Best Practices for labeling images
Polygons was chosen over bounding boxes to reduce excess noise that is often associated with bounding boxes, this in turn increases the accuracy of the model.
Best Practices We Follow
- Only create polygons for each fish that you can clearly see
- Look for bogus polygons that may have been autogenerated and delete them (i.e.: rocks, seaweed/kelp, animals that are not fish- marine mammals, cephalopods, star fish, etc-, humans, boats, etc.)
- Meta data settings
- If “Fish Art or Fish Drawing or Fish Memes” make image as “Not a real fish (drawing)”
- Turn on “Not a real fish (drawing)” flag under meta data on top level image
- If image if of a underwater fish, mark "Photo taken Under Water" in the meta data
- If photo is a xray of a fish, mark "X-ray image"
- When cleaning up image polygons (polylines)
- “zoom in” on the polygon and make sure that it’s very accurately drawn around fish species body shape (tight to body)
- If you know the name of the fish, tag it with the species name.
- You can try using the polygon level fish detection AI model to see if the fish species in the current AI model
- Otherwise, mark as UNKNOWN species
- Turn on "Dead or Mutilated" if fish is mutilated or decayed!
- How to deal with human body parts such as fingers, hands, etc.
- Draw around fishes head and remainder of body stopping at obscured areas.
|Obscured area's of the fish
|Keep the polylines tight to the body shape
Best Practices for Labeling Fish Traits
Each fish species has a unique set of identifying attributes, and these attributes will be tagged with smaller polygons. Training the AI model to recognize the unique identifying attributes of species we will increase the accuracy of the model.
Best Practices we follow
- The traits can only be labeled after the polygon has been tagged with a species name
- Add fish traits for each clearly visible trait in body polygon
- If trait is not clearly shown in image, skip and do the next trait