A Machine Vision System for the Rotational Positioning and Verification of Oil Filters

A machine vision system that has been designed for the precise positioning and the reliable verification of industrial parts is described. This system has been integrated into an assembly line for producing various types of oil filters. The primary goal was to design a flexible system that could be easily adapted for dealing with various types of oil filters. The flexibility was provided by visual learning using an appearance-based eigenspace approach. For computational efficiency, a modified Karhunen–Loeve (K-L) algorithm was developed. The required robustness and accuracy were achieved by using a closed loop, such that all operations can be evaluated and corrected when required. The results obtained show that the system works reliably, and meets the requirements for flexibility, accuracy and speed.     

基于机器视觉的智能定位与检测技术研究

基于机器视觉工作原理,对机器视觉的智能定位与检测等相关技术进行了研究,设计了一种基于机器视觉智能定位与检测系统,对视觉跟踪精度进行测试,跟踪精度约为（-0.2°～＋0.2°）,并将该技术应用于LED球泡灯自动化装配线,可满足装配线的实际生产需求.     

Development of machine vision system and dimensional analysis of the automobile front-chassis-module

In the present research work, an automated machine vision system and a new algorithm to interpret the inspection data has been developed. In the past, the control of tolerance of front-chassis-module was done manually. In the present work a machine vision system and required algorithm was developed to carryout dimensional evaluation automatically. The present system is used to verify whether the automobile front-chassis-module is within the tolerance limit or not. The directional ability parameters related with front-chassis-module such as camber, caster, toe and king-pin angle are also determined using the present algorithm. The above mentioned parameters are evaluated by the pose of interlinks in the assembly of an automobile front-chassis-module. The location of ball-joint center is important factor to determine these parameters. A method to determine the location of ball-joint center using geometric features is also suggested in this paper. In the present work a 3-D best fitting method is used for determining the relationship between nominal design coordinate system and the corresponding feature coordinate system.     

Robotic automatic assembly system for random operating condition

A vision guided tabletop robotic assembly system is constructed for the random environment automatic assembly purpose. The distributed control structure is designed for this assembly system. A PC is specified as the central control unit for machine vision operation and the harmonization of At89c51 distributed control unit for each axis. The machine vision is introduced to search the locations and measure the size of the assembly parts in the robotic workspace. Then, the central controller commands the robot to pick the part sequentially based on the specified program to complete the automatic assembly process. The model-free fuzzy sliding mode control scheme is embedded in each joint micro controller for simplifying the model based control problem. The central control unit PC communicates with each joint controller by network communication. This assembly system is implemented on a 3 DOF SCARA robot. The experimental results show that the robotic motion control performance is good enough for assembling work; the vision pattern classification and assembly motion paths planning and monitoring are exactly executed by the central control unit. The goal of this robotic random assembly operation is achieved with the appropriate integration of robotic motion control, machine vision, sensor and assembly force detecting techniques.     

基于机器视觉的激光植球系统标定

构建了一个基于机器视觉辅助定位的激光植球系统。通过试验发现，整个系统的定位误差主要由机器视觉中透镜畸变和X—Y定位平台的装配误差引起。对于透镜畸变，引入了基于直线方法的图像畸变校正算法进行校正；对于非垂直的X—Y定位平台，通过基于机器视觉的软件补偿算法来校正X—Y定位平台的装配误差。最后，给出了试验对比数据和试验结果，验证了整个系统的标定效果。     

Fault diagnosis system for series compensated transmission line based on wavelet transform and adaptive neuro-fuzzy inference system

This paper proposes a new fault diagnosis approach based on combined wavelet transform and adaptive neuro-fuzzy inference system for fault section identification, classification and location in a series compensated transmission line. It performs an effective feature extraction approach based on norm entropy in order to obtain the features represented main frequency, harmonic and transient characteristics of the fault signals. The proposed method uses the samples of fault voltages and currents for one cycle duration from the inception of fault. The feasibility of the proposed method has been tested on a 400 kV, 300 km series compensated transmission line for all the ten types of faults using MATLAB/Simulink for a large data set of 23,436 fault cases comprising of all the 10 types of faults. Fault signals varying with fault resistance, fault inception angle, fault distance, load angle, percentage compensation level and source impedance are applied to the proposed algorithm. The results also indicate that the proposed method is robust to wide variation in system conditions and has higher fault diagnosis accuracy with regard to the other approaches in the literature for this problem.     

