Performance Evaluation for Motif-Based Patterned Texture Defect Detection

This paper carries an extensive evaluation on the performance of a generalized motif-based method for detecting defects in 16 out of 17 wallpaper groups in 2-D patterned texture. The motif-based method evolves from the concept that every wallpaper group is defined by a lattice, which contains a further constituent-motif. It utilizes the symmetry properties of motifs to calculate the energy of moving subtraction and its variance among motifs. Decision boundaries are determined by learning the distribution of those values among the defect-free and defective patterns in the energy-variance space. In this paper, shape transform for irregular motif has been demonstrated according to the three basic motif shapes: rectangle, triangle, and parallelogram. An error analysis for the misclassifications has also been delivered. In the database of fabrics and other patterned textures, a total of 381 defect-free lattices are used for formulation of boundaries while further 340 defect-free and 233 defective lattices are for testing. The motif-based method has a consistent result and reaches a detection success rate of 93.86%.     

Motif-based defect detection for patterned fabric

This paper proposes a generalized motif-based method for detecting defects in 16 out of 17 wallpaper groups in 2D patterned texture. It assumes that most patterned texture can be decomposed into lattices and their constituents—motifs. It then utilizes the symmetry property of motifs to calculate the energy of moving subtraction and its variance among different motifs. By learning the distribution of these values over a number of defect-free patterns, boundary conditions for discerning defective and defect-free patterns can be determined. This paper presents the theoretical foundation of the method, and defines the relations between motifs and lattice, from which a new concept called energy of moving subtraction is derived using norm metric measurement between a collection of circular shift matrices of motif and itself. It has been shown in this paper that the energy of moving subtraction amplifies the defect information of the defective motif. Together with its variance, an energy-variance space is further defined where decision boundaries are drawn for classifying defective and defect-free motifs. As the 16 wallpaper groups of patterned fabric can be transformed into three major groups, the proposed method is evaluated over these three major groups, from which 160 defect-free lattices samples are used for defining the decision boundaries, with 140 defect-free and 113 defective samples used for testing. An overall detection success rate of 93.32% is achieved for the proposed method. No other generalized approach can achieve this success rate has been reported before, and hence this result outperforms all other previously published approaches.     

Automated fabric defect detection—A review

This paper provides a review of automated fabric defect detection methods developed in recent years. Fabric defect detection, as a popular topic in automation, is a necessary and essential step of quality control in the textile manufacturing industry. In categorizing these methods broadly, a major group is regarded as non-motif-based while a minor group is treated as motif-based. Non-motif-based approaches are conventional, whereas the motif-based approach is novel in utilizing motif as a basic manipulation unit. Compared with previously published review papers on fabric inspection, this paper firstly offers an up-to-date survey of different defect detection methods and describes their characteristics, strengths and weaknesses. Secondly, it employs a wider classification of methods and divides them into seven approaches (statistical, spectral, model-based, learning, structural, hybrid, and motif-based) and performs a comparative study across these methods. Thirdly, it also presents a qualitative analysis accompanied by results, including detection success rate for every method it has reviewed. Lastly, insights, synergy and future research directions are discussed. This paper shall benefit researchers and practitioners alike in image processing and computer vision fields in understanding the characteristics of the different defect detection approaches.     

基于图像显著性特征的铸坯表面缺陷检测

针对钢铁铸坯表面检测的缺陷复杂性问题,从图像处理及图形特征角度提出一种基于显著性区域特征的算法.该算法首先对源图像进行显著性特征区域处理和Gabor小波滤波处理,得到了对应的特征图像;然后再将2幅图像中的特征区域进行融合,得到可信度较高的缺陷特征区域图像;最后在缺陷区域中用训练好的Adaboost分类器检测缺陷,得到最终的缺陷定位结果.该算法结合了显著性特征和Gabor小波特征,既缩小了Adaboost分类器的搜索范围,也提高了排除伪缺陷的能力,具有较快的定位速度和较高的准确率.实验结果表明,该算法能获得较好的效果,具有较高的实用价值.     

Implementation and Evaluation of Decision Trees with Range and Region Splitting

We propose an extension of an entropy-based heuristic for constructing a decision tree from a large database with many numeric attributes. When it comes to handling numeric attributes, conventional methods are inefficient if any numeric attributes are strongly correlated. Our approach offers one solution to this problem. For each pair of numeric attributes with strong correlation, we compute a two-dimensional association rule with respect to these attributes and the objective attribute of the decision tree. In particular, we consider a family R of grid-regions in the plane associated with the pairof attributes. For R R, the data canbe split into two classes: data inside R and dataoutside R. We compute the region R_opt R that minimizes the entropy of the splitting,and add the splitting associated with R_opt (foreach pair of strongly correlated attributes) to the set of candidatetests in an entropy-based heuristic. We give efficient algorithmsfor cases in which R is (1) x-monotone connected regions, (2) based-monotone regions, (3) rectangles, and (4) rectilinear convex regions. The algorithm has been implemented as a subsystem of SONAR (System for Optimized Numeric Association Rules) developed by the authors. We have confirmed that we can compute the optimal region efficiently. And diverse experiments show that our approach can create compact trees whose accuracy is comparable with or better than that of conventional trees. More importantly, we can grasp non-linear correlation among numeric attributes which could not be found without our region splitting.     

