 The proposed convolutional neural networks, CNN, combined feature engineering forecasting model, SFCNN, improves the accuracy of NDVI predictions by integrating image characteristics learned from CNN and statistic characteristics calculated by historical data related to the forecast period, resulting in better generalization ability, reliability, and robustness for predicting NDVI through simple statistical characteristics while reducing uncertainties. SFCNN can learn seasonal and southern changes in four different and complex study areas with considerable accuracy without extra data. This article was authored by Chang Lu Kui, Wen Zhong, Ji Ming Hong, and others.