search

sdtaylor / PhenocamCNN2

Other
1 stars 0 forks source link

Journal to submit to #4

Closed sdtaylor closed 2 years ago

sdtaylor commented 3 years ago

Note: submitting Oct 1 to put it in FY22

Environmental Research Letters - https://iopscience.iop.org/journal/1748-9326 classification stuff may be too specific for here, but maybe worth a shot. Scope seems to emphasize policy relevant stuff, but also management. Agriculture is well within scope.

Science of remote sensing - fits scope but would need to justify novelty. ie "beyond Gcc with phenocams"

Agricultural and Forest Meteorology - absolutely fits here, $3750

Applications in Plant Sciences - $1500 https://bsapubs.onlinelibrary.wiley.com/journal/21680450 EIC sent invite based on preprint

MDPI Sensors - $2400 - several special issues related to machine learning/AI in AG.
machine learning in ag special issue, due oct 2021
Application of Artificial Neural Network and Sensing in Advanced Agriculture - due march 2022
Deep Learning Methods for Remote Sensing - due march 2022

sdtaylor commented 3 years ago

MDPI Remote Sensing special issues - $2600
Advances in Remote Sensing for Crop Monitoring and Yield Estimation - due dec 31, 2021
Digital Agriculture with Remote Sensing - due sep 1, 2021
Applications of Deep Learning in Smart Agriculture - due aug 31, 2021
Crop Growth Monitoring Using Remote Sensing: Progress, Challenges and Opportunities - due june 2022
Applications of Remote Data Capture Systems in Agriculture and Vegetation - due feb 2022

This one Advances in Remote Sensing for Crop Monitoring and Yield Estimation - due dec 31, 2021

sdtaylor commented 2 years ago

potential cover letter

Greenness indices have been the de-facto standard for land surface phenology since the 70s. Near surface cameras have essentially coped this technique, which discards large amounts of contextual information from the images. Here we address that in one of the more difficult land cover types: croplands. We use a deep learning model to classify phenocam images simultaneously into the dominant field cover, crop type, and crop phenological status. Image classification models do not have a temporal component, so we accounted auto-correlation by using a hidden markov model in the post processing.

sdtaylor commented 2 years ago

published