软件缺陷预测模型间的比较实验: 问题、进展与挑战

近年来,研究者提出了大量的软件缺陷预测模型,新模型往往通过与过往模型进行比较实验来表明其有效性.然而,研究者在设计新旧模型间的比较实验时并没有达成共识,不同的工作往往采用不完全一致的比较实验设置,这可能致使在对比模型时得到误导性结论,最终错失提升缺陷预测能力的机会.对近年来国内外学者所做的缺陷预测模型间的比较实验进行系统性的总结:首先,阐述缺陷预测模型间的比较实验的研究问题;然后,分别从缺陷数据集、数据集划分、基线模型、性能指标、分类阈值这5个方面对现有的比较实验进行总结;最后,指出目前在进行缺陷预测模型间比较实验时面临的挑战,并给出建议的研究方向.     

一种特征转移和域自适应的异质缺陷预测方法

软件缺陷预测是软件工程领域中的一个研究热点.跨项目缺陷预测(Cross-Project Defect Prediction, CPDP)采用源项目的缺陷数据来预测目标项目的缺陷倾向性.然而,源项目和目标项目的特征可能会有所不同.因此,研究人员提出了异质缺陷预测(Heterogeneous Defect Prediction, HDP).为了提高HDP模型的性能,本文提出了一种基于特征迁移和域自适应的异质缺陷预测(Feature Transfer and Domain Adaptation, FTDA)方法.首先,FTDA使用基于相关性的特征选择算法从源项目的特征集中选择最优的特征子集.随后,使用欧氏距离进行匹配特征,并为每个目标项目选择最合适的源项目.再次,使用TCA算法解决不同项目之间的分布差异问题.最后,使用SMOTETomek算法处理类不平衡问题.为了验证FTDA方法的有效性,本文对AEEEM,PROMISE,NASA和Relink数据集中的24个项目进行了实证研究.实证研究表明,FTDA显著提高了异质缺陷预测的性能.     

Defect Prediction Using Akaike and Bayesian Information Criterion

Data available in software engineering for many applications contains variability and it is not possible to say which variable helps in the process of the prediction. Most of the work present in software defect prediction is focused on the selection of best prediction techniques. For this purpose, deep learning and ensemble models have shown promising results. In contrast, there are very few researches that deals with cleaning the training data and selection of best parameter values from the data. Sometimes data available for training the models have high variability and this variability may cause a decrease in model accuracy. To deal with this problem we used the Akaike information criterion (AIC) and the Bayesian information criterion (BIC) for selection of the best variables to train the model. A simple ANN model with one input, one output and two hidden layers was used for the training instead of a very deep and complex model. AIC and BIC values are calculated and combination for minimum AIC and BIC values to be selected for the best model. At first, variables were narrowed down to a smaller number using correlation values. Then subsets for all the possible variable combinations were formed. In the end, an artificial neural network (ANN) model was trained for each subset and the best model was selected on the basis of the smallest AIC and BIC value. It was found that combination of only two variables’ ns and entropy are best for software defect prediction as it gives minimum AIC and BIC values. While, nm and npt is the worst combination and gives maximum AIC and BIC values.     

Numerical simulation of molding-defect formation during resin transfer molding

The present study investigated a numerical simulation of molding-defect formation during resin transfer molding using boundary element method and line dynamics. The proposed method enables to simulate small molding defects by increasing the node for required position during time evolution; thereby, the method computes high-resolution flow front without being affected by the initial mesh geometry. The method was applied to the radial injection RTM with single inlet, and it was confirmed by comparison with theoretical value based on Darcy’s law that the flow advancement was computed with high accuracy. In addition, the method was also applied to the flow advancement for inclusion problem with cylinder, and four-point injection problem. The simulated flow behavior, void formation, and shrinkage agreed with the results in references. Finally, the method was compared with experiments using two-point injection problem. The computed configuration of the flow front and weld line agreed well with the experimental results.     

