基于无人机倾斜摄影技术的多源遥感影像变化检测并行系统设计
2023,31(2):91-96
摘要：为使无人机遥感图像的影像残差值得到有效控制，提升多源遥感影像变化特征的检测精度水平，设计基于无人机倾斜摄影技术的多源遥感影像变化检测并行系统。在C/S框架体系中，设置并行运作电路、像素点检测主机、HBase存储结构与遥感影像显示器，完成对多源遥感影像变化检测并行系统的硬件设计。根据联合平差指标的数值水平，计算密集度指标，联合已知影像数据，求解无人机倾斜摄影过程中的纹理映射条件，实现对多源遥感影像的建模处理。按照影像特征提取结果，完善影像检测金字塔的构型模式，将无人机数字影像与并行检测节点匹配起来，再结合各级硬件应用结构，完成基于无人机倾斜摄影技术的多源遥感影像变化检测并行系统设计。实验结果表明，设计的系统在无人机倾斜摄影技术的作用下，遥感图像在x轴、y轴、z轴方向上的影像残差指标均出现明显下降的数值变化状态，能够有效提升多源遥感影像变化特征的检测精度。
关键词：无人机倾斜摄影;多源遥感影像;联合平差;纹理映射;金字塔构型;影像残差;
Design of parallel system for multi-source remote sensing image change detection based on UAV tilt photography technology
Abstract：In order to effectively control the image residual value of UAV remote sensing image and improve the detection accuracy level of multi-source remote sensing image change characteristics, a multi-source remote sensing image change detection parallel system based on UAV tilt photography technology is designed. In the C/S framework system, the parallel operation circuit, pixel detection host, HBase storage structure and remote sensing image display are set up to complete the hardware design of the multi-source remote sensing image change detection parallel system. According to the numerical level of the joint adjustment index, the density index is calculated, and the known image data are combined to solve the texture mapping conditions in the process of UAV tilt photography, so as to realize the modeling and processing of multi-source remote sensing images. According to the image feature extraction results, improve the configuration mode of image detection pyramid, match the UAV digital image with the parallel detection node, and then complete the design of multi-source remote sensing image change detection parallel system based on UAV tilt photography technology in combination with the hardware application structure at all levels. The experimental results show that under the action of UAV tilt photography technology, the image residual indexes of remote sensing images in the x-axis, Y-axis and z-axis directions are significantly reduced, which can effectively improve the detection accuracy of change characteristics of multi-source remote sensing images.
Key words：UAV tilt photography; Multi-source remote sensing image; Joint adjustment; Texture mapping; Pyramid configuration; Image residual;
收稿日期：2022-06-29
基金项目：杨凌职业技术学院自然科学基金项目：“基于无人机倾斜摄影测量技术的快速智慧校园三维建模”（编号：A2019023）
