SCECA U-Net Crop Classification for UAV Remote Sensing Image

Accurately identifying crop distribution is beneficial for assessing crop growth,yield,and disasters. Crop identification is influenced by terrain and complex ecological environments,considering the challenges posed by traditional remote sensing methods such as cost,sensitivity to westher conditions, and limitations in accurately extracting crop festures.Therefore,we selected unmanned aerial vehicle(UAV) remote sensing images of Majiaoba Town, Mianyang City,Sichuan Province,China,aiming to identify and classify various crops in the region,including soybeans,corn,rice,ginger and walunts.We propose using theVGG-16 (Visual Geography Group Network-16) architecture as the main framework,enhancing it by integrating efficient channel ateention(ECA) modules into the U-Net model.It is worth noting that the improved model achieved an impressive F1-Score of 0.78,emphasizing its effectiveness in crop classification. Experimental results show that the method achieves a comprehensive classification accuracy of 79.44% and an average cross-combination rate of 0.64.Our approach outperforms traditional U-Net,PSPNet(Pyramid Scene Parsing Network),and deeplabv3+ models in terms of classification accuracy and metrics, highlighting its advantages in crop identification.