Interpreting the outcome data

[rodrign2002]

I was able to run my script with different images, I have got the results and now I want to estimate the FW biomass of my plants. I used the equation described in :A Versatile Phenotyping System and Analytics Platform Reveals Diverse Temporal Responses to Water Availability in Setaria [https://www.sciencedirect.com/science/article/pii/S1674205215002683#mmc1], the equation uses the side-view area, the sum of the above-ground plant pixel area from all four side-view VIS images

Mfw=3.755*10ˆ-5 Asv-0.2704
where Mfw is fresh-weight biomass and Asv is side-view area.
I did all this and then plotted the real biomass from my plants (g) vs the fresh-weight biomass from the images, but the correlation is too low R2=0.17.
I plotted as well the Asv vs the real biomass from my plants (g) and the correlation is the same=0.17

I was wondering if someone could give me a hand, I don't know what I am doing wrong

Thanks

[nfahlgren]

Hi @rodrign2002,

The linear model in the paper will not work for other datasets (or it would at least be very unlikely). The reason for this is that the model depends explicitly on the resolution, field of view, and other camera settings in our system that affect what pixel measurements represent, and likely also depends on other factors not explicitly defined in the model (e.g. the plant species).

The general methodology of using linear modeling to predict biomass using one or more image-based phenotypic measurements will work however if you have training data for your dataset.