Skip to content

Question for R-sig-ecology

3 messages · Anshika Kulshrestha, Lucas Heber Mariano Dos Santos, Sarah Goslee

#
Hello everyone,

I have limited experience with R and have only been using it for GIS tasks.
Now I have data from camera traps which take pictures when there is some
movement and at the specific time of the day everyday. What I am planning
to with this data is rather simple, I just want my script to identify the
images where the lens was covered in snow and couldn't capture anything and
eliminate those images. I think I am not able to look for the right
material to read, since with all the research I did till now, I can only
find blur image detection which is not exactly what I am looking for.

I would really appreciate it if anyone can help with what exactly I should
be reading to learn this.

Thank you
#
Hello
You will need a image classification model and implement it yourself
Here some resource to get you started
https://towardsdatascience.com/a-laymans-guide-to-building-your-first-image-classification-model-in-r-using-keras-b285deac6572

https://rpubs.com/spalladino14/653239

Cheers
#
Hi,

I'm not sure you even need image classification.

Use list.files() to get the image filenames.
Read the image into R.
Discard it if the pixel values have a very low variance, indicating
that they didn't detect anything (or are mostly black? depending on
what your camera records when covered in snow).

There are R packages such as jpg and tiff that allow you to read
common image formats into R.

Sarah

On Mon, Jan 4, 2021 at 11:02 AM Anshika Kulshrestha
<akulshrestha22 at gmail.com> wrote: