With the extensive application of unmanned aerial vehicles (UAVs), there is an increasing demand for fast processing of coloured UAV images. The coloured UAV image pixels are usually represented by quaternion vectors with three bands of visible light corresponding to the three imaginary parts of the pure imaginary quaternion. Accordingly, the colour image edge points can be determined based on the quaternion polar coordinating the rotation principle. Here, a quaternion-based improved cuckoo algorithm is proposed to perform fast processing for UAVs images. In particular, a novel guiding equation is used to optimize the positions of the improved cuckoo algorithm before the Levi flight. Furthermore, a novel disturbance equation is used to obtain a varied location for the next location after the Levi flight. Comprehensive experiments are conducted to evaluate the performance of the proposed solution. The experimental results showed that the proposed method significantly reduces the image processing time and remarkably improves the quality.

In this paper, we propose a novel feature extraction method for pattern classification problem. We propose to map the original data to subspaces for feature extraction and hope the mapped data can reconstruct the original data. The motive is to avoid of losing of information of the original data in the process of subspace mapping. We assume that if the original data can be reconstructed from the subspace, the critical information can be preserved. Moreover, we also observed that the reconstruction error is a low-rank matrix if the reconstruction is performed well. We propose to measure the reconstruction error matrix rank by the nuclear norm and minimize it to learn the optimal subspace transformation matrix. Meanwhile, the classification is also used to regularize the learning to improve the discriminate ability of the subspace representations. Experiments over several benchmark data sets show the advantage of the proposed method over the existing subspace learning methods.

在传统网络巡检方法中,网络异常发现主要基于单一参数进行阈值触发,误报率较高,效率低.为了高效准确地发现网络异常,提出了一种基于BPNN的网络异常预测模型.首先对采集系统采集的数据进行特征提取和初始化处理;然后,将初始化后的数据作为神经网络样本进行训练,根据误差阈值调整网络参数,确定网络结构;最后,在Matlab环境下进行仿真实验,将提出的BP神经网络模型用于网络异常预测,结果表明本文提出的方法对网络异常预测有较高的预测率.

Aiming at the low robustness of the existing image registration methods,a noise-insensitive multi-distorted image registration method was proposed.The rotation angle was estimated by Radon transform, and the translation and expansion were solved by fast Fourier transform.Experimental results showed that this method could effectively improve the accuracy of all kinds of image registration,and had high robustness,which could effectively resist the interference of noise signals and had high practical value.

We prepared the amorphous selenium (a-Se) thin films via thermal evaporated method with different deposition time, revealing that the increase of deposition time was propitious to improve the quality of a-Se thin films. In detail, the optical gap of a-Se thin films enhanced from 2.08 to 2.15 eV, and the transmittance showed the blue shift of absorption edges with prolonging deposition time. Further, the dark conductivity was analyzed systematically around three main physics issues. (i) The initial dark conductivity was high and decayed with test time. (ii) The dark conductivity showed a dependence of applied electric field. (iii) The dark conductivity decreased with extending deposition time. Moreover, a-Se thin films showed the improved X-ray photoconductivity gain from 1.5 to 3.1 times under exposure dose rate of 14.83 × 10−4 Gy/s of X-radiation, with a rapid photoresponse and a small applied electric field requirement.

Environmental air quality prediction plays an important role in the prevention of environmental pollution.Because the prediction ambient air quality is affected by many factors,the accuracy of prediction can not meet the needs of the development. The artificial bee colony algorithm(ABC)is improved and introduced into the back propagation neural network(BP).The reciprocal of the training error is used as the fitness function,and the initial value of the ABC is assigned as the initial weight and the threshold of BP.The global optimal solution obtained by the improved artificial bee colony algorithm(IABC)is the global optimal weight and threshold of the BP.The optimized BP neural network is used to predict the ambient air quality,by comparing the traditional BP neu-ral network,the traditional artificial bee colony optimization back propagation neural network.Experimental results show the opti-mized BP neural network proposed in this paper has achieved ideal results in ambient air quality prediction,and can be used in practice.

Reasonable warehouse storage planning and assignment is the key technology to reduce the product storage and retrieve time and improve warehouse operation efficiency.In the intelligent warehouse environment which was developed based on the technology of Internet of manufacturing things,to solve the problem of storage assignment at intelligent multi-product warehouse,this paper proposed a multi-objective intelligent warehouse storage assignment model with many constrain rules.Then it developed an improved genetic algorithm to solve the model.Experimental results show that the proposed model and the improved genetic algorithm can produce effective storage assignment schemes,and verify the effectiveness of the model.

为提高环境空气质量预测的精度,提出一种由改进人工蜂群算法和反向传播神经网络相结合的环境空气质量预测方法(KABC-BP).对人工蜂群算法中雇佣蜂、跟随蜂的搜索空间提出一种随迭代次数递减的搜索公式,以随机初始化此改进人工蜂群算法的不同初始解作为不同组反向传播神经网络权值,以蜂群算法迭代代替人工神经网络的梯度下降修正迭代,以蜂群个体的对应权值下训练误差倒数作为适应度函数,该改进人工蜂群算法所求全局最优解就是所求反向神经网络最优权值.通过基于改进蜂群算法的反向传播神经网络算法、传统蜂群算法的反向传播神经网络算法(ABC-BP)及反向传播神经网络算法(BPNN)的环境空气质量预测的仿真实验表明,该算法的环境空气质量预测精度是最高的.

随着全景图技术的快速发展及广泛应用,古建筑文物全景图生成方法备受关注并得到研究和发展.本文从全景图照相设备、全景图生成工具和全景图拼接算法三方面来对目前古建筑文物全景图生成方法进行概述、比较和分析,最后得出相关结论并给出展望,希望为从事古建筑文物全景图生成、图像处理方面的设计人员和研究人员提供一定的参考.

