机械视觉疲劳裂纹扩展试件自动定位安装方法

针对现有疲劳裂纹扩展试验中试件的安装定位仍采用操作繁琐、效率低下的手动安装问题,本文提出了一种基于机械视觉的疲劳裂纹扩展试件安装定位方法,首先,对试件及夹具图像进行采集、处理和分析,实现夹具、试件圆心坐标及半径的自动快速识别、孔差距离的精确测量;然后利用模糊PID算法来控制直流伺服电机运动定位,从而实现试件的正确安装定位;最后采用工业数字显微镜对不同时间点的试件位置偏差进行停机测量对比,实验结果表明：所提出方法能够对试件进行精确安装定位,最大偏差为0.122 mm,具有重要的理论和应用价值,并为其它类似基于机器视觉的检测定位方法提供了有价值的参考。     

基于DSP的驾驶员疲劳检测系统

为了能够有效地、实时地对驾驶员进行疲劳检测,构建了以ICETEK-DM6437-B模块为核心、以近红外发光二极管为光源和以电荷耦合器摄像头为图像采集设备的驾驶员疲劳检测系统.提出了以人脸区域定位为检测主体的、在PER-CLOS方法原理基础上改进的PER-NOFACE方法结合多种简单高效的图像处理算法的疲劳检测方案,可有效地检测出驾驶员的疲劳状态.为了保证系统检测的实时性,在DM6437达芬奇处理器上对疲劳检测算法进行了代码优化.实验结果表明,该系统能够较为准确地、实时地对驾驶员进行疲劳状态检测.     

飞机强度试验中结构健康监测采集控制系统

针对结构健康监测在全机疲劳试验中需要采用多种监测系统的特点,结合飞机强度疲劳试验对数据采集的要求,设计了一种将光栅光纤传感器、压电传感器和声发射传感器等集成为一体的飞机结构健康监测系统.设计了多类型传感器采集控制系统,其中采用了分布式、多任务的结构设计,解决了多台设备采集数据的同步性、有效性和一致性.针对多类传感器的数据源多、数据类型多的特点,设计了通用数据结构解决了多类传感器的数据合并处理问题.通过在某全机疲劳试验中的运用,验证了本系统是一种稳定、高效、易扩展的多传感器健康监测集成系统,能够获取更加可靠、全面的健康监测数据,为健康监测在全机疲劳试验中损伤的准确识别,提供了强有力的支撑.     

基于机器视觉的成熟番茄自动识别技术

为实现番茄收获机械自动化,提出了利用计算机视觉代替人眼对成熟番茄进行自动识别的方法。首先对番茄图像进行了各种去除噪声方法的比较,提出用HSI颜色系统的H色调分量作为识别的颜色特征参数。利用最大方差自动取阈法及数学形态运算等构成的识别算法对采集的番茄图像进行自动识别,采用Matlab软件平台来构建算法。为验证自动识别方法的可行性,用CCD摄像头组成的图像采集系统对不同光线条件下的番茄图像进行了识别。试验证明,该算法及程序设计能够较快的识别出自然光下的成熟番茄,满足自动收获的需要。     

基于TMS230DM643的驾驶疲劳实时视觉检测系统

针对驾驶疲劳检测系统小型化、廉价和低功耗的实际需求,提出一个基于TMS230DM643的驾驶疲劳实时视觉检测系统。系统通过CCD摄像头获取驾驶员的人脸图像,采用TMS230DM643作为核心处理器,对获取的图像进行人脸检测、眼睛定位等处理,并提取眼睛的闭合度作为PERCLOS疲劳检测的依据。实验表明,系统能够准确实时地检测驾驶疲劳,具有较好的实用性。     

基于ARM-linux的显微镜自动对焦系统研究

本文介绍了一种在ARM-linux平台下实现显微镜自动对焦的方法.给出了以s3c2410处理器为核心,采用通用型USB摄像头进行采集图像、步进电机模块作为传动装置的系统软硬件方案及其实现步骤,并对如何在Linux平台下编写CMOS图像采集、图像清晰度评价函数等核心程序进行了讨论.     

显微视觉系统对柱状微零件自动聚焦技术研究

针对显微视觉中的柱状物体图像的清晰度问题,提出了一种实时检测图像特征并跟踪特征区域进行自动聚焦的方法。该聚焦算法包括聚焦评价函数、聚焦搜索算法和聚焦区域选择。该聚焦搜索算法实现了显微视觉系统下快速准确的聚焦,克服了爬山搜索算法的缺点,有效避免搜索结果陷入局部极大值。显微视觉中对聚焦区域的选择尤为重要,以图像特征区域作为聚焦区域,实时检测该特征区域进行聚焦,实现了在景深小于柱状物体半径的情形下对柱状物体边缘的精确聚焦。粗聚焦后电机位置的标准差为205μm,精聚焦后电机位置的标准差为37μm。实验结果表明,该自动聚焦方法能够满足微装配系统对显微视觉的聚焦需求。     

