Building Area Measurement of UAV Remote Sensing Image Based on Parallel Convolution Neural Network
UAV remote sensing images are applied in various fields,it is difficult to distinguish between building and background areas,which leads to a decrease in the reliability of UAV remote sensing image building area measurement results.To solve this problem,a UAV remote sensing image building area measurement method based on parallel convolutional neural network is proposed.UAV remote sensing images are obtained and preprocessed through static output,image fusion,image dehazing and other steps.A parallel convolutional neural network is constructed to extract the edge features of the UAV remote sensing image building area in the preprocessed UAV remote sensing image through the network training and propagation,and the building area of the UAV remote sensing image is recognized through the feature matching.The measurement result of the building area is obtained by combining the area calculation result.After the precision performance testing experiments,it is concluded that compared with the traditional area measurement method,the proposed method reduces the measurement error of the building area by 0.505 km2 and 0.305 km2 in foggy and non-foggy environments,respectively,indicating that the measurement result reliability of this method is higher and can be widely used in the field of UAV remote sensing image building area measurement.
parallel convolution neural networkUAV measurementremote sensing imagebuilding area measurement
万方数据
维普
