Research on UAV Image Recognition Based on Convolutional Neural Network

As an important visual function of drones, image recognition has helped drones accomplish a variety of tasks in complex environments. At present, most of drone image detection uses the traditional method of artificial feature extraction. It is difficult to meet the increasing accuracy requirements. Based on this, this paper applies convolutional neural networks to UAV image recognition to meet high-precision requirements. This article first introduced the development process of convolutional neural networks to facilitate their overall integration. And then summarized the “visual function” of the drone, then analyzed the future development trend of the drone image recognition technology, expounded the key technology of the drone image recognition, and finally explained the image recognition technology based on convolutional neural network, discussed its recognition process and recognition method, and it has a certain significance for deepening the research of UAV image recognition.