Proposed Vision Sensor Research Articles

Dynamic response and tension estimation of stay cables with disturbed images acquired from vision sensors

ABSTRACT This study introduces a novel approach to measure cable responses of such bridges using a vision sensor comprising a digital camcorder mounted on a tripod. Prioritizing simplicity and efficiency, this setup requires minimal equipment. The sensor is strategically positioned to ensure the target cable remains within the camcorder’s field of view. However, operational bridge imagery is frequently disturbed by vehicular traffic, leading to defocusing and blurring due to the camcorder’s automatic focus adjustments. These disturbances complicate the identification of reference points essential for accurate cable response measurements, potentially skewing results. To address this, we apply an image quality evaluation method to distinguish reference points from disturbed images, mitigating the impact of these disturbances on measurement accuracy. We evaluated the method’s effectiveness in estimating cable tension under disturbances by capturing images of an operational cable-stayed bridge experiencing ambient vibrations. Cable tension estimations were derived from the analysed responses using the vibration method and validated against data from a long-term monitoring accelerometer. This comparison highlights the potential of the proposed vision sensor method to enhance the accuracy of cable tension measurements in cable-stayed bridges, even in the presence of disturbed images.

A High-Speed Vision-Based Sensor for Dynamic Vibration Analysis Using Fast Motion Extraction Algorithms.

The development of image sensor and optics enables the application of vision-based techniques to the non-contact dynamic vibration analysis of large-scale structures. As an emerging technology, a vision-based approach allows for remote measuring and does not bring any additional mass to the measuring object compared with traditional contact measurements. In this study, a high-speed vision-based sensor system is developed to extract structure vibration signals in real time. A fast motion extraction algorithm is required for this system because the maximum sampling frequency of the charge-coupled device (CCD) sensor can reach up to 1000 Hz. Two efficient subpixel level motion extraction algorithms, namely the modified Taylor approximation refinement algorithm and the localization refinement algorithm, are integrated into the proposed vision sensor. Quantitative analysis shows that both of the two modified algorithms are at least five times faster than conventional upsampled cross-correlation approaches and achieve satisfactory error performance. The practicability of the developed sensor is evaluated by an experiment in a laboratory environment and a field test. Experimental results indicate that the developed high-speed vision-based sensor system can extract accurate dynamic structure vibration signals by tracking either artificial targets or natural features.

Vision Based Localization for Multiple Mobile Robots Using Low-cost Vision Sensor

This paper presents an efficient approach for a vision based localization of multiple mobile robots in an indoor environment by using a low cost vision sensor. The proposed vision sensor system that uses a single camera mounted over the mobile robots field takes advantages of small size, low energy consumption, and high flexibility to play an important role in the field of robotics. The nRF24L01 RF transceiver is connected to the vision system to enable wireless communication with multiple devices through 6 different data pipes. The downward-facing camera provides excellent performance that has the ability to identify a number of objects based on color codes, which form colored landmarks that provide mobile robots with useful image information for localization in the image view, which is then transformed to real world coordinates. Experimental results are given to show that the proposed method can obtain good localization performance in multi-mobile robots setting.

Vision-based multipoint displacement measurement for structural health monitoring

Summary A novel noncontact vision sensor for simultaneous measurement of structural displacements at multiple points using one camera is developed based on two advanced template matching techniques: the upsampled cross correlation (UCC) and the orientation code matching (OCM). While existing studies on vision sensors are mostly focused on the time-domain performance evaluation, this study investigates the performance in both time and frequency domains through a shaking table test of a three-story frame structure, in which the displacements at all the floors are measured by using one camera to track either high-contrast artificial targets or low-contrast natural targets on the structural surface such as bolts and nuts. Excellent agreements are observed between the displacements measured by the single camera and those measured by high-performance laser displacement sensors. The results of structural modal analysis based on the measurements by the vision sensor and reference accelerometers also agree well. Moreover, the identified modal parameters are used to update the finite element model of the structure, demonstrating the potential of the vision sensor for structural health monitoring applications. This study further examines the robustness of the proposed vision sensor against ill environmental conditions such as dim light, background image disturbance, and partial template occlusion, which is important for future implementation in the field. Significant advantages of the proposed vision sensor include its low cost (a single camera to remotely measure structural displacements at multiple points without installing artificial targets) and flexibility to extract structural displacements at any point from a single measurement. Copyright © 2015 John Wiley & Sons, Ltd.

Application of circular laser vision sensor (CLVS) on welded seam tracking

Area-array CCD vision sensor based on circular laser trajectory used for weld seam location and seam tracking was developed. In order to meet the requirement of 3D recovery, order-statistic filtering method based on one-third value instead of mean value was presented to make the circular laser trajectory perfect. Two-step segmentation (TSS) technique under GTAW (gas tungsten arc welding) arc condition or without welding arc condition was investigated. Using proposed 3D calculation algorithm, seam tracking for ramp welded seam was studied. The results showed, this vision sensor could measure the 3D change of welded seams at the same time, and 3D seam tracking could be realized. Proposed vision sensor manifested a promising result with high precision and high efficiency.

