The invention discloses a water hyacinth recognition system and method based on low-altitude remote sensing images and deep learning. The system comprises an unmanned aerial vehicle, a positioning device, a communication device, a control system and a calculation processing unit, and the positioning device is connected to a Beidou satellite positioning system and is used for obtaining latitude andlongitude coordinate information of the unmanned aerial vehicle. The communication device is used for wirelessly communicating with the unmanned aerial vehicle and receiving a ground remote sensing image sent by the unmanned aerial vehicle; and the calculation processing unit is used for adding a unique geographic coordinate identifier to each pixel of each ground remote sensing image. Accordingto the invention, the unmanned aerial vehicle carries out low-altitude remote sensing and high-resolution aerial photography; the water hyacinth is automatically identified based on Tensorflow computer deep learning and a Beidou high-precision positioning technology by using an unmanned aerial vehicle high-resolution image, the space-time dynamic identification and the positioning of the invaded species of the water hyacinth are constructed, and the real-time and effective monitoring of the water hyacinth is facilitated.