共查询到20条相似文献，搜索用时 281 毫秒
1.
《中国科技信息》2017,(22)
为了减少地铁车站设备故障导致的人员伤亡,本文提出了粒子群算法结合BP神经网络对屏蔽门系统的故障进行预测。利用BP神经网络结构作为粒子群算法的适应度函数对BP网络的权值与阈值进行优化。在确定神经网络结构之后,该模型以权值和阈值作为粒子,利用粒子群算法的寻找全局最优的思想为BP网络寻找最优权值和阈值。减少了BP神经网络的训练结果出现较大偏差的概率。该算法可以适用于地铁站内受多种不定因素影响的设备,本文采用屏蔽门系统故障较为频繁的门锁机构来分析模型,得到的预测结果相差不到一天范围内,因此该算法具有理想的预测精度。最后利用MATLAB仿真验证该算法的可用性。 相似文献
2.
3.
针对图书馆流通量预测问题．提出了基于遗传神经网络的预测模型。该模型采用遗传算法作为神经网络权值全局搜索算法,BP算法作为局部搜索算法。结合实例进行计算,结果表明,该算法用于预测图书馆流通量是可行和有效的。 相似文献
4.
5.
网络并发式流量特征具有信号时间可预测性,通过对网络流量的解卷积测度特征提取,提高对网络流量的预测性能。传统法方法采用粒子群优化算法实现对网络流量的特征测度盲解卷积分析,对原始信号的统计信息提取效果不好。提出一种基于粒子群退化重采样的网络流量解卷积测度提取算法,构建并发式网络流量序列采集模型,设计粒子退化重采样技术,将每个粒子的当前适应度值与其自身的个体最优值进行比较,如果优于个体最优值,得到粒子当前最优位置。仿真实验表明,采用该算法,收敛速度很快,在粒子群进化50代以内就可以实现成功收敛,对流量序列的测度特征提取结果准确,预测精度较高,展示了算法的优越性能。 相似文献
6.
7.
提出一种改进粒子群的垃圾短信信息过滤算法,将垃圾短信特有特征权值搜索改造为一个非线性权值递减搜索,加快短信中存在的"噪声"权值的递减速度,使得算法排除"噪声"干扰,尽快进入有用信息的过滤。实验证明,该方法提高了垃圾短信过滤的准确度。 相似文献
8.
9.
提出一种自主感应分析算法,针对通信小区域内的断点,设计最优路径选取方法。计算路径中的相关断点权值,为每个路径的选择设计相关的阀值标准,选取最优路径,避免断点区域路径的选择,同时也避免了传统方法逐个遍历网络节点带来的通信效率过低的问题。实验证明,利用该方法能够快速选取通信最优路径,有效提高了P2P网络通信效率,取得了满意的结果。 相似文献
10.
为实现数字水印的隐蔽嵌入,提高水印鲁棒性和信息量,基于小波离散变换(DWT)、非下采样Contourlet变换(NSCT)和奇异值分解(SVD)提出了一种数字水印算法。利用Arnold变换实现水印图像的置乱加密;通过NSCT处理载体图像得到大小相同的低频子带,对该低频子带进行小波分解,然后进行SVD;确定嵌入强度,并将置乱的水印信息嵌入到奇异值中,重构低频分量;通过小波逆变换和NSCT逆变换实现水印嵌入。实验结果表明,在保证水印嵌入信息量前提下,该算法能够满足水印的隐蔽性;同时可抵抗噪声、滤波、压缩、剪切、旋转等攻击,鲁棒性较强;PSNR值在30 d B以上,NMSE值均在0.2以下。所述图像数字水印算法具有很好的鲁棒性、不可见性以及抵抗各种攻击的能力,对数字产品的版权保护具有促进作用。 相似文献
11.
We present a lossless Deoxyribonucleic Acid (DNA) sequence hiding method that can be used for ensuring authenticity of DNA sequence in the context of Mobile Cloud based healthcare systems. Hiding data within DNA sequence results in permanent information loss in DNA sequence. Therefore, providing DNA sequence authenticity using data hiding is challenging. Moreover, existing works on DNA data hiding require a reference DNA sequence data to retrieve hidden data. Hence, current methods are not blind approaches and inappropriate for ensuring authenticity of DNA sequence in the Mobile Cloud. The proposed method hides authentication data within DNA sequence, extracts authentication data, and reconstructs the DNA sequence without any loss of information. From there, our proposed approach guarantees DNA sequence authenticity and integrity in Mobile Cloud based healthcare systems. We present a security analysis of our method to show that the method is secured. We conduct several experiments to demonstrate the performance of our proposed method. 相似文献
12.
Kangrok Oh Zhengguo Li Beom-Seok Oh Kar-Ann Toh 《Journal of The Franklin Institute》2018,355(4):1614-1637
In this paper, an optimization problem is formulated for stable binary classification. Essentially, the objective function seeks to optimize a full data transformation matrix along with the learning of a linear parametric model. The data transformation matrix and the weight parameter vector are alternatingly optimized based on the area above the receiver operating characteristic curve criterion. The proposed method improves the existing means via an optimal data transformation rather than that based on the diagonal, random and ad-hoc settings. This optimal transformation stretches beyond the fixed settings of known optimization methods. Extensive experiments using 34 binary classification data sets show that the proposed method can be more stable than competing classifiers. Specifically, the proposed method shows robustness to imbalanced and small training data sizes in terms of classification accuracy with statistical evidence. 相似文献
13.
