Abstract: Accurate quantification of crop residue cover (CRC) is crucial for monitoring and evaluating conservation tillage practices, yet it poses a significant image segmentation challenge. The subtle visual distinctions between fragmented residue and soil, compounded by variable illumination and shadows in field imagery, often lead to poor segmentation performance. To overcome these limitations, we introduce RCTUnet, a novel deep learning architecture designed for robust crop-residue-soil segmentation and precise CRC estimation. RCTUnet’s architecture synergistically integrates three key components: (1) a ResNet50 backbone for deep, multi-scale feature extraction; (2) a convolutional block attention module (CBAM) to adaptively focus on salient residue features across both channel and spatial dimensions; and (3) a transformer-based global context fusion module (GCFM) to model long-range spatial dependencies, which is critical for interpreting heterogeneous residue patterns. We evaluated RCTUnet on a dataset of 1220 field-acquired images spanning four typical crop rotations. Experimental results show that, compared to traditional models: (1) RCTUnet achieves significantly higher crop-residue-soil segmentation accuracy than classic models including Unet, Unet++, DeepLabV3, segmentation network (SegNet), and fully convolutional network (FCN), with improvements of 3.24%, 3.42%, 4.88%, 8.28%, and 6.05% in overall accuracy, respectively; (2) RCTUnet yields superior residue-soil segmentation performance, with increases in residue recall of 7.67%, 7.37%, 14.09%, 27.05%, and 16.91%, respectively; (3) RCTUnet shows enhanced CRC estimation accuracy, achieving a root mean square error (RMSE) of 4.875, representing a 45.5% improvement over Unet (RMSE=8.941). These results demonstrate the efficacy of our hybrid approach, which combines deep hierarchical features, dual-domain attention, and global context modeling. RCTUnet provides a robust and reliable tool for automated CRC assessment, advancing the capabilities of in-field agricultural monitoring.

RCTUnet：一种用于作物-残茬-土壤图像分割与秸秆覆盖度提取的深度学习模型

[1]BeecheC, SinghJP, LeaderJK, et al., 2022. Super U-Net: a modularized generalizable architecture. Pattern Recognit, 128:108669.

[2]BroschT, TangLYW, YooY, et al., 2016. Deep 3D convolutional encoder networks with shortcuts for multiscale feature integration applied to multiple sclerosis lesion segmentation. IEEE Trans Med Imaging 35(5):1229-1239.

[3]DelandmeterM, ColinetG, PierreuxJ, et al., 2024. Combining field measurements and process-based modelling to analyse soil tillage and crop residues management impacts on crop production and carbon balance in temperate areas. Soil Use Manage, 40(3):e13098.

[4]de Obade Paul ObadeV, GayaC, 2020. Mapping tillage practices using spatial information techniques. Environ Manage, 66(4):722-731.

[5]DingYL, ZhangHY, WangZQ, et al, 2020. A comparison of estimating crop residue cover from Sentinel-2 data using empirical regressions and machine learning methods. Remote Sens, 12(9):1470.

[6]DuL, LuZC, LiDL, 2022. Broodstock breeding behaviour recognition based on Resnet50-LSTM with CBAM attention mechanism. Comput Electron Agric, 202:107404.

[7]FengYG, HanB, WangXC, et al., 2024. Self-supervised transformers for unsupervised SAR complex interference detection using canny edge detector. Remote Sens, 16(2):306.

[8]GaoGF, ZhangSX, ShenJN, et al., 2024. Segmentation and proportion extraction of crop, crop residues, and soil using digital images and deep learning. Agriculture, 14(12):2240.

[9]GaoLL, ZhangC, YunWJ, et al., 2022. Mapping crop residue cover using Adjust Normalized Difference Residue Index based on Sentinel-2 MSI data. Soil Tillage Res, 220:105374.

[10]GaoPP, SongY, SongMH, et al., 2022. Extract nanoporous gold ligaments from SEM images by combining fully convolutional network and Sobel operator edge detection algorithm. Scr Mater, 213:114627.

[11]HanXJ, ChengQB, ChenQZ, et al., 2025. Deep learning-based multi-category disease semantic image segmentation detection for concrete structures using the Res-Unet model. J Civil Struct Health Monit, 15(5):1369-1380.

[12]HivelyWD, LambBT, DaughtryCST, et al., 2018. Mapping crop residue and tillage intensity using WorldView-3 satellite shortwave infrared residue indices. Remote Sens, 10(10):1657.

[13]JinZY, HongWJ, WangYR, et al., 2025. A transformer-based symmetric diffusion segmentation network for wheat growth monitoring and yield counting. Agriculture, 15(7):670.

[14]LaamraniA, Pardo LaraR, BergAA, et al., 2018. Using a mobile device “app” and proximal remote sensing technologies to assess soil cover fractions on agricultural fields. Sensors, 18(3):708.

[15]LiK, ZhangYJ, WangTF, et al., 2025. FreqUNet: a lightweight dual-branch network with frequency-aware decomposition for retinal vessel segmentation. Expert Syst Appl, 287:128124.

[16]LiL, LiJ, LvCX, et al., 2021. Maize residue segmentation using Siamese domain transfer network. Comput Electron Agric, 187:106261.

[17]LiYW, ZhaoHS, QiXJ, et al., 2023. Fully convolutional networks for panoptic segmentation with point-based supervision. IEEE Trans Pattern Anal Mach Intell, 45(4):4552-4568.

[18]LiuJ, QiuTY, PeñuelasJ, et al., 2023. Crop residue return sustains global soil ecological stoichiometry balance. Global Change Biol, 29(8):2203-2226.

[19]LuoCH, ChenJP, GuoSX, et al., 2022. Development and application of a remote monitoring system for agricultural machinery operation in conservation tillage. Agriculture, 12(9):1460.

