共查询到20条相似文献,搜索用时 15 毫秒
1.
2.
目的对于人工进行压缩机滑片计数烦琐的问题,基于机器视觉技术和PLC设计一种对压缩机滑片自动计数的算法和装置。方法采用工业相机及大功率LED灯设计视觉装置,并利用PLC作为流水线控制器,与视觉计算机通过TCP通信。基于图像处理技术设计自动计数算法,首先对相机采集的箱装滑片图像裁剪出感兴趣区域并进行滤波、直方图均衡化、形态学处理、自适应阈值分割来增强图像对比度;然后对图像空洞进行填充、较大的噪声斑块去除后使用thin算法对粘连紧密的滑片细化处理,通过水平投影法绘制直方图找出切割位置;最后进行连通域分析确定滑片数目,并设计用户界面显示计数结果和系统状态。结果通过现场多次测试,整体系统运行流畅,不受外界光照影响,计数准确高效。在一天内不同时间段测试,平均每次计数时间为7.6 s,准确率达到99.83%。结论投影法的应用解决了密集粘连的分割难题,该方法对环境适应性强,对滑片可稳定准确地进行自动计数,在滑片实际批量生产包装中具有明显应用优势。 相似文献
3.
There is a real need to develop a sensory system that can monitor and track the behavioral status of a person in real time while protecting his/her privacy. In addition, it is important to have the ability to detect stationary and moving objects in real time to support dynamic navigation needs for mobile robots. This paper describes the development of intelligent fiber-grating (FG)-based 3-D vision sensory system that enables real-time object detection, monitoring, and tracking. This system uses the richness and the strength of the vision while reducing the data load and computational cost by encoding the working space using a limited number of spatially interrelated 2-D laser spots. The sensor system generates the laser spots by projecting a coherent laser beam on two perpendicularly overlaid layers of FGs. The presence of a target within the visible operational view of the sensor disturbs the part of the projected laser spots' pattern where the target appeared. By calculating the displacement of the disturbed spots within the CCD image plane, the sensor system can detect, in a short time, an object/human and provide relevant information in terms of range, approximate shape, and position. To facilitate faster detection and real-time tracking, the author has developed two strategies that decompose the working space into zones and virtual planes 相似文献
4.
Ali Raza Samia Allaoua Chelloug Mohammed Hamad Alatiyyah Ahmad Jalal Jeongmin Park 《计算机、材料和连续体(英文)》2023,75(2):3275-3289
Pedestrian detection and tracking are vital elements of today’s surveillance systems, which make daily life safe for humans. Thus, human detection and visualization have become essential inventions in the field of computer vision. Hence, developing a surveillance system with multiple object recognition and tracking, especially in low light and night-time, is still challenging. Therefore, we propose a novel system based on machine learning and image processing to provide an efficient surveillance system for pedestrian detection and tracking at night. In particular, we propose a system that tackles a two-fold problem by detecting multiple pedestrians in infrared (IR) images using machine learning and tracking them using particle filters. Moreover, a random forest classifier is adopted for image segmentation to identify pedestrians in an image. The result of detection is investigated by particle filter to solve pedestrian tracking. Through the extensive experiment, our system shows 93% segmentation accuracy using a random forest algorithm that demonstrates high accuracy for background and roof classes. Moreover, the system achieved a detection accuracy of 90% using multiple template matching techniques and 81% accuracy for pedestrian tracking. Furthermore, our system can identify that the detected object is a human. Hence, our system provided the best results compared to the state-of-art systems, which proves the effectiveness of the techniques used for image segmentation, classification, and tracking. The presented method is applicable for human detection/tracking, crowd analysis, and monitoring pedestrians in IR video surveillance. 相似文献
5.
基于嵌入式机器视觉的多瓶口缺陷检测系统设计 总被引:1,自引:1,他引:0
目的 为了解决瓶口质量人工检测效率低且不稳定的问题。方法 采用HD-SDI的高清视频图像技术设计一种基于频率为70 MHz的STM32硬件检测系统,通过对视频读入、256 MB存储、29.97帧的视频输入输出和报警等硬件电路的设计,基于此搭建视频处理主流程、通信程序和图像处理的一系列软件算法来实现瓶喉口的缺陷检测,通过单张相片中塑料瓶口边缘轮廓来对比分析系统的可靠性。结果 实验证明,系统可以在0.26 s内识别1个瓶口,生产线产出速度为2.4 s/个。结论 该系统可以稳定、实时地检测多个瓶体口缺陷效果。 相似文献
6.
