Method for detecting abnormal state of flotation machine based on marker and visual algorithm

The invention discloses a method for detecting the abnormal state of a flotation machine based on a marker and a visual algorithm, and the method comprises the steps: additionally installing a mechanical marker at the edge of the flotation machine, and judging the state of the flotation machine according to whether the bending part of the marker is exposed or not, so that the manual judgment of the state of the flotation machine has better objectivity and consistency; the process is automated through machine vision in combination with the position information of the image features and the detection features of the markers, and the detection accuracy higher than that of a pure vision algorithm is obtained, so that the foam frame of the flotation machine can be more accurately detected under the condition that the foam frame of the flotation machine is polluted. And stable and effective detectable visual features can still be provided, and the method has higher accuracy and consistency than a pure visual algori