Convolutional neural network-based unmanned air vehicle to-ground specific target recognition method

Abstract

The invention relates to a convolutional neural network-based unmanned air vehicle to-ground specific target recognition method. The method comprises the following steps of: (1) labeling to-ground specific target data sets acquired by an unmanned air vehicle according to categories, and dividing the to-ground specific target data sets into training sets, verification sets and test sets according to proportions; (2) setting convolutional neural network model training parameters, starting training and obtaining an optimal solution of a convolutional neural network model according to the trainingcondition; (3) changing a depth of the convolutional neural network model, restarting the training in the step (2) to obtain an optimal convolutional neural network model; (4) testing the test set soas to obtain recognition correctness; and (5) applying convolutional neural network model parameters, correctness of which satisfies requirements, to a practical scene of an unmanned air vehicle to-ground specific target, so as to recognize an image target acquired by the unmanned aerial vehicle.

Description

technical field [0001] The invention relates to a method for recognizing a specific target by an unmanned aerial vehicle based on a convolutional neural network, and belongs to the cross field of aviation and computer vision information processing. Background technique [0002] UAV (Umanned Air Vehicle, UAV) is an unmanned aircraft operated by radio remote control equipment and self-contained program control device, or operated completely or intermittently by the on-board computer. UAVs can be divided into military and civilian use according to the application field. In terms of military use, UAVs are divided into reconnaissance aircraft and target aircraft. In terms of civilian use, drones + industry applications are the real rigid needs of drones; currently, they are used in aerial photography, agriculture, plant protection, micro selfies, express delivery, disaster relief, wildlife observation, infectious disease monitoring, surveying and mapping, news reports, power pat...

AI Technical Summary

Problems solved by technology

[0006] The technology of the present invention solves the problem: Overcoming the problem of insufficient recognition accuracy of the traditional method for ground target recognition of UAVs and the difficulty in satisfying the training data set of the deep learning method, it provides a ground-specific UAV based on convolutional neural network. The target recognition method adjusts the depth of the network model according to the size of the ground target data set collected by the UAV to achieve higher recognition accuracy, thus providing an effective way for the UAV to identify specific targets on the ground

Images