【摘要】 对于山水林田湖草沙等重要的自然资源，传统的管理和监测模式往往存在效率低下、管理滞后的缺点，且容易忽略一些隐蔽的隐患区域，存在不容忽视的监测风险。在这一背景下，铁塔视频监测被认为具有广泛的视角和高度的远程观察优势具备实时、高清等特点，成为实现智能感知的关键手段。尤其是其中的自动定位技术，作为视频技术的核心，需要深入开展广泛的应用。自动定位技术的提高，不仅可以有效提高监管效率，降低人工投入，还不受气候影响。因此，在各行业的应用前景越来越广泛。本文以湖南省铁塔视频为例，提出了一种正射遥感影像辅助的监控视频地理定位方法，可以实现铁塔视频图像与正射影像之间的高精度坐标转换，解决了自然资源监管过程中精准定位困难的问题，使视频画面2 km范围内平原地区定位精度优于1 m,地形起伏较小的丘陵地区定位精度优于8 m。本文方法为铁塔视频监测的广泛应用提供了可行性方案。更多还原

【Abstract】 In the realm of essential natural resources such as “mountains, rivers, forests, fields, lakes, grass, and sands, ”conventional management and monitoring approaches often suffer from inefficiencies and lagging oversight. Moreover, these methodologies are prone to neglecting concealed hazard zones, thereby confronting an incontrovertible risk in monitoring. Against this backdrop, tower-mounted video surveillance is deemed to possess a comprehensive perspective and a heightened advantage in remote observation. Endowed with features such as real-time and high-definition capabilities, it emerges as a pivotal means for achieving intelligent perception. Particularly noteworthy is the automatic positioning technology within serving as the core of video technology, which necessitates extensive applications. Overcoming challenges associated with automatic positioning technology not only proves instrumental in significantly enhancing supervisory efficiency and reducing human intervention but also serves to transcend climatic impacts. Consequently, its application prospects span across various industries. Taking the case of tower-mounted video surveillance in Hunan province as an illustrative example, this paper proposes a video geo-localization method based on orthorectified remote sensing image. This method enables the accurate conversion of coordinates between tower-mounted video images and orthophoto images, effectively addressing the difficulty in precise positioning of natural resource regulation. The positioning accuracy of the plain area within 2 km of the video is better than 1 m, and the positioning accuracy of the hilly area with lees relief is better than 8 m.The proposed video geo-localization method outlined in this paper provides a feasible solution for the widespread application of tower-mounted video surveillance.更多还原