基于DSP的疲劳驾驶检测系统的研究

针对汽车驾驶员疲劳驾驶检测的要求,设计了一种基于图像处理DSP芯片DM6437的疲劳驾驶视觉检测系统。通过DSP的GPIO口用软件控制摄像头轴上和轴外的两种不同波长的近红外光源(850nm/950nn)交替采集驾驶员图像,根据亮瞳效应两帧图像差分后粗定位人眼,用模板检测提取人眼的边界,根据PERCLOS值方法判断驾驶员是否疲劳。根据人眼的不同状态分别处理,用蜂鸣器作为报警系统提醒驾驶员。实验表明,该系统简单实用,而且能够全天候准确快速地判断驾驶员是否疲劳。     

基于ARM的人脸识别门禁系统设计

针对现有门禁系统因采用IC卡、射频卡等考勤方法存在"认卡不认人",采用指纹虹膜等生物识别技术存在指纹容易损伤等问题,设计了一套基于ARM的人脸识别门禁系统。该系统采用S3C2440B芯片为处理器、USB摄像头作为视频采集设备,自动搜索人脸进行拍照;采用基于PCA的人脸识别算法对采集的人脸图像进行识别,若为非法入侵人员则发出报警声,并通过GPRS技术发短信通告系统管理员采取安全防范措施。测试结果表明,该系统在严格的测试环境下能够准确高效地识别出人脸,平均识别率达95%以上。     

计算机视觉在人脸识别领域中的应用研究

随着人工智能的发展,需要自动人脸识别的场景会越来越多,采用计算机视觉快速准确地检测并识别出目标人脸成为必然趋势。使用摄像头采集到的人脸图像,通过灰度化的方法进行预处理,经过预处理后的图像更容易进行特征提取。该方法能够快速检测到人脸,有效提取人脸面目特征,并且准确识别出相应的人脸,主要应用在静态图像的人脸识别中,基本上满足了人脸检测和识别的实时性要求,应用前景广阔。     

基于数字图像处理的表面裂纹检测算法

本文提出了一种基于数字图像处理技术的表面裂纹检测算法。运用这种算法能精确的检测裂纹的位置、长度等特征，将这种裂纹检测算法运用到裂纹自动检测系统以及裂纹扩展行为监测中不仅大大降低了劳动时间和劳动强度，而且提高了测量的准确性。     

Pavement crack detection using hessian structure propagation

Pavement images are widely used in transportation agencies to detect cracks accurately so that the best proper plans of maintenance and rehabilitation could be made. Although crack in a pavement image is perceived because the intensity of crack pixels contrasts with that of the pavement background, there are still challenges in distinguishing cracks from complex textures, heavy noise, and interference. Unlike the intensity or the first-order edge feature of crack, this paper proposes the second-order directional derivative to characterize the directional valley-like structure of crack. The multi-scale Hessian structure is first proposed to analytically adapt to the direction and valley of cracking in the Gaussian scale space. The crack structure field is then proposed to mimic the curvilinear propagation of crack in the local area, which is iteratively applied at every point of the crack curve to infer the crack structure at the gaps and intersections. Finally, the most salient centerline of the crack within its curvilinear buffer is exactly located with non-maximum suppression along the perpendicular direction of crack. The experiments on large numbers of images of various crack types and with diverse conditions of noise, illumination and interference demonstrate the proposed method can detect pavement cracks well with an average Precision, Recall and F-measure of 92.4%, 88.4%, and 90.4% respectively. Also, the proposed method achieves the best performance of crack detection on the benchmark datasets among methods that also require no training and publicly offer the detection results for every image.     

机器视觉系统的自动对焦技术研究

为了解决在工业生产现场复杂环境下机器视觉系统中定焦相机的失焦问题,提出了一种机器视觉定焦相机的自动对焦技术,该技术通过上位机实时对相机采集到的图像进行感兴趣区域(Region of Interest,ROI)裁剪、离散傅里叶变换(Distributed Fourier Transformation,DFT)滤波等预处理过程,再由改进的图像清晰度评价函数对图像清晰度进行评价,最后由运动机构带动相机自动搜寻焦点位置并完成对焦功能。实验结果表明,经过傅里叶变换滤波后,提升了评价函数的灵敏度。通过与几种典型的评价函数对比,提出的方案能有效地排除环境干扰,自动对焦的结果与标准样张最为接近,且在不同物距情况下仍能够高效运行。     

显微镜下两层物体的快速自动对焦算法

针对显微镜下两层物体的大规模图像采集问题，提出了基于图像边缘和邻域预测快速自动对焦算法。通过提取成像主体的边缘区域作为对焦窗口，结合线性预测方法，实现了既保持上层成像目标清晰，又有效抑制下层背景干扰的自动对焦系统。实验表明该算法在对焦速度、对焦精度以及抵抗背景干扰方面有突出优点。     

