• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.12
• /
• pp.1109-1114
• /
• 2015

This paper presents an improved learning-based visual inspection method for auto parts inspection in severe lighting changes. Automobile sunroof frames are produced automatically by robots in most production lines. In the sunroof frame manufacturing process, there is a quality problem with some parts such as volts are missed. Instead of manual sampling inspection using some mechanical jig instruments, a learning-based machine vision system was proposed in the previous research[1]. But, in applying the actual sunroof frame production process, the inspection accuracy of the proposed vision system is much lowered because of severe illumination changes. In order to overcome this capricious environment, some selective feature vectors and cascade classifiers are used for each auto parts. And we are able to improve the inspection accuracy through the re-learning concept for the misclassified data. The effectiveness of the proposed visual inspection method is verified through sufficient experiments in a real sunroof production line.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 1996.06c
• /
• pp.607-614
• /
• 1996

A computer vision based automatic intelligent sorting system for dried oak mushrooms has been developed. The developed system was composed of automatic devices for mushroom feeding and handling, two sets of computer vision system for grading , and computer with digital I/O board for PLC interface, and pneumatic actuators for the system control. Considering the efficiency of grading process and the real time on-line system implementation, grading was done sequentially at two consecutive independent stages using the captured image of either side. At the first stage, four grades of high quality categories were determined from the cap surface images and at the second stage 8 grades of medium and low quality categories were determined from the gill side images. The previously developed neuro-net based mushroom grading algorithm which allowed real time on-line processing was implemented and tested. Developed system revealed successful performance of sorting capability of approximate y 5, 000 mushrooms/hr per each line i.e. average 0.75 sec/mushroom with the grading accuracy of more than 88%.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.5
• /
• pp.490-498
• /
• 2011

In this paper, we present a two-stage vision-based approach to detect multi views of pedestrian in road scene images. The first stage is HG (Hypothesis Generation), in which potential pedestrian are hypothesized. During the hypothesis generation step, we use a vertical, horizontal edge map, and different colors between road background and pedestrian's clothes to determine the leg position of pedestrian, then a novel symmetry peaks processing is performed to define how many pedestrians is covered in one potential candidate region. Finally, the real candidate region where pedestrian exists will be constructed. The second stage is HV (Hypothesis Verification). In this stage, all hypotheses are verified by Support Vector Machine for classification, which is robust for multi views of pedestrian detection and recognition problems.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.1
• /
• pp.54-60
• /
• 2009

In this paper, a precision landing approach is implemented based on real-time image processing. A full-scale landmark for automatic landing is used. canny edge detection method is applied to identify the outside quadrilateral while circular hough transform is used for the recognition of inside circle. Position information on the ground landmark is uplinked to the unmanned helicopter via ground control computer in real time so that the unmanned helicopter control the air vehicle for accurate landing approach. Ground test and a couple of flight tests for autonomous landing approach show that the image processing and automatic landing operation system have good performance for the landing approach phase at the altitude of $20m{\sim}1m$ above ground level.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1995.10a
• /
• pp.1124-1128
• /
• 1995

This paper presents the appplication of computer vision for the purpose of determing the position of the unknown point in the plane. The presented contrik method is estimate the six view parameters reqresenting the relationship between the image plane coordinates and the real physical coordinates. The estimation of six parameters is indispensable for transforming the 2-dimensional camera coordinates to the 3-dimensional spatial coordinates. Then, the position of unknown point is estimated based on the estimated parameters depending on the cameras. The suitability of this control scheme is demonstrated experimentally by determining of position the unknown point in the plane.

• Journal of the Korean Society for Precision Engineering
• /
• v.32 no.2
• /
• pp.191-198
• /
• 2015

In this paper, six-degree-of-freedom (DoF). Displacement measurement technique using a compact stereo-vision system is proposed. The measuring system consists of a camera, an optical prism, two plane mirrors, and a planar marker on a target. The target was attached on an object so that its six-DoF displacement can be calculated using a proposed coordinates estimating algorithm and stereo images of the marker. A prototype was designed and fabricated for performance test. From the test results, it can be confirmed that the proposed measuring technique can be applied to monitoring and control of various manipulators.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.4
• /
• pp.1683-1691
• /
• 2013

Concrete surface grinding is frequently used for flatness of concrete surface, concrete pavement rehabilitation, and adhesiveness in pavement construction. The procedure is, however, labor intensive and has a hazardous work condition. Also, the productivity and the quality of concrete surface grinding highly depend on the skills of worker. Thus, the development of remote controlled concrete surface grinding equipment is necessary to prevent the environmental pollution and to protect the workers from hazardous work condition. However, it is difficult to evaluate the grinded surface objectively in a remote controlled system. Also, The machine vision system developed in this study takes the images of grinded surface with the network camera for image processing. Then, by representing the quality test results to the integrated program of the remote control station, the quality control system is constructed. The machine vision algorithm means the image processing algorithm of grinded concrete surface and this paper presents the objective quality control standard of grinded concrete surface through the application of the suggested algorithm.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.608-611
• /
• 1997

The soccer robot system consists of multi agents, with highly coordinated operation and movements so as to fulfill specific objectives, even under adverse situation. The coordination of the multi-agents is associated with a lot of supplementary work in advance. The associated issues are the position correction, prevention of communication congestion, local information sensing in addition to the need for imitating the human-like decision making. A control structure for soccer robot is designed and several behaviors and actions for a soccer robot are proposed. Variable zone defense as a basic strategy and several special strategies for fouls are applied to SOTY2 team.

• Proceedings of the KIPE Conference
• /
• 2004.07a
• /
• pp.155-158
• /
• 2004

This paper describes a lane following control of vision based mobile robot that follows guidline. Summation of binarization conversion and image data of vertical axis was used in image processing. As an extraction of specific parameters of lane image, the raw image was converted to the binary data, and the binary data was summerized to the specific data vertically. The specific parameters were made to the inputs of neural network. Summation of image data was used for input of the net, and optimized value of turn angles of learned mobile robot was output. By using neural network algorithm, possibility of mobile robot moving to the target point and following the guidlines quickly and effectively was proved.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.2
• /
• pp.164-170
• /
• 2011

This paper proposes a novel monocular vision-based SLAM (Simultaneous Localization and Mapping) method using both position and orientation information of ceiling lamps. Conventional approaches used corner or line features as landmarks in their SLAM algorithms, but these methods were often unable to achieve stable navigation due to a lack of reliable visual features on the ceiling. Since lamp features are usually placed some distances from each other in indoor environments, they can be robustly detected and used as reliable landmarks. We used both the position and orientation of a lamp feature to accurately estimate the robot pose. Its orientation is obtained by calculating the principal axis from the pixel distribution of the lamp area. Both corner and lamp features are used as landmarks in the EKF (Extended Kalman Filter) to increase the stability of the SLAM process. Experimental results show that the proposed scheme works successfully in various indoor environments.