摘要
采用当前方法重建园林地形景观叠加区域时,重建所用的时间较长,得到的重建结果与实际园林地形景观叠加区域之间的拟合度较低,存在重建效率低和精准度低的问题。现提出复杂地貌园林地形景观叠加区域三维仿真方法,利用灰度梯度图描述园林地形景观叠加区域图像中像素点之间的灰度差,采用Vincent分水岭算法分割园林地形景观叠加区域灰度梯度图,获取园林地形景观叠加区域的轮廓边缘。根据局部极值点的相对位置关系提取园林地形景观叠加区域轮廓边缘中的特征向量,计算特征向量与基准图中特征向量之间的距离,选择距离做小的特征向量作为匹配特征,通过特征点匹配完成复杂地貌园林地形景观叠加区域的三维重建。仿真结果表明,所提方法的重建效率高、精度高。
In the current methods, the reconstruction time is long and the reconstruction result has low fitting degree with the actual superimposed region of garden landscape. Therefore, this paper proposed a three-dimensional simulation method for superimposed region of garden landscape in complex landform. The grayscale gradient map was used to describe the grayscale difference between pixels in the image of superimposed region in garden landscape topography. Then, Vincent watershed algorithm was used to segment the gray gradient graph of the superimposed region of garden landscape topography, so that the contoured edge of the superimposed region was obtained. According to the relative position relationship of local extremum points, the feature vectors in the contour edge of superimposed region of garden landscape were extracted. Moreover, the distance between the feature vector and the feature vector in benchmark graph was calculated. The feature vector with the minimum distance was selected as the matching feature. Finally, 3 D reconstruction for superimposed region of garden topography landscape in complex landform was completed by matching feature points. Simulation results show that the proposed method has high reconstruction efficiency and high precision.
作者
叶津
YE Jin(Wuhan College,Hubei Wuhan 430212,China)
出处
《计算机仿真》
北大核心
2019年第11期207-210,共4页
Computer Simulation
关键词
杂地貌
园林地形景观
三维重建
Complex landform
Garden landscape
Three-dimensional reconstruction