J. Pospíšil 《Journal of Modern Optics》2013,60(12):2479-2490
Abstract The role of super-Poisson quantum fluctuations of stimulating thermal light photons on the human scotopic (rod) vision at threshold is theoretically considered in this contribution. The considerations are based on the theoretical photon-counting and neural-counting probability distributions under negligible dead-time effect and on the corresponding theoretical probability-of-detection curves. The linearly polarized thermal light and the influence of the multiplicative and additive visual noise is assumed. Also the role of Bernoulli random deletion is considered. The theoretical results, obtained for some assumed illumination and observation conditions, are presented in analytical and graphical forms and also compared and evaluated. They are of interest principally when utilization of stimulating thermal light cannot be avoided. 相似文献
7.
8.
9.
目的研究设计焊条自动包装机上的焊条在线计数装置。方法在调查现有焊条生产线生产现状的基础上,结合现在使用的包装方式,采用伺服电机控制焊条计数,由旋转编码器进行计数设定,用红外线探测头控制被追踪的焊条,到达分离位置进行分离,完成计数。结果设定150根焊条为1个包装,伺服电机的追踪运动,能准确找到开始计数的被追踪焊条,计数没有偏差,焊条包装每包的质量误差不超过企业标准要求。结论设计制造的焊条在线计数装置,能很好地完成输送带上大量焊条的分包计数,能在实际生产中代替人工,提高效率。 相似文献
10.
11.
12.
13.
14.
对生产过程中的成品进行计数和分组控制常常是十分重要的。讨论对吸管包装机生产的吸管进行计数和分组控制的一些问题,如光电转换、信号处理、加法和减法计数、出料方向控制等。它不仅适用于吸管的生产,其结构和原理还可广泛用于其它产品的生产包装过程的相应控制。 相似文献
15.
A. Cañas 《Journal of Modern Optics》2013,60(5):583-584
Measurements of the energy spectra of infra-red images of stellar point sources were made at the Coudé focus of the 3·8 m telescope at Kitt Peak National Observatory, U.S.A. A scanning slit was used to record the irradiance distribution in the image plane. Experimental results are compared with theoretical models, assuming either a circular gaussian or a log-normal statistic for the perturbation of the complex amplitude of the incoming wavefront. In both cases the telescope central obscuration is taken into account. A good fit is obtained with the log-normal model for the highest scanning rate. 相似文献
16.
17.
光子计数式四象限跟踪探测器的噪声分析 总被引:4,自引:1,他引:3
为满足对弱光目标的跟踪探测需要,提出采用光地数式四象限探测器作为跟踪探测器。分析了四象限波探测器的探测灵敏度和噪声,推导了开环和闭环的噪声公式,同时还给出了系统在室内实测的开环和闭环噪声。 相似文献
18.
The past 10 years has seen an explosion of interest for the area of science and technology labelled “nanotechnology.” Although
at an early stage, nanotechnology is providing a space for the creation of new alliances and the forging of new ties in many
actor arenas, initiated based on promises and high expectations of the fruits that could be harvested from development and
investment into nanotechnology. Those trying to characterise the dynamics of emerging ties and networks within this field
are faced with a number of complexities which are characteristic of the nanotechnology umbrella term, which covers many technologies,
various mixes of disciplines and actors, and ongoing debates about definitions of fields and terminology.
In this paper we explore an approach for capturing dynamics of emergence of a particular area of nanotechnology by investigating
visions of possible futures in relation to molecular mechanical systems (molecular machines). The focus of this text is to
outline an approach used to map and analyse visions in an emerging field by taking as the unit of analysis linkages made in
statements in texts, and the agglomeration of linkages around certain nodes. Taking the linkage, rather than node, allows
one to probe deeper into the dynamics of emergence at early stages when definitions and meanings of certain words/nodes are
in flux and patterns of their use change dramatically over short periods of time.
As part of a larger project on single and macromolecular machines we explore the dynamics of visions in the field of molecular
machines with the eventual aim to elucidate the shaping strength of visions within nanotechnology. 相似文献
19.
20.
Modules that perform photorefractive two-beam coupling operations have been built, characterized, and tested. These portable modules, interconnected by fiber optics, dispense with the need for repeated alignment and greatly facilitate the prototyping of complex signal- or image-processing photorefractive circuits. To evaluate the performance of the modules in a photorefractive circuit, we interconnected them in the feature extractor configuration: a ring configuration composed of two modules that selects the strongest signal within the signals presented on its input. With two signals at the input, an output contrast ratio of 45.4 dB is obtained for an input contrast ratio of 5 dB. 相似文献