Author(s)

Yue Deng, KaiMing Zeng^*

Affiliation(s)

Jiangxi Institute of Territorial Space Surveying and Planning, Nanchang 330029, Jiangxi, China.

Corresponding Author

KaiMing Zeng

ABSTRACT

As the agricultural cornerstone of Ganzhou, Jiangxi Province, China, citrus production necessitates rapid and precise mapping of orchard spatial distribution for agricultural management, resource assessment, ecological conservation, and science-driven industry development. Remote sensing has emerged as a vital agricultural informatics tool in China due to its non-contact, large-scale data acquisition capabilities. Recent advances in deep neural networks enable state-of-the-art solutions for image processing and computer vision tasks. Capitalizing on this, we propose a multispectral remote sensing framework integrating a Pyramid-Shaped Residual Network with a decision tree algorithm customized for citrus cultivation patterns. The results indicate that the new method demonstrates extremely high accuracy, with an estimation precision exceeding 80% when compared to the statistical yearbook. This approach demonstrates significant potential for citrus planting area identification, providing a valuable reference for precision agriculture applications.

KEYWORDS

Citrus plantation area identification; Multi-spectral; Deep neural network; Pyramid residual

CITE THIS PAPER

Yue Deng, KaiMing Zeng. Recognition of citrus planting area by integrating pyramid bottleneck residual network and decision tree algorithm. Journal of Computer Science and Electrical Engineering. 2025, 7(8): 15-21. DOI: https://doi.org/10.61784/jcsee3107.

REFERENCES

[1] Li SY, Zhou XJ, Yin JM et al. Comprehensive division of climate, topography and soil for navel orange planting in Xunwu County. Acta Agriculturae Jiangxiensis, 2007, 19(5): 40–43.

[2] Xu H, Qi S, Gong P, et al. Long-term monitoring of citrus orchard dynamics using time-series Landsat data: a case study in southern China. International Journal of Remote Sensing, 2018, 39(22): 8271-8292.

[3] Chen Y, Hu J, Cai Z, et al. A phenology-based vegetation index for improving ratoon rice mapping using harmonized Landsat and Sentinel-2 data. Journal of Integrative Agriculture, 2024, 23(4): 1164-1178.

[4] Parente L, Taquary E, Silva A P, et al. Next generation mapping: Combining deep learning, cloud computing, and big remote sensing data. Remote Sensing, 2019, 11(23): 2881.

[5] Tu A, Xie S, Mo M, et al. Water budget components estimation for a mature citrus orchard of southern China based on HYDRUS-1D model. Agricultural Water Management, 2021, 243: 106426.

[6] Xiao C, Li P, Feng Z, et al. Sentinel-2 red-edge spectral indices (RESI) suitability for mapping rubber boom in Luang Namtha Province, northern Lao PDR. International Journal of Applied Earth Observation and Geoinformation, 2020, 93: 102176.

[7] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks. Communications of the ACM, 2017, 60(6): 84-90.

[8] Russakovsky O, Deng J, Su H, et al. Imagenet large scale visual recognition challenge. International journal of computer vision, 2015, 115(3): 211-252.

[9] Zhong Z, Li J, Luo Z, et al. Spectral–spatial residual network for hyperspectral image classification: A 3-D deep learning framework. IEEE transactions on geoscience and remote sensing, 2017, 56(2): 847-858.

[10] Paoletti M E, Haut J M, Fernandez-Beltran R, et al. Deep pyramidal residual networks for spectral–spatial hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing, 2018, 57(2): 740-754.

[11] Han D, Kim J. Deep Pyramidal Residual Networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition, 2017: 5927-5935.