Internet of things (IoT) coupled with mobile cloud computing has made a paradigm shift in the service sector. IoT-assisted mobile cloud based e-healthcare services are making giant strides and are likely to change the conventional ways of healthcare service delivery. Though numerous approaches for preventing unauthorized access to information exchanged between a mobile phone and cloud platform do exist, but there is no security mechanism to prevent unauthorized access by the cloud administrators. With an aim to ensure security of client data such as Electronic Patient Records (EPR), we propose a novel high-capacity and reversible data hiding approach for securely embedding EPR within the medical images using Optimal Pixel Repetition (OPR). OPR converts every pixel of the input image to a 2 × 2 block to facilitate reversibility by ensuring all the pixels in a 2 × 2 block to have different values. Since a 2 × 2 block is comprised of 4-pixel elements, which could be arranged in sixteen possible ways; we generate a lookup table corresponding to sixteen possible positions of pixels. EPR hiding in each block is achieved by permuting the pixels of a block according to the four-bit word of secret data, resulting in a histogram invariant stego image. The histogram invariance improves the robustness of the proposed scheme to statistical attacks. A stego image is said to hide embedded data securely, when it provides better imperceptivity for an appreciably high payload. Thus, while using information embedding approach for securing client data on a mobile-cloud platform, high imperceptivity is a desirable feature. Experimental results show that average PSNR obtained is 42 dB for payload 1.25 bpp by our scheme, showing its effectiveness for preventing unauthorized access to client’s sensitive data. 相似文献
14.
提出一种对基于扩频信息隐藏图像进行隐写分析的方法.应用马尔科夫链模型,根据图像的相邻像素点之间的相关性,结合阈值判别方法和改进的模式识别方法,判断一幅图像中是否存在隐写信息.在Corel图像库上的实验表明,本方法降低了虚警率,且正检率的性能优于已有方法. 相似文献
15.
一种基于视频的无损信息隐藏方法 总被引：2，自引：0，他引：2
描述了一种基于MPEG-Ⅱ彩色视频中的无损信息隐藏方法，该方法在Alattar算法的基础上，使宿主矢量类型的判别由四个不等式，减小到两个不等式，降低了算法的时间复杂度。数据嵌入过程中采用简单数据链路SDL成帧。在局域网上进行了隐秘传输模拟实验，从嵌入信息的视频中正确提取出所嵌入汉字或二值指纹图像后，原宿主视频可无损恢复。该方法较适用于视频中的隐秘传输及产权保护等领域。 相似文献
16.
《Information processing & management》2023,60(2):103222
Making adversarial samples to fool deep neural network (DNN) is an emerging research direction of privacy protection, since the output of the attacker's DNN can be easily changed by the well-designed tiny perturbation added to the input vector. However, the added perturbation is meaningless. Why not embed some useful information to generate adversarial samples while integrating the functions of copyright and integrity protection of data hiding? This paper solves the problem by modifying only one pixel of the image, that is, data hiding and adversarial sample generation are achieved simultaneously by the only one modified pixel. In CIFAR-10 dataset, 11 additional bits can be embedded into the host images sized 32 × 32, and the successful rate of adversarial attack is close to the state-of-the-art works. This paper proposes a new idea to combine data hiding and adversarial sample generation, and gives a new method for privacy-preserved processing of image big data. 相似文献
17.
《Information processing & management》2020,57(5):102255
Collaborative frequent itemset mining involves analyzing the data shared from multiple business entities to find interesting patterns from it. However, this comes at the cost of high privacy risk. Because some of these patterns may contain business-sensitive information and hence are denoted as sensitive patterns. The revelation of such patterns can disclose confidential information. Privacy-preserving data mining (PPDM) includes various sensitive pattern hiding (SPH) techniques, which ensures that sensitive patterns do not get revealed when data mining models are applied on shared datasets. In the process of hiding sensitive patterns, some of the non-sensitive patterns also become infrequent. SPH techniques thus affect the results of data mining models. Maintaining a balance between data privacy and data utility is an NP-hard problem because it requires the selection of sensitive items for deletion and also the selection of transactions containing these items such that side effects of deletion are minimal. There are various algorithms proposed by researchers that use evolutionary approaches such as genetic algorithm(GA), particle swarm optimization (PSO) and ant colony optimization (ACO). These evolutionary SPH algorithms mask sensitive patterns through the deletion of sensitive transactions. Failure in the sensitive patterns masking and loss of data have been the biggest challenges for such algorithms. The performance of evolutionary algorithms further gets degraded when applied on dense datasets. In this research paper, victim item deletion based PSO inspired evolutionary algorithm named VIDPSO is proposed to sanitize the dense datasets. In the proposed algorithm, each particle of the population consists of n number of sub-particles derived from pre-calculated victim items. The proposed algorithm has a high exploration capability to search the solution space for selecting optimal transactions. Experiments conducted on real and synthetic dense datasets depict that VIDPSO algorithm performs better vis-a-vis GA, PSO and ACO based SPH algorithms in terms of hiding failure with minimal loss of data. 相似文献
18.
本文对信息隐藏的基本模型、主要特征、研究方法及应用领域进行了综述。对信息隐藏技术的发展现状和未决问题进行了分析和评述,指出了今后该领域的研究方向。 相似文献
19.
信息隐藏是20世纪90年代逐步兴起的研究课题。语音信号的不特定的静音间隔使得它比较于音乐等其他音频信号缺少了很大的隐藏空间,而语音信号的信息隐藏在Internet和有线与无线电话信道又有着很好的应用前景。提出一种语音信号的信息隐藏算法,能够在MFCC参数中隐藏秘密消息,语音的短时能量具有的较强的稳定性,可以保证隐藏和提取时的帧的同步,使得对应的提取算法可以准确地从隐藏的语音中恢复出信息。本方法可以适用于Internet信道和局域网络或高速网络中的语音应用。 相似文献