Advanced Materials Research Vols. 317-319

Abstract: Because the structural and motion parameters errors, when the robot move in the ideal situation, actual posture of the robot is likely to deviate from its required position. Based on the discussion of forward kinematics, which is calculating the influence of error sources on the posture error of robot, is researched in this paper, and an example is used to explain. The effect of each parameter on posture error can be evaluated by analyzing the results of example. The designer can make different error control method for each parameter according to the influence of them.

Abstract: The high order equation causing by analytic method in multi-legged robot forward kinematics analysis may have imaginary root, repeated root, extraneous root or even lost solution. A system based on the theory of computer aided geometric method is proposed. Consideration with the internal structural constraint relations of multi-legged walking robots, the solidworks model was constructed and Visual Basic develop platform was adopted to fulfill the secondary development of solidworks. A system of forward kinematics analysis of multi-legged walking robots is established. The example validates that the system is simple and effective for all reptiles-like quadruped walking robot.

Abstract: Non-contact Measurement of circular hole, which based on machine vision is already used for industrial applications. In this paper, a novel practical method to obtain parameters of a circular hole from an image is presented. The main idea of the proposed method is that we use gradient information to find a region that contains possible centers of the circle (the hole) and use evidence-collecting process to obtain the reasonable radius. This method fits the situation that the hole object in a image is totally brighter or darker than its background. Experimental results demonstrate that our method can detect the hole with the acceptable precision and stability for practical applications.

Abstract: For acquisition of complete 3D models, this paper uses a rotational device to capture a set of image sequences. A direct projective reconstruction method is proposed by linear transformation, which can avoid getting corresponding points in more than two images. Actually, projective reconstructions are obtained from two neighboring images and the reconstructions are combined with the common 3D points. Finally, all reconstructions are merged into the initial one to construct a complete model. Several practical experiments have been carried out to validate the accuracy of the method.

Abstract: The detection of foreign substances in injection so far is still achieved artificially, which result in low accuracy and low efficiency. This paper focuses on developing a novel vision-based approach for detection of foreign substances. Foreign substances are classified into two categories, subsiding-slowly object and subsiding-fast object. A relative movement caused by a motor helps to distinguished foreign substances from ampoule surface scratches. Moving objects in injection are divided from static ones by a background image derived from two frames. The Mean Shift Embedded Particle Filter (MSEPF) is proposed to detect moving-slowly object while Frame Distance is defined to detect moving-fast object. 200 ampoule samples filled with injection are tested. The integrated detection accuracy with this approach is 98.00%, with 97.56% accuracy for subsiding-slowly objects and 96.67% accuracy for subsiding-fast ones. The result shows that the system can detect foreign substances effectively.

Abstract: Machine vision is widely used in chip pin automation detection of defects, the accuracy of image edge extraction for detection result has very big effect. Canny edge detection algorithm is a kind of edge detection algorithm which have good comprehensive evaluation, but the Gaussian filter algorithm it used may cause image too smooth and fuzzy of the edge, and it is very sensitive for impulse noise. This paper discusses the improved methods of Canny, proposes an improved switch median filter algorithm, which is applied into Canny algorithm, makes the edge of IC Chip pin more complete and remove noise better.

Abstract: Localization of barcode is the key technology for barcode recognition. This paper presents an automatic localization algorithm for multiple barcodes in complex background. The proposed algorithm includes dividing the image into a plurality of tiles, and scanning each of the tiles so as to detect a pattern of strips associated with the barcode in at least one of the tiles. The pattern of strips is analyzed so as to determine the candidate region. Then the effective affine region of barcode is computed by the method of Blob analysis and the barcode is only contained in it. Experimental results indicate that the proposed approach has certain accuracy, the efficiency and robustness.

Abstract: Visual positioning system is of great importance in the structure of PDP lighting inspection machine. This paper describes the overall structure of the system, the process of the interaction with the PLC system, the hard structure and continuous positioning principle. A point correspondence method is also studied.

Abstract: In order to reduce the position error produced by single board assembly and partial uneven surface in automated optical inspection, a new position method based on three layers of MARK point is developed. the first layer is the whole board alignment which compensate the position error of the PCB loading process, the second layer is the single board alignment which compensate the position error of the single board assembly, the last layer is the partial FOV(field of view) alignment which compensate the position error of the partial uneven surface. The experiment result proved the proposed position method improved the position accuracy obviously.

Abstract: Pedestrian detection, which has a wide application in surveillance, advanced robotics, and especially intelligent vehicles, is an important area in computer vision. This paper applies a detection approach based on improved Adaboost algorithm. We use a dataset to train the weak classifiers (with different numbers) to cascade to be strong classifiers, in which we employ optimized strategy of sample weight adjustment to reduce the over-fit. After constructing a strong classifier, we apply different scale of sliding widow to shift and calculate the corresponding features to classify them as pedestrians or non-pedestrians. The experiments show that different numbers of weak classifiers layer and different scale of sliding windows can give different performance in detecting.