Image Recognition of Rice Growth Period Based on Convolutional Neural Network

In the field of crop image detection, traditional detection methods for the growth period of related crops rely heavily on humans, resulting in relatively low accuracy. At present, the mainstream spectral detection technology based on remote sensing images of crops has high hardware requirements for image acquisition and a complex system. In order to realize a crop growth period image detection method with high accuracy, convenient operation, and fast speed. A crop convolutional neural network (Crop-Net) is proposed to train and test rice images at different growth stages. Experimental results show that the proposed detection algorithm has higher detection accuracy in the detection task of ten growth periods of rice.