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, https://doi.org/10.1016/j.jhydrol.2013.05.011).

中文翻译：

---

技术说明：具有校准目标的图像中水位测量的开源软件

基于图像的水位测量通过视觉验证提供其他方法无法提供的数据质量保证。GaugeCam Remote Image Manager-Educational 2 (GRIME2) 是一款成熟的开源商业友好型软件应用程序，可自动检测和测量实验室和现场环境中的水位。该软件依靠专用的目标背景进行水线检测和图像校准。系统检测与目标表面水位交叉点相关的像素灰度值变化。目标背景上的基准用于精确地创建像素到现实世界的坐标传递矩阵并校正相机移动。所提出的软件包实现了算法并使水位测量过程自动化，使用结果叠加对图像进行注释、创建动画以及将结果输出到文件中，这些文件可以在电子表格中或使用 R 或 Python 进行进一步分析。这些 GRIME2 特征使用来自沿海沼泽场地的图像进行说明。讨论了工作流和算法复杂性和易用性之间的权衡，并确定了未来的改进，目的是让用户社区可以采用、修改和改进这种可查找、可访问、可互操作和可重复使用的软件。虽然与现场部署相关的图像分辨率、质量和其他因素（例如，水面粗糙度、太阳眩光、阴影和生物污染）将对测量质量产生影响，但之前没有表现出这些问题的受控实验室测试显示出潜在的±3 mm 的精度（Gilmore 等人，