Optimization of Urban Landscape Planning and Layout under Multicriteria Constraints

Abstract

The layout and planning of urban landscape has a strong correlation with urban land utilization rate and ecological environment index. Urban landscape architects have a hard time dealing with these interrelated factors. This study uses a multicriteria constraint algorithm to optimize the relevant factors in urban landscape layout and planning. The convolutional long short-term memory (ConvLSTM) method was used to extract temporal features for urban landscape layout and planning tasks. Compared with the multicriteria algorithm without constraints, the multicriteria algorithm with constraints can better optimize the layout and planning tasks of urban landscape, and the maximum error of this method is only 1.96%. At the same time, the distribution of errors is more uniform under the multicriteria constraints, and it is all within 2%. The fusion of the multicriteria constraint algorithm and the ConvLSTM algorithm can better predict the relevant factors of the urban landscape layout, and the linear correlation coefficients of the three relevant factors have reached a high standard.