Cultivated Land Extraction from High Resolution Remote Sensing Image Based on Convolutional Neural Network

被引：0
作者：
Chen L. [1 ,2 ]
Shi Z. [1 ,2 ]
Liao K. [1 ,2 ]
Song Y. [1 ,2 ]
Zhang H. [3 ]
机构：
[1] Jiangxi Academy of Water Science and Engineering, Nanchang
[2] Jiangxi Key Laboratory of Soil Erosion and Prevention, Nanchang
[3] School of Hydraulic and Ecological Engineering, Nanchang Institute of Technology, Nanchang
来源：
Nongye Jixie Xuebao/Transactions of the Chinese Society for Agricultural Machinery | 2022年 / 53卷 / 09期
关键词：
convolutional neural network; cultivated land; extraction; high resolution remote sensing image; LWIBNet;
D O I：
10.6041/j.issn.1000-1298.2022.09.017
中图分类号：
学科分类号：
摘要：
It is of great significance for agricultural resources monitoring to accurately extract cultivated land map information from remote sensing images. To improve the defects of traditional models for extracting cultivated land and solve the problem that most FCN model pays more attention to accuracy but ignores the consumption of time and computing resources, a lightweight model for extracting cultivated land map spots was established based on FCN (L W I B N e t), and post-processing combined with mathematical morphology algorithm were used to carry out automatic extraction of cultivated land information. LWIBNet drew on the advantages of lightweight convolutional neural network and U - Net model, and it was built with the core of Inv - Bottleneck (composed of deep separable convolution, compression - excitation block and inverse residual block) and the skeleton of efficient coding - decoding structure. Compared LWIBNet with the cultivated land extraction effect of traditional model, and the computational resources and time consumption of classical FCN model. The results showed that LWIBNet was 12. 0% higher than the Kappa coefficient of the best traditional model, and compared with U - Net, LWIBNet had 96. 5%, 87.1%, 78. 2% and 75% less parameters, calculation, training time and split time-consuming, respectively. Moreover, the segmentation accuracy of LWIBNet was similar to that of the classical FCN model. © 2022 Chinese Society of Agricultural Machinery. All rights reserved.
引用
收藏
页码：168 / 177
页数：9
相关论文
共 38 条
  • [21] acomparative study, 2021 International Conference on Digital Futures and Transformative Technologies (ICoDT2), pp. 1-6, (2021)
  • [22] NING Jifeng, NI Jing, HE Yijia, Et al., Recognition ol film-covered farmland by UAV multispectral remote sensing images based on convolution attention, Transactions of the Chinese Society for Agricultural Machinery, 52, 9, pp. 213-220, (2021)
  • [23] IANDOLA F N, HAN S, MOSKEWICZ M W, Et al., SqueezeNet
  • [24] AlexNet-level accuracy with 50x fewer parameters and < 0. 5 MB model size, pp. 1-13, (2016)
  • [25] CHOLLET F., Xception
  • [26] deep learning with depthwise separable convolutions, 2017 IEEE Conference on Computer Vision and Pattern Recognition (C V P R), pp. 1251-1258, (2017)
  • [27] ZHANG X, ZHOU X, LIN M, Et al., ShuffleNet: an extremely efficient convolutional neural network for mobile devices [J], pp. 1-9, (2017)
  • [28] MA N, ZHANG X, ZHENG H T, Et al., ShuffleNet V 2: practical guidelines for efficient CNN architecture design, European Conference on Computer Vision, pp. 122-138, (2018)
  • [29] HOWARD A G, ZHU M, CHEN B, Et al., Mobilenets
  • [30] efficient convolutional neural networks for mobile vision applications [J], pp. 1-9, (2017)
1234