En
Ru
En | Ru

A neural network that determines salad weight using photo images

Discuss your project
ABOUT
ABOUT
ABOUT
ABOUT
ABOUT
ABOUT

Customer

01/
iFarm is a Russian IT startup that builds vertical software-controlled farms for growing berries, vegetables and herbs.
02/

The task

Нейросеть
To teach a neural network the ability to distinguish Romano salad and determine a estimated weight using photo images
This will allow companies to monitor their grow rooms 24/7 and help detect which plants are stunted. This means:
- Notice undesirable plants earlier and economizing grow houses
- Automatically regulate the microclimate inside greenhouses
03/

What have we done

Interesting fact

When determining the weight of the salad, the neural network calculates the area of the green area in the image.

How it works

04/
Stationary cameras take pictures of salad on the shelves
Images are sent to the AI system EORA Vision Hub
The neural network calculates the parameters of the plant and compares them with the forecast
The neural network will alert is dimensions and forecast are very different
IFarm system regulates the microclimate
in the greenhouse

How it works

04/
Stationary cameras take pictures of salad on the shelves
Images are sent to the AI system EORA Vision Hub
The neural network calculates the parameters of the plant and compares them with the forecast
The neural network will alert is dimensions and forecast are very different
IFarm system regulates the microclimate
in the greenhouse
1
2
3
4
5

Our model vs reality

05/
Predicted weight
Maximum daily weight
Average daily weight
Minimum daily weight
Weight in grams
Growth day

The problems we faced

05/
Data markup
Decision
Time spent
We taught the neural network to see plant outlines but lost ability when plants grew and overlapped
We enclosed the bushes in rectangles and reduced marking to only two minutes
Up to 15 minutes was being spent to contour one photo

What is the project's value?

07/
Ecology
Saving
Plants need fewer pesticides and fertilizers.
Lower salad price: partially automated care.

Project team

08/
Techlead
Konstantin Kubrak
Computer Vision Engineer
Aelita Shaikhutdinova
Project manager
Nadezhda Zagvozkina
Backend developer
Andrey Chertkov

Our plans

06/
Automate the collection of plant data
Start processing photos on the spot, without sending them to the server
Use a drone with a camera to shoot plants.
Contact us