Unmanned aerial vehicle small target detection method

• Main Authors: NIU BEN, XIN GAOPENG, HE LONG, MAO XIWANG, SHANG PENGHUI, AI BAIYUN, LIU KE
• Format: Patent
• Language: Chinese; English
• Published: 17.11.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention relates to an unmanned aerial vehicle small target detection method. The method comprises the steps of collecting an image set inspected by an unmanned aerial vehicle as a training set; preprocessing the data set; target objects in the data set are labeled, label text files are stored, and the label files comprise normalized coordinate values of target frame coordinates and target categories; dividing the labeled data set into a training set, a verification set and a test set in proportion; building a target detection model of the small target, training the model by using the training set, comparing a real result with a prediction result, calculating a loss function, and updating a network weight; and performing target detection on the data acquired by the unmanned aerial vehicle by using the target detection model. The method has the beneficial effects that a special model which is more suitable for small target detection of the visual angle of the unmanned aerial vehicle is designed by providi