Udemy
    •  
    •  
    •  
    •  
    •  
    •  
    •  
    •  
Subscribe
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
A hands-on introduction to Optical Character Recognition
Rating: 3.9 out of 5(37 ratings)
1,196 students

A hands-on introduction to Optical Character Recognition

Understanding and Implementing OCR Techniques with Real-world Applications
Created byVenkatapathy Subramanian, Dishant Padalia, Gauri Sharma, Saurabh Baghel
Last updated 7/2023
English

What you'll learn

  • Understand the basic concepts of Optical Character Recognition
  • Walk through the steps in OCR and different types of methods available for each step
  • Understand how Computer Vision, Machine Learning and Deep-learning techniques are applied for OCR
  • Have hands-on walk-through for each concept

Course content

5 sections • 12 lectures • 1h 59m total length

  • What is OCR?, and Why OCR?7:22

    This part of the lecture provides a comprehensive explanation of Optical Character Recognition (OCR), a technology that converts images of text into a machine-readable format. It begins with a brief history and description of OCR and then discusses why OCR is important, particularly in terms of reducing storage space, improving the accuracy and efficiency of data interpretation, and enhancing data security. The video then highlights various applications of OCR, ranging from its role in assistive technology for visually impaired users to its use in data entry, digitization of books and medical records, and recognition of traffic signs and number plates. It ends by emphasizing OCR's transformative potential in governmental and private industry processes.

  • A basic demo of OCR in Google Colab6:50

    In this lecture, we will run a basic, simple, end-to-end demo of Optical Character Recognition using a tool called doctr. You can find the link to the Google Colab notebook in the resources section.

  • Types of OCR6:26

Requirements

  • A basic understanding of Deep learning
  • Intermediate experience in Python

Description

This course provides a comprehensive, hands-on introduction to the field of Optical Character Recognition (OCR). Aimed at students and professionals with a foundational knowledge of computer science and basic programming skills, this course offers an in-depth exploration of the principles, techniques, and applications of OCR technology.

The curriculum begins with a brief history and overview of OCR, where learners will gain insight into how the field has evolved over time. Participants will then delve into the core techniques used in OCR, such as image pre-processing, binarization, segmentation, feature extraction, and character recognition. The course also incorporates detailed explanations of various machine learning and deep learning models used in contemporary OCR systems, such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs).

Participants will have opportunities to apply these concepts in practical, hands-on labs where they will develop their own basic OCR systems. These lab exercises will cover various real-world applications, such as document digitization, automatic license plate recognition, and handwriting recognition. They will also learn how to use popular OCR tools and libraries, such as Tesseract and Pytesseract.

The course also addresses challenges faced in OCR such as handling noise in images, dealing with different fonts and sizes, recognizing cursive handwriting, and understanding the implications of these challenges on OCR accuracy.

By the end of the course, participants will have a solid understanding of the principles and methodologies used in OCR. They will have developed the skills necessary to implement, optimize, and troubleshoot OCR systems, and will be equipped with the knowledge to explore more advanced topics in the field.

Course Prerequisites: This course assumes familiarity with basic computer science principles and programming, particularly in Python. Prior knowledge of machine learning concepts will be beneficial but is not required. All necessary mathematical concepts will be reviewed as part of the course material.

Who this course is for:

  • Beginners who wanted to venture in the domain of Computer Vision and Language model topics
  • Professionals who are looking for job and research opportunities in the field of Document AI

Report abuse