Waterfront Recreational Landscape Planning and Ecological Protection Based on Cloud Computing and Neural Network Evaluation

Abstract

With the improvement of material quality of life and the acceleration of daily rhythm, more and more people hope to enjoy the rare leisure. Since then, a number of dynamic prediction models have been put forward by scholars at home and abroad, including representative models, such as numerical model, grey system model, catastrophe theory model, time series model, neural network model and so on. In the case of the adaptation of the test parameters, all kinds of prediction models can play a good performance. Cloud computing and neural network analysis can effectively provide evaluation accuracy. This paper puts forward the principles, strategies and methods of landscape planning based on the knowledge of plant community, ecology and aesthetics. This paper analyses the ecological allocation of waterfront recreational sites in the Pearl River Delta region and makes SBE evaluation. The results show that the bigger the SBE average value of landscape allocation is, the more popular it is. It can reflect the best ecological model which is in line with both ecology and aesthetics. It can provide scientific basis and technical reference for plant landscape construction of waterfront recreational sites.