Published October 21, 2021 | Version v2
Other Open

Analysis code for: Classification of daily crop phenology in PhenoCams using deep learning and hidden markov modelss

Description

This is the analysis code repository for the following study:

Taylor, SD and Browning, DM, 2022. Classification of daily crop phenology in PhenoCams using deep learning and hidden markov modelss. bioRxiv. https://doi.org/10.1101/2021.10.20.465168

Note the initial version was titled "Deep learning models for identifying crop and field attributes from near surface cameras" and was changed during the review process.

The original repo is also here: https://github.com/sdtaylor/PhenocamCNN2

See the readme.md file in the primary PhenocamCNN.zip for descriptions. PhenocamCNN.zip contains all code in the analysis, and data to reproduce the manuscript figures. The only thing not included is the fitted keras model file, vgg16_v4_20epochs.h5, which is available by itself since it's quite large.

If you'd just like to use the final predictions of crop and field status you can download the final_predictions.zip file.

 

Notes

Version 2: a few analysis scripts were finalized. The manuscript was switched to a new latex template, had a new figure added, and some minor text changes. Version 3: title updated.

Files

final_dataset.zip

Files (4.5 GB)

Name Size Download all
md5:78c987aaa5a2094cb775a2781dfde057
4.0 MB Preview Download
md5:437c30b293576807b08aaf6c7821c21b
37.7 MB Preview Download
md5:f7c9f24f8f4fd54b302a17bdc90d1dfe
4.5 GB Download

Additional details

Related works

Is supplement to
Preprint: 10.1101/2021.05.21.445173 (DOI)