ID 原文 译文
4553 与先签名后加密的方式相比，所提方案私钥数据量更少，效率更高。 Compared with the method of signing-then-encrypting method, the proposed scheme has thesmaller private key and higher efficiency.
4554 针对传统暗通道先验易在高亮度区域失真和产生光晕效应的不足，提出一种基于补偿透射率和自适应雾浓度系数的雾天图像复原算法。 Aiming at the drawbacks of traditional dark channel prior, which was prone to distortion and Halo effects inthe bright areas, a haze image restoration algorithm based on compensated transmission and adaptive haze concentrationcoefficient was proposed.
4555 首先利用高斯函数拟合有雾和无雾图像间的衰减关系，通过修正透射率对高亮区域进行补偿。 First of all, a Gaussian function was used to fit the attenuation relationship between the hazeand haze-free image, and the compensation transmission was set to correct the initial transmission.
4556 然后分析雾气特性，提出亮度熵概念，对原图亮通道进行逐像素处理求取熵值，结合高斯金字塔提取纹理特征，得到雾气分布图； Then the characteris-tics of haze was analyzed, the concept of brightness entropy was introduced and the bright channel operation was per-formed to acquire entropy value with pixel by pixel. Combined with the Gaussian pyramid to extract texture features, thehaze distribution map was obtained.
4557 同时建立一种线性变换来自适应求取雾浓度系数，并获得优化透射率。 An adaptive transformation was established to seek the haze concentration coeffi-cient and get the accurate transmission.
4558 最后改进局部大气光的获取方法，结合大气散射模型得到复原结果。 Finally, the recovery results were restored by improved atmospheric light valueand atmospheric scattering model.
4559 实验表明，所提算法可以有效复原出降质图像的颜色与细节，明亮度适宜，去雾程度彻底，效果清晰自然。 Experimental results show that the recovered image has better color and detail, the de-gree of dehazing is thorough, the brightness is appropriate, and the effect is clear and natural.
4560 基于现实推荐系统数据集非常稀疏，导致传统的协同过滤算法往往无法提供高质量推荐的问题，提出了一种基于粗糙集规则提取的协同过滤算法。 To address the problem that in a practical recommendation system (RS), because of the datasets are often verysparse, the traditional collaborative filtering (CF) approach cannot provide recommendations with higher quality, a novelCF based on rough set rule extraction was proposed.
4561 首先利用用户/物品属性和用户−物品评分矩阵构建决策表， Firstly, the attributes of user/item and the user-item rating matrixwere used to construct a decision table.
4562 然后通过决策表约简算法得到每一条规则的核值， Then, the core value of each rule in the table was extracted through using the de-cision table reduction algorithm.