Machine Vision for tree

Community General Discussions
1

Hi,

I want to develop machine vision algorithm for identifying fruit from a tree say mango.  I want to obtain location and coordinates of the fruit  for further decision making. Task is to be carried out in day light .Which type and how many  of cameras , sensor do I need and also do I need any processor like arduino for it? Can I do it in Python . 

Thanks

2

OpenCV

Checkout OpenCV (Open Computer Vision) software. A lo of that functionality is inherent. It needs  asingle board computer running Linux, perhaps a Raspberry Pi 3 plus webcam.

3

Trying to identifiy the

Trying to identifiy the mango on the tree will be an interesting and time consuming problem to say the least.  Inconsistent lighting will make it difficult to identify what is and what is not a mango, ripeness etc.

This is an example of a cucumber farmer in Japan who has things in a highly controlled environment.  

https://www.youtube.com/watch?v=MtQmqaSxszU

Their approach with generating a model for Tensorflow and then using it to identify things is probably the way to go.  If you get enough data, it will be smart enough to identify it no matter the lighting.  It will need a LOT of data.  Let us know how you make out.  You are on to some cool stuff and a fun problem to solve.

Regards,

Bill Schreiber