谷歌Chrome浏览器插件
订阅小程序
在清言上使用

Technical Note: Open‐source software for water‐level measurement in images with a calibration target

Water Resources Research(2022)

引用 3|浏览10
暂无评分
摘要
Image-based water level measurements offer data quality assurance through visual verification that no other method can provide. GaugeCam Remote Image Manager-Educational 2 (GRIME2) is a mature, open-source commercial friendly software application that automatically detects and measures water level in laboratory and field settings. The software relies on a dedicated target background for water line detection and image calibration. The system detects the change in pixel gray scale values associated with the intersection of the water level at the target surface. Fiducials on the target background are used to precisely create a pixel to real world coordinate transfer matrix and to correct for camera movement. The presented software package implements the algorithms and automates the water level measurement process, annotation of images with result overlays, creation of animations, and output of results to files that can be further analyzed in a spreadsheet or with R or Python. These GRIME2 features are illustrated using imagery from a coastal marsh field site. Tradeoffs between workflow and algorithm complexity and ease of use are discussed and future improvements are identified with the intention that this Findable, Accessible, Interoperable, and Reusable-inspired software can be adopted, modified and improved by the user community. While image resolution, quality and other factors associated with field deployment (e.g., water surface roughness, sun glare, shadows, and bio-fouling) will have an impact on measurement quality, previous controlled laboratory testing that did not manifest these issues showed potential for accuracy of +/- 3 mm (Gilmore et al., 2013, ).
更多
查看译文
关键词
stage measurement, image processing, open source software, hydrology, pixel to world calibration, waterline detection
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要