Scribd
Upload
287 views11 pages

Intro to Computer Vision with OpenCV

Computer vision is a field that deals with how computers can understand digital images and videos. It is inspired by biological vision systems. Computer vision is used in many areas including automotive, defense, agriculture, robotics, and healthcare. Machines understand images by performing tasks like image classification, object detection, and semantic segmentation using algorithms. OpenCV is a popular open source library for computer vision that contains optimized algorithms. It supports languages like C++ and Python and can be used for real-time applications on devices like Raspberry Pi. For example, information can be extracted from printed documents using computer vision with requirements of being scale, rotation, and translation invariant and efficient on portable code.

Uploaded by

Stephen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
287 views11 pages

Intro to Computer Vision with OpenCV

Computer vision is a field that deals with how computers can understand digital images and videos. It is inspired by biological vision systems. Computer vision is used in many areas including automotive, defense, agriculture, robotics, and healthcare. Machines understand images by performing tasks like image classification, object detection, and semantic segmentation using algorithms. OpenCV is a popular open source library for computer vision that contains optimized algorithms. It supports languages like C++ and Python and can be used for real-time applications on devices like Raspberry Pi. For example, information can be extracted from printed documents using computer vision with requirements of being scale, rotation, and translation invariant and efficient on portable code.

Uploaded by

Stephen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

A Gentle Introduction to Computer Vision and OpenCV

Stephen Pilli
What is Computer Vision?

● Computer vision is an interdisciplinary scientific field that deals with how


computers can gain high-level understanding from digital images or videos.
● Inspired by biological vision system by evolution.
● In simple terms - Machines making sense of things that they look at.
Automotive

Where is Computer Vision?

Defence

Astronomy Agriculture

Domestic
Robotics
Health Care

Palmistry
What and how does Machine Understand?

● Machines don’t understand like we do. Because we don’t yet understand how
we understand. But at-least we understand how machines understand.

● Image Classification
● Object Detection
● Semantic Segmentation
● Instance Segmentation
● Object Tracking
● Pose Estimation
● Action Recognition
● Super Resolution
● Image Restoration
● Visual Anomaly detection
● Scene Reconstruction
● Optical Character
Recognition
How do we do Computer Vision? - OpenCV
● Open Source Computer Vision Library
○ https://opencv.org/
● Originally developed by Intel
● Has more than 2500 optimized algorithms
● Supports
○ C, C++, Python, Java
○ Linux, Windows, MacOS, IOS, Android, Any custom OS ( Source code - Compilation)
● Leverages performance with GPU Acceleration
● Realtime
○ 30fps
● Embedded Systems
○ BeagleBone Black, Raspberry Pi, Jetson Nano
○ Even it can be installed on your router, provided your router runs linux.
○ CPU MT7620A 580 MHz - MIPS arch.
○ 128MB DDR2 16MB SPI Flash
○ OS - OpenWRT
Language Selection
Case Study - Information Extraction from Printed Documents

Requirement

● Extract the information printed on


the document.

Constraints

● Scale, Rotation and Translation


Invariant (RST)
● Highly Efficient [ Whatever I’m trying to explain right now is to give you an Idea
on how to work on a simple Computer Vision Project at industry
● Portable Code level, directly copying these techniques that I’m about to discuss
will lead you nowhere as these are outdated. ]
Thank You
● https://github.com/901stephen
● https://www.linkedin.com/in/stephen-philip-pilli-7636a255
● [email protected]

You might also like