Search

This paper proposes a novel method that significantly improves the accuracy of the fisheye stereo camera using multiple disparity offset maps that contain disparity error information. The disparity offset map is constructed based on disparity errors that are calculated from the feature points extracted from an object with a known distance. Modified stereo matching process based on the maps is carried out to greatly reduce the disparity errors that cannot be completely removed by general calibration scheme. Furthermore, the our method is possible to achieve high accuracy in both short and long range by combining the multiple maps. The proposed method is verified by experiments.

The use of high frequencies is essential in detecting deterioration in a physical function, but expert evaluation is expensive, and there are various other problems such as the need for many measurement devices. The aim of this study was to automate the evaluation of fall risk by video recording the gaits of elderly people using a general camera. We took videos of 12 subjects performing a 5-m walking test in a nursing home. In this paper, we evaluate (1) the use of the feature extraction method on walking behavior obtained from video recordings and (2) the accuracy of fall prediction using machine learning.

In integrated circuit (IC) manufacturing, wafer-map analysis has been essential for yield improvement. In this study, we focused on wafer-map failure pattern recognition. We proposed a deep learning-based failure pattern recognition framework. The proposed framework needs only wafer-maps with and without target failure patterns to recognize, and ascertains the features of the target failure patterns automatically. Conventional deep learning methods need a large amount of wafer-maps with the target failure patterns as training data for achieving high recognition accuracy. In the proposed framework, a data augmentation technique with noise reduction is proposed, and it is the key to achieve high recognition accuracy if the number of wafer-maps with the target failure patterns is small. Experimental results using a benchmark dataset showed that the proposed framework achieves high recognition accuracy with a failure pattern recognition problem and also multiple failure pattern recognition problem, and we confirmed the effectiveness of the proposed data augmentation technique.

This paper deals with improving the accuracy of a fisheye stereo camera by correcting images using a disparity offset map, which is a map of disparity errors that are calculated from the disparity of the feature points obtained from an object with a known distance. By correcting the disparity errors and aligning the images, errors, such as those due to the insufficient calibration of the camera parameters, are reduced. A disparity offset map is constructed, and the accuracy of the distance measurement of a fisheye stereo camera using the disparity offset map is evaluated by experiments.

This paper proposes a method of stereo rectification for stereo camera using two fisheye cameras. Rectification parameters are estimated from feature points obtained from an image of an arbitrary environment. Errors due to deviation of the position and rotation of the left and right cameras are removed by stereo rectification. The performance of the fisheye stereo camera using equirectangular images is evaluated by simulation and experiments. It is shown that accuracy of range measurement is improved by the proposed rectification.

Traffic congestion in urban areas leads to huge economic loss. Smart transportation systems are needed for efficient operations of the core economic activities. To develop such systems, we focus on multimodal urban transport systems, particularly on the optimization of dedicated bus lanes which is regarded as one of the most promising traffic demand management measures. Although there exists a great body of literature on the operation and planning of the dedicated bus lanes, quantitative approach is still lacked for evaluating the relationship between lane space and traffic performance, especially at the system level. In this context, [8] proposed to decide the allocation of the dedicated bus lanes based on the three-dimensional macroscopic fundamental diagram (3D-MFD). The proposed bus lane optimization framework was applied and tested utilizing the Geneva network via simulation. This paper follows the direction of the 3D-MFD based macroscopic approach, and it implements this approach, with slight modification, to the Tokyo CBD area which has a complete different and more complex network structure, to verify the robustness and the universality of the approach. We illustrate the 3D-MFD for the current network. It is found that the results of 3D-MFD are consistent with the results of [9]. That is, it is crucial to ensure certain number of buses running in the network, in order to optimize the network performance in terms of passenger service throughput.

This paper proposes making a stereo camera using two fisheye cameras. The measurement range of the stereo camera is increased by using fisheye cameras. Fisheye images are deformed by using an equirectangular projection to simplify the stereo correspondence search in the images. Perspective transformation is applied to images so that feature points of far objects coincide in the images of the two cameras in order to remove mismatches due to the errors of intrinsic and extrinsic parameters. The corresponding points are calculated by using stereo matching in the deformed image, and the distance is measured from the disparity. The accuracy of the constructed stereo camera's distance measuring is evaluated by experiments. The performance of the constructed stereo camera is evaluated by indoor and outdoor experiments.

Chemical Mechanical Polishing (CMP)-aware design has become important for reliability and yield. Recent work on predictive models for wafer surface planarity of Cu CMP has proven that the variation of wafer surface planarity is impacted by the metal perimeter in addition to the pattern density. Dummy fill insertion has been widely adopted to improve the CMP planarity in industrial design flows. However, conventional dummy fill insertion has been derived mainly to optimize the pattern density uniformity, which may worsen the CMP planarity because of missing impacts due to metal perimeter. In this paper, we propose; 1) a design of experiment (DOE) based method of evaluating the quality of fill insertion by using a CMP simulator which considers the impacts due to both pattern density and metal perimeter, and 2) a condition-based dummy fill insertion using the results of the proposed DOE method. Compared to the conventional pattern density driven rule-based fill insertion, the proposed method reduces the post-CMP Cu surface height variation by 24.3%. The metric of the metal perimeter may be applied to the model-based fill insertion methods, which will improve the planarity in the practical fill insertion flow.