[20]MahmoodMT, UcanON, 2025. Data and image processing for intelligent glaucoma detection and optic disc segmentation using deep convolutional neural network architecture. Discov Comput, 28:73.

[21]PachecoA, McNairnH, 2010. Evaluating multispectral remote sensing and spectral unmixing analysis for crop residue mapping. Remote Sens Environ, 114(10):2219-2228.

[22]QiangJ, LiuWJ, LiXX, et al., 2023. Detection of citrus pests in double backbone network based on single shot multibox detector. Comput Electron Agric, 212:108158.

[23]SerbinG, Hunt JrER, DaughtryCST, et al., 2009. An improved ASTER index for remote sensing of crop residue. Remote Sens, 1(4):971-991.

[24]ShahiTB, DahalS, SitaulaC, et al., 2023. Deep learning-based weed detection using UAV images: a comparative study. Drones, 7(10):624.

[25]ShangCJ, ZhangD, YangY, 2021. A gradient-based method for multilevel thresholding. Expert Syst Appl, 175:114845.

[26]SheQS, SunSK, MaYL, et al., 2025. LUCF-Net: lightweight U-shaped cascade fusion network for medical image segmentation. IEEE J Biomed Health Inform, 29(3):2088-2099.

[27]ShiJF, JiSS, JinHY, et al., 2025. Multi-feature lightweight DeeplabV3+ network for polarimetric SAR image classification with attention mechanism. Remote Sens, 17(8):1422.

[28]SongHH, WangJQ, BeiJL, et al., 2024. Modified snake optimizer based multi-level thresholding for color image segmentation of agricultural diseases. Expert Syst Appl, 255:124624.

[29]SongWY, NieFX, WangC, et al., 2024. Unsupervised multi-scale hybrid feature extraction network for semantic segmentation of high-resolution remote sensing images. Remote Sens, 16(20):3774.

[30]TaoWC, XieZX, ZhangY, et al., 2021. Corn residue covered area mapping with a deep learning method using Chinese GF-1 B/D high resolution remote sensing images. Remote Sens, 13(15):2903.

[31]WangFY, LvCX, JiangHL, et al., 2025. Efficient detection of corn straw coverage in complex agricultural scenarios. Comput Electron Agric, 235:110338.

[32]WangGD, BaiD, LinHF, et al., 2024. FireViTNet: a hybrid model integrating ViT and CNNs for forest fire segmentation. Comput Electron Agric, 218:108722.

[33]WangGZ, WangJP, ZouXY, et al., 2019. Estimating the fractional cover of photosynthetic vegetation, non-photosynthetic vegetation and bare soil from MODIS data: assessing the applicability of the NDVI-DFI model in the typical Xilingol grasslands. Int J Appl Earth Obs Geoinf, 76:154-166.

[34]WangYY, GaoXB, SunY, et al., 2025. Semantic segmentation-based conservation tillage corn straw return cover type recognition. Comput Electron Agric, 229:109792.

[35]XieEZ, WangWH, YuZD, et al., 2021. SegFormer: simple and efficient design for semantic segmentation with Transformers. Advances in Neural Information Processing Systems, 34:12077-12090.

[36]XuJY, ZhouSY, XuAJ, et al., 2022. Automatic scoring of postures in grouped pigs using depth image and CNN-SVM. Comput Electron Agric, 194:106746.

[37]YangH, SunH, WangK, et al., 2025. Enhanced farmland extraction from Gaofen-2: multi-scale segmentation, SVM integration, and multi-temporal analysis. Agriculture, 15(10):1073.

[38]YuFH, BaiJC, FangJY, et al, 2024. Integration of a parameter combination discriminator improves the accuracy of chlorophyll inversion from spectral imaging of rice. Agric Commun, 2(3):100055.

[39]YueJB, TianQJ, 2020. Estimating fractional cover of crop, crop residue, and soil in cropland using broadband remote sensing data and machine learning. Int J Appl Earth Obs Geoinf, 89:102089.

[40]YueJB, TianQJ, TangSF, et al., 2019. A dynamic soil endmember spectrum selection approach for soil and crop residue linear spectral unmixing analysis. Int J Appl Earth Obs Geoinf, 78:306-317.

[41]YueJB, TianQJ, LiuY, et al., 2023. Mapping cropland rice residue cover using a radiative transfer model and deep learning. Comput Electron Agric, 215:108421.

[42]YueJB, LiT, FengHK, et al., 2024. Enhancing field soil moisture content monitoring using laboratory-based soil spectral measurements and radiative transfer models. Agric Commun, 2(4):100060.

[43]ZhangTT, HuDN, WuCX, et al., 2023. Large-scale apple orchard mapping from multi-source data using the semantic segmentation model with image-to-image translation and transfer learning. Comput Electron Agric, 213:108204.

[44]ZhangWQ, LiWJ, WangC, et al., 2025. A novel index for mapping crop residue covered cropland using remote sensing data. Comput Electron Agric, 231:109995.

[45]ZhaoJL, LiZ, LeiY, et al., 2023. Application of UAV RGB images and improved PSPNet network to the identification of wheat lodging areas. Agronomy, 13(5):1309.

[46]ZhengXX, CaoF, OuJY, et al., 2024. RSU-Net: a new method for fine classification of corn residue coverage in black soil area using Chinese GF-1B PMS image. Ecol Front, 44(6):1259-1268.

[47]ZhouDY, LiM, LiY, et al., 2020. Detection of ground straw coverage under conservation tillage based on deep learning. Comput Electron Agric, 172:105369.

[48]ZhuQL, WangK, LiangD, et al., 2025. WLUSNet: a lightweight wheat lodging segmentation network based on UAV image. Comput Electron Agric, 237:110587.

Similar articles