Paddy field segmentation method combining attention mechanism and spatial feature fusion algorithm

Abstract

The invention relates to a paddy field segmentation method combining an attention mechanism and a spatial feature fusion algorithm, and the method comprises the steps: constructing a low-altitude paddy field image data set, marking the data in the data set, and dividing the data into a training set and a verification set; building a semantic segmentation network model, and training the semantic segmentation network model based on the training set to obtain a trained semantic segmentation network model; and verifying the trained semantic segmentation network model based on the verification set to obtain a verification result. The paddy field prediction result is more efficient and accurate, an important basis is provided for further obtaining high-precision paddy field boundary positioning information and constructing a high-precision map of a plurality of paddy fields in a large area, and the method plays a positive role in promoting efficient and accurate paddy field cultivated land informatization management.

Description

technical field [0001] The invention relates to the technical field of smart agriculture, in particular to a paddy field segmentation method combining an attention mechanism and a spatial feature fusion algorithm. Background technique [0002] Smart agriculture is an inevitable trend in the development of modern agriculture. In order to help agricultural machinery better understand the farmland environment and realize smart agricultural machinery operations, it plays an important role in promoting the development of smart agriculture. At present, there are many farmland image data based on high-resolution UAV low-altitude remote sensing It is a large-scale farmland extraction in plain areas, and it is difficult to extract paddy fields in hilly areas and the interpretation accuracy is relatively low. The present invention provides a deep neural network structure of SA-DeepLabv3+ that combines attention mechanism and spatial feature fusion algorithm to realize the extraction o...

AI Technical Summary

Problems solved by technology

[0002] Smart agriculture is an inevitable trend in the development of modern agriculture. In order to help agricultural machinery better understand the farmland environment and realize smart agricultural machinery operations, it plays an important role in promoting the development of smart agriculture. At present, there are many farmland image data based on high-resolution UAV low-altitude remote sensing For the extraction of large-scale farmland in plain areas, it is difficult to extract paddy fields in hilly areas and the interpretation accuracy is relatively low

Images