Improved digital photogrammetry technique for crack monitoring

Inspections to evaluate the safety, durability, and service life of aging infrastructure play an important role in determining the countermeasures that need to be taken, such as reinforcement, repair, and reconstruction. In infrastructure containing concrete, such as bridges and tunnels, crack widths and patterns on surfaces are two of the most important signs used to estimate durability. Current conventional techniques used for this purpose suffer from challenges such as tediousness, subjectivity, and high cost. Consequently, a new measurement technique that overcomes these challenges while measuring crack displacement with high accuracy and precision in aging civil engineering structures is needed. In this paper, we proposed a technique for measuring crack displacement using a digital camera image. In the proposed technique, reflective targets are established around both sides of a crack as gauges, and subsequent digital camera images of the targets are subjected to image processing to determine the displacements of the targets. These displacements can be measured using images captured from any arbitrary camera position. The results of experiments conducted to verify the efficacy of the proposed method show that crack displacements of less than 0.10 mm can be measured with high accuracy and precision using digital images captured at a distance of 10.0 m from the target, while less than 0.20 mm changes in the tensile displacement of the crack can be measured from an image captured at 25.0 m from the crack. Measurement results obtained from a tunnel are also presented to show that cracks in the walls of an actual tunnel can be identified through simple measurements. These measurements, taken over a period of one year, indicate that the tendency of crack displacement and slide movements are in close agreement.     

基于图像清晰度的自动聚焦算法

在数字监控和机器视觉中，摄像头的焦距可以由计算机根据图像的清晰度自动控制。本文针对监控系统在目标跟踪时对自动聚焦的速度和精度的要求，提出了两次定位、单向搜索的快速精确的聚焦算法。在聚焦精度方面首先针对全局图像进行第一次粗略的聚焦，图像的清晰度判断采用计算速度较快的相邻象素差分法的判断函数；然后再对图像中的目标区域进行第二次精确的聚焦，图像清晰度判断采用对清晰度变化敏感的基于Laplacian边缘检测的判断函数。调焦过程采用最佳焦距位置的单向搜索法，以提高聚焦的速度。实验证明该算法具有较好的实时性和精确性。     

Simulation of three-dimensional fatigue crack propagation using enriched finite elements

A three-dimensional methodology for simulation of fatigue crack propagation is presented. The method is leveraged by the use of enriched crack tip elements to compute the mixed-mode stress intensity factors. The crack growth model used and the crack propagation life calculation are also described. As examples, fatigue crack propagation of a mode-I surface crack and crack advancements of mixed-mode surface cracks are simulated. The predicted results showed excellent agreement with experimental data from the literature. Thus, it is concluded that the crack propagation method developed allows efficient and accurate simulation of three-dimensional fatigue crack propagation problems.     

Study of TBM cutterhead fatigue crack propagation life based on multi-degree of freedom coupling system dynamics

Cutterhead is the core component of TBM tunneling equipment, which endures strong, multi-point distributed impact loads when the TBM tunnels, owing to the extreme surrounding rock environment of high hardness, high temperature and high quartz content. For this reason, the cutterhead works in an extremely severe vibration environment, which leads to engineering fault by a large area crack damage before the service life. Hence, the study on life prediction of TBM cutterhead under the impact loads is a core part of cutterhead design. This paper combines with the technology of system dynamics, linear elastic fracture mechanics and cumulative theory of fatigue damage, for the first time, proposes a method of fatigue crack propagation life prediction for the large and complex structures. In this paper, the TBM cutterhead of an actual project is taken as an example, to predict the fatigue crack propagation life of cutterhead piece and analyze the influences of plate thicknesses on fatigue life, then a new improved scheme of cutterhead structure is presented. The results show that the fatigue crack propagation life of actual cutterhead is 26.6 km, which is able to meet the requirement of 20 km service life. Moreover, the upper cover plate thickness has the greatest influence on cutterhead fatigue crack propagation life, with the thickness increasing 10%, the life increases nearly by 1.24 times. Then, the other influencing factors are as follows: thickness of the main support plate, thickness of the annular support plate and thickness of the support plate, whereas the influence of the lower cover plate thickness on fatigue life is minimal. Furthermore, the plate thickness limit sizes meeting the life requirement are obtained, and a new structure modified scheme of cutterhead is proposed. Compared with the original scheme, the new cutterhead scheme meets the requirements of structural strength and service life with 8.08% weight decrease, which achieves life determination design and lightweight design. The proposed method of fatigue crack propagation life prediction is feasible in the design and application stage of TBM cutterhead, besides, it is flexible enough and can also be applied in damage strength assessment, dynamic parameters optimization and establishment of nondestructive inspection cycle for the other large and complex structure, and takes on stronger project value and generality.     

A dedicated hardware architecture for real-time auto-focusing using an FPGA

The auto-focus is a fundamental function of a camera system which is required to photograph a clear image of an object. To obtain the optimal focus of a specific region within an image, the sharpness of the region must be measured. Since the sharpness represents the difference between a pixel and its neighbors, multiple pixel references occur while evaluating each pixel. To compensate for the processing bottleneck caused by this repetitive memory reference, this paper presents a dedicated hardware architecture for real-time auto-focusing. The proposed system processes the incoming pixel simultaneously with its neighboring pixels based on its parallelized window processing architecture. In addition, the proposed system performs an adaptive thresholding-based sharpness function with multiple windows to achieve accuracy and robustness. The proposed system is compared to several conventional pixel-based auto-focusing systems under various environmental conditions.