 The study proposes a novel object-based hybrid classification model named GMNN that combines grasshopper optimization algorithm, GOA, and multiple-class neural network, MNN, for urban pattern detection in Hanoi, Vietnam. The model uses four bands of spot-7 image and a rideable NDVI, NDWI to classify urban surface types into water, impervious surface, vegetation, and bare soil with an overall accuracy of approximately 87%. The results show that GMNN outperforms established methods in all comparable indicators, suggesting that the hybrid model can be used as an alternative classification method for urban land cover studies. This article was offered by Kwong Ton Bui, Man Pham Van, Nguyen Thi Tui Hang, and others.