一种基于权重融合的JPEG隐写分析方法

针对JPEG图像通用隐写检测中检测效率低、训练时间长的问题,提出一种基于集成分类器的新检测方法。算法以CC-PEV为特征对图像进行描述并作为隐写分析特征;然后,随机构造若干个特征子空间,用bootstrap方法构造图像训练子集,分别进行训练得到数个基分类器;根据基分类器的分类结果赋予基分类器不同的权重,将基分类器的结果按照其权重进行融合得到最终的结果。本文对该算法进行了测试,对它的集成性、检测准确率和训练时间进行分析。实验结果表明,相对于传统的集成方法,本文方法用自举方法构造训练集、随机方法构造子特征空间、赋予基分类器不同权重进行融合能够显著地提高算法准确率。本文方法相对于SVM和传统的集成分类方法,具有更高的检测率,对于特征维数更大的图像检测,具有更好的拓展性和一般适用性。     

Real‐Time License Plate Detection in High‐Resolution Videos Using Fastest Available Cascade Classifier and Core Patterns

We present a novel method for real‐time automatic license plate detection in high‐resolution videos. Although there have been extensive studies of license plate detection since the 1970s, the suggested approaches resulting from such studies have difficulties in processing high‐resolution imagery in real‐time. Herein, we propose a novel cascade structure, the fastest classifier available, by rejecting false positives most efficiently. Furthermore, we train the classifier using the core patterns of various types of license plates, improving both the computation load and the accuracy of license plate detection. To show its superiority, our approach is compared with other state‐of‐the‐art approaches. In addition, we collected 20,000 images including license plates from real traffic scenes for comprehensive experiments. The results show that our proposed approach significantly reduces the computational load in comparison to the other state‐of‐the‐art approaches, with comparable performance accuracy.     

改进特征和时间融合框架的区分式跟踪算法

文章基于多示例学习的跟踪框架,引入改进型的分布场特征并融合目标时间维度信息,提出了一种新的跟踪算法。新的特征能够更为有效地描述目标的空间结构信息,对于目标模糊、局部遮挡以及细微形变有良好的鲁棒性。加入的目标时间维度信息融合方法,包含了目标的历史信息,同时也能响应目标的外观变化,提高了跟踪器从跟踪异常中恢复的能力。通过对比新算法与其他先进算法在多组测试视频上的跟踪结果,可以发现本文提出的算法具有更为优异的性能,能够在各种复杂情况下对目标进行稳定的跟踪。     

一种适用于WAMS量测数据的系统暂态功角稳定评估方法

考虑WAMS量测数据刷新速率快,数据量大的特点,提出了一种适用于WAMS量测数据的暂态功角稳定评估方法。选取初始特征量集并用核主成分分析法对特征集进行降维,过滤冗余特征并降低分类器输入向量的维度。构建训练样本集,计算各样本的初始特征量集并进行降维。通过训练ECVM分类器对暂态功角稳定进行评估,并用测试数据集验证分类器的准确率。在新英格兰10机39节点系统中的仿真表明,所提算法有较高的分类准确率,与传统分类算法相比降低了单个样本评估所需的时间,具有工程使用价值。     

Visualisierung der Partikelbewegung in einem Abweiseradsichter

In spite of its great importance for the dry classification of fine powders, the processes in deflector wheel classifiers are to a large extent unknown. Therefore, in the present work a commercial available classifier was modified to gain an optical access to the deflector wheel. For the first time, the obtained photographs enable an observation of the gas flow and the particle motion between the blades of the deflector wheel. Especially the importance of particle‐particle‐ and particle‐blade‐collisions could be shown.     

Classification and assessment of power system static security using decision tree and random forest classifiers

The power system static security classification and assessment is essential in order to identify the post‐contingency problems and take corrective measures and to protect the system from blackout. In this paper, application of two data mining classifiers have been proposed for the security classification and assessment of a multiclass security problem. To design the security problem, contingency analysis is carried out under N‐1 line outage, and static severity index (SSI) is computed, which is a function of the line overload and the voltage deviation using Newton–Raphson load flow method, considering the variable load and generating conditions. Corresponding to the computed values of SSI, the voltage, phase angle, Mega Volt Ampere line flow and so on, a 1 × 7 pattern vector is generated. The generated pattern vectors are used to design a multiclass security problem. The designed security pattern vectors are given as inputs to the decision tree (DT) and random forest (RF) model in order to classify the security status of the power system. The proposed classifiers are investigated on an IEEE 30‐bus test system. The classification accuracy of the DT and the RF are compared with state‐of‐the‐art classifier models, namely, multilayer perceptron (MLP), radial basis function (RBF), and support vector machine (SVM).The simulation results clearly indicate that the proposed DT and RF classifiers are more efficient, reliable, and out performs MLP, RBF, and SVM classifiers for the assessment of the security status of the power system. Hence, DT and RF classifiers are found to be suitable for online implementation. Copyright © 2015 John Wiley & Sons, Ltd.     

FG-12型螺旋式分级机的螺旋轴连接方式改进

针对在工作状态下的FG-12型螺旋分级机出现的主轴振动过大,主轴和螺旋叶片连接处频繁断裂的状况,经过研究计算,对其连接方式作出技术改进。     

涡流分级机的分级轮研究

在分级系统中,分级轮对于物料的分散、分级起着非常重要的作用。通过分析气体在分级轮中所受的作用力,推导出分级粒径公式,并得出分级轮的风量、转速与分级粒度及分级精度的关系。     

Forgetting Exceptions is Harmful in Language Learning

We show that in language learning, contrary to received wisdom, keeping exceptional training instances in memory can be beneficial for generalization accuracy. We investigate this phenomenon empirically on a selection of benchmark natural language processing tasks: grapheme-to-phoneme conversion, part-of-speech tagging, prepositional-phrase attachment, and base noun phrase chunking. In a first series of experiments we combine memory-based learning with training set editing techniques, in which instances are edited based on their typicality and class prediction strength. Results show that editing exceptional instances (with low typicality or low class prediction strength) tends to harm generalization accuracy. In a second series of experiments we compare memory-based learning and decision-tree learning methods on the same selection of tasks, and find that decision-tree learning often performs worse than memory-based learning. Moreover, the decrease in performance can be linked to the degree of abstraction from exceptions (i.e., pruning or eagerness). We provide explanations for both results in terms of the properties of the natural language processing tasks and the learning algorithms.     

旋流器分级磨矿回路测控系统

本文介绍了旋流器分级磨矿回路测控系统的原理配置及特点 ,结合实际应用 ,论述了该测控系统的功能及使用结果