Autoencoder-based anomaly detection for surface defect inspection

In this paper, the unsupervised autoencoder learning for automated defect detection in manufacturing is evaluated, where only the defect-free samples are required for the model training. The loss function of a Convolutional Autoencoder (CAE) model only aims at minimizing the reconstruction errors, and makes the representative features widely spread. The proposed CAE in this study incorporates a regularization that improves the feature distribution of defect-free samples within a tight range. It makes the representative feature vectors of all training samples as close as possible to the mean feature vector so that a defect sample in the evaluation stage can generate a distinct distance from the trained center of defect-free samples. The proposed CAE model with regularizations has been tested on a variety of material surfaces, including textural and patterned surfaces in images. The experimental results reveal that the proposed CAE with regularizations significantly outperforms the conventional CAE for defect detection applications in the industry.     

基于光电传感器的机器视觉系统

对光电传感器在视觉系统上的应用进行了研究。通过学习光电传感器的理论知识,利用其将可见光信号转化为电信号,便于人们识别的原理,在采摘果实机器人上做了研究。运用SolidWorks制图技术清晰呈现出光电传感器的光信号识别功能。本文通过软件模拟,描述光电传感器对果实成熟度的辨识,从而保证农产品的销售质量,大幅度提升采摘农产品时所消耗的人力物力财力。     

Automatic Quality Inspection of Bakery Products Based on Shape and Color Information

基于形状和颜色信息的烘焙产品的自动质量检测

Babar Khan,韩芳*,王直杰,Ather Iqbal

（东华大学 信息科学与技术学院,上海 201620）

创新点说明：

利用支持向量机和基于生物启发式HMAX模型的形状辨识器与RGB对抗颜色通道辨识器集成而形成一种有效的智能分类系统,并将其用于烘焙产品的质量检测。

研究目的：

本文要对烘焙产品进行更有效的质量检测,需要提出一种智能方法。

研究方法：

本文采用了支持向量机、基于生物启发式HMAX模型的形状辨识器与RGB对抗颜色通道辨识器集成,形成一种高级自动分类系统,对烘焙产品根据其形状和颜色（烘焙时间）进行对应分类。

结果：

利用本文方法对烘焙产品进行质量检测的结果与其它方法相比较,发现本文提出的基于生物启发的机器视觉系统运行更加精确和有效,同时发现采用One Versus One支持向量机和Directed Acyclic Graph支持向量机可以得到最大的分类精确率。本文方法对饼干形状和颜色的辨识分别达到了95%和100%的分类精确率。同时,算法稳定。

结论：

本文方法因其更加可靠和精确可用于代替现有的烘焙产品检测方法。

关键词：烘焙产品质量检测,计算机视觉,HMAX, 对抗颜色通道,RGB颜色辨识器,支持向量机

     

全启发式教学法在“金属切削机床与刀具”课程中的应用

针对当代大学生的特点,阐述了来自于建构主义的全启发式教学方法的理论,对“金属切削机床与刀具”课程内容进行问题分解,采用以启发诱导、共同协作为核心的全启发式教学方法,对该课程的教学环节进行了合理的设计。     

独立电源系统中的无刷双馈电机动态运行特性研究

针对独立电源系统中的无刷双馈发电机的特点,考虑到功率绕组和控制绕组的端电压随转速和负载的变化而变化,在无刷双馈电机d-q轴数学模型的基础上,建立了无刷双馈发电机在带载情况下的数学模型。并基于该数学模型,在matlab／simulink下建立了转速变化和负载变化的仿真模型,完成了其动态响应特性的研究。     

民办高职院校党员服务中心构建对策

构建民办高职院校党员服务平台,可以充分发挥其服务党员、服务群众、服务师生的作用,促进学校健康稳定发展.分析了民办高职院校党员服务中心的建设动因,明确了服务功能、服务项目,提出了建设路径及意义.     

离心铸造铸铁排水管生产中的一些问题

针对离心铸造铸铁排水管生产过程中存在的问题，如设备选型、涂料的影响，麻坑、针孔等铸造缺陷的产生原因等进行了分析和探讨。只要铸造生产工艺合理、涂料性能合适，即使是简易型离心铸管机也可生产出无铸造缺陷的高质量铸管。涂料性能与涂料使用方法强烈影响铸管表面质量和管模使用寿命。铸管外表面麻坑、针孔等缺陷的主要影响因素是涂料本身的性能和涂料使用工艺，铸管内表面龟纹等铸造缺陷的主要影响因素是铁水质量。     

镁合金压铸机动模板极限载荷的三维有限元分析

压铸机动模板在合模状态下的变形程度对所加工零件的精度具有重要影响，针对这一问题，利用有限元分析手段对30000kN压铸机动模板极限载荷下的力学行为进行了分析，比较了不同设计方案的位移场以及应力场等计算结果。在依据计算结果对设计方案进行分析的基础上，对相应的加工制造提出了一定的意见和建议。分析结果表明，减重孔对动模板刚度具有一定影响．目前的设计方案中传力铰耳承载面积偏小，存在设计缺陷。     

New approach in tool wear measuring technique using CCD vision system

This paper suggests a reliable direct measuring procedure for measuring different tool wear parameters. Modern image processing techniques and machine vision systems today enable direct tool wear measurement to be accomplished in-cycle. The presented system is characterized by its measurement flexibility, high spatial resolution and good accuracy. The system consists of a light source to illuminate the tool, CCD camera, laser diod (used in conjunction with profile deepness assessment) with linear projector, grabber for capturing the picture, and a PC. The technique is specially characterized by its determination of profile deepness with the help of projected laser raster lines on a tool surface. So it has advantage comparing with other techniques, which can measure only 2D profiles. With the technique presented in this paper a 3D image of relief surface can be obtained without having need to employ a very complicated measuring system. All indirect methods like acoustic emission, force measurement, spindle current measurement, vibration sensors, etc. are very time consuming and demand very expensive subsidiary measurement equipment, compared with the method presented here.