AI-automation
PROJECT EXODUS
TEAM TRITONS (JAIN UNIVERISTY)
Introduction:
- The never-ending technological advances have brought us here today to live in a digital world instead of caves. Artificial Intelligence is now used in every field from Netflix to self- driving cars it is everywhere. AI goes one step further by automating tasks such as driving a car, providing medical advice or playing chess games, which we thought could never be automated. AI can spur growth by replacing labor, which is in finite supply, by capital, which is in unbounded supply, both in the production of goods and services and in the production of ideas. Minimizing or optimizing the work processes, business processes reengineering shifted the industrial age towards the digital age by the help of e-business environments would help us create an era of AI.
ARTIFICIAL INTELLIGENCE
- We can implement AI with robotics to give robots a 100% increase in efficiency. Just like robotics the field of AI is emerging from the ashes, though it is not something new to the world, it wasn’t implemented or ‘DEPLOYED’ effectively.
- Now AI, can access your phone, can control your home appliances, can give you better suggestions (Netflix), and many more. The field of machine vision and computer vision is giving new insights to the people, with the pandemic in the big picture, robots with AI are outperforming humans, for example: Face mask detection, classifying images, object detection and also self-autonomous vehicles.
- AI can be used in the manufacturing sector to detect any faulty equipment. In the food and beverage industry AI can be equipped with sensors to make sure the quality of food, the storage, and filling them in containers. These not only create more jobs but also make the employee equipped with add-on skills.
ROBOTICS:
-
Robotics Is one of the emerging sectors in technology. Technology has played a role in making work more efficient for thousands of years, from simple farming tools to current-day assembly-line robots in factories. Robots are becoming present in more and more situations in business. They work right alongside human workers or completely replace them.
-
Amazon Inc Uses a variety of robots in its warehouses to stock inventory, and retrieve and package items. Tesla Motors Inc. boasts robotic and automated assembly lines for its electric cars and batteries.
ARTIFICIAL INTELLIGENCE IN AUTOMATION:
- An Intelligent Automation system functions using these three components of artificial intelligence. Depending upon the need, they can be combined or used to separately to create a fully automated solution:
- Machine Vision
- Natural Language Processing
- Machine Learning
EXODUS
About the project
- Project Exodus is a computer vision based application which classifies a given product as an OK image or a defect one. Reason for collecting this data is casting defects! - So what is casting defects?
Casting defect is an undesired irregularity in a metal casting process. There are many types of defects in casting like blow holes, pinholes, burr, shrinkage defects, mould material defects, pouring metal defects, metallurgical defects, etc.
- Defects are an unwanted thing in the casting industry. For removing this defective product, all industries have their quality inspection department. But the main problem is this inspection process is carried out manually. It is a very time-consuming process and due to human accuracy, this is not 100% accurate. This can because of the rejection of the whole order. So it creates a big loss in the company.
- We decided to make the inspection process automatic and for this, we have build an AI to solve this problem.
ABOUT THE DATASET
- We have used around 7348 images from KAGGLE PUBLIC DATASET for the model, and the division is as follows
- Train: - defect-front have 3758 and ok-front have 2875 images
- Test: - defect-front have 453 and ok-front have 262 images
- We have used DEEP LEARNING IMAGE CLASSIFCATION technique, we get 98% accuracy on train set and 99% accuracy on Validation set with no overfitting of the model.
IMAGES OF PRODUCT:
RESULTS:
AREA UNDER THE CURVE
- Why AUC is considered as an important metric?
- AUC - ROC curve is a performance measurement for classification problem at various thresholds settings. ROC is a probability curve and AUC represents degree or measure of separability. It tells how much model is capable of distinguishing between classes. Higher the AUC, better the model is at predicting 0s as 0s and 1s as 1s. By analogy, Higher the AUC, better the model is at distinguishing between DEFECT and OK product.
EXODUS: AUC
PHASE 2:
PLAN OF ACTION:
- Build a robot to classify the images as DEFECTIVE OR OK
- deploy them in real time scenario
- change the product according to the user requirements
If we win this competition it would be easy to complete the project and take it to PHASE 3
1 - DEFECT PRODUCT
2 - OK PRODUCT
TEAM TRITONS (JAIN UNIVERISTY): 1. Siddarth.D 2. Athish Anand Kumar 3. Chevula Haarvish 4. Dennis McLeaord