技术领域technical field
本发明涉及一种基于大律法的布面疵点检测及分类方法，应用于纺织品的质量检测，属于图像处理技术领域。 The invention relates to a cloth defect detection and classification method based on Dalu Law, which is applied to the quality detection of textiles and belongs to the technical field of image processing. the
背景技术Background technique
在纺织品生产、评级以及销售各环节中，布料疵点的检测与识别是织物质量控制的关键环节，具有十分重要的意义。传统的人工裸眼检测易出现漏检和误检，因而布料疵点自动检测成为近年来国内外研究的热门课题。目前国内学者已经提出了一些布料疵点自动检测方法。 In the process of textile production, grading and sales, the detection and identification of fabric defects is a key link in fabric quality control, which is of great significance. Traditional manual naked eye detection is prone to missed detection and false detection, so automatic detection of fabric defects has become a hot topic of research at home and abroad in recent years. At present, domestic scholars have proposed some automatic detection methods for fabric defects. the
申请号为201310119051.1，公开号为CN103207186A，发明名称为“自动验布机疵点检测识别方法及其系统”的中国专利，提供了一种自动验布机疵点检测识别方法，首先通过电荷耦合器件(CCD)传感器线阵相机采集布匹图片，然后利用疵点识别算法进行正常布与布匹疵点的准确区分，被机器判别为织物疵点的图片首先存储到验布机工控机的硬盘内，然后该主机经由网络将疵点图片及疵点位置等信息传输到办公室内远程终端，疵点图片通过图像增强算法使得疵点具有更好地人眼分辨率，再由验布工人根据该处理后的疵点图片进行疵点类别的判断。该发明虽然一定程度上提高了验布效率，但其采用CCD线阵相机采集图片，一方面CCD相比接触式图像传感器(CIS)，CCD相机的光学系统需要开阔的现场，因而需要占用较大的空间，不利于对现有设备的改造利用，增加了装置设计复杂度且无法消除光源对疵点识别率的影响；另外线阵CCD相机尤其是彩色线阵CCD相机的价格昂贵，使得设备的制造成本上升，不利于专利技术的推广应用。在疵点分类方法上该发明仍主要依赖于人工，没有完全取缔人工验布。 The application number is 201310119051.1, the publication number is CN103207186A, and the Chinese patent titled "Automatic Cloth Inspection Machine Defect Detection and Recognition Method and System" provides a method for automatic cloth inspection machine defect detection and recognition. ) sensor line array camera collects cloth pictures, and then uses the defect recognition algorithm to accurately distinguish between normal cloth and cloth defects. Information such as defect pictures and defect locations are transmitted to the remote terminal in the office. The defect pictures are made to have better human eye resolution through image enhancement algorithms, and then the fabric inspectors judge the defect categories based on the processed defect pictures. Although this invention improves the cloth inspection efficiency to a certain extent, it uses a CCD line array camera to collect pictures. On the one hand, compared with a contact image sensor (CIS), the optical system of a CCD camera needs a wider field, so it needs to occupy a larger area. The space is not conducive to the transformation and utilization of existing equipment, which increases the complexity of device design and cannot eliminate the influence of light sources on the defect recognition rate; in addition, the expensive price of line array CCD cameras, especially color line array CCD cameras, makes the equipment manufacturing Rising costs are not conducive to the popularization and application of patented technologies. In the defect classification method, this invention still mainly relies on manual work, and has not completely banned manual cloth inspection. the
申请号为201010222997.7，公开号为CN 101957326A，发明名称为“一种纺织品表面质量多光谱的监测方法与装置”的中国专利，由接触式图像传感器所组成的CIS阵列、多光谱LED通断控制、图像数据采集、电机控制器、主控制器、触摸屏或键盘、显示器、喷墨式指示器、通信接口和打印机组成。主控制器通过电机控制器控制验布机里的电机运转，纺织品在验布机的动力装置带动下经过CIS阵列，电机每前进一步，就进行一次各种光谱照明下的图像数据采集，主控制器对图像数据进行识别，将疵点、异纤、类别、等级等信息位置保存下来，并 通过通信接口发送到监控中心。该发明是布料疵点线下检测系统，需要布料拉卷装置，相比在线检测增加了检测工序，人工成本增加。该发明装置设计没有利用现有织布机，而且采用多光谱检测，对于单一布料检测需求的用户来讲，该发明装置比较复杂，成本较高，算法效率低。 