Tensor-based land use simulation method, system, device and storage medium

Abstract

The invention discloses a tensor-based land use simulation method, system, equipment and storage medium. The method includes: acquiring land use data, driving factor data, conversion cost table and inertial intensity table; training a convolutional neural network to obtain a fitting model, And further obtain the conversion probability tensor; perform the neighborhood influence calculation to obtain the neighborhood influence tensor; perform the conversion cost calculation to obtain the conversion cost tensor; perform the inertia coefficient calculation to obtain the inertia coefficient tensor; according to the conversion probability tensor and the neighborhood influence tensor Quantity, conversion cost tensor and inertial coefficient tensor to obtain the combination probability tensor; perform tensor roulette selection on the combination probability tensor to obtain the simulation result. The present invention formalizes the various components in the land use simulation process based on cellular automata into tensors, and uses the tensor frame to accelerate the operation, while avoiding complicated parallel algorithm design and CUDA programming, and realizing efficient Land use simulation process.

Description

[0001] technology neighborhood [0002] The invention relates to the field of geographical information science, in particular to a tensor-based land use simulation method, system, device and storage medium. Background technique [0003] With the further development of society, the concept of land planning also changes with the change of industrial structure; therefore, the study of the spatial and temporal change trend of land in a specific area has important reference significance for future planning. [0004] Cellular automata, also known as CA, as a space-time discrete dynamic model with strong spatial computing capabilities, plays an important role in complexity science; the multi-class land use simulation method based on cellular automata, according to Neighborhood state and a series of conversion rules to simulate the complex temporal and spatial changes of land use have become an important tool for studying land use change and its socio-economic and natural environmenta...

AI Technical Summary

Problems solved by technology

These methods can indeed improve the performance of the CA method, but the realization of these advanced algorithms requires more computer knowledge and experience, which is usually lacked by geographers

Images