The chart below shows the three most likely mushroom species according to the recognizer.
Make a photo of a mushroom and see which species it is. Determine over 200 different species.
In order to classify your photo correctly, your photo needs to match some conditions. The mushroom has to be the main object in it without disturbing elements. It has to fill a big part of the photo. Secondly, you get the best results if you shoot the mushroom laterally. Lateral shoots show the most details what is critical to determine the specie. You even get better results if you shoot slightly from the bottom to the top so that the bottom of the mushroom head is visible.
Why so far away?
Not from the top
Share this project Tweet me
Follow me on Twitter @SeverinBuhler
Sign up the mailing list to hear when I will push this project further.
I'm a passionate software engineer with a huge interest in machine learning and artificial intelligence. My main focus lies on reinforcement learning and visual recognition. I formerly worked at OSMDeepOD, a satellite object detection software.
After struggling to identify mushroom species two years running, I have tried to ease my life with an automatic mushroom recognizer.
This mushroom recognizer is an experimental software that tries to determine the specie. There is no guarantee that the prediction of this software is true. Don't rely on the prediction if you want to eat a mushroom. I will not be liable for any false, inaccurate, inappropriate or incomplete information presented on the website. I specifically disclaim liability for incidental or consequential damages. You use this software at your own risk.