决策域分布保持的启发式属性约简方法

在决策粗糙集中,由于引入了概率阈值,属性增加或减少时,正域或者非负域有可能变大、变小或者不变,即属性的增减与决策域（正域或非负域）之间不再具有单调性.分析结果表明,现有的基于整个决策域的属性约简定义可能会改变决策域.为使决策域保持不变,引入了正域分布保持约简与非负域分布保持约简的概念.此外,决策域的非单调性使得属性约简算法必须检查一个属性集合的所有子集.为了简化算法设计,提出了正域和非负域分布条件信息量的定义,并证明其满足单调性,从而为设计决策域分布保持约简的启发式计算方法提供了理论基础.为了进一步获得最小约简,提出一种基于遗传算法的决策域分布保持启发式约简算法,并在两种单调的决策域分布条件信息量基础上构造了新算子,即修正算子,确保遗传算法找到的是约简而不是约简的超集.对比实验从分类正确率与误分类代价两个方面都反映了决策域分布保持约简定义的合理性,并且,所提出的遗传算法在大多数情况下都找到了最小约简.     

基于Siamese-YOLOv4的印刷品缺陷目标检测

在印刷工业生产中,针对直接使用YOLOv4网络进行印刷缺陷目标检测精度低、所需训练样本数量大的问题,提出了一种基于Siamese-YOLOv4的印刷品缺陷目标检测方法。首先,使用了一种图像分割和随机参数变化的策略对数据集进行增强;然后,在主干网络中增加了孪生相似性检测网络,并在相似性检测网络中引入Mish激活函数来计算出图像块的相似度,在此之后将相似度低于阈值的区域作为缺陷候选区域;最后,训练候选区域图像,从而实现缺陷目标的精确定位与分类。实验结果表明：Siamese-YOLOv4模型的检测精度优于主流的目标检测模型,在印刷缺陷数据集上,Siamese-YOLOv4网络对卫星墨滴缺陷的检测准确率为98.6%,对脏点缺陷的检测准确率为97.8%,对漏印缺陷的检测准确率为93.9%;检测的平均精度均值（mAP）达到了96.8%,相较于YOLOv4算法、Faster R-CNN算法、SSD算法、EfficientDet算法分别提高了6.5个百分点、6.4个百分点、14.9个百分点、10.6个百分点。所提Siamese-YOLOv4模型一方面在印刷品缺陷检测中有较低的误检率和漏检率,另一方面通过相似性检测网络计算图像块的相似度从而提高了检测的精度,表明所提缺陷检测方法可应用于印刷质检以提高印刷企业的缺陷检测水平。     

面向属性归纳下的多层次决策规则获取算法

针对信息系统中容错能力差、样本量小以及条件相同而决策结果不一致等问题,提出了一种在面向属性归纳下基于变精度粗糙集模型的多层次决策规则获取算法.首先,在条件属性的概念层次下分析高低层次决策表在变精度模型中下近似、正域、边界域和负域间关系.基于各层次决策表关系图,先由最高层决策表自顶向下按经典粗糙集模型获取确定性规则,然后再由最底层决策表自底向上获取更抽象的规则.实例分析说明了该算法的可行性.     

Stabilizable regions of receding horizon predictive control with input constraints

Stabilizable regions of receding horizon predictive control (RHPC) with input constraints are examined. A feasible region of states, which is spanned by eigenvectors of the closed-loop system with a stabilizing feedback gain, is derived in conjunction with input constraints. For states in this region, the feasibility of state feedback is guaranteed with the corresponding feedback gain. It is shown that an RHPC scheme with adequate finite terminal weights can guarantee stability for any initial state which can be steered into this region using finite number of control moves in the presence of input saturation. This methodology results in feasible regions which are infinite (in certain directions) even in the case of open-loop unstable systems. It is shown that the proposed feasible regions are larger than the ellipsoidal regions which were suggested in earlier works. We formulated the optimization problem in LMI so that it can be solved by semidefinite programming.     