Low-valence ion addition induced more compact passive films on nickel-copper nano-coatings

Ni-Cu nano-coatings were prepared by pulsed electroplating technique in the baths containing various amount of boric acid. Their microstructure, morphologies and corrosion resistance were characterized in detail. The addition of boric acid strongly influences on the microstructure of the Ni-Cu coatings. The coating with a grain size of 130 nm, obtained from the bath containing 35 g L⁻¹ boric acid, shows the highest corrosion resistance. This is attributed to the low-valence Cu ion (Cu⁺) additions in nickel oxide, which could significantly decrease the oxygen ion vacancy density in the passive film to form a more compact passive film. The higher Cu⁺ additions and the lower diffusivity of point defects (D₀) are responsible for the formation of more compact passive film on the coating obtained from the bath with 35 g L⁻¹ boric acid.     

Study of polymer‐network structure on polymer‐stabilized blue phase

Abstract— To understand the mechanism of the disclination defect of the liquid‐crystal (LC) phase, this study was conducted to directly analyze the polymer‐network (PN) structure of polymer‐stabilized blue phase (PSBP), which is minutely formed on all LC layers. The PN was examined after first removing the glass decap and then the LC. Important to note is that the removal of the glass decap did not affect or damage the PN structure. The PN was determined to be a stable structure without any change to the thickness of the layer. When removing the LC, both hexane and acetone solutions were used. Moreover, there was no structural deformation to the PN when using the hexane solution. The results of the study show that the actual size of the polymer chain is in fact 50–60 nm, five times larger than previous theories which estimated the size to be only 10 nm. In addition, this study confirmed that the pores between the PN are 100–200 nm. The PN structure was shown to be susceptible to change based on different heating temperatures. In summation, now that defect lines of a LC display (LCD) could be directly measured, further progress and development in the theoretical interpretations of the Kerr effect on PSBP can be realized.     

低对比度下水印图像缺陷检测

在主成分分析(PCA)及核主成分分析(KPCA)进行特征提取基本原理的基础上,提出了一种改进的提取非线性的图像特征来重建图像方法,应用于嵌入式防伪水印图案缺陷的检测。该方法使得图像协方差矩阵维数大幅下降,且有效地保留了嵌入式防伪水印图案的信息,通过比较检测出图像的缺陷。实验结果表明,该方法对输入数据实现了有效的降维,缩短了计算时间,提高了检测效果和精确度。KPCA算法相比原有的PCA算法具有更高的性能指标,适用范围更广。     

A new strategy for defect inspection by the virtual inspection in semiconductor wafer fabrication

The defect of process equipments is a major factor that impairs the yields in the mass production of semiconductor wafer fabrication and it is a main supervision means to use high-resolution defect inspection tools to detect and monitor the defect damage. Due to the high investment costs of these inspection tools and the resulting decrease in the throughput, how to improve the sampling rate is an important issue for the associated inspection strategy. This paper proposes a new concept and implementation of virtual inspection (VI) to enhance the detection and monitoring of defect in semiconductor production process. The underlying theory of the VI concept is that the state variables identifications (SVIDs) of process equipments can reflect the process quality effectively and loyally. The approach of VI is to combine the application of the fault detection and classification (FDC), and the defect library and the re-engineering of inspection procedure to reach the full-scope of strategic objective. VI enables the defect monitoring to enter a new era by promoting the monitoring level of defect inspection from the previous lot-sampling basis to the wafer-sampling level, and hence upgrades the sampling strategy from random-sampling to full and right-sampling. In this study, various typical defect cases are utilized to illustrate how to create VI models and verify the reliability of the proposed approach. Furthermore, a feasible architecture of the VI implementation for mass production in semiconductor factory is presented in the paper.     

AOI技术在PCB缺陷检测中的设计与实现

随着计算机技术和数字图像处理技术的快速发展,传统PCB缺陷检测技术因技术落后已经无法满足现代PCB质量的要求,而AOI技术以其准确、高效的特点在PCB缺陷检测中的应用越来越受到重视.主要研究了AOI技术在PCB缺陷检测中的实际应用,分别进行了硬件选择和软件系统研究,建立了基于AOI技术的PCB缺陷检测系统,采用基于轮廓对比的数据处理方法进行缺陷检测.实验证明了该系统的可行性与实用性.