The automated video-oriented plankton tracking workflow (AVPTW) utilizes deep learning for efficient and continuous monitoring of plankton abundance in water environments. It is sensitive to environmental changes and remains robust when tested with contaminated and uncontaminated water samples. This integration of imaging devices and deep learning provides a replicable paradigm for long-term online environmental monitoring and understanding enviromental indicators.