The invention relates to the technical field of
electric power operation and maintenance, in particular to an unmanned aerial vehicle self-positioning
bird nest target locking method based on
machine vision recognition, and aims to solve the problems of high manual dependence, low recognition precision, inaccurate positioning and high
safety risk in traditional bird damage prevention and control. The method comprises the following six steps: S1, planning a
route; s2, the unmanned aerial vehicle autonomously flies according to a
route and synchronously collects multi-view images; s3, image preprocessing and
distortion correction are carried out, a Faster R-CNN model is adopted to identify a suspected
nest, and secondary
verification is carried out in combination with morphology and thermal imaging features; s4, similar triangles, multi-view geometry and a PnP
algorithm are fused to solve three-dimensional coordinates of the
bird nest; s5, automatically locking the target, and returning information to the
ground control station; and S6, recording data and updating the model, and generating a bird damage risk thermodynamic diagram. The
bird nest recognition, positioning and locking full-process
automation is realized, the recognition precision and the
flight safety are improved, an intelligent operation and maintenance
closed loop is constructed, and the method is suitable for bird damage prevention and control of
power transmission lines of various
voltage grades.