The application number is 201010222997.7, the publication number is CN 101957326A, and the Chinese patent titled "a multi-spectral monitoring method and device for textile surface quality" consists of a CIS array composed of contact image sensors, multi-spectral LED on-off control, It consists of image data acquisition, motor controller, main controller, touch screen or keyboard, display, inkjet indicator, communication interface and printer. The main controller controls the operation of the motor in the cloth inspection machine through the motor controller. The textiles pass through the CIS array driven by the power device of the cloth inspection machine. Every time the motor advances one step, image data collection under various spectral lighting is performed. The main control The sensor recognizes the image data, saves the position of defect, foreign fiber, category, grade and other information, and sends it to the monitoring center through the communication interface. This invention is an offline detection system for fabric defects, which requires a fabric roll-up device. Compared with online detection, the detection process is increased, and the labor cost is increased. The design of the device of the invention does not utilize the existing loom, and adopts multi-spectral detection. For users who need a single cloth detection, the device of the invention is relatively complicated, the cost is high, and the algorithm efficiency is low. the
发明内容Contents of the invention
本发明要解决的技术问题是提供一种实时、有效、可取代人工肉眼验布的布面疵点检测和识别的方法。 The technical problem to be solved by the present invention is to provide a real-time and effective cloth surface defect detection and identification method that can replace the cloth inspection with the naked eye. the
为了解决上述技术问题，本发明的技术方案是提供一种基于大律法的布面疵点检测及分类方法。其特征在于：疵点的检测和分类由下位机和上位机两部分进行：下位机通过图像传感器采集布面图像信息，然后进行布面疵点的检测，上位机根据下位机上传的疵点检测数据，对疵点进行分类；其中，下位机疵点检测步骤如下：(1)对采集的布面图像信息进行滤波、插值、方差采样预处理；(2)进行大律法图像分割及二值化；(3)进行空洞填充、小块处理操作；(4)进行二值图像通道连通处理；(5)提取各个疵点区域信息并保存；上位机对疵点分类的步骤如下：(1)根据疵点长短轴之比将疵点分为区域类和非区域类疵点；(2)根据疵点主轴倾斜角度将疵点分为经类疵点和纬类疵点；(3)根据疵点的整体灰度值将疵点分为暗类疵点和亮类疵点。 In order to solve the above technical problems, the technical solution of the present invention is to provide a fabric defect detection and classification method based on Dalu law. It is characterized in that: the defect detection and classification are carried out by two parts, the lower computer and the upper computer: the lower computer collects the image information of the cloth surface through the image sensor, and then detects the defects of the cloth surface, and the upper computer performs the defect detection data uploaded by the lower computer. The defects are classified; among them, the defect detection steps of the lower computer are as follows: (1) filter, interpolate, and variance sampling preprocessing on the collected cloth image information; (2) perform image segmentation and binarization of Dalufa; (3) Carry out hole filling and small block processing operations; (4) carry out binary image channel connection processing; (5) extract and save the information of each defect area; the steps of the host computer to classify the defects are as follows: (1) divide the Defects are divided into regional and non-regional defects; (2) According to the inclination angle of the main axis of the defects, the defects are divided into warp defects and weft defects; (3) According to the overall gray value of the defects, the defects are divided into dark defects and bright defects. class defects. the
优选地，所述布面疵点检测方法具体步骤如下： Preferably, the specific steps of the cloth surface defect detection method are as follows:
步骤1：读入数字图像原始数据并保存于变量I； Step 1: read in the original digital image data and save it in variable I;
步骤2：对I进行一次均值滤波处理； Step 2: Carry out a mean filtering process on I;
