Building recognition and automated monitoring of occupying farmland for construction based on deep learning：Taking X Village in Changsha City, Hunan Province as an example

doi:10.14088/j.cnki.issn0439-8114.2024.01.035

Abstract

Abstract: In response to the common phenomenon of occupying farmland for building houses in rural areas, an automated monitoring method based on deep learning and image analysis was proposed. By preprocessing high-resolution remote sensing image data, an automated monitoring model based on convolutional neural networks was constructed to effectively determine whether each pixel cell in the target image occupied farmland for building houses. Taking X Village in Changsha City, Hunan Province as an example, horizontally compared the recognition capabilities of U-Net, SegNet, and DeepLabV3p models were. The results showed that when the learning rate was 0.01, the batch size was 2, and the number of iterations was 100, the U-Net model had the best recognition results for buildings;the model found a total of 66 cases of potential occupation of farmland for building houses, with high recognition accuracy and less time consumption;this model fully utilized modern information technology and methods, which could improve the efficiency of land law enforcement and supervision to a certain extent and save work time and resources.

Key words: deep learning, U-Net model, automated monitoring, building recognition, occupation of farmland, land enforcement, Changsha City,Hunan Province

CLC Number:

P237

SHI Shan, HU Bing, YANG Cong-rui. Building recognition and automated monitoring of occupying farmland for construction based on deep learning：Taking X Village in Changsha City, Hunan Province as an example[J]. HUBEI AGRICULTURAL SCIENCES, 2024, 63(1): 195-198.

References

[1] 牛善栋,方斌.中国耕地保护制度70年:历史嬗变、现实探源及路径优化[J].中国土地科学, 2019, 33(10): 1-12.
[2] 卢绍婷,周俊晖.农村乱占耕地建房整治方法研究与系统实现[J].国土资源信息化, 2021(2): 48-53.
[3] LI M,STEIN A,KMD B.A bayesian characterization of urban land use configurations from VHR remote sensing images[J]. International journal of applied earth observation and geoinformation, 2020, 10:102175.
[4] 陈琳,刘清华.城镇和农村住宅用地使用现状及整治建议——以湖北省钟祥市为例[J]. 中国土地, 2020(5): 52-53.
[5] 吴家杰,黄霞,刘锟铭.农村乱占耕地建房疑似违法图斑自动提取研究[J].江西测绘, 2020(4):37-38, 41.
[6] 夏晓鸿,潘俊国,董海潮,等.全省农村土地综合监管机制创新调研报告[J].浙江国土资源, 2018(11): 35-37.
[7] 周建国. 农村耕地非法转为宅基地的问题与对策[J].贵州农业科学, 2010, 38(2): 183-188.
[8] 王丹,张璐,柴燕妮.基于深度学习的乱占耕地建房疑似图斑自动提取方法研究[J].测绘与空间地理信息,2022,45(6): 51-53,57.
[9] YU H, YANG Z, TAN L, et al.Methods and datasets on semantic segmentation: A review[J]. Neurocomputing, 2018, 304, 82-103.
[10] CROMMELINCK S,KOEVA M,YANG M Y,et al.Application of deep learning for delineation of visible cadastral boundaries from remote sensing imagery[J]. Remote sensing, 2019,11(21): 2505.
[11] LECUN Y, BENGIO Y, HINTON G.Deep learning[J]. Nature, 2015, 521:436-444.
[12] LONG J,SHELHAMER E,DARRELL T.Fully convolutional networks for semantic segmentation[J]. IEEE transactions on pattern analysis and machine intelligence, 2015,39(4): 640-651.
[13] PERSELLO C, STEIN A.Deep fully convolutional networks for the detection of informal settlements in VHR images[J]. IEEE geoscience and remote sensing letters, 2017,14(12): 2325-2329.
[14] 范荣双,陈洋,徐启恒,等.基于深度学习的高分辨率遥感影像建筑物提取方法[J].测绘学报, 2019, 48(1):34-41.
[15] 季顺平,田思琦,张驰.利用全空洞卷积神经元网络进行城市土地覆盖分类与变化检测[J].武汉大学学报(信息科学版), 2020, 45(2):233-241.
[16] 王舒洋,慕晓冬,杨东方,等.融合高阶信息的遥感影像建筑物自动提取[J].光学精密工程, 2019, 27(11):2474-2483.
[17] LIU R, KUFFER M, PERSELLO C.The temporal dynamics of slums employing a CNN-based change detection approach[J]. Remote sensing, 2019, 11(23):2844.
[18] XIA X, PERSELLO C, KOEVA M.Deep fully convolutional networks for cadastral boundary detection from UAV images[J]. Remote sensing, 2019, 11(14):1725.