Efficient and accurate analysis of large amounts of images is critical for water monitoring studies. This study proposes the use of open-source deep learning, RGB and HSL-based OpenCV Python code as a solution to overcome the limitations of manually analyzing large numbers of images by humans. Previous studies have already utilized these technologies for water monitoring and analysis, but which technology is most suitable and efficient is not yet clear. In this study, RGB-based images and HSL images, and we plan to analyze them using RGB-based OpenCV Python code, HSL-based OpenCV Python code, and deep learning techniques. Through this, we will evaluate which technology produces the most accurate results and suggest a direction for developing efficient program tools for water monitoring research. This will support more accurate and reliable data acquisition in the field of water monitoring. And if this technology develops further, it will be possible to detect not only natural river flows but also urban floods using various crowdsource within the city.