SlideShare a Scribd company logo
AI-as-a-Service
State-of-Art
Optical Character Recognition
Optical Character Recognition
OCR IS
- > that can be processed by MASHINES.
method of transferring TEXT from IMAGE form
(such as scan, video or photo) -> into TEXTUAL FORM
Optical Character Recognition - Global
Source: Transparency Market Research, 2017
• Namely legal
• Recruitment
• Banking
• Education
• Healthcare
• Finance
• Government
Sectors
Global Optical Character Recognition Market, By Type
Software Service
Desktop based OCR
Mobile based OCR
Cloud based OCR
Others (Batch OCR, Server based OCR etc.)
Multi-tenant cloud (Public cloud)
Private Cloud
Consulting
Outsourcing
Implementation & Integration
OCR - Implementation
• Store it in a more compact form
• Display or print the material without
losing quality
• Analyzing information
• Apply to the text an electronic translation
• Formatting or converting to speech
• Convert books and docs to e – format
• Automation of accounting systems in
business
• Text publishing on a web page
• Text’s editing
• Search words and phrases
Current OCR problem - Global
Current solutions are ‘heavy’ in
computational cost, could easily
take tens of seconds or even
minutes to process single page.
Handwritten or rare fonts could
pose another problem for
solutions currently on the market.
• Deep learning based solution
• Uses GPUs to speed up computations
• Takes 400-500 milliseconds per page
OUR Solution
FAST
ADAPTIVE
• Trained just by observing the data
• Adding new language takes only few weeks
• Solves hard cases, like handwritten text
recognition
• Can be trained on client’s side
• Any image source will work, like web-camera
OUR Solution
• Automatically groups related text blocks
• Classifies blocks with complex relationships for
business purposes
• Doesn’t require retraining when templates change as
long as the new template structure is more or less
semantically the same
SMART
OUR Solution
Team
• Our team structure enables each member to
grow safari-fix as a professional.
• An employee creativity, deep and varied
background are essential for our business.
• We have an open and safe communication
space to create positive and productive
collaborations.
• XOresearch is always in search of passionate,
creative and self-motivated people. Join us and
be a part of international and constantly
evolving team.
Contacts
Headquarters
3, Republikas square, office 307
Riga LV-1010
Latvia
Development office
5b, Dilova street, office 306
Kyiv 03150
Ukraine
Latvia: Tel.: +371 67 305 084
kp@sapiensapi.com
www
LinkedIn
Facebook

More Related Content

PPTX
Presentation on OCR
PPTX
Offline Omni Font Arabic Optical Text Recognition System using Prolog Classif...
 
PDF
A brief introduction to OCR (Optical character recognition)
PPTX
OCR (Optical Character Recognition)
PPTX
Optical Character Recognition( OCR )
PPTX
Machine learning
PDF
Handwriting recogntion slides boeing
Presentation on OCR
Offline Omni Font Arabic Optical Text Recognition System using Prolog Classif...
 
A brief introduction to OCR (Optical character recognition)
OCR (Optical Character Recognition)
Optical Character Recognition( OCR )
Machine learning
Handwriting recogntion slides boeing

What's hot (20)

PDF
Design and implementation of optical character recognition using template mat...
PPTX
OCR 's Functions
PPT
PPTX
Ocr algorithm for ge’ez characters
DOC
Ocr abstract
PPT
Text reader [OCR]
PDF
Optical Character Recognition (OCR) System
PPTX
Optical Character Recognition (OCR)
PPTX
PPTX
OCR speech using Labview
DOCX
Optical character recognition IEEE Paper Study
PPTX
Optical Character Recognition (OCR) based Retrieval
PPTX
Final Report on Optical Character Recognition
PPTX
Basics of-optical-character-recognition
PPT
optical character recognition system
PDF
Optical Character Recognition Using Python
PPTX
Handwriting Recognition
PPTX
Optical Character Recognition
PPTX
Optical Character Recognition
DOCX
Project report of OCR Recognition
Design and implementation of optical character recognition using template mat...
OCR 's Functions
Ocr algorithm for ge’ez characters
Ocr abstract
Text reader [OCR]
Optical Character Recognition (OCR) System
Optical Character Recognition (OCR)
OCR speech using Labview
Optical character recognition IEEE Paper Study
Optical Character Recognition (OCR) based Retrieval
Final Report on Optical Character Recognition
Basics of-optical-character-recognition
optical character recognition system
Optical Character Recognition Using Python
Handwriting Recognition
Optical Character Recognition
Optical Character Recognition
Project report of OCR Recognition
Ad

Similar to State-of-Art Optical Character Recognition case (20)