复杂条件下多个指示色标块的检测

提出一种在不同光照和不同背景条件下多个指示色标块检测的算法。在Lab颜色空间的L通道中,在连续帧中根据图像的相似性确定背景图像,当前帧图像与背景图像差分确定运动区域以去除背景干扰和缩小指示色标块查找的区域。在运动区域中进行区域生长,计算运动区域凸包,在凸包中进行采样,并根据权值表赋予权值,通过聚类确定种子,使用基于最小错误率的贝叶斯决策作为生长准则进行生长。实验结果表明,该方法与传统的颜色阈值向量方法相比,应用场景更广,颜色块的检测效果更好。     

基于空间重叠度的DTBSVM多类分类算法

针对传统的DTBSVM算法中判断类间的可分的难易程度时可能造成的错误判断,提出了基于空间重叠度的DTSVM多类分类方法。该方法通过计算已知的类别样本在空间中的重叠度,合并有重叠的类,组合为一个新的类,再基于一种有效的类间可分性准则进行划分,使得容易划分的类能从决策树的根节点开始逐层分割出来,再划分有类间重叠的类,这样就可以尽量地避免“误差累积”的风险,构造出分类效果好的决策树结构。实验结果表明,该方法大大提升了DTSVM多类分类算法的分类正确率。     

一种基于重构相容决策表的属性约简算法

基于正区域的属性约简是目前最常用的一类约简算法。现实中的决策表有可能存在不一致的对象。另外,在约简过程中随着属性个数的减少,也有可能产生新的不一致对象。对于基于正区域的约简算法来说,不一致的对象并没有提供任何有用的信息,删除不一致的对象不会改变正区域的计算结果以及最终的约简结果,而且可以显著提高算法的效率。然而现有的基于正区域的约简算法并没有考虑到这个问题,它们采用论域中的所有对象来计算正区域并得出约简结果。针对这一问题,定义了重构相容决策表和重构相容决策子表的概念。引入这两个概念的目的是在约简过程中删除初始决策表中的不一致对象,从而获得一个相容决策表。借助于这两个概念,提出了一种新的基于正区域的属性约简算法。在真实数据集上的实验表明,与传统的算法相比,该算法能够获得较小的约简结果和较高的分类精度,并且具有相对较低的时间复杂度。     

代价敏感的Boosting软件缺陷预测方法

软件缺陷预测可以有效提高软件的可靠性,修复系统存在的漏洞。Boosting重抽样是解决软件缺陷预测样本数量不足问题的常用方法,但常规Boosting方法在处理领域类不平衡问题时效果不佳。为此,提出一种代价敏感的Boosting软件缺陷预测方法CSBst。针对缺陷模块漏报和误报代价不同的问题,利用代价敏感的Boosting方法更新样本权重,增大产生第一类错误的样本权重,使之大于无缺陷类样本权重与第二类错误样本的权重,从而提高模块的预测率。采用阈值移动方法对多个决策树基分类器的分类结果进行集成,以解决过拟合问题。在此基础上,通过分析给出模型构建过程中权重和阈值的最优化设置。在NASA软件缺陷预测数据集上进行实验,结果表明,在小样本的情况下,与CSBKNN、CSCE方法相比,CSBst方法的BAL预测指标分别提升7%和3%,且时间复杂度降低一个数量级。     

运动区域提取和闪频分析并行的火焰检测算法

在火焰检测中对火焰运动区域提取和闪烁特征分析大都分开进行,本文在提取运动区域的同时分析该区域的闪频特性,即将火焰的运动特征和闪烁特征同时提取。首先基于Ohta颜色空间找出图像中具有火焰颜色的疑似区域,其次根据视频图像某个位置在一段时间内变化的程度和次数是否都达到一定程度提取具有闪烁特性的运动区域,最后根据具有火焰颜色的连通区域是否包含这种运动区域,且颜色区域与运动区域的面积比例是否达到一定比值,来判断该连通区域是否为火焰。实验结果表明该方法在提取运动区域的同时能排除不具火焰闪烁特征的前景,且能在运动区域提取不完整的情况下保持较高的火焰检测率和较低的误检率。     

基于空间聚集特征的沥青路面裂缝检测方法

沥青路面裂缝自动检测是制约公路养护科学决策的最主要瓶颈.针对现有裂缝检测算法在大规模应用特别是广地域、多路况等复杂环境下算法稳定性、可靠性及实时性等方面存在严重不足问题.本文在观察大量实际工程路面图像基础上, 对路面裂缝特征进行全新定义, 提出了一种基于空间聚集特征的沥青路面裂缝检测方法, 参考裂缝的空间分布、灰度、几何等特征, 以子块图像为处理单元, 采用逐步求精的策略对子块图像进行分割, 快速定位空间聚集区域, 再对聚集区域进行评估得到信度高的裂缝候选区域; 最后以裂缝候选区域为种子区域, 在准确估算裂缝发展趋势的基础上, 结合裂缝片段聚集及相似性等特性, 去除噪声同时合并连接断裂的裂缝, 实现了裂缝区域较为完整的检测.通过测试多路况、多采集环境下近万样本, 并采用不同的方法对测试结果进行评估, 结果显示, 算法对不同类型路面图像中具有不同特征的裂缝区域均具有良好的检测性能, 裂缝定位准确性达到95%以上, 裂缝区域检测的完整性达到90%以上.     

