The University of Adelaide
Browse
1/1
29 files

Machine vision solutions for monitoring pest snails in Australian no-till cropping fields: An exploration of spectral characteristics and detectability

journal contribution
posted on 2024-04-03, 05:18 authored by Huajian LiuHuajian Liu

Data and code description for the journal paper “Machine vision solutions for monitoring pest snails in Australian no-till cropping fields: An exploration of spectral characteristics and detectability” published in the Journal of Agriculture and Food Research.

The data and the Python codes written in version 3.11 are saved in the folder “published_data_codes_snails” which includes two sub-folders, “data” and “python_codes”. The “data” folder contains the pre-processed reflectance data of snails and different backgrounds in .sav format. The folder “python_codes” contains the Python scripts to analyse the reflectance data and to do cross-validation of the machine learning models. The script “analysis_all_snails_background.py” and “analysis_different_snails_bk.py” conduct Spectral and principal component analysis. The script “train_2_5_nn_model_snail.py”, “train_2_5_rf_model_snail.py” and “train_2_5_svm_model_snail.py” conduct cross-validation for neural networks, random forest and support vector machine models respectively.

Funding

This work was supported by the University of Adelaide Faculty of Sciences Roadmap grant "Improving detection and monitoring of biosecurity threats using drones, field robots and machine learning” and Grains Research and Development Corporation project (PROC-9176394) "More effective control of pest molluscs (snails and slugs) in Australian grain crops”. We acknowledge the use of the facilities, and scientific and technical assistance of the Australian Plant Phenomics Facility, which is supported by the Australian Government’s National Collaborative Research Infrastructure Strategy (NCRIS). We also acknowledge the investment from the Government of South Australia Department for Industry, Innovation and Science.

History

Usage metrics

    The University of Adelaide

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC