WLTOG: An optimization approach for wild large-range target omnidirectional geolocation based on monocular PTZ camera

As an omnidirectional monocular camera, the pan-tilt-zoom (PTZ) camera has been widely employed in urban governance, forest fire protection, traffic supervision due to its omnidirectional coverage and inexpensive cost. Traditional monocular camera geolocation research focuses on small-range geolocation with a central projection camera under ideal conditions, making it difficult to meet the requirements of omnidirectional coverage and large-range geolocation for PTZ cameras. In view of this, this paper proposed a wild large-range target omnidirectional geolocation (WLTOG) algorithm. Firstly, the WLTOG algorithm transforms the target in a single image to the camera's visual center, obtaining the camera posture value. Then, it fits the nonlinear mapping relationship between multiple pairs of PTZ camera postures and the geospatial coordinates by using high-dimensional overdetermined equation, and finally realizes the bidirectional calculation between pixel coordinates and geospatial coordinates. The experimental PTZ camera is mounted on a communication tower in wild hilly environment. This method can be used to construct an omnidirectional geolocation relationship that meets the requirements of omnidirectional geolocation within 945m based on unknown accurate internal and external orientation elements, and low requirements for wild hardware installation. The WLTOG algorithm is conducive to the geometric measurement of a single image, GIS-augmented video surveillance, and 3D object recognition.

作为一种全方位单目相机，云台变焦（pan-tilt-zoom, PTZ）相机凭借其全方位覆盖特性与低廉的部署成本，已被广泛应用于城市治理、森林消防、交通监管等领域。传统单目相机地理定位研究多聚焦于理想条件下中心投影相机的小范围定位场景，难以满足PTZ相机对全方位覆盖与大范围地理定位的需求。鉴于此，本文提出了野外大范围目标全方位地理定位（wild large-range target omnidirectional geolocation, WLTOG）算法。该算法首先将单幅图像中的目标转换至相机视场中心，以此获取相机位姿参数；随后通过高维超定方程拟合多组PTZ相机位姿与地理空间坐标间的非线性映射关系，最终实现像素坐标与地理空间坐标的双向计算。本次实验的PTZ相机部署于野外丘陵环境中的通信铁塔之上。该方法无需精确的内外方位元素先验知识，且对野外硬件安装的要求较低，可构建945米范围内满足全方位地理定位需求的定位关系。WLTOG算法可应用于单幅图像几何测量、地理信息系统（GIS）增强型视频监控以及三维目标识别等场景。