基于机器视觉的阀芯自动装配系统

针对球阀阀芯自动装配过程中存在装配柔性低、时常无法精确定位的问题,基于机器视觉技术,构建了一套在线机器视觉系统.采用NI Vision Assistant模块对系统及图像进行快速、准确的标定,并在其中对零件图像进行预处理、模式匹配以及边缘检测等修正处理.在LabVIEW中调用Vision Assistant节点,得出零...     

Assembly Auxiliary System for Narrow Cabins of Spacecraft

Due to the narrow space and complex structure of spacecraft cabin, the existing asssembly systems can not well suit for the assembly process of cabin products. This paper aims to introduce an assembly auxiliary system for cabin products. A hierarchical-classification method is proposed to re-adjust the initial assembly relationship of cabin into a new hierarchical structure for efficient assembly planning. An improved ant colony algorithm based on three assembly principles is established for searching a optimizational assembly sequence of cabin parts. A mixed reality assembly environment is constructed with enhanced inforamtion to promote interaction efficiency of assembly training and guidance. Based on the machine vision technology, the inspection of left redundant objects and measurement of parts distance in inner cabin are efficiently performed. The proposed system has been applied to the assembly work of a spacecraft cabin with 107 parts, which includes cabin assembly planning, assembly training and assembly quality inspection. The application result indicates that the proposed system can be an effective assistant tool to cabin assembly works and provide an intuitive and real assembly experience for workers. This paper presents an assembly auxiliary system for spacecraft cabin products, which can provide technical support to the spacecraft cabin assembly industry.     

面向航天器舱体的智能装配系统研究

针对航天器舱体的装配,运用计算智能、人类智能和机器智能构建了一个智能装配系统。采用蚁群算法对舱体内零件的装配顺序进行了规划和优化,降低了复杂结构体的装配规划难度。采用混和现实技术给出了虚拟装配验证和训练的解决方案,从视觉、信息和操作三个方面来提高装配的人机交互效率;基于视觉的装配质量检测方法可为检查人员提供有效的提示信息,协助检查人员快速发现多余物,减小遗漏多余物的几率。     

Support vector machine based data processing algorithm for wear degree classification of slurry pump systems

Classification is a useful tool in identifying fault patterns. Generally, a good classification implementation is closely related to the effectiveness of data used. The word “effectiveness” implies that the data should be clean and the features indicating fault patterns should be properly selected. Unfortunately, data cleaning is not often implemented in reported work of fault pattern classifications. In this paper, a data processing algorithm is developed to achieve the effectiveness, which includes data cleaning followed by feature selection. A data cleaning algorithm is developed based on support vector machine and random sub-sampling validation. Candidate outliers are selected based on fraction values provided by the proposed data cleaning algorithm and final outliers are determined based on their removal impacts on classification performance. The feature selection algorithm adopts the classical sequential backward feature selection. The performance of the data cleaning algorithm is tested using three benchmark datasets. The tests show good capability of the data cleaning algorithm in identifying outliers for all datasets. The proposed data processing algorithm is adopted in the classification of the wear degree of pump impellers in a slurry pump system. The results show good effectiveness of sequentially using data cleaning and feature selection in addressing fault pattern classification problems.     

