Spatially Heterogeneous Cropland Characteristic Extraction Based on Improved Semi-supervised Models

CHEN Li, HAN Yi, YANG Guang, LAI Youchun, ZHENG Yongjun, ZHOU Yuguang

Abstract

Extracting cropland accurately and efficiently from high-resolution remote sensing images is of great significance to agricultural production and agricultural resource investigation. Cropland with different areas, ground covers and cultivation types in remote sensing images have large differences in features, whereas the insufficient generalization ability of traditional supervised learning models also leads to poor extraction of heterogeneous cropland with the above features. To solve this problem, the semi-supervised semantic segmentation with mutual knowledge distillation (SSS-MKD) model as the base model and incorporated an online hard example mining strategy based on a weighted loss function. The proposed model was evaluated on the Vaihingen dataset and achieved the highest overall accuracy of 87.1% and an average F1 score of 85.0%, The model had the best extraction accuracy compared with other semi-supervised models. In addition, for the task of large-area cropland extraction, the feature information of the unannotated images that were heterogeneous and homogeneous with the annotated images were added to the training of the semi-supervised learning model by designing two sets of experiments using Jilin-1 cropland image dataset, respectively, in order to improve the cropland extraction accuracy in the proposed extraction area. The experimental results showed that the proposed model could achieve the highest overall accuracy of 84.0% by using the cropland images to be extracted for assisted training. In addition, by using unlabeled images with strong similarity to the cropland in the target region, the overall accuracy could be further improved by 2.1~6.1 percentage points. The maximum overall accuracy achieved using unlabeled images with strong similarity to the cropland features in the training set was 81.6%, which was 6.6~8.5 percentage points higher than the accuracy achieved by using conventional supervised learning. Taking Xian County area in Hebei Province as an example, the model used the Jilin-1 cropland image dataset (part 1) as the labeled training set, and achieved the highest accuracy of 88.7% overall accuracy after training with both the Jilin-1 cropland image dataset (part 2) and the unlabeled images from the Xian County areas Gaofen-2 image dataset, compared with the extraction accuracy by using these two datasets alone and without the unlabeled images, respectively 1.0 percentage points, 4.9 percentage points, and 8.1 percentage points. Under the realistic background of abundant cropland remote sensing data and insufficient corresponding labeled information. The semi-supervised model learned the feature information in the images of cultivated land in the proposed extraction area and the training set area, and improved the extraction accuracy of heterogeneous cultivated land. A method that maximized the learning of heterogeneous cropland unlabeled images and a minimum amount of labeled data using a semi-supervised learning model for the task of wide-area cropland extraction was proposed, and the extraction results were effective.


Keywords: cropland characteristic extraction, semantic segmentation, semi-supervised learning, spatially heterogeneous cropland, remote sensing image

 

Download Full Text:

PDF


References


DEBATS S R, LUO D, ESTES L D, et al. A generalized computer vision approach to mapping crop fields in heterogeneous agricultural landscapes [ J ]. Remote Sensing of Environment, 2016, 179: 210 -221.

BELGIU M, CSILLIK 0. Sentinel-2 cropland mapping using pixel-based and object-based time-weighted dynamic time warping analysis [J] . Remote Sensing of Environment, 2018, 204; 509 -523.

SYKAS D, SDRDAKA M, ZOGRAFAKIS D, et al. A Sentinel — 2 multiyear, multicountry benchmark dataset for crop classification and segmentation with deep learning [ J ]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15: 3323 -3339.

RAEI E, ASANJAN A A, NIKOO M R, et al. A deep learning image segmentation model for agricultural irrigation system classification[ J ]. Computers and Electronics in Agriculture, 2022, 198; 106977.

ZHANG Chao, GAO Lulu, YUN Wenju, et al. Research progress on obtaining cultivated land quality evaluation indexes by remote sensing [J] . Transactions of the Chinese Society for Agricultural Machinery, 2022,53( 1 ) : 1 - 13. ( in Chinese)

WANG Pengxin, TIAN Huiren, ZHANG Yue, et al. Crop growth monitoring and yield estimation based on deep learning; state of the art and beyond [ J]. Transactions of the Chinese Society for Agricultural Machinery, 2022,53(2) ;1 - 14. (in Chinese)

CHEN Li, YANG Guang, LIU Mingyang, et al. Extraction of field and road information based on remote sensing image and convolution neural network [ J ]. Journal of China Agricultural University, 2022,27(6) : 182 - 191. (in Chinese)

LU Heng, FU Xiao, HE Yi'nan, et al. Cultivated land information extraction from high resolution UAV images based on transfer learning [J ]. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46( 12) :274 -279,284. ( in Chinese)

YANG Siqi. Scene classification of high resolution remote sensing images based on transfer learning [ D] . Beijing: China University of Geosciences, 2021. (in Chinese)

ZHANG Shuyu. High-resolution remote sensing image land cover classification based on deep learning and multi-scale and multi-feature fusion [D] . Hangzhou; Zhejiang University,2020. (in Chinese)

YUAN J, LIU Y, SHEN C, et al. A simple baseline for semi-supervised semantic segmentation with strong data augmentation [С] // Proceedings of the 18th IEEE/CVF International Conference on Computer Vision, 2021 : 8229 -8238.

CUYPERS S, BASSIER M, VERGAUWEN M. Deep learning on construction sites: a case study of sparse data learning techniques for rebar segmentation[ J]. Sensors, 2021 , 21 ( 16) ; 5428.

CHEN X, YUAN Y, ZENG G, et al. Semi-supervised semantic segmentation with cross pseudo supervision [С] // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021 ; 2613 -2622.

LEE D H. Pseudo-label; the simple and efficient semi-supervised learning method for deep neural networks[ J]. Workshop on Challenges in Representation Learning, 2013, 3(2) : 896 -902.

TENG Wenxu, W ANG Ni, CHEN Taisheng, et al. Deep adversarial domain adaptation method for cross-domain classification in high-resolution remote sensing images [j]. Laser & Optoelectronics Progress ,2019, 56( 1 1 ) -.236 -246. (in Chinese)

WEN Yimin, YUAN Zhe, YU Hang, et al. A new semi-supervised inductive transfer learning framework; Co —Transfer [J] . Journal of Computer Research and Development ,2023 .60 (7 ) ; 1603 - 1614. (in Chinese)

W ANG W . Semi-supervised semantic segmentation network based on knowledge distillation [ С ] / 2021 IEEE 4th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC). IEEE, 2021; 1900 -1905.

TARVAINEN A, VALPOLA H. Mean teachers are better role models; weight-averaged consistency targetsimprove semi-supervised deep learning results[J]. Advances in Neural Information Processing Systems, 2017, 30;30 -40.

CHEN X, Y UAN Y, ZENG G, et al. Semi-supervised semantic segmentation with cross pseudo supervision [С] // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021 ; 2613 -2622.

HUANG Hai. Research on extraction methods of agricultural land based on high-resolution remote sensing images [ D ]. Guangzhou; Guangzhou University, 2021. (in Chinese)

TAGHANAKI S A, ZHENG Y, ZHOU S К, et al. Combo loss; handling input and output imbalance in multi-organ segmentation[ J ]. Computerized Medical Imaging and Graphics, 2019, 75; 24 -33.

SZEGEDY C, VANHOUCKE V, IOFFE S, et al. Rethinking the inception architecture for computer vision [С] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016; 2818 -2826.


Refbacks

  • There are currently no refbacks.