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

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.