An Efficient Arabic Text Spotting from Natural Scenes Images

AN EFFICIENT
ARABIC TEXT
SPOTTING FROM
NATURAL SCENES
IMAGES
CONTENTS
• EAST Text detection
• Pipeline of text detection from natural scene
• Results of text detection on our data set
• The Optical character Recognition using Tesseract
• Results of text recognition our data set
• Comparison between Arabic & English recognition
• Character segmentation using MSER Using EASTR DATA
• Training TESSERACT with Arabic Data
EAST TEXT DETECTOR
• An Efficient and Accurate Scene Text Detector by Zhou (2017)
The core of text detection is the design of features to distinguish text from backgrounds
Fully Convolutional Network
FCN
Non Maximum Suppression
• Method consists of two stages: a Fully Convolutional Network and an NMS merging stage.
• The pipeline is flexible to produce either word level or line level predictions.
COMPARISON OF PIPELINES OF SEVERAL MODELS
STRUCTURE
The model is a fully-convolutional neural network adapted for text detection
that outputs dense per-pixel predictions of words or text lines.
This eliminates intermediate steps such as candidate proposal, text region
formation and word partition.
The post-processing steps only include thresholding and NMS on predicted
geometric shapes
Adopted DenseBox Method for Object detection
DENSEBOX
The single convolutional network simultaneously output multiple predicted bounding boxes and class
confidence.
All components of object detection in DenseBox are modeled as a Fully Convolutional Network FCN except
the non-maximum suppression step, so region proposal generation is unnecessary.
The system takes an image (at the size of m × n) as input, and output a m/ 4 × n/ 4 feature map with 5 channels.
Finally every pixel in the output map is converted to bounding box with score, and non-maximum suppression is
applied to those boxes whose scores pass the threshold
DENSEBOX
PIPELINE
The model can be decomposed in to three parts:
1. Feature extractor stem (convolutional network pre-trained
on ImageNet )
Four levels of feature maps, denoted as fi, are extracted from
stem, whose sizes are 1/ 32, 1 /16, 1 /8 and
1 /4 of the input image, respectively
2. Feature-merging branch
Fed to an unpooling layer to double its size
concatenated with the current feature map
conv1×1 bottleneck cuts down the number of channels and
reduces computation
conv3×3 that fuses the information to finally produce the
of this merging stage.
3. Output layer
From 32 channel
to one channel
of score map
U-
Shape
NON MAXIMUME SUPPRESSION
 Each output prediction have certain probability Pc
 Disregard All boxes whose Pc < 0.6
 For the remaining box
 Pick the box with the largest Pc
 Disregard all the remaining boxes IoU (Intersection over Union) > 0.5 with the output box in the previous step
RESULTS ON ARABIC
RESULTS ON ARABIC
RESULTS ON ARABIC
An Efficient Arabic Text Spotting from Natural Scenes Images
INPUT IMAGE
TESSERACT TEXT
RECOGNITION
EAST DETECTION EXTRACT TEXT ROI
OUTPUT
TEXT RECOGNITION
TESSERACT OCR
Tesseract is an open source OCR engine originally developed by Hewlett-Packard Laboratories, Bristol and
Hewlett-Packard Co.
It is considered one of the most accurate OCR engines that is available. It can read a wide variety of image
formats and can convert text written in more than 60 languages.
CODE
determine the ratio of the original image
dimensions to new image dimensions
we resize the image
The first layer is our output sigmoid activation which
gives us the probability of a region containing text or not
The second layer is the output feature map that
represents the “geometry” of the image
Extract Region of interest ROI
Text recognition using Tesseract
text_recognition.py --east frozen_east_text_detection.pb --image images/offer.jpg
text_detection.py --image images/offer.jpg --east frozen_east_text_detection.pb
CONVERTING Image into blob
passing the path to the EAST detector
•The output geometry map used to derive the bounding box coordinates of
text in our input images
•And similarly, the scores map, containing the probability of a given region
containing text
•rects : Stores the bounding box (x, y)-coordinates for text
regions
•confidences : Stores the probability associated with each
of the bounding boxes in rects
Measure geometric score and
probabilities and add
bounding box
An Efficient Arabic Text Spotting from Natural Scenes Images
EXTERMAL REGIONS
Extremal regions are connected areas that are characterized by uniform intensity and surrounded by a
contrast background.
The stability of a region can be measured by calculating how resistant the region is to the thresholding
variance.
This variance can be measured with a simple algorithm:
1. Applying the threshold generates an image A. Detect its connected pixel regions (extremal regions).
2. Increasing the threshold by a delta amount generates an image B. Detect its connected pixel regions
(extremal regions).
3. Compare image B with A. If a region in image A is similar to the same region in image B, then add it to the
same branch in the tree. The criteria of similarity may vary from implementation to implementation, but it's
usually related to the image area or general shape. If a region in image A appears to be split in image B,
create two new branches in the tree for the new regions, and associate them with the previous branch
MSER
The probability of each ER being a character is estimated using novel features calculated with complexity
O and only ERs with locally maximal probability are selected for the second stage,
ETERMAL REGION
DETECTOR
TRAIN DATA
CHARACTER RECOGNITION OF EASTR USING MSER
An Efficient Arabic Text Spotting from Natural Scenes Images
An Efficient Arabic Text Spotting from Natural Scenes Images
An Efficient Arabic Text Spotting from Natural Scenes Images
An Efficient Arabic Text Spotting from Natural Scenes Images
1 de 28