步骤3：对I进行双线性插值处理； Step 3: Perform bilinear interpolation processing on I;
步骤4：对I进行一次方差采样处理； Step 4: Perform a variance sampling process on I;
步骤5：对I进行双线性插值处理； Step 5: Perform bilinear interpolation processing on I;
步骤6：对I进行Otsu大律法图像分割处理并得到图像分割阀值T； Step 6: Carry out the image segmentation processing of Otsu Daitsu to I and obtain the image segmentation threshold T;
步骤7：对I通过Otsu大律法进行二值化处理； Step 7: Binarize I through the Otsu Great Law;
步骤8：对I进行空洞填充处理； Step 8: Carry out hole filling processing on I;
步骤9：对I进行小块处理操作； Step 9: Perform small block processing operations on I;
步骤10：对I进行二值图像八通道连通处理； Step 10: Carry out binary image eight-channel connection processing to I;
步骤11：提取I的图像属性并保存于多结构变量stats； Step 11: Extract the image attribute of I and save it in the multi-structure variable stats;
步骤12：提取变量stats中疵点个数Num1； Step 12: Extract the number of defects Num1 in the variable stats;
优选地，所述布面疵点分类方法具体步骤如下： Preferably, the specific steps of the cloth surface defect classification method are as follows:
步骤a：设定Y3Y2Y1Y0为疵点判定结果的四位数变量，判断Num1是否小于或等于0，若Num1小于或等于0，则Y3＝0判定此织物图像为无瑕疵图像，否则Y3＝1跳转到步骤b； Step a: Set Y3Y2Y1Y0 as the four-digit variable of the defect judgment result, judge whether Num1 is less than or equal to 0, if Num1 is less than or equal to 0, then Y3=0 determines that the fabric image is a flawless image, otherwise Y3=1 jump to step b;
步骤b：将疵点与区域具有相同标准二阶中心矩的椭圆比较，判断区域具有相同标准二阶中心矩的椭圆的长轴长度与区域具有相同标准二阶中心矩的椭圆的短轴长度的比值是否小于疵点长短轴之比且大于疵点长短轴之比的倒数，若是，则Y2＝1判定此图像疵点为区域类疵点并且跳转到步骤d，否则Y2＝0判定此图像疵点为非区域类疵点且跳转到步骤c； Step b: Compare the defect with the ellipse with the same standard second-order central moment in the area, and judge the ratio of the length of the major axis of the ellipse with the same standard second-order central moment in the area to the length of the minor axis of the ellipse with the same standard second-order central moment in the area Whether it is smaller than the ratio of the defect’s long-short axis and greater than the reciprocal of the ratio of the defect’s long-short axis, if so, then Y2=1 determines that the image defect is a regional defect and jumps to step d, otherwise Y2=0 determines that the image defect is a non-regional defect defect and jump to step c;
步骤c：判断疵点主轴与x轴的夹角Q的绝对值是否大于主轴倾斜度An，若Q的绝对值大于主轴倾斜度An，则Y1＝1判定此图像疵点为经类疵点并且跳转到步骤d，否则Y1＝0判定此图像疵点为纬类疵点且跳转到步骤d； Step c: Determine whether the absolute value of the angle Q between the main axis of the defect and the x-axis is greater than the inclination An of the main axis, if the absolute value of Q is greater than the inclination An of the main axis, then Y1=1 determines that the image defect is a meridian defect and jumps to Step d, otherwise Y1=0 determines that the image defect is a weft defect and jumps to step d;
步骤d：判断疵点整体灰度值P是否大于疵点灰度平均值H，若P值大于H值，则Y0＝1判定此图像疵点为亮类疵点并且跳转到步骤e，否则Y0＝1判定此图像疵点为暗类疵点且跳转到步骤e； Step d: Determine whether the overall gray value P of the defect is greater than the average gray value H of the defect. If the P value is greater than the H value, then Y0=1 determines that the image defect is a bright defect and jumps to step e, otherwise Y0=1 determines This image defect is a dark defect and jump to step e;
步骤e：表1为分类结果状态输出表，其中X表示该位未经判断； Step e: Table 1 is the classification result status output table, wherein X indicates that this bit has not been judged;
表1 分类结果状态输出表 Table 1 Classification result status output table
