1 浙江大学信息与电子工程学院, 浙江 杭州 310027
2 浙江省信息处理与通信网络重点实验室, 浙江 杭州 310027
针对目前单目同步定位与建图(SLAM)初始化方法鲁棒性差、精度低的问题,提出了一种基于点线特征的单目SLAM初始化方法。首先,在两帧图像之间提取线特征,并进行特征匹配。然后,通过最大化投影线特征的重叠长度优化两帧图像之间的初始旋转和平移矩阵。最后,用滑动窗口增加初始化的图像帧数,利用多帧图像的信息和约束关系以及全局光束平差法优化初始地图和估计的相机关键帧位姿。在TUM和OpenLORIS数据集上的测试结果表明,相比传统初始化方法,本方法的鲁棒性更好、精度更高,能在具有挑战性的场景中快速完成高精度初始化。
图像处理 运动估计和光流 多帧图像处理 单目视觉初始化 定位与建图 点线特征 光学学报
2021, 41(12): 1215002
Author Affiliations
Abstract
School of Aeronautics and Astronautics, Shanghai Jiao Tong University, Shanghai 200240, China
By improving the long-term correlation tracking (LCT) algorithm, an effective object tracking method, improved LCT (ILCT), is proposed to address the issue of occlusion. If the object is judged being occluded by the designed criterion, which is based on the characteristic of response value curve, an added re-detector will perform re-detection, and the tracker is ordered to stop. Besides, a filtering and adoption strategy of re-detection results is given to choose the most reliable one for the re-initialization of the tracker. Extensive experiments are carried out under the conditions of occlusion, and the results demonstrate that ILCT outperforms some state-of-the-art methods in terms of accuracy and robustness.
100.4999 Pattern recognition, target tracking 110.4155 Multiframe image processing 330.4150 Motion detection Chinese Optics Letters
2019, 17(3): 031001
Author Affiliations
Abstract
1 Institute for Digital Communications, University of Edinburgh, EH9 3JL Edinburgh, UK
2 Innovation Works Germany, EADS Deutschland GmbH, 81663 Munich, Germany
In this paper, a new approach for wireless data transmission within an aircraft cabin is presented. The proposed application enables the transmission of data to a passenger’s user device. As wireless in-flight applications are subject to strict frequency and electromagnetic compatibility (EMC) regulations, the data is transferred by optical wireless transmission, specifically by two-dimensional visual codes. To this end, black-and-white or colored visual code sequences are displayed on the in-flight entertainment screen. These visual codes are captured by the built-in camera of the passenger’s mobile device and are decoded to reconstruct the transmitted data. In order to compensate for frame losses caused by effects like occlusion and motion blur, a temporal forward error correction coding scheme is applied. Transmission experiments within an Airbus A330 cabin mock-up demonstrate the functionality of the implemented system under realistic conditions such as ambient illumination and geometric configuration. Representative user devices are used for evaluation; specifically, a low-cost and a high-end smartphone are employed as receivers. Performance evaluations show that the proposed transmission system achieves data rates of up to 120 kbit/s per individual passenger seat with these user devices. As the user device has no physical connection to the sensitive on-board system, the proposed transmission system provides an intrinsic safety feature.
Image detection systems Optical interconnects Optical devices Multiframe image processing Free-space optical communication Optical effects on vision Photonics Research
2014, 2(5): 05000150
Author Affiliations
Abstract
This paper proposes the post-integration technology based on sub-pixel image registration and image fusion to improve the signal-to-noise ratio (SNR) of remote sensing images without motion degradation caused by satellite vibration. A two-dimensional vibration system is set up to simulate satellite disturbance. Image sequences with different exposure times are captured using a high-speed CMOS camera. The displacement plots are compared with the motion data measured by the grating linear encoder. These plots indicate that the accuracy of the registration algorithm is better than 0.1 pixels. The sub-pixel image fusion shows an improvement in image quality, thus indicating that this technology is powerful for staring imaging systems in geostationary orbit.
280.4788 Optical sensing and sensors 110.4153 Motion estimation and optical flow 110.4155 Multiframe image processing Chinese Optics Letters
2014, 12(4): 042802
Author Affiliations
Abstract
Face-centered orthorhombic (FCO) sampling can be implemented more easily on CMOS image sensors than on other video acquisition devices. The sampling efficiency of FCO is the highest among all three-dimensional (3D) sampling schemes. However, interpolation of FCO-sampled data is inevitable in bridging human perception and machine-vision algorithms. In this letter, the concept of motion compensation is borrowed from deinterlacing, which displays interlaced videos on progressively scanned devices. The combination of motion estimation based on intrafield interpolated frames and motion-compensated interfield interpolation is found to provide the best performance by evaluating different combinations of motion estimation and interpolation.
100.3010 Image reconstruction techniques 110.3010 Image reconstruction techniques 110.4155 Multiframe image processing 100.2000 Digital image processing Chinese Optics Letters
2012, 10(6): 061001
Author Affiliations
Abstract
School of Electronics and Information Engineering, Beihang University, Beijing 100191, China
We present an image recognition method to distinguish targets with cat-eye effect from the dynamic background based on target shape and modulation frequency. Original image sequences to be processed are acquired through an imaging mechanism that utilizes a pulsed laser as active illuminator and an industrial camera as detection device. There are two criterions to recognize a target: one exploits shape priors and the other is the active illuminator’s modulation frequency. The feasibility of the proposed method and its superiority over the single criterion method have been demonstrated by practical experiments.
110.2970 Image detection systems 100.3008 Image recognition, algorithms and filters 110.4155 Multiframe image processing 100.2000 Digital image processing Chinese Optics Letters
2011, 9(4): 041101
Author Affiliations
Abstract
School of Optoelectronics, Beijing Institute of Technology, Beijing 100081, China
By analyzing the error distribution rule of the boundary recursive reconstruction algorithm in controlled micro-scanning, a sub-pixel image processing algorithm is proposed to reduce the error. The gray statistical principle is used in the algorithm to optimize the error and acquire the sub-pixel image that approximates the original image. The simulation result shows that the e?ect of this algorithm is better than the over-sample and simple boundary recursive algorithm (BRA), and it results in a good effect both in those of visible light and infrared imaging systems. Therefore, the application of this algorithm will enhance the performance of optoelectronic imaging systems.
可控微扫描 边框回代算法 灰度统计 亚像元成像 110.4155 Multiframe image processing 100.3008 Image recognition, algorithms and filters 100.3020 Image reconstruction-restoration 100.2000 Digital image processing Chinese Optics Letters
2010, 8(2): 146
Author Affiliations
Abstract
School of Electronics and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, ChinaE-mail: jq_ma@126.com
Traditional color-based mean shift tracking algorithm is unable to accurately track the object. To address this problem, we present an improved tracking algorithm. The improved tracker integrates the color and motion cues which characterize the appearance and motion information of the object, respectively. These two visual cues can complement each other and make for more precise target localization. Experiments show that the proposed tracking algorithm has better performance than the traditional mean shift tracker.
视觉跟踪 颜色线索 运动线索 100.4999 Pattern recognition, target tracking 110.4155 Multiframe image processing 330.4150 Motion detection 330.1710 Color, measurement Chinese Optics Letters
2009, 7(5): 05400
