Research Of Identification Of Drought In Early Maize Based On Image Features

Image processing technology is used in image computing and processing technology.Under normal circumstances,the real image signal will be transferd into digital image signal by image processing technology,which uses the computer in related processing.At present,image processing has been widely used in many fields.The applications in agriculture mainly include pest and disease identification,the shape recognition of agricultural crops,the recognition of the maturity of crops and the detection of the state of crop growth.Some achievements have been made.At present,agriculture is facing a big bottleneck in terms of drought monitoring in both theory and technology.In order to overcome the drought of maize plants and the shortcomings of the existing technologies,this paper presents a drought recognition method based on multi-feature fusion.In this paper,the experiments on the five different drought level samples(normal,mild drought,moderate drought,heavy drought and Particularly drought)were conducted.The main work is:(1)In this paper,we use the clustering K-means ++ algorithm to segment the original maize image,and extract the maize leaf well,which lays a solid foundation for the subsequent feature extraction.(2)In order to describe the features of images more fully,this paper extracts three image features including color,SVD and texture,and combines the three features to construct the feature database.(3)In order to reduce the impact of the direct integration of the three features on the construction of the classification model,genetic algorithm is used to optimize the selection of the feature dataset.(4)Based on the optimal feature subset,a classification model based on least squares support vector machine is constructed.It can be seen from the recognition results that the method in this paper can effectively identify the drought situation of maize images.The results show that the proposed algorithm has a good effect on the discrimination of different grades of drought in pre-corn plants and provides a new idea for the identification of agricultural drought.