Aiming at the low performance problem in sentiment analysis related fields,this paper proposed a neuro-fuzzy model-based opinion detection method.The novel model firstly used a lexical-knowledge-free strategy to achieve language independent feature selection.Secondly,it used a fuzzy theory based neural network method,implemented the sentence level subjectivity detection.Experimental results on several datasets reveal that the method has ideal precision in opinion detection.It is a outstanding opinion detection method and has significance in cross language sentiment analysis.

Because of the complexity of the prediction of network emergencies,the prediction results of traditional network emergencies are not satisfactory. In response to this situation,the traditional search algorithm of artificial bee colony algorithm (ABC)has been improved greatly.The traditional single search formula is extended to a search formula for the employed bees and the follower bees,respectively.According to the swarm intelligence emergence principle of the artificial bee colony algorithm,the improved artificial bee colony algorithm(IABC)is introduced into the network group environment for predicting the occurrence of network emergencies.Based on this,a prediction algorithm for network emergencies based on improved artificial bee colony is pro?posed.The K nearest neighbor classifier(KNN),ABC,GABC、IABC four algorithms are used to evaluate the corpus.The experi?mental results show that the proposed algorithm is optimal in the prediction accuracy,recall rate and comprehensive evaluation in?dex of the network emergency prediction.So it can be used in practice.

为更好地解决遗传算法在智能组卷过程中出现的早收敛问题,以及组卷质量和组卷速度呈负相关的问题,提出一种基于分段整数编码、多点交叉的遗传算法.通过大量实验,有针对性地对该算法中的编码结构、选择算子、交叉算子和变异算子进行优化设计;对相关控制参数进行合理调整,实验结果表明,该算法不仅有效地提高了组卷质量和组卷速度,而且具有很好的收敛性.

In order to improve the speed and quality of color image segmentation, aiming at the limitation of the cuckoo algorithm, every time after the end of the Lévy flight, a new optimization seeking equation is proposed, and the discovery probability and the pace factor are respectively proposed a new operating equation. Based on this, proposed an enhanced cuckoo algorithm (ECS),and the ECS algorithm is based on the multi threshold segmentation of color image. Through the comparison of the proposed algorithm (ECS), the standard PSO algorithm and the standard CS algorithm,the ECS algorithm is the best of both subjective and objective results, fully able to be applied to the actual multi threshold segmentation.

The behaviors of employed bees,onlooker bees in artificial bee colony algorithm are introduced into every time after the end of flight Levi of cuckoo search(CS)algorithm to optimized guiding,the discovery probability and the pace factor are also using the corresponding new variation factors,and change with cuckoo algorithm operation.Based on this,a hybrid artificial bee colony algorithm and cuckoo search algorithm (HACS)is proposed,and the HACS algorithm is used for color image segmentation.Experimental results show,the HACS algorithm can effectively solve the cuckoo algorithm long convergence time,lower accuracy.Good results have been achieved in the color image multi threshold segmentation.

随着遥感图像大数据的出现,常见的彩色遥感图像边缘检测方法运算量大、速度慢、效果差等缺点越来越明显。以四元数表示彩色像素为基础,改进人工蜂群算法的单一搜索方程,加大雇主蜂搜索范围,加入跟随蜂莱维飞行因子,提出了基于双搜索方程的人工蜂群算法。实验结果表明,该算法具有计算量小、去噪能力强、边缘检测效果好等优点。该算法能有效地应用于从遥感图像中获取识别目标。

A Rough Set and Information Gain based on sentiment feature selection method is proposed for building a solid foundation in sentiment analysis .The novel method firstly uses Information Gain to select a feature subset which has high relativity with the class attribute .Secondly ,the features which have high redundancy will be eliminated by Rough Set .Experimental results on several datasets reveal the method makes accuracy increase 4-9 percentages than other methods .It is an outstanding feature selection method and has significance in sentiment analysis .

The training and shaping of engineering applied talents in local colleges and universities is achieved through courses,the depth of curriculum reform is the overall optimization of the curriculum system. Comply with local characteristics,serve the regional economic and social development,meet the needs of the market,optimize and integrate the curriculum system from the perspective of vocational ability development,complete the organic integration of knowledge and ability,achieve the cultivation and improving of comprehensive knowledge,engineering ability and pro-fessional ethics.

为了进一步提高双聚类结果的性能,提出了一种基于变分贝叶斯的半监督双聚类算法.首先,在双聚类过程中引入了行和列的辅助信息,并提出了相应的联合分布概率模型;然后基于变分贝叶斯学习方法对联合概率分布中的参数进行估计;最后,通过合成数据集和真实的基因表达式数据集对提出的算法性能进行评估.实验表明,提出的算法在进行双聚类分析时,其归一化互信息量明显优于相关的双聚类算法.

As the color remote sensing image has the most notable features such as huge amount of data, rich image details, and the containing of too much noise, the edge detection becomes a grave challenge in processing of remote sensing image data. To explore a possible solution to the urgent problem, in this paper, we first introduced the quaternion into the representation of color image. In this way, a color can be represented and analyzed as a single entity. Then a novel artificial bee colony method named improved artificial bee colony which can improve the performance of conventional artificial bee colony was proposed. In this method, in order to balance the exploration and the exploitation, two new search equations were presented to generate candidate solutions in the employed bee phase and the onlookers phase, respectively. Additionally, some more reasonable artificial bee colony parameters were proposed to improve the performance of the artificial bee colony. Then we applied the proposed method to the quaternion vectors to perform the edge detection of color remote sensing image. Experimental results show that our method can get a better edge detection effect than other methods.