Welded seam 3D calculation and seam location for welding robot system

Using laser scanning techniques, scanning lens rotating around a laser diode and a circular laser trajectory was projected onto the surface of a weldment to detect it. Furthermore, one novel vision sensor based on the circular laser is developed. The three-dimensional (3D), circular laser based seam location sensor is investigated. On the basis of the light path system, 3D calculation algorithm is brought forward and used to locate the welded joint. After image denoise, filter, segmentation and thinning, characteristic points of the welded joint could be detected real time using a proposed vision sensor and confirmed by real experiments of butt welded joints with I groove and V groove, and lap, and fillet and ramp welded joints. The results show that circular laser based vision sensor can be used in seam locating and its calculation precision meets the requirement of seam tracking.

An active vision sensing method for welded seams location using “circle–depth relation” algorithm

In order to meet the requirement of welded seams location and seam tracking, vision sensor based on circular laser trajectory has been developed and three-dimensional information of welded seams can be calculated in real- time, escaping from shortcomings of less information, explain ambiguity and single tracking direction for “spot” or “line” laser vision sensors. We put forward three vital angles (cone angle α, deflection angle β, and off-axis angle γ) so that welded seams could be described using three-dimensional information. On this basis, a mathematical model of depth recovery according to “circle–depth relation” algorithm is constructed. The relation of depth values and off-axis angle γ under gas tungsten arc welding (GTAW) experiment condition is addressed by real experiment and the characteristic points of welded seams is described and located. Mean square error (MSE) was used to analyze the recovery precision. The results show: (1) Depth recovery for butt joint with V-groove (60°), butt joint with gap, lap joint, corner joint (120°), circular welded seams and ramp welded seams are realized and the recovery accuracy is compulsory for GTAW process; (2) Welded seams can be located very well using proposed vision sensor based on circular laser trajectory; and proposed vision sensor manifested a promising result with high precision and high efficiency.

Simplifying the kinematic calibration of parallel mechanisms using vision-based metrology

In this paper, a vision-based measuring device is proposed and experimentally demonstrated to be an accurate, flexible, and low-cost tool for the kinematic calibration of parallel mechanisms. The accuracy and ease of use of the proposed vision sensor are outlined, with the suppression of the need for an accurate calibration target, and adequacy to the kinematic calibration process is investigated. In particular, identifiability conditions with the use of such an exteroceptive sensor are derived, considering the calibration with inverse or implicit models. Extensive results are given, with the evaluation of the measuring device and the calibration of an H4 robot. Using the full-pose measurement, an experimental analysis of the optimal calibration model is achieved, with study of the kinematic behavior of the mechanism. The efficiency of the provided method is thus evaluated, and the applicability of vision-based measuring devices to the context of kinematic calibration of parallel mechanisms is discussed.

Obstacle avoidance using time-to-contact information

Machine vision can give rich information about the environment. However, image processing usually demands high computational effort. This paper proposes a framework for a vision-based row-data range sensor for obstacle detection using the concept of time-to-contact. Differently from usual approaches, here this quantity is computed directly, without involving a previous motion field estimation (via optic flow calculation, for example). Theoretical and practical comparisons between the direct approach and the optic-flow-based one for computing time-to-contact are presented. An experience that uses the proposed vision sensor to control a real robot was performed with good results.

A Real-Time Motion Stereo Vision Based on Brightness Gradient.

To realize autonomous manipulators which work outdoors, 3-D vision sensors having capability of real-time processing, robustness against sunlight and compactness are required to measure objects and environments. In this paper, a simple real-time motion stereo vision is proposed, which utilizes direct calculation of the integrated gradient constraint equation for sequential images taken by a motion camera. The vision sensor can measure 3-D coordinates at every pixel without huge computation. The basic characteristics are investigated through indoor measuring experiments. The measuring accuracy of ± 10% in a range of 0.7-1.5 m and the estimated measuring time of 0.1 s for 512 pixels × 100 scanning lines are obtained. Then the performance of the proposed vision sensor is experimentally investigated under natural conditions including outdoors.

中心窩を有する新しい視覚センサの開発

In this paper a new concept of a vision sensor with fovea is presented and the possibility of the application for on-line machine control is demonstrated. The proposed vision sensor is designed approximate to the visual mechanism of the human eye, which has a fovea (high resolution attention zone) in its field of view and executes the necessary movements of viewpoint. A distortion lens plays an important role in the system, which enlarges the image in the attention zone and compresses the one in the peripheral zone. The lens has been designed in order to realize the system. The specification and characteristics are presented in detail. The results show that the proposed system will work well to recognize the environments of mobile robots.