作　　者： (蒋缠文）; (王晓艳）; (董治宝）;

机构地区： 中国科学院西北生态环境资源研究院沙漠与沙漠化重点实验室,甘肃兰州730000 渭南师范学院农商学院陕西省河流湿地生态与环境重点实验室,陕西渭南714099

出　　处： 《干旱区地理》 2017年第4期746-753,共8页

摘　　要： 通过相邻图像减法获得了清晰的沙粒运动图像。在此基础上,分别提出了通过人工目视解译与计算机追踪相结合进行跃移颗粒数字轨迹追踪的多帧图像匹配算法以及更适用于计算跃移沙粒群运动瞬时速度场的两帧图像匹配算法。结果表明:与前人研究采用的单纯的人工匹配计算相比,多帧图像匹配算法在保证数据准确性的同时,极大的提高了工作效率。两帧图像匹配算法克服了传统的PTV匹配算法对流场内粒子群运动特征的要求,更加适宜于跃移沙粒群的速度测量,不仅拥有较高的匹配率,而且全过程实现全自动计算,具有较高的计算速度,能够为跃移相整体运动特性研究提供具有代表性的数据。因此,此方法有助于高速摄影技术在研究跃移沙粒运动中的优势更加明显。 The velocity of saltating sand is a key parameter in aeolian movement,but remains poorly understanding because of limitations in the available measurement technology.High-speed photography illuminated by intensive and continuous laser provides an efficient method to track the trajectories of saltating particles.From a digital recording of the trajectory,particle velocity and other movement parameters can be derived.However,accurate and quick reconstruction of particle trajectories from video images is not yet possible,as this requires the use of prohibitively laborious manual methods.In addition,light reflection from the bed decreases the contrast between near-surface particles and the bed.This makes saltation information near the surface difficult to obtain,even though this data is vitally important for studying particle entrainment processes and the mechanisms by which particles leave the surface.In this paper,we used motion-detection algorithms based on the subtraction of consecutive images and developed a new series of high-contrast images from the original images.We then describe two novel methods for tracking the trajectories of saltating particles and determining the velocity field in a cloud of blowing sand,respectively.We tracked particles by combining a manual operation with a computerized operation,and then the movement parameters of saltating sand particles in air and near the bed surface,such as the lift-off velocity and angle,impact velocity and angle,characteristic length and height,and instantaneous velocity and accelerated velocity vector at each point in the trajectory were quickly and accurately calculated by polynomial fitting method.Compared with traditional manual track method,this novel technique can easier to obtain a large number of continuous points within one or more trajectories.We evaluated the velocity field by a novel two-frame PTV algorithm based on the concept of maximizing the match efficiency.This approach overcomes some disadvantages of the conventional PTV methods,such as simi

关 键 词： 风沙跃移 运动状态参数 颗粒速度 高速摄影技术 颗粒追踪算法