Papers by Keyword: Vision System

Abstract: With the development of the concept of Industry 4.0, research relating to robots is being paid more and more attention, among which the humanoid robot is a very important research topic. The humanoid robot is a robot with a bipedal mechanism. Due to the physical mechanism, humanoid robots can maneuver more easily in complex terrains, such as going up and down the stairs. However, humanoid robots often fall from imbalance. Whether or not the robot can stand up on its own after a fall is a key research issue. However, the often used method of hand tuning to allow robots to stand on its own is very inefficient. In order to solve the above problems, this paper proposes an automatic learning system based on Particle Swarm Optimization (PSO). This system allows the robot to learn how to achieve the motion of rebalancing after a fall. To allow the robot to have the capability of object recognition, this paper also applies the Convolutional Neural Network (CNN) to let the robot perform image recognition and successfully distinguish between 10 types of objects. The effectiveness and feasibility of the motion learning algorithm and the CNN based image classification for vision system proposed in this paper has been confirmed in the experimental results.

Abstract: For general dynamic pick and place tasks that the objects are transferred with high speed by the conveyor belt, the capability of a delta robot to track the traveling objects is very important for the efficiency. To meet the needs of precision and smooth control, a computed-torque control scheme for conveyor tracking is implemented in this paper. For higher efficiency and accuracy, computer vision system, encoder and conveyor belt region are incorporated into the control scheme. Dividing the conveyor belt into three regions, the robot is commanded to track, pick and give up according to the subregions. Conveyor belt is equipped with an encoder that provides the controller with real-time position and speed of the belt. Based upon those informations, the controller automatically compensates the end positions with respect to the belt to adjust for the position of the conveyor. Then, the conveyor tracking problem is converted to a subregional tracking problem.

Abstract: Security systems employ the RFID method to protect objects and data and to ensure public safety and the safety of commercial activity, e.g. through the control over technological processes. The research activity aimed at improving the level of safety of electronic and technical protection needs to be verified through the manufacture of prototype IDs with RFID inlays.The authors presents an original experimental device enabling the implementation of individual chips in RFID inlays, which were made on a flexible base to which an RFID antenna is attached. The collection of the chip from the dispenser and an unambiguous orientation of the soldered tips in relation to the antenna are provided by a four-axis vacuum manipulator cooperating with a vision system identifying the location of the chip on the manipulator and its final application area. A computer control system enables the operation in two modes, i.e. a semi-automatic mode in which it is possible to manually control the order and the way in which individual tasks are performed, and an automatic mode conducted according to the designed algorithm. The developed device is intended for research on the ways to improve the manufacturing techniques and electronic document protection by means of RFID technologies.

Abstract: Nowadays, the utilization of cameras as an inspection tool has been increasing. The flexibility functions of camera fits to get different kind of information. This research is focusing on developing a robust visual inspection system for corrosion detection that is able to detect corrosion in any environment, and the corrosion detection will be using visual data as primary tools. A review on current pipeline inspection would give a brief detail on the improvement of the proposed inspection system. Furthermore, the inadequacies of the proposed visual corrosion detection are identified and discussed from the reviewing process on existing researches and analysis on preliminary data obtained. It is expected that the output of the proposed system will be a new method of corrosion detection and pioneer for the inspection system on robust environment.

Abstract: The article presents a selected area of on-going research on the possibility of using intelligent computer-aided design systems for surface inspection of products manufactured in large batch processes. Systems based on machine vision are used wherever it is essential to obtain high efficiency, reproducibility, and where there is the need to use non-contact measurement methods. The IT solution significantly improves the design processes by automatically generating optimal configuration of machine vision systems intended for the detection of surface defects in manufactured products. General structure of the model and its individual modules, performing functions such as automatic component selection of machine vision systems, setting the angle of the light emission towards the surface of the analysed objects, and the selection of the range of light radiation are discussed. The individual configuration steps of vision systems for surface inspection, in which processes are implemented using expert systems making inferences based on both classical bivalent logic, as well as multi-valued fuzzy logic, are shown. The author presents the original methodology for the optimisation of structures forming vision systems intended for the reduction of components and costs associated with their implementation in physical structures, designed for use in production lines. The results of empirical studies of the calculation model are shown.

Abstract: Unmanned ground platforms are increasingly used to perform tasks under direct threat to the operator’s life and health. Separating an operator from a controlled machine places high demands to its steering system. The requirement of reliability put to such platforms in terms of potential threats requires their steering systems to be very precisely controlled and tested at the stage of construction. The key element in the steering process of unmanned ground platforms are surroundings display systems that determine the success of conducting rescue operations. This paper presents an unmanned ground platform controlled by remote operator.

Abstract: The edge etcher machine is fully automated with very dangerous chemical circumstance; the shortcoming is low capacity for production, it spends too much time on the centering process, we use vision system to reduce the time of centering process and increase the capacity of production.

Abstract: The type inspection of surface mounted devices (SMD) components is an important part of the numerical control (NC) placement machine’s vision system. In order to improve the detection speed, accuracy rate and versatility, a detection method based on moment features and neural network is proposed. Firstly, component images are preprocessed in order to eliminate the influence of non-uniform illumination and simplify the calculation, so component lead images can be obtained, and then seven invariant moments and a zero-order Zernike moment of the lead images are extracted. Next, the moment features are corrected and normalized. Finally, back propagation (BP) neural network based on the Levenberg-Marquardt algorithm is taken as a classifier for training and testing the 8-dimensional mixed moment feature vectors, 0 and 1 are used to represent the degree of belonging of each image. The experimental results show that this method doesn’t need complex lighting system and has good versatility, and the correct rate can be up to 100%.

Abstract: In recent years, vision system in robotics have a major impact on research field and industry. Many industrial applications have a huge requirement of sorting out the things as per different color or label. This paper deals with an automated material handling system, which synchronizes the movement of robotic arm to pick a single object from the maximum numbers. This paper aims to sort out different kind of objects by their color occupied pin code and place them to its respective pre-programmed place.

Abstract: This paper describes computer vision algorithms for detection, identification, and tracking of moving objects in a video file. The problem of multiple object tracking can be divided into two parts; detecting moving objects in each frame and associating the detections corresponding to the same object over time. The detection of moving objects uses a background subtraction algorithm based on Gaussian mixture models. The motion of each track is estimated by a Kalman filter. The video tracking algorithm was successfully tested using the BIWI walking pedestrians datasets [. The experimental results show that system can operate in real time and successfully detect, track and identify multiple targets in the presence of partial occlusion.