SCRYLENE - Open source Ethylene analyzer

Building an open source analyzer to quantify ethylene gas and mechanism to map rotting behavior for fruits/vegetables etc.

Similar projects worth following
Recently, I started working on a project to study & predict the rotting behavior of fruits and vegetables in different environment conditions. It is a common technique in households to detect rotting via smelling the fruit. That smell can be attributed to ETHYLENE which is continuously secreted by the such raw food items.
The problem is that an ethylene sensor is not a readily available item as is the case with other gases like carbon dioxide, carbon monoxide or formaldehyde.

The project objectives were the following :--

1. Determine the ethylene concentration in a closed contained environment.
2. Understand how the surface rotting (of fruits) is dependent on the ethylene secretion.
3. Learning from the data, the system which can possibly predict the surface rotting .
4. Should be a raspberry pi based system and can be developed using some readily available chemicals + electronics.

When the fruits start maturing, they give out different gases, the composition of which is based on the composition of the respective food- fruit,vegetable etc. But ethylene is the gas which has the highest concentration amongst all of them.

Objective :-
The objective of this device is to develop an open source method to monitor the rotting process - primarily in fruits/vegetables and develop an analyzer that can detect and quantify the ethylene concentrations.

Concept :-

Ethylene sensors are not readily available as is the case with other gas sensors like carbon dioxide, carbon monoxide, formaldehyde etc. You can buy ethylene analyzers which are bulky and expensive like felix instruments gas analyzer.

This is because there are different techniques to determine ethylene concentrations :-

  1. Gas chromatography
  2. Non dispersive infrared spectroscopy
  3. Electrochemical sensors

In this project I will try to use NIR spectroscopy method & if funds permit develop an electrochemical sensor. Additionally, I will be using computer vision to detect the type of food and use deep learning techniques to learn about the rotting mechanisms and automatically predict the time for consumability plus other parameters on which i will expand in the next few weeks.

(more details to be added with time)

P.S. : I will be glad to have any kind of feedback on this project.

  • 1 × Raspberry Pi A normal raspberry pi . I am using pi2
  • 1 × Pi camera

  • First step : Image segmentation

    Aman Garg07/30/2016 at 23:47 0 comments

    I divided the project into two phases -- first one to identify automatically the objects inside the closed container and second to establish the ethylene secretion levels inside the container.

    For the first step, i.e. the recognition of the fruits/vegetables, I did image segmentation to optimize the detection mechanism. I deployed it using raspberry pi and opencv with contours detection. I initially took a default image to check the results :-

    source : Google images

    I ran the image segmentation techniques for such a condition with the following resulting contours :-

    You can find the code on the github link that I have shared.

    The image segmentation is dependent on a lot of things with light/brightness of the original image of the major issue.

    I will be modifying the algorithm to take into consideration low light conditions inside a closed chamber with the raspberry pi mounted inside the top.

    Also, my next steps will include machine learning to identify the fruits and isolating them individually.

View project log

Enjoy this project?



Rahil wrote 07/21/2016 at 20:21 point

Interesting thoughts. It may find a lot of potential in future. 

  Are you sure? yes | no

Aman Garg wrote 07/30/2016 at 23:37 point

The objective is to develop an analyzer and food recognition tool. Hopefully it does achieve it.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates