基于机器视觉的印刷标签检测系统的改进

针对图像采集的非同步性和传送平台存在的抖动等因素造成采集图像质量降低的问题,提出了一种简单可行的高可靠性机器视觉印刷标签检测系统。通过比较几种边缘检测算法,采用Canny这种相对高可靠性的边缘检测算法制作边缘掩膜,通过在模板图像上加盖边缘掩膜,在差影比较后对差影图像进行形态学去噪来消除轮廓伪影和人眼难以识别的微小缺陷。该方法运用在印刷标签质量检测系统中,有效地降低了印刷标签误检率,并且符合人眼识别特性。     

Process color watermarking: the use of visual masking and dot gain correction

Subtractive-Color-Reproduction is the fundamental mechanism in 4-color process printing, which employs CMYK (cyan, magenta, yellow and black) dots for color reproduction. Watermarks embedded in RGB images are distorted when the image is transformed from what to process color for printing usage. Based on the feature of CMYK color space and human visual system, this paper presents a robust CMYK watermarking algorithm for printing images, which embeds an informative watermark and a structure template signal in DCT domain of the Yellow information channel and DFT domain of the black information channel respectively. The embedded template enables robustness against geometric distortions in the print-scanning process. Another spatial visual mask is proposed to reshape the embedded energy after it is inverted to the spatial domain, which significantly improves the color fidelity and the image quality. In addition, a dot gain compensation model is put forward for color correction. A series of proof tests have been carried out on an offset press, which show that the watermark preserves the fidelity of the image and can be extracted with a high quality; moreover, a remarkable fidelity of color and tones on printed copies is well preserved.     

自动孔板印刷机中图像检测系统的设计与实现

设计与实现了自动孔板印刷机中印刷图像质量检测系统。该系统主要包括孔板印刷中多路印刷图像的获取、特征的提取和匹配以及质量监控处理等功能模块。在RG Chromaticity空间中采用模板匹配的方法设计了相关的算法,实现了对印刷图像的错位、漏印、渗溢以及颜色不均等质量因素的快速实时检测,并成功地应用于自动丝网印刷设备中。经过测试和运行取得了良好的质量监控效果和经济效益。     

Deep learning-based optimal segmentation of 3D printed product for surface quality improvement and support structure reduction

Large-sized product cannot be printed as one piece by a 3D printer because of the volume limitation of most 3D printers. Some products with the complex structure and high surface quality should also not be printed into one piece to meet requirement of the printing quality. For increasing the surface quality and reducing support structure of 3D printed models, this paper proposes a 3D model segmentation method based on deep learning. Sub-graphs are generated by pre-segmenting 3D triangular mesh models to extract printing features. A data structure is proposed to design training data sets based on the sub-graphs with printing features of the original 3D model including surface quality, support structure and normal curvature. After training a Stacked Auto-encoder using the training set, a 3D model is pre-segmented to build an application set by the sub-graph data structure. The application set is applied by the trained deep-learning system to generate hidden features. An Affinity Propagation clustering method is introduced in combining hidden features and geometric information of the application set to segment a product model into several parts. In the case study, samples of 3D models are segmented by the proposed method, and then printed using a 3D printer for validating the performance.     

Preview control of web position in roll-to-roll printing using alignment patterns

In printed electronics systems, the roll-to-roll process is a next-generation process technology that may be applied to the production of various electronic products such as flexible electronic devices, solar cells, and display panels. The process allows for production at low unit cost and high speed. As a factor for improving the print quality, there are ink viscosity, printing speed, printing pressure, and alignment precision. Among these factors, alignment precision, a critical factor in the performance of an electronic circuit, is the accurate alignment of layers in a multi-layered structure. In our previous study, we suggested a measurement method for improving the alignment precision by using an optical pre-measuring device and an alignment pattern that is printed by roll-to-roll printing process. In the measuring system, when a web is transported in the web feed direction, the optical pre-measuring device measures the quantity of light reflected by the alignment pattern. A lateral position can be aligned accurately based on the measured signal. However, because the position is measured in advance in the web feed direction, it is not applicable to a real-time printing system, and when the lateral position is measured using the optical pre-measuring device, a lateral position error, that is generated by disturbances induced by tension acting on the web and the vibration of the motor, can be estimated. A motor driven linear translation stage for lateral direction is used to minimize the lateral position errors in roll-to-roll systems. In this paper, we propose a preview control system to reflect the measured lateral position error to the printing result at the appropriate time for improving the alignment precision in roll-to-roll printing. As a control method, using roll-to-roll model, the system of disturbance, lateral angle and lateral position can be expressed as a state-space equation. Based on this equation, a preview controller to find the optimal value of state and control error is designed in a similar method to the conventional preview control method used for the active suspension system of a vehicle. Then, the alignment results are verified using another optical measuring device of a web transport system.

     

基于DSP的印刷品质量在线检测系统设计

研究一种基于高速处理芯片TMS320C5416的印刷品质量在线检测系统.对印刷品质量在线检测系统的图像采集部分做了分析,确定了以线阵CCD为中心的印刷品图像采集系统;分析TMS320C5416的特点,介绍了各硬件设备的原理及在系统中的应用.对图像定位、图像匹配算法进行比较,实现了缺陷产品的自动检测技术.     

基于YOLOv5和生成对抗网络的塑料标签缺陷检测

塑料标签物的缺陷检测与识别是工业过程控制和质量控制的关键;为了克服现有塑料标签缺陷检测方法的局限性,使用了单阶段目标检测模型YOLOv5对其瑕疵进行实时检测与分类;此外,为解决由于样本缺陷数量不足造成的模型识别准确率低等问题,采用了一种基于Defect-GAN的生成对抗网络对小样本进行数据增强和扩增;该方法通过模拟缺陷生成和缺陷图像重建的过程,可以高效合成大量具有高保真度和多样性的缺陷样本,尤其适用于形状不规则、分布随机且尺寸不同的瑕疵生成;实验结果表明,通过使用扩增数据集训练目标检测器,并对网络的超参数进行优化,可以显著提高目标检测器的准确率和精度,其平均精度mAP可达99.5%;此外,为了模拟该方法在实际生产中的应用场景,设计并定制了一台半自动的图像采集机械平台用于采集圆柱样品表面的印刷标签,以及一个自主开发的图像处理和统计分析软件用于样本采集、图像处理及统计分析;该方法和平台可以很容易地推广并应用到其他工业质量控制和缺陷检测系统中。     

基于小波系数比较的抗打印扫描数字水印算法

提出了一种在打印监控系统中应用的图像数字水印算法。将待打印的文档图像进行相同大小的分块,依据分块数量确定嵌入容量。对每一小块图像进行三级离散小波变换(DWT),通过调整子带中不同位置系数的大小关系在中频和高频分量中重复嵌入水印信息;依据择大判决原则从各子带提取的水印中确定最终水印,提取过程不需要原始图像,扫描分辨率在75dpi以上均可正确提取。实验表明,该算法能够有效抵抗几何攻击,对打印、复印和扫描具有鲁棒性,可实现对打印文件进行版权保护和来源鉴别。     

骨组织模型3D打印建模中CT图像序列选择的策略分析

骨组织模型的精确3D打印对骨科手术前进行手术方案规划、术中定位和术后评估手术效果均有不可替代的作用.然而,打印等比例的、与患者完全匹配的、精确的骨组织模型涉及术前CT影像扫描、3D建模、3D打印及打印后处理一系列流程.其中,对于3D打印前的重建,CT扫描图像序列的选择直接影响打印模型的质量.通过比较骨窗序列和标准序列C...     

Psychovisual Evaluation of Lossy CMYK Image Compression for Printing Applications

In the digital prepress workflow, images are represented in the CMYK colour space. Lossy image compression alleviates the need for high storage and bandwidth capacities, resulting from the high spatial and tonal resolution. After the image has been printed on paper, the introduced visual quality loss should not be noticeable to a human observer. Since visual image quality depends on the compression algorithm both quantitatively and qualitatively, and since no visual image quality models incorporating the end-to-end image reproduction process are satisfactory, an experimental comparison is the only viable way to quantify subjective image quality. This paper presents the results from an intensive psychovisual study based on a two-alternative forced-choice approach involving 164 people, with expert and non-expert observers distinguished. The primary goal is to evaluate two previously published adaptations of JPEG to CMYK images, and to determine a visually lossless compression ratio threshold for typical printing applications. The improvements are based on tonal decorrelation and overlapping block transforms. Results on three typical prepress test images indicate that the proposed adaptations are useful and that for the investigated printing configuration, compression ratios up to 20 can be used safely.     

基于图像模式识别与PLC控制的印刷套准自适应系统*

针对印刷过程的不稳定性,通过对套印标记的图像采集和处理,适时调整纸张张力大小,构建了一种印刷套准无模型自适应系统。该系统通过图像、PLC和组态软件的联合控制,实现了套准标记的自动检测、图像识别和自动校准。利用该系统中的PLC和组态软件还可实现印刷过程的实时监控,便于观察印刷过程的动态特性。该系统用于各类多色印刷设备上,可提高印刷设备的自动化程度。     

基于机器视觉的印刷质量检测研究（英文）

为了实现对物品印刷质量的高效检测,通过机器视觉技术,采用图像处理的方法来实现印刷品质量的检测。根据印刷检测的功能需求,设计了三个具体的功能模块,并在VS2008和OpenCV1.0平台下实现了印刷质量检测的功能,支持印刷品图像的输入、人机交互处理、划痕和偏色检测、结果输出等功能。与传统采用的手工方法相比,提高了印刷质量检测的效率,检测速度快,准确率高,检测结果客观,能够迅速而精确地检测出目标印刷品的外观缺陷程度。     

