Downloading images from iNaturalist

I have to use Alliaria-Petiolata images for a Deep Learning project and the problem is, I am unable to download the images. I have data in .CSV format and I need images for my task. Any help will be appreciated.

welcome to the forum.

see:

just for reference, this would get all observations with CC BY-NC, CC BY, and CC0 licenses: https://www.inaturalist.org/observations?photo_license=CC-BY-NC,CC-BY,CC0&place_id=any&taxon_id=56061

5 Likes

@pisum Thank you for the reply. I have the data in the .CSV file but I don’t know how to extract the images using API. Can you help me in that? do you know any API that I can use. Thank you.

other variations of this same thread have been added by the same person as:

Hello,
I need the images of Alliaria-Petiolata plant for deep learning task to find out the effect of climate change on this particular plant. I need to download the images but being researching and trying it from the last couple of week, I am unable to do it. If anyone can provide me an API through which I will be able to download the images will be great and a real help.
Thank you.

The links that @pisum replied with on your first post (https://forum.inaturalist.org/t/downloading-images-from-inaturalist/33529/2) include info on how to use the API.

1 Like

I am downloading images of Alliaria Petiolata (Garlic Mustard) from this link
https://www.inaturalist.org/taxa/56061-Alliaria-petiolata/browse_photos
but the problem is, I need Flowering, Flowering Budding and Fruiting images in separate folders. The API I am using downloads all the images only into a single folder and it’s not practical to separate these images manually. There is a filter where I can choose Flowering, Flowering Budding and Fruiting on the page with the link provided above under Plant Phenology but I want to find these images like Flowering, Flowering Budding and Fruiting in the search in the following link.
https://www.inaturalist.org/observations/export?flow_task_id=252768
Any help will be really appreciated.
Thank you!

why do you need all these images of Garlic Mustard? and why do you need to separate them between different Plant Phenology annotations (Flowering, etc.)? what would you do with images from observations that are not annotated?

Thank you for the reply. I have to develop 2 models, 2_stage_model and 4_stage_model using CNN which can classify the images whether flowering or non flowering and flowering, flowering and budding, vegetative and fruiting. Only experts in this field can do it manually and it takes too much time. The goal is to develop such a model which can do this classification even if there is not an expert. I will need to remove the images that are not annotated.

only experts can classify whether the plant is flowering or not? what happens after the experts classify the images? is that data used for some particular purpose?

I am using this data for research. My only task is to develop a model which can find whether a given plant is flowering or not and compare it with a .csv file where the original data about that image is stored. In this way I have to check the accuracy of my model. I am pasting here the link of the work formerly done on this and I have to check whether any further improvement is possible or not.
https://github.com/naziz4/DeepPhenology
The link of the paper is:
https://github.com/naziz4/DeepPhenology
And I got the .csv data from here
https://www.frontiersin.org/articles/10.3389/fpls.2021.787407/full#supplementary-material
But now I need images for this in separate folders.

ok. i’m going to assume that you’re doing this for some sort of class.

that said, here are some links that you should point you in the right direction:

you should probably keep in mind that iNaturalist images are licensed separately from the observations themselves. so you should probably make sure that the images you use are properly licensed for whatever you’re working on here.

1 Like

I merged the 3 topics into one topic.

1 Like

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.