PPTX
Ephesoft & Linux Webinar: Smart Capture™ is Now Even Smarter!
PPTX
Osource Company Profile
PPTX
Machine Teaching for workflow automation RIGA COMM 2020
PPTX
How to plan your Modern Workplace Project - SPS Denver October 2018
PDF
Nuts and Bolts of Building Compliance Process with Nintex and SharePoint 2013
PPTX
web design & development services
PDF
PPT
CrowningTechnologies1
PPTX
Maximizing Team Productivity with Microsoft Office 365
PDF
Nintex presentation Building forms and Workflows
PPTX
Global M365 Developer Bootcamp 2020 Hyderabad: KEYNOTE
PDF
SharePoint as a Business Platform Why, What and How? – No Code
PDF
Groundbreaking and Game-changing Enterprise Search Webinar
PPTX
Maximizing Your Office 365 Investments With OneDrive
PDF
Coexist or Integrate? Manage Unstructured Content from Diverse Repositories a...
PDF
Microsoft teams.pdf
PPT
Softengi Software Development Company Profile
PPTX
ESPC16-HybridWS.pptx
PDF
Netwoven nintex hawkeye webinar presentation
PDF
KLO VENTURE BPO
Ephesoft & Linux Webinar: Smart Capture™ is Now Even Smarter!
Osource Company Profile
Machine Teaching for workflow automation RIGA COMM 2020
How to plan your Modern Workplace Project - SPS Denver October 2018
Nuts and Bolts of Building Compliance Process with Nintex and SharePoint 2013
web design & development services
CrowningTechnologies1
Maximizing Team Productivity with Microsoft Office 365
Nintex presentation Building forms and Workflows
Global M365 Developer Bootcamp 2020 Hyderabad: KEYNOTE
SharePoint as a Business Platform Why, What and How? – No Code
Groundbreaking and Game-changing Enterprise Search Webinar
Maximizing Your Office 365 Investments With OneDrive
Coexist or Integrate? Manage Unstructured Content from Diverse Repositories a...
Microsoft teams.pdf
Softengi Software Development Company Profile
ESPC16-HybridWS.pptx
Netwoven nintex hawkeye webinar presentation
KLO VENTURE BPO
Ad

Recently uploaded (20)

PDF
WOOl fibre morphology and structure.pdf for textiles
PDF
How ambidextrous entrepreneurial leaders react to the artificial intelligence...
PPTX
TLE Review Electricity (Electricity).pptx
PDF
DASA ADMISSION 2024_FirstRound_FirstRank_LastRank.pdf
PDF
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
PDF
Getting started with AI Agents and Multi-Agent Systems
PDF
Hindi spoken digit analysis for native and non-native speakers
PPTX
Tartificialntelligence_presentation.pptx
PDF
A comparative study of natural language inference in Swahili using monolingua...
PPTX
OMC Textile Division Presentation 2021.pptx
PDF
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
PPTX
Modernising the Digital Integration Hub
PDF
Microsoft Solutions Partner Drive Digital Transformation with D365.pdf
PDF
A contest of sentiment analysis: k-nearest neighbor versus neural network
PDF
Architecture types and enterprise applications.pdf
PDF
2021 HotChips TSMC Packaging Technologies for Chiplets and 3D_0819 publish_pu...
PDF
August Patch Tuesday
PPTX
O2C Customer Invoices to Receipt V15A.pptx
PDF
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
PDF
DP Operators-handbook-extract for the Mautical Institute
WOOl fibre morphology and structure.pdf for textiles
How ambidextrous entrepreneurial leaders react to the artificial intelligence...
TLE Review Electricity (Electricity).pptx
DASA ADMISSION 2024_FirstRound_FirstRank_LastRank.pdf
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
Getting started with AI Agents and Multi-Agent Systems
Hindi spoken digit analysis for native and non-native speakers
Tartificialntelligence_presentation.pptx
A comparative study of natural language inference in Swahili using monolingua...
OMC Textile Division Presentation 2021.pptx
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
Modernising the Digital Integration Hub
Microsoft Solutions Partner Drive Digital Transformation with D365.pdf
A contest of sentiment analysis: k-nearest neighbor versus neural network
Architecture types and enterprise applications.pdf
2021 HotChips TSMC Packaging Technologies for Chiplets and 3D_0819 publish_pu...
August Patch Tuesday
O2C Customer Invoices to Receipt V15A.pptx
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
DP Operators-handbook-extract for the Mautical Institute

State-of-Art Optical Character Recognition case

  • 2. Optical Character Recognition OCR IS - > that can be processed by MASHINES. method of transferring TEXT from IMAGE form (such as scan, video or photo) -> into TEXTUAL FORM
  • 3. Optical Character Recognition - Global Source: Transparency Market Research, 2017 • Namely legal • Recruitment • Banking • Education • Healthcare • Finance • Government Sectors
  • 4. Global Optical Character Recognition Market, By Type Software Service Desktop based OCR Mobile based OCR Cloud based OCR Others (Batch OCR, Server based OCR etc.) Multi-tenant cloud (Public cloud) Private Cloud Consulting Outsourcing Implementation & Integration
  • 5. OCR - Implementation • Store it in a more compact form • Display or print the material without losing quality • Analyzing information • Apply to the text an electronic translation • Formatting or converting to speech • Convert books and docs to e – format • Automation of accounting systems in business • Text publishing on a web page • Text’s editing • Search words and phrases
  • 6. Current OCR problem - Global Current solutions are ‘heavy’ in computational cost, could easily take tens of seconds or even minutes to process single page. Handwritten or rare fonts could pose another problem for solutions currently on the market.
  • 7. • Deep learning based solution • Uses GPUs to speed up computations • Takes 400-500 milliseconds per page OUR Solution FAST
  • 8. ADAPTIVE • Trained just by observing the data • Adding new language takes only few weeks • Solves hard cases, like handwritten text recognition • Can be trained on client’s side • Any image source will work, like web-camera OUR Solution
  • 9. • Automatically groups related text blocks • Classifies blocks with complex relationships for business purposes • Doesn’t require retraining when templates change as long as the new template structure is more or less semantically the same SMART OUR Solution
  • 10. Team • Our team structure enables each member to grow safari-fix as a professional. • An employee creativity, deep and varied background are essential for our business. • We have an open and safe communication space to create positive and productive collaborations. • XOresearch is always in search of passionate, creative and self-motivated people. Join us and be a part of international and constantly evolving team.
  • 11. Contacts Headquarters 3, Republikas square, office 307 Riga LV-1010 Latvia Development office 5b, Dilova street, office 306 Kyiv 03150 Ukraine Latvia: Tel.: +371 67 305 084 kp@sapiensapi.com www LinkedIn Facebook