基于机器视觉的芯片编号识别算法

针对芯片检测过程中芯片编号形状小、人眼无法识别的现状,开发了一套芯片编号识别系统。系统以机器视觉技术为基础,配备高分辨率Basler相机、高倍率镜头以及高性能点光源,能实现高分辨率图像采集;利用Hough直线检测对图像进行水平旋转,利用图像轮廓查找算法对兴趣区域进行提取分割,利用帧差法对兴趣区域与标准库进行比较,找到最匹配的图像,其标签为识别结果。实验结果表明,运用系统的图像处理算法可以代替人眼通过高倍放大镜识别,改进了芯片检测工艺,大大提高了芯片检测的效率。     

On-machine 3D vision system for machining setup modeling

In computer numerical control machine tools, using machining simulation to prevent collision becomes more and more popular due to its efficiency and low cost. However, if the entire digital model of the machining setup does not exist, the simulation is not applicable. As a result, the operator has to manually check the numerical control program, which is a time-consuming and error-prone process. In this paper, an on-machine vision system is presented to quickly construct the digital model based on the actual machining setup. The total construction for a complex setup can be done within a few minutes. Several key technologies have been developed. First, a 2D edge feature detection algorithm has been designed which will extract the edges of the object of interest by processing both the real and virtual images. Second, a stereo vision system is developed which will obtain the three-dimensional (3D) edge data of the object of interest. A new algorithm is presented to solve correspondence, which is the key problem of the stereovision system. Furthermore, the 3D object recognition algorithm is developed to let the system intelligently search for the matched solid model in the database and import it into the virtual environment with an accurate pose. Finally, experiments are carried out to test the developed system.     

Monitoring and classification of dimensional faults for automotive body assembly

Sudden process changes occurring during automobile body assembly processes will influence the downstream assembly process and the functionality and final appearance of the vehicle. Furthermore, these faults could result in a decreased production rate and an increase in the cost if sudden process changes are so serious that the production line has to be stopped for investigation. Thus, sudden process changes should be detected and eliminated as soon as possible to prevent defective products from being produced and to reduce the cost of repairs/reworks. A monitoring algorithm is developed to detect, classify, and group process changes by analyzing the dimensional data of car bodies. The results of this monitoring algorithm can help diagnose the root causes of variation according to the locations of measurement points, body structure, assembly sequence, and tooling layout. Measurement data obtained from an optical coordinate measuring machine (OCMM) are used to demonstrate the monitoring technique.     

In-cycle monitoring of tool nose wear and surface roughness of turned parts using machine vision

Tool wear has been extensively studied in the past due to its effect on the surface quality of the finished product. Vision-based systems using a CCD camera are increasingly being used for measurement of tool wear due to their numerous advantages compared to indirect methods. Most research into tool wear monitoring using vision systems focusses on off-line measurement of wear. The effect of wear on surface roughness of the workpiece is also studied by measuring the roughness off-line using mechanical stylus methods. In this work, a vision system using a CCD camera and backlight was developed to measure the surface roughness of the turned part without removing it from the machine in-between cutting processes, i.e. in-cycle. An algorithm developed in previous work was used to automatically correct tool misalignment using the images and measure the nose wear area. The surface roughness of turned parts measured using the machine vision system was verified using the mechanical stylus method. The nose wear was measured for different feed rates and its effect on the surface roughness of the turned part was studied. The results showed that surface roughness initially decreased as the machining time of the tool increased due to increasing nose wear and then increased when notch wear occurred.     

基于机器视觉的全自动灯检机关键技术研究

设计了一种基于机器视觉技术的全自动智能灯检机,以实现药液中可见异物的在线检测.首先对采集到的各帧图像通过形态学滤波、帧差等预处理抑制大部分背景噪声.然后提取图像中目标的不变特征,通过各目标不变特征间的匹配,实现相邻帧点迹的关联,得到若干假设的目标运动轨迹.最后根据异物运动轨迹的连续性和平滑特点来确定是否存在可见异物.研制的全自动灯检机样机已经在某药厂的注射液生产线上试验运行.实验表明,该机器的识别准确率和速度均高于熟练的灯检工.     

