Idea - FICS


Team and Contact Details

Student Name School Degree Year Email

Inter School Idea ? No
Do you need expertises from another area: No
If Yes please provide details of expertises you need:

Idea Details

Slogan: Good Soil Good Luck
Supervisor Name: Muhammad Tariq Saeed
Supervisor Designation: Assistant Professor of Computational Science & Engineering (CS&E)
Supervisor School: RCMS
Supervisor Department: RCMS
Contact number: 03028671230
Email ID:
    A programme based on machine learning that detects the composition of soil and predict what type of soil it is through use of images. It enables an opportunity for testing at a cheaper way.
What is the unmet need in society that your idea will fulfill ?
    Basically, our target industry and business is those that have a use of soil before initialising their projects such as construction, oil drilling, farming etc. The need to test soil is a must in order for a positive yield.
Who needs it ? How many would benefit ?
   Developing companies need cheaper ways of producing products. The image testing can be seen to be the simplest method to acquire substantial results in order to move on and give positive feedback.
How will the solution works
    An image is taken and compared with the collected data to determine what type of soil is identified. Images are converted into pixels (every soil colour has its own pixel) on which classifiers are applied to get precise results as to what type of soil we are dealing with. Multiple classifiers from machine learning is applied to produce values i.e. precision, recall data etc. that allows us to distinguish the soil type we are dealing as a sample.
Who are your competitors ? How is your solution different
    There are businesses that work on testing soil samples for companies such as oil drilling, marble making, construction. They take samples to be tested and analyse. The process is expensive and hectic. In our process, image of the soil would be taken to analyse its type from site survey at same day.
Status: new
Entry Date & Time: 2021-01-14 (1731)