A Fine Crop Classification Model Based on Multitemporal Sentinel-2 Images
Tengfei Qu,Hong Wang,Xiaobing Li,Dingsheng Luo,Yalei Yang,Jiahao Liu,Yao Zhang
DOI: https://doi.org/10.1016/j.jag.2024.104172
2024-01-01
Abstract:Information on the sowing areas and yields of crops is important for ensuring food security and reforming the agricultural modernization process, while crop classification and identification are core issues when attempting to acquire information on crop planting areas and yields. Obtaining information on crop planting areas and yields in a timely and accurate manner is highly important for optimizing crop planting structures, formulating agricultural policies, and ensuring national economic development. In this paper, a fine crop classification model based on multitemporal Sentinel-2 images, CTANet, is proposed. It comprises a convolutional attention architecture (CAA) and a temporal attention architecture (TAA), incorporating spatial attention modules, channel attention modules and temporal attention modules. These modules adaptively weight each pixel, channel and temporal phase of the given feature map to mitigate the intraclass spatial heterogeneity, spectral variability and temporal variability of crops. Additionally, the auxiliary features of significant importance for each crop category are identified using the random forest-SHAP algorithm, enabling the construction of classification datasets containing spectral bands, spectral bands with auxiliary features, and spectral bands with optimized auxiliary features. Evaluations conducted on three crop classification datasets revealed that the proposed CTANet approach and its key CANet component demonstrated superior crop classification performance on the classification dataset consisting of spectral bands and optimized auxiliary features in comparisons with the other tested models. Based on this dataset, CTANet achieved higher validation accuracy and lower validation loss than those of the other methods, and during testing, it attained the highest overall accuracy (93.9 %) and MIoU (87.5 %). When identifying rice, maize, and soybeans, the F1 scores of CTANet reached 95.6 %, 95.7 %, and 94.7 %, and the IoU scores were 91.6 %, 91.7 %, and 89.9 %, respectively, significantly exceeding those of some commonly used deep learning models. This indicates the potential of the proposed method for distinguishing between different crop types.