Method for identifying ecological flow discharge through video

The invention discloses a method for identifying ecological flow discharge through a video. The method comprises the following steps: extracting a section of ecological flow discharge video, performing frame extraction on the video to obtain n images, further comparing the images of the front and back frames to obtain a difference index of the front and back frames, and judging whether continuous moving water flow exists in the video according to the difference index so as to judge whether ecological flow discharge exists. The main algorithm is high in universality, precision and robustness, a fuzzy verification function is added, and the abnormal conditions of shadow, dam overflow, water mist, sunlight irradiation and video packet loss can be well eliminated. For a small number of special and inapplicable scenes, such as weed interference and water ripple interference, convolutional neural network training can be adopted to obtain a sub-model, and ecological flow discharge identification is performed through