将判定结果Y3Y2Y1Y0对比表1，最终判定并输出七种疵点类型：无疵点、亮经类疵点、暗经类疵点、亮纬类疵点、暗纬类疵点、亮区域类疵点、暗区域类疵点，同时已分类疵点数计数值加1，若已分类疵点数小于等于存在的疵点数则跳转到步骤b继续对下一个疵点进行分类，否则分类结束。 Compare the judgment results Y3Y2Y1Y0 with Table 1, and finally judge and output seven defect types: no defect, bright warp defect, dark warp defect, bright weft defect, dark weft defect, bright area defect, dark area defect, At the same time, the count value of the number of classified defects is increased by 1. If the number of classified defects is less than or equal to the number of existing defects, then jump to step b and continue to classify the next defect, otherwise the classification ends. the
优选地，所述布面疵点检测方法步骤2中，均值滤波处理模板范围为：8*8窗到10*10窗。 Preferably, in step 2 of the cloth surface defect detection method, the range of the mean value filter processing template is: 8*8 windows to 10*10 windows. the
本发明提供的方法克服了现有技术的不足，用机器视觉来取代人眼检测，对上位机和下位机进行合理任务划分，具有快速、正确、高效、实时性好的特点，可极大地降低误检、漏检率，在提高生产率的同时，可以有效降低废次品造成的人力、物力、财力和能源的浪费和损失，同时减小了设备的体积，降低了制造成本。 The method provided by the present invention overcomes the deficiencies of the prior art, uses machine vision to replace human eye detection, and divides tasks reasonably between the upper computer and the lower computer. False detection and missed detection rate, while improving productivity, can effectively reduce the waste and loss of manpower, material resources, financial resources and energy caused by waste and defective products, and at the same time reduce the volume of equipment and reduce manufacturing costs. the
附图说明Description of drawings
图1为本发明提供的基于大律法的布面疵点检测及分类方法总流程图； Fig. 1 is the general flow chart of the cloth surface defect detection and classification method based on the big law provided by the present invention;
图2为下位机示例图； Figure 2 is an example diagram of the lower computer;
图3为布面图像采集及图像预处理和疵点提取流程图； Fig. 3 is a flow chart of cloth surface image acquisition, image preprocessing and defect extraction;
图4为布面疵点分类方法流程图。 Fig. 4 is a flow chart of the classification method for cloth surface defects. the
具体实施方式Detailed ways
为使本发明更明显易懂，兹以一优选实施例，并配合附图作详细说明如下。 In order to make the present invention more comprehensible, a preferred embodiment is described in detail below with accompanying drawings. the
图1为本发明提供的基于大律法的布面疵点检测及分类方法总流程图，所述的基于大律法的布面疵点检测及分类方法将布面图像信息采集和疵点的识别交给下位机完成，疵点的分类及人机交互交给上位机完成，合理的任务划分使得对布面疵点的识别分类更加高效，满足在线实时疵点检测的要求。 Fig. 1 is the general flow chart of the cloth surface defect detection and classification method based on the Dalu law provided by the present invention, and the cloth surface defect detection and classification method based on the Dalu law assigns cloth surface image information collection and defect recognition to The lower computer is completed, and the defect classification and human-computer interaction are handed over to the upper computer. Reasonable task division makes the identification and classification of fabric defects more efficient and meets the requirements of online real-time defect detection. the
结合图2，下位机通过CIS采集布面图像信息，由数字信号处理器(DSP)对图像信息进行布面疵点识别，并将识别结果传输到上位机，由上位机负责疵点的分类及人机交互。其中CIS、DSP均连接现场可编程门阵列(FPGA)模块和AD模块，FPGA模块负责CIS图像采集的时序控制，AD模块则对CIS输出的模拟信号进行模数转换。 Combined with Figure 2, the lower computer collects cloth image information through the CIS, and the digital signal processor (DSP) recognizes the defects of the cloth on the image information, and transmits the recognition results to the upper computer, which is responsible for the classification of defects and man-machine interact. Among them, CIS and DSP are connected with Field Programmable Gate Array (FPGA) module and AD module. The FPGA module is responsible for the timing control of CIS image acquisition, and the AD module performs analog-to-digital conversion on the analog signal output by CIS. the
