Are there more efficient techniques available, rather than ImageJ, for quantifying herbivory damage from photos of leaves? I have a basic understanding of machine learning but don't want to spend too much time trying to learn another complicated technique.
I haven't used :
http://www.plant-image-analysis.org/software/lamina
https://play.google.com/store/apps/details?id=upvision.bioleaf
Expensive : PhotoShop
What's wrong with ImageJ ??
Dear Heather Lea Slinn,
Look the link, maybe useful.
https://alternativeto.net/software/imagej/
https://imagej.en.softonic.com/windows/alternatives
https://imagej.net/Getting_Started
Regards, Shafagat
Heather,
ImageJ is not a technique, it is a collection of image processing routines, with some high level routines developed especially for medical imaging analysis purposes. Most routines in ImageJ are low-level and either very general or medical-image-processing oriented and you'll need Computer Vision knowledge to use it appropriately.
I suppose you're not intending to learn CV techniques, so the development of specific CV techniques for Foliar Damage estimation would not be in your list of objectives and you are looking for ready-to-run application software.
Foliar Damage Quantification is a very specific area of Computer Vision. Most software would be proprietary applications from different companies doing research for big grain multinationals.
There are a few applications that could be affordable. I would suggest both below:
BioLeaf (already suggested - https://arxiv.org/pdf/1609.08004.pdf - http://www.plant-image-analysis.org/software/bioleaf) It was developed by a public Brazilian University: There are chances they are willing to help you using it. Seems to be simple and the Bezier approximation they use could be error-prone.
WinFOLIA: http://regent.qc.ca/assets/winfolia_about.html This is a product. It probably will cost.
-
For those that are interested, I am now collaborating with a software company called MIPAR to automate herbivory pattern recognition. In the study system that I work with, we know which patterns are caused by specific caterpillar or ant genera. We are working to get output data from the software that identifies how many genera were feeding and what percentage of the leaf was eaten by them. If this is useful for folks, I can follow up with my thoughts on the final product.
You could try LeafByte. It's a free and open source app for iPhones and iPads. It measures leaf area and herbivory levels much more quickly than ImageJ and related software, and it automatically saves your data for you in a spreadsheet. You can download it in the app store or read more about it at the website below.
https://zoegp.science/leafbyte
Disclaimer-I was involved in making this app.
- Badges
- Science topic