微型铣刀外径视觉测量的不确定度

建立了基于机器视觉的测量系统,实现了微型铣刀外径的在线测量,并对测量不确定度进行了分析.测量系统由背光源、夹头与旋转平台、旋转编码器、摄像机和相应的测量软件组成.系统运行时,铣刀匀速旋转,由旋转编码器触发相机对其进行采样拍照并完成测量.从照明光源平行性、摄像机的镜头畸变与CCD噪声、调焦效果、待测物体表面几何特性、算法设计5个方面分析了测量结果不确定度来源及其影响.实验表明,当系统采样率不足时,对于铣刀外径测量不确定度影响最大的是待测物体表面几何特性;在采样率足够大的情况下,外径测量不确定度主要由光源平行性确定.本文系统采样率为33 frame/circle,系统测量不确定度为4μm,可满足生产要求.     

自适应神经模糊推理系统在倒立摆控制中的应用

针对单级倒立摆系统具有多变量、非线性、绝对不稳定的特点,应用Matlab/Simulink设计了用于倒立摆系统的、基于自适应神经模糊推理系统的ANFIS控制器,采用反向传播算法和最小二乘算法的混合算法对倒立摆控制样本数据进行学习,调整各变量的隶属度函数,自动产生模糊规则.仿真结果表明,ANFIS控制器对倒立摆系统的摆杆角度和小车位置的控制过程具有良好的动态性能和稳态性能.     

Compensation of installation errors in a laser vision system and dimensional inspection of automobile chassis

Laser vision inspection systems are becoming popular for automated inspection of manufactured components. The performance of such systems can be enhanced by improving accuracy of the hardware and robustness of the software used in the system. This paper presents a new approach for enhancing the capability of a laser vision system by applying hardware compensation and using efficient analysis software. A 3D geometrical model is developed to study and compensate for possible distortions in installation of gantry robot on which the vision system is mounted. Appropriate compensation is applied to the inspection data obtained from the laser vision system based on the parameters in 3D model. The present laser vision system is used for dimensional inspection of car chassis sub frame and lower arm assembly module. An algorithm based on simplex search techniques is used for analyzing the compensated inspection data. The details of 3D model, parameters used for compensation and the measurement data obtained from the system are presented in this paper. The details of search algorithm used for analyzing the measurement data and the results obtained are also presented in the paper. It is observed from the results that, by applying compensation and using appropriate algorithms for analyzing, the error in evaluation of the inspection data can be significantly minimized, thus reducing the risk of rejecting good parts.     

An automated assembly line for wrist watches using robots with vision

This paper deals with an automated assembly line using Seiko high precision robots (D-TRAN XY-3000) and incorporating a Seiko Vision-I system exclusively developed for the assembly of wrist watch casings. The line has the capability of instant changeover to cater for various types of watch casing. It discusses the reasons for the development of this extremely flexible assembly line, which is now in operation at the Takatsuka plant of Seiko Instrument & Electronics Ltd. The assembly of wrist watch casings is achieved by analysing the features of casing assembly so as to integrate the vision system and robots into the line.     

Toward an automatic wheat purity measuring device: A machine vision-based neural networks-assisted imperialist competitive algorithm approach

Wheat product quality is closely related to wheat seed purity. Purity is an important factor that has a considerable impact on wheat product prices in grain storage silos. The aim of this paper was to introduce a machine vision based approach as a primarily step for fabricating an automatic wheat purity determination and grading device. Experimental data consists of 52 color, morphology, and texture characteristic parameters, extracted from images of samples, including four local wheat grades and eight common weed seeds growing in wheat fields of Iran, were used to build the classification models. A new algorithm that combines Imperialist Competitive Algorithm (ICA) and Artificial Neural Networks (ANNs) has been used for two purposes: to find the best characteristic parameters set and to create robust classification models. Based upon the results obtained from this study, the total classification rate of ICA–ANN approach for wheat grains vs. non-wheat seeds, wheat grain classes, and non-wheat seed classes was 96.25%, 87.50%, and 77.22%, respectively.