DSP基于大律法算法对布面图像信息进行疵点识别，并将识别结果传输到上位机，上位机负责对疵点进行分类及分类结果的显示。 The DSP recognizes the defects of the cloth image information based on the big law algorithm, and transmits the recognition results to the host computer, which is responsible for classifying the defects and displaying the classification results. the
图3为本实施例中布面图像采集及图像预处理与疵点提取流程图，具体步骤如下： Fig. 3 is the flowchart of image acquisition, image preprocessing and defect extraction in this embodiment, and the specific steps are as follows:
步骤1：采集布面图像信息； Step 1: Collect image information of cloth surface;
步骤2：读入图像原始数据并保存于变量I； Step 2: Read in the original image data and save it in variable I;
步骤3：对I进行一次均值滤波处理，使图像模糊化；均值滤波处理模板范围为：8*8窗到10*10窗； Step 3: Carry out an average value filtering process on I to blur the image; the range of the average value filter processing template is: 8*8 windows to 10*10 windows;
步骤4：对I进行双线性插值处理，由于模糊化处理图像后，图像大小会发生一定改变，采用双线性插值恢复原图像大小； Step 4: Perform bilinear interpolation processing on I, since the size of the image will change to a certain extent after the image is blurred, use bilinear interpolation to restore the original image size;
步骤5：对I进行一次方差采样处理，增强布料图像疵点信息； Step 5: Perform a variance sampling process on I to enhance the defect information of the cloth image;
步骤6：对I进行双线性插值处理，增强疵点信息后图像大小发生变化，再采用一次双线性插值恢复原始图像大小； Step 6: Perform bilinear interpolation processing on I, the size of the image changes after enhancing the defect information, and then use bilinear interpolation to restore the original image size;
步骤7：对I进行Otsu大律法图像分割处理，通过对比布料图像疵点信息得到图像二值化分割阀值T； Step 7: Carry out the image segmentation processing of Otsu Daitsu to I, and obtain the image binarization segmentation threshold T by comparing the defect information of the cloth image;
步骤8：对I通过Otsu大律法进行二值化处理； Step 8: Binarize I through the Otsu Great Law;
步骤9：对I进行空洞填充处理，消除疵点内部存在的空洞； Step 9: Carry out hole filling processing on I to eliminate the holes existing inside the defect;
步骤10：对I进行小块处理操作，消除小于预定面积的疵点； Step 10: Perform small block processing on I to eliminate defects smaller than the predetermined area;
步骤11：对I进行二值图像八通道连通处理，方便图像疵点的属性提取； Step 11: Perform binary image eight-channel connection processing on I to facilitate attribute extraction of image defects;
步骤12：提取I的图像属性并保存于多结构变量stats； Step 12: Extract the image attribute of I and save it in the multi-structure variable stats;
步骤13：提取变量stats中疵点个数Num1。 Step 13: Extract the number of defects Num1 in the variable stats. the
图4为本实施例中对采集到的布料疵点图像进行布料疵点分类的方法流程图，其具体实施方案步骤如下： Fig. 4 is the flow chart of the method for classifying the cloth defect image collected in this embodiment, and its specific implementation steps are as follows:
步骤a：判断Num1是否小于或等于0，若Num1小于或等于0，则Y3＝0判定此织物图像为无瑕疵图像，否则Y3＝1跳转到步骤b； Step a: Determine whether Num1 is less than or equal to 0, if Num1 is less than or equal to 0, then Y3=0 determines that the fabric image is a flawless image, otherwise Y3=1 and jump to step b;
步骤b：判断疵点与区域具有相同标准二阶中心矩的椭圆的长轴长度与区域具有相同标准二阶中心矩的椭圆的短轴长度的比值是否小于疵点长短轴之比且大于疵点长短轴之比的倒数，若在范围内则Y2＝1判定此图像疵点为区域类疵点并且跳转到步骤d，否则Y2＝0判定此图像疵点为非区域类疵点且跳转到步骤c； Step b: Determine whether the ratio of the length of the major axis of the ellipse with the same standard second-order central moment of the defect to the area and the length of the minor axis of the ellipse with the same standard second-order central moment of the area is less than the ratio of the major and minor axes of the defect and greater than the ratio of the major and minor axes of the defect The reciprocal of the ratio, if it is within the range, Y2=1 determines that the image defect is a regional defect and jumps to step d, otherwise Y2=0 determines that the image defect is a non-regional defect and jumps to step c;
