Machine Learning India bio photo

Machine Learning India

Fostering science and machine learning in India

Email Mailing List Twitter Github

Are you a student, professor, CEO or Maschinenmensch? Subscribe to ml-india's google group to join the discussion and recieve updates, news and resources about India's ml-ecosystem. Click here

We are into quality testing of agricultural products using Machine Learning and Image Processing

Products and Services

We are using deep learning to analyse the quality of food grains. Our product is an amalgam of image processing and Deep Learning. We capture image of each grain from multiple sides. First we have to map these images and then perform segmentation to remove the background and identify touching kernels in individual. Once it is done, these grains are passed through the classifier to detect their category. This all exercise runs on a local system with normal computing power. And this all has to be completed in a span of less than 20 seconds for more than 2000 grains.

Subfields of AI Being Used

  • Computer Vision
  • Deep Learning

Uses AI for

  • Agriculture

Applies AI on

  • Images

Team

Our Machine Learning team is headed by Yogesh Kumar Gupta who is a graduate from IIT Jodhpur with a BTech in Computer Science. He has previously worked in various organisation ranging from ARM Embedded System to OYO Rooms. He has vast experience in data science and data modelling work. He heads the entire tech development at Nebulaa. One of the senior member of the team is Tapish Rathore who is currently pursuing his Masters from University of California, Merced. He is doing his masters in Machine Learning & Computer Vision

Nebulaa Innovations’s AI team has 3 members.

Key Innovations

We have optimized the segmentation and classification algorithm such that it can operate on normal computation devices very efficiently and accurately. Our segmentation algorithm has been completely developed in-house and attains an accuracy of more than 99.5%.

Technology Stack

  • AWS

Hyderabad

Website

*The data is as provided by the company.