Anthill Inside 2018

On the current state of academic research, practice and development regarding Deep Learning and Artificial Intelligence.

Industrial Vision & Deep Learning for Manufacturing Quality Inspection

Submitted by Dr. Chiranjiv Roy (@drroy) on Tuesday, 24 October 2017

videocam_off

Technical level

Intermediate

Section

Full talk

Status

Submitted

Vote on this proposal

Login to vote

Total votes:  -1

Abstract

Quality Inspection in Manufacturing Industries is of great importance and incurs spends of about $ 3 Billion USD, which is a complex pipeline of tasks which are rigourous, time-consuming and a lot of manual work. Complex testing processes takes weeks to conclude the impact of tests on Parts/Subject where there is a lot of scope for faster inspection using Analytics. The proposed talk elaborates the power of Artificial Intelligence through usage of Advanced Algorithms like Deep Learning, Reinforcement Learning and Computer Vision to Automatically Find “interesting” features from images and videos to inspect parts and spares on a Augmented Reality software. Corrosion recognition is one of the most complex and impactful in Warranty for Automotives. Can AI help?

Outline

Outline of the Talk as per presentation Uploaded:
S1: Introduction of Speaker
S2: Quality Inspection process in Manufacturing
S3: Testing for Corrosion
S4: Architecture
S5: Traditional Vision & Initial Hybrid ML Classifier - For CPUs
S6: Deep Learning Optimization - Next level with GPU - part 1
S7: Deep Learning Optimization - Next level with GPU - part 2
S8: Auto-Summary with Spark
S9: Final Product Design Framework, Novel Framework
s10: Thanks

Requirements

Open mind, knowledge of Machine Learning (basic)

Speaker bio

Principal Data Scientist with PhD in Machine Learning with over 15 years of experience in Analytics Consulting, Big Data Strategy and Applied Research in Computer Vision and Machine Learning for Consumer and Enterprise Businesses for Manufacturing, Automobile, IT & Financial Services organisations of repute. Techno-Functional expertise in transforming Research to Applied Research through Proof Concepts and then develop Big Data Solution Development and Professionalization.

Comments

  • 1
    chiderasa swaroopini (@chidekrasa) 12 months ago

    Amazing idea

    • 1
      Dr. Chiranjiv Roy (@drroy) Proposer 12 months ago

      Thanks

  • 1
    Dr. Chiranjiv Roy (@drroy) Proposer 12 months ago

    People working in manufacturing and engineering domains would be the beneficiary. A clear example of actual Deep learning work in a Professionalized “production” is the essence of this Use Case which focusses on practisioner’s view of Artificial Intelligence rather than theoritical presentations

  • 1
    Hari C M (@haricm) Reviewer 11 months ago

    We won’t be covering computer vision at The Fifth Elephant. So I’m moving this proposal to Anthill Inside 2018. Both these conferences are happening back to back.

  • 1
    Hari C M (@haricm) Reviewer 9 months ago

    You have to share slides and preview video for us to evaluate this talk.

  • 1
    Hari C M (@haricm) Reviewer 9 months ago

    Have you given same talk at any other conference recently?

    • 1
      Dr. Chiranjiv Roy (@drroy) Proposer 9 months ago

      No. As the head of Engineering Analytics and Big Data COE my role is to drive multiple Use Cases other than Autonomous Driving. I would share the concept of ManTech 4.0 and descrive through one of the Use Cases thereby explaining high level of the process being applied. ManTech is my coined concept and extremely novel in nature. How to mix the complex world of Engineering, Physics, Deep Learning and Computer Vision is the crux of Artificial Intelligence ManTech 4.0. Without sharing any data I will describe a Use Case [keeping confidentiality in mind]

  • 0
    Robin Jasin (@robinjasin) 12 months ago

    Is this talk really necessary?

    • 1
      Dr. Chiranjiv Roy (@drroy) Proposer 12 months ago (edited 12 months ago)

      It focussess on Deep Learning Application rather than theory on an Enterprise World which has impacted production.

Login with Twitter or Google to leave a comment