步骤c：判断疵点主轴与x轴的夹角Q的绝对值是否大于主轴倾斜度An，若|Q|值大于主轴倾斜度An，则Y1＝1判定此图像疵点为经类疵点并且跳转到步骤d，否则Y1＝0判定此图像疵点为纬类疵点且跳转到步骤d； Step c: Determine whether the absolute value of the included angle Q between the main axis of the defect and the x-axis is greater than the inclination of the main axis An, if the |Q| Step d, otherwise Y1=0 determines that the image defect is a weft defect and jumps to step d;
步骤d：判断疵点整体灰度值P是否大于疵点灰度平均值H，若P值大于H值，则Y0＝1判定此图像疵点为亮类疵点并且跳转到步骤e，否则Y0＝1判定此图像疵点为暗类疵点、跳转到步骤e。 Step d: Determine whether the overall gray value P of the defect is greater than the average gray value H of the defect. If the P value is greater than the H value, then Y0=1 determines that the image defect is a bright defect and jumps to step e, otherwise Y0=1 determines This image defect is a dark defect, skip to step e. the
步骤e：表1为分类结果状态输出表，其中X表示该位未经判断； Step e: Table 1 is the classification result status output table, wherein X indicates that this bit has not been judged;
表1 分类结果状态输出表 Table 1 Classification result status output table
将判定结果Y3Y2Y1Y0对比表1，最终判定并输出七种疵点类型：无疵点、亮经类疵点、暗经类疵点、亮纬类疵点、暗纬类疵点、亮区域类疵点、暗区域类疵点，同时已分类疵点数计数值加1，若已分类疵点数小于等于存在的疵点数则跳转到步骤(b)继续对下一个疵点进行分类，否则分类结束。 Compare the judgment results Y3Y2Y1Y0 with Table 1, and finally judge and output seven defect types: no defect, bright warp defect, dark warp defect, bright weft defect, dark weft defect, bright area defect, dark area defect, At the same time, the count value of the number of classified defects is increased by 1. If the number of classified defects is less than or equal to the number of existing defects, jump to step (b) and continue to classify the next defect, otherwise the classification ends. the
本发明提供的方法克服了现有自动验布机疵点检测识别系统中CCD线阵相机体积大、成本高，系统装置复杂，算法效率低的不足，用机器视觉来取代人眼检测，具有快速、正确、高效的特点，可极大地降低误检、漏检率，在提高生产率的同时，可以有效降低废次品造成的人力、物力、财力和能源的浪费和损失；采用CIS图像传感器采集图像数据，消除CCD线阵相机对光源的要求，同时减小了设备的体积和降低了制造成本，有利于本发明产品的推广应用；本发明是将检测探头安装在织布机上，充分利用了织布机的电机拖动装置，实现织检一体，一方面省去验布装置成本，另一方面节省了验布时间，减少了织布厂的人工成本，具有广阔的应用前景和巨大的市场经济效益。 The method provided by the invention overcomes the shortcomings of the existing automatic cloth inspection machine defect detection and identification system, such as large volume, high cost, complex system devices, and low algorithm efficiency of the CCD line array camera, and uses machine vision to replace human eye detection, which has fast, The characteristics of correctness and high efficiency can greatly reduce the rate of false detection and missed detection. While improving productivity, it can effectively reduce the waste and loss of manpower, material resources, financial resources and energy caused by waste and defective products; CIS image sensor is used to collect image data , eliminate the requirements of the CCD line array camera on the light source, reduce the volume of the equipment and reduce the manufacturing cost, which is conducive to the popularization and application of the product of the present invention; the present invention installs the detection probe on the loom, making full use of the The motor drive device of the machine realizes the integration of weaving and inspection. On the one hand, it saves the cost of the cloth inspection device, on the other hand, it saves the time of cloth inspection and reduces the labor cost of the weaving factory. It has broad application prospects and huge market economic benefits. . the
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410200849.3ACN104048966B (en) | 2014-03-14 | 2014-05-13 | The detection of a kind of fabric defect based on big law and sorting technique |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2014100958148 | 2014-03-14 | ||
