Replies: 1 comment 4 replies
-
@ArtNd this is a pretty standard object detection task, nothing special is required. Visit Train Custom Data tutorial to get started: YOLOv5 Tutorials
|
Beta Was this translation helpful? Give feedback.
4 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello everyone, I'm trying to use YOLOv5 to detect some pollens.
My dataset is composed of large images with some little pollens everywhere (it's a microscopic image) :
![large_image_pollens](https://user-images.githubusercontent.com/61831659/141752911-45ba662b-08bd-4163-999b-785669ae64d7.jpg)
And I have some cropped images representing individual pollens :
![cropped_pollen](https://user-images.githubusercontent.com/61831659/141753694-93c554c5-51fe-492d-ab8a-da24c7518bf9.jpg)
The goal is to classify the pollens shown on the large picture.
My issue
I'm asking myself what is the best way to train my model ?
I think one of the best options would be to train my model to create bounding boxes and detect pollens through the image and then pass it through a classifier to get the type of every pollen on the picture.
For the last option, do I have to train the model to detect the pollen object in general or to detect already pollens by species with the same classes as for the classifier ?
I hope you will understand my concerns. I'm a bit lost in the way of get it done.
Thank you !
Beta Was this translation helpful? Give feedback.
All reactions