结合丰度特征的决策树及其土地覆盖分类

将混合像元分解的丰度加入特征集,结合光谱信息和DEM数据生成决策分类规则。运用陆地卫星TM影像对黄河源区的玛多县进行土地覆盖分类试验。通过特征提取、决策分类和后处理,得到该县的土地覆盖类型图。采用1∶10万土地覆盖类型图和实地考察数据进行精度评价,结果表明:结合丰度的决策树与最大似然分类和普通决策树分类(不加丰度信息)相比,分类精度分别提高了17.3%和9.5%。     

区间值决策表的正域增量式属性约简算法

实际应用中存在大量动态增加的区间型数据，若采用传统的非增量正域属性约简方法进行约简，则需要对更新后的区间值数据集的正域约简进行重新计算，导致属性约简的计算效率大大降低。针对上述问题，提出区间值决策表的正域增量属性约简方法。首先，给出区间值决策表正域约简的相关概念；然后，讨论并证明单增量和组增量的正域更新机制，提出区间值决策表的正域单增量和组增量属性约简算法；最后，通过8组UCI数据集进行实验。当8组数据集的数据量由60%增加至100%时，传统非增量属性约简算法在8组数据集中的约简耗时分别为36.59 s、72.35 s、69.83 s、154.29 s、80.66 s、1498.11 s、4124.14 s和809.65 s，单增量属性约简算法的约简耗时分别为19.05 s、46.54 s、26.98 s、26.12 s、34.02 s、1270.87 s、1598.78 s和408.65 s，组增量属性约简算法的约简耗时分别为6.39 s、15.66 s、3.44 s、15.06 s、8.02 s、167.12 s、180.88 s和61.04 s。实验结果表明，提出的区间值决策表的正域增量式属性约简算法具有高效性。     

带权决策表的属性约简

属性约简是粗糙集理论的重要应用。考虑将决策表中的每行都作为一条决策规则时，若把表中出现相同决策规则的次数作为权，可得到带权决策表。提出了关于带权决策表的正域约简相应的辨识矩阵并给出了证明，从而得到了约简算法。相比于决策表中的正域约简时发现，通过将决策表转化为带权决策表后，再利用算法1进行约简时，其在一定程度上优于前者。提出了近似分类精度约简相应的辨识矩阵并给出了证明。对于2个算法，在选取的UCI数据集上进行了实验验证。通过实验进一步说明了所提出算法的可行性和有效性。     

A homogeneous ensemble method for predicting gastric cancer based on gastroscopy reports

Gastroscopy is important for finding suspicious stomach lesions, screening for gastric cancer, and providing early diagnoses. Due to the differences in the levels of diagnosis and treatment among gastroscope doctors, clinical diagnosis based on gastroscopy is limited by low diagnostic sensitivity and specificity to gastric cancer. An assistive system for gastroscopy report analysis can be helpful to improve the success rate of gastric cancer detection. In this study, a homogeneous ensemble decision support system for gastric cancer screening (Endo-GCS) that performs word segmentation, feature extraction, and gastric cancer screening on text-based gastroscopy reports is proposed. The proposed Endo-GCS method establishes a progressive local weighting algorithm that improves the overall prediction performance of the homogeneous ensemble model in gastric cancer screening. An optimal threshold estimation algorithm is developed to minimize the negative impact of misdiagnosis and missed diagnoses. Through a comparative experimental study using real gastroscopy report data, the pathological examination conclusion is the gold standard. The sensitivity of the proposed Endo-GCS method is 88.27%, the specificity is 77.84%, and the accuracy is 82.11%, which significantly improved the sensitivity 65.49% and the accuracy 80.5% of the gastroscopic diagnosis results, respectively.     

基于多分类器融合的未知嵌入率图像隐写分析方法

提出一种基于多分类器融合的未知嵌入率图像隐写分析方法.通过建立多个不同嵌入率下的训练分类器模型,得到对测试图像的多个局部决策值;然后将得到的局部决策值转化为证据,并根据各分类器的漏检率和虚警率,对各局部决策值分配权重;最后由基于权重系数的D-S(Dempster-Shafer)证据理论推理得到最终的决策.针对LSB匹配隐写的实验结果表明,本文方法改善了未知嵌入率下的隐写检测性能.