| CN201410095814.8 | 2014-03-14 | ||
| CN201410095814 | 2014-03-14 | ||
| CN201410200849.3ACN104048966B (en) | 2014-03-14 | 2014-05-13 | The detection of a kind of fabric defect based on big law and sorting technique |
| Publication Number | Publication Date |
|---|---|
| CN104048966Atrue CN104048966A (en) | 2014-09-17 |
| CN104048966B CN104048966B (en) | 2016-08-03 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410200849.3AExpired - Fee RelatedCN104048966B (en) | 2014-03-14 | 2014-05-13 | The detection of a kind of fabric defect based on big law and sorting technique |
| Country | Link |
|---|---|
| CN (1) | CN104048966B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105699386A (en)* | 2016-02-29 | 2016-06-22 | 东华大学 | Automatic cloth inspection marking method adopting contact image sensor |
| CN105738376A (en)* | 2016-02-29 | 2016-07-06 | 东华大学 | Automatic cloth inspection machine using contact image sensor |
| CN105784712A (en)* | 2016-02-29 | 2016-07-20 | 东华大学 | Automatic cloth inspecting method adopting contact type image sensor |
| CN107169961A (en)* | 2017-05-15 | 2017-09-15 | 中烟追溯（北京）科技有限公司 | A kind of cigarette sorting detecting system and method based on CIS IMAQs |
| CN107220649A (en)* | 2017-05-27 | 2017-09-29 | 江苏理工学院 | A kind of plain color cloth defects detection and sorting technique |
| CN107248154A (en)* | 2017-05-27 | 2017-10-13 | 江苏理工学院 | A kind of cloth aberration real-time on-line detecting method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5815198A (en)* | 1996-05-31 | 1998-09-29 | Vachtsevanos; George J. | Method and apparatus for analyzing an image to detect and identify defects |
| CN101866427A (en)* | 2010-07-06 | 2010-10-20 | 西安电子科技大学 | Fabric defect detection and classification method |
| CN102175692A (en)* | 2011-03-17 | 2011-09-07 | 嘉兴学院 | System and method for detecting defects of fabric gray cloth quickly |
| CN102879401A (en)* | 2012-09-07 | 2013-01-16 | 西安工程大学 | Method for automatically detecting and classifying textile flaws based on pattern recognition and image processing |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5815198A (en)* | 1996-05-31 | 1998-09-29 | Vachtsevanos; George J. | Method and apparatus for analyzing an image to detect and identify defects |
| CN101866427A (en)* | 2010-07-06 | 2010-10-20 | 西安电子科技大学 | Fabric defect detection and classification method |
| CN102175692A (en)* | 2011-03-17 | 2011-09-07 | 嘉兴学院 | System and method for detecting defects of fabric gray cloth quickly |
| CN102879401A (en)* | 2012-09-07 | 2013-01-16 | 西安工程大学 | Method for automatically detecting and classifying textile flaws based on pattern recognition and image processing |
| Title |
|---|
| 姚芳: "织物纹理的表征和自动识别的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 8, 15 August 2010 (2010-08-15), pages 40 - 48* |
| 李建福: "织物疵点在线检测理论与应用的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 3, 15 March 2009 (2009-03-15)* |
| 王永灿: "工业流水线织物疵点检测及分类算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 6, 15 June 2009 (2009-06-15)* |
| 石美红等: "基于Blob 算法的织物疵点检测算法的研究", 《现代电子技术》, no. 24, 31 December 2010 (2010-12-31)* |
| 郭攀峰: "基于下采样的改进的织物图像预处理方法", 《工业控制计算机》, vol. 24, no. 1, 31 January 2011 (2011-01-31)* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105699386A (en)* | 2016-02-29 | 2016-06-22 | 东华大学 | Automatic cloth inspection marking method adopting contact image sensor |
| CN105738376A (en)* | 2016-02-29 | 2016-07-06 | 东华大学 | Automatic cloth inspection machine using contact image sensor |
| CN105784712A (en)* | 2016-02-29 | 2016-07-20 | 东华大学 | Automatic cloth inspecting method adopting contact type image sensor |
