Land-use/land-cover classification method under spatial constraint

A classification method under spatial constraint was proposed. Based on the prior knowledge such as digital elevation model and land use database, spatial constraint relationships were constructed; prior knowledge and spectral information were syncretized. The results showed that the method was resistant to the problem of having apparent same spectrum but different ground covers and was capable of removing salt and pepper noise. The result also proves the importance of spatial information in remote sensing image classification and the effectiveness of classification method under spatial constraint.