding leads to a new method to measure the image edges in focus deblurring. We employed an all-in-focus measure in the frequency domain, based on the energy level of frequency components. We also used a multi-circle enhancement model to analyze this spatial information to provide a more accurate method for measuring images. We compared our results with those using other methods in similar studies. Findings demonstrate the effectiveness of our new method. 相似文献
20.
基于特征匹配的亚像素级全景图像配准算法 总被引:1,自引:0,他引:1
为了提高全景图全自动拼接中图像配准的速度和匹配稳健性,提出了一种基于相对距离法去除外点的亚像素级图像配准算法,并给出了分析和实验结果,亚像素级像素定位误差在0.01~0.1之内。在对12组图像匹配的实验结果表明,该匹配算法的匹配正确率达到100%,且匹配的时间小于目前通用的RANSAC匹配算法。 相似文献