| CN105784712B (en)* | 2016-02-29 | 2018-07-17 | 东华大学 | A kind of automatic cloth inspection method using contact-type image sensor |
| CN105738376B (en)* | 2016-02-29 | 2018-07-17 | 东华大学 | A kind of automatic cloth inspecting machine using contact-type image sensor |
| CN107169961A (en)* | 2017-05-15 | 2017-09-15 | 中烟追溯（北京）科技有限公司 | A kind of cigarette sorting detecting system and method based on CIS IMAQs |
| CN107220649A (en)* | 2017-05-27 | 2017-09-29 | 江苏理工学院 | A kind of plain color cloth defects detection and sorting technique |
| CN107248154A (en)* | 2017-05-27 | 2017-10-13 | 江苏理工学院 | A kind of cloth aberration real-time on-line detecting method |
| Publication number | Publication date |
|---|---|
| CN104048966B (en) | 2016-08-03 |
| Publication | Publication Date | Title |
|---|---|---|
| CN104048966B (en) | The detection of a kind of fabric defect based on big law and sorting technique | |
| CN107643295B (en) | A method and system for online detection of cloth defects based on machine vision | |
| CN106841209B (en) | One kind being based on big data self study chemical fiber wire ingot intelligence appearance detection system and method | |
| CN110308151A (en) | A method and device for identifying surface defects of metal workpieces based on machine vision | |
| CN107886500A (en) | A kind of production monitoring method and system based on machine vision and machine learning | |
| CN107486415A (en) | Thin bamboo strip defect on-line detecting system and detection method based on machine vision | |
| CN102592286A (en) | Automatic identification method of color fabric color mold pattern image based on image processing | |
| CN104360501A (en) | Visual detection method and device for defects of liquid crystal display screen | |
| CN104574389A (en) | Battery piece chromatism selection control method based on color machine vision | |
| CN101555661A (en) | Measuring method and system for weight of cotton foreign fiber based on machine vision technique | |
| CN102175692A (en) | System and method for detecting defects of fabric gray cloth quickly | |
| CN102980892B (en) | On-line examination system and method for steel pipe | |
| CN108802052A (en) | A kind of detecting system and its detection method about slide fastener defect | |
| CN104535589A (en) | Online detection method and device for low-voltage current mutual inductor | |
| CN107749057A (en) | A kind of method of solar battery sheet outward appearance spillage defects detection | |
| CN107256549B (en) | A method of bamboo strip defect detection based on machine vision | |
| CN102507008B (en) | Multi-template automatic optical color detection method | |
| CN106780464A (en) | A kind of fabric defect detection method based on improvement Threshold segmentation | |
| CN113820322B (en) | Detection device and method for appearance quality of seeds | |
| CN104297439A (en) | Computer vision-based automatic identification detection method for double-yolked duck eggs | |
| CN108362708A (en) | Pen core scribing line standardized detection system based on machine vision and method | |
| CN106093051A (en) | Paper roll tangent plane burr detection method based on machine vision and device | |
| CN103926255A (en) | Method for detecting surface defects of cloth based on wavelet neural network | |
| CN203711336U (en) | Sorting device | |
| CN112304876A (en) | Cotton color degree detection device and detection method based on machine vision |
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee | Granted publication date:20160803 Termination date:20190513 |