基于FPGA+GPU的图像采集处理系统设计

随着嵌入式图像处理系统的快速发展,对于前端图像采集模块的需求越来越高。图像采集的速度、分辨率、可靠性以及集成度对后续设计的准确度由极大的影响。通过对数字图像采集系统进行研究,设计出了基于FPGA和GPU架构的图像采集处理系统,重点研究了图像采集处理系统的硬件设计过程和软件设计过程。在基于FPGA+GPU的图像采集处理系统中,让具有强大运算处理能力的GPU专注于数据存储、用户交互以及后续的图像处理。系统中,FPGA则负责图像的采集、外设控制、任务调度。GPU与FPGA之间通过高速PCIE总线进行通信,分别设计编写基于Linux系统的驱动程序和FPGA端PCIE程序。实验结果表明,所设计基于FPGA+GPU的图像采集处理系统可实现437.5Mbps的实时图像采集存储速度,传输过程实时稳定,数据传输完整。     

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。     

Digital lithographic processing for large‐area electronics

Abstract— A non‐contact jet‐printed mask‐patterning process is described. By combining digital imaging with jet printing, digital lithography was used to pattern a‐Si:H‐based electronics on glass and plastic substrates in place of conventional photolithography. This digital lithographic process is capable of layer‐to‐layer registration of ±5 μm using electronic mask files that are directly jet printed onto a surface. Aminimum feature size of 50 μm was used to create 180 × 180 element backplanes having 75‐dpi resolution for display and image‐sensor applications. By using a secondary mask process, the minimum feature size can be reduced down to ～15 μm for fabrication of short‐channel thin‐film transistors. The same process was also used to pattern black‐matrix wells in fabricating color‐filter top plates in LCD panels.     

Estimating the amount of cyan,magenta, yellow,and black inks in arbitrary colour pictures

This paper is concerned with the offset lithographic colour printing. To obtain high quality colour prints, given proportions of cyan (C), magenta (M), yellow (Y), and black (K) inks (four primary inks used in the printing process) should be accurately maintained in any area of the printed picture. To accomplish the task, the press operator needs to measure the printed result for assessing the proportions and use the measurement results to reduce the colour deviations. Specially designed colour bars are usually printed to enable the measurements. This paper presents an approach to estimate the proportions directly in colour pictures without using any dedicated areas. The proportions—the average amount of C, M, Y, and K inks in the area of interest—are estimated from the CCD colour camera RGB (L*a*b*) values recorded from that area. The local kernel ridge regression and the support vector regression are combined for obtaining the desired mapping L*a*b* ⇒ CMYK, which can be multi-valued.     

低成本便携式多光谱成像系统的研发及优化

目的 针对现有多光谱成像系统存在成本高、结构复杂、操作难度大和响应速度慢等问题。因此，本文提出了一种基于脉冲调制的低成本便携式多光谱成像系统，并采用客观图像质量评估（image quality assessment，IQA）的方法对其系统参数进行优化。方法 该系统主要由光源模块、控制模块、图像采集模块和图像分析模块4部分组成。光源模块采用9个波长的LED （light emitting diode）阵列，其中心波长为365 nm、390 nm、460 nm、515 nm、585 nm、620 nm、650 nm、730 nm和840 nm；控制模块主要包括LED驱动电路和USB （universal serial bus）电源，可以通过发送一定时间间隔的脉冲波来分时点亮LED，并通过一定阻抗匹配使LED发光强度达到最大值；图像采集模块主要使用去除红外截止滤波片的高清红外工业相机，该相机的最佳光谱感应范围包含所选的9个LED灯珠的中心波长；图像分析模块主要执行客观图像质量评估算法。系统执行时，STC89C51单片机发射周期为T的脉冲波来驱动9种不同波长的LED分时点亮。然后，计算机平台调用高清红外相机模组，以相匹配的间隔捕获多光谱图像。在系统拍摄参数优化实验中，本文采用模糊度和清晰度评价指标对所获得的多光谱图像从相机拍摄时间间隔、相机拍摄距离和光照强度3个角度进行质量评估，进而获得较优的系统成像参数。结果 通过改变系统拍摄参数，对3个场景下的不同拍摄条件所获取的多光谱图像质量进行评估，结果显示：对于本文所搭建的多光谱成像系统，相机拍摄时间间隔与LED灯珠频闪周期同步，拍摄距离为25 mm，光照强度为45 Lux下成像质量相对较好。结论 本文设计并搭建的基于脉冲调制的低成本便携式多光谱成像系统成本低、操作难度小、结构简单、成像质量较好、成像速度较快，可以满足多光谱成像系统大规模推广使用的要求。此外，本文的系统设计方法、设计思路和实验方案等可以为后续研究提供借鉴。     

基于多色集合理论的印刷工艺规划建模研究

在研究印刷工艺的加工需求和加工工序的基础上,将计算机辅助工艺设计引入印刷制造系统,并应用多色集合理论建立印刷工艺规划的数学模型,提出了一种基于此模型的印刷工艺路径求解算法。此数学模型包括围道矩阵和多色图两部分。围道矩阵描述了印刷工序实现印刷加工需求的关系,多色图描述了印刷工序的相互作用。该算法便于计算机编程的实现,能根据印刷加工需求搜索所需的工艺算子,然后求解出印刷品的印刷工艺路线。最后,通过简例分析,对所提出的模型与算法进行了验证。