Recomendados

MATLAB IMPLEMENTATION OF SELF-ORGANIZING MAPS FOR CLUSTERING OF REMOTE SENSIN... por
MATLAB IMPLEMENTATION OF SELF-ORGANIZING MAPS FOR CLUSTERING OF REMOTE SENSIN...MATLAB IMPLEMENTATION OF SELF-ORGANIZING MAPS FOR CLUSTERING OF REMOTE SENSIN...
MATLAB IMPLEMENTATION OF SELF-ORGANIZING MAPS FOR CLUSTERING OF REMOTE SENSIN...Daksh Raj Chopra
2.5K vistas64 diapositivas
Introduction to image processing and pattern recognition por
Introduction to image processing and pattern recognitionIntroduction to image processing and pattern recognition
Introduction to image processing and pattern recognitionSaibee Alam
153 vistas25 diapositivas
face recognition system using LBP por
face recognition system using LBPface recognition system using LBP
face recognition system using LBPMarwan H. Noman
11.4K vistas20 diapositivas
Beyond Bag of Features: Adaptive Hilbert Scan Based Tree for Image Retrieval por
Beyond Bag of Features: Adaptive Hilbert Scan Based Tree for Image RetrievalBeyond Bag of Features: Adaptive Hilbert Scan Based Tree for Image Retrieval
Beyond Bag of Features: Adaptive Hilbert Scan Based Tree for Image RetrievalAssociation of Scientists, Developers and Faculties
92 vistas5 diapositivas
Texture descriptor based on local combination adaptive ternary pattern por
Texture descriptor based on local combination adaptive ternary patternTexture descriptor based on local combination adaptive ternary pattern
Texture descriptor based on local combination adaptive ternary patternProjectsatbangalore
171 vistas1 diapositiva
Ajay ppt region segmentation new copy por
Ajay ppt region segmentation new   copyAjay ppt region segmentation new   copy
Ajay ppt region segmentation new copyAjay Kumar Singh
8.6K vistas20 diapositivas

Más contenido relacionado

La actualidad más candente

Image processing and alignment with RNiftyReg and mmand por
Image processing and alignment with RNiftyReg and mmandImage processing and alignment with RNiftyReg and mmand
Image processing and alignment with RNiftyReg and mmandJonathan Clayden
1K vistas16 diapositivas
Computer vision,,summer training programme por
Computer vision,,summer training programmeComputer vision,,summer training programme
Computer vision,,summer training programmePraveen Pandey
1.3K vistas5 diapositivas
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS por
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMSCLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMSsipij
11 vistas11 diapositivas
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS por
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMSCLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMSsipij
10 vistas11 diapositivas
Image processing with open cv,regular training programme in waayoo.com por
Image processing with open cv,regular training programme in waayoo.comImage processing with open cv,regular training programme in waayoo.com
Image processing with open cv,regular training programme in waayoo.comPraveen Pandey
579 vistas3 diapositivas
Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Pr... por
Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Pr...Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Pr...
Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Pr...MLAI2
950 vistas11 diapositivas

La actualidad más candente(20)

Image processing and alignment with RNiftyReg and mmand por Jonathan Clayden
Image processing and alignment with RNiftyReg and mmandImage processing and alignment with RNiftyReg and mmand
Image processing and alignment with RNiftyReg and mmand
Jonathan Clayden1K vistas
Computer vision,,summer training programme por Praveen Pandey
Computer vision,,summer training programmeComputer vision,,summer training programme
Computer vision,,summer training programme
Praveen Pandey1.3K vistas
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS por sipij
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMSCLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS
sipij11 vistas
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS por sipij
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMSCLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS
CLASSIFICATION AND COMPARISON OF LICENSE PLATES LOCALIZATION ALGORITHMS
sipij10 vistas
Image processing with open cv,regular training programme in waayoo.com por Praveen Pandey
Image processing with open cv,regular training programme in waayoo.comImage processing with open cv,regular training programme in waayoo.com
Image processing with open cv,regular training programme in waayoo.com
Praveen Pandey579 vistas
Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Pr... por MLAI2
Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Pr...Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Pr...
Learning to Extrapolate Knowledge: Transductive Few-shot Out-of-Graph Link Pr...
MLAI2950 vistas
Learning Graph Representation for Data-Efficiency RL por lauratoni4
Learning Graph Representation for Data-Efficiency RLLearning Graph Representation for Data-Efficiency RL
Learning Graph Representation for Data-Efficiency RL
lauratoni4135 vistas
SEGMENTATION AND CLASSIFICATION OF POINT CLOUDS FROM DENSE AERIAL IMAGE MATCHING por ijma
SEGMENTATION AND CLASSIFICATION OF POINT CLOUDS FROM DENSE AERIAL IMAGE MATCHINGSEGMENTATION AND CLASSIFICATION OF POINT CLOUDS FROM DENSE AERIAL IMAGE MATCHING
SEGMENTATION AND CLASSIFICATION OF POINT CLOUDS FROM DENSE AERIAL IMAGE MATCHING
ijma1.6K vistas
Graph Signal Processing for Machine Learning A Review and New Perspectives - ... por lauratoni4
Graph Signal Processing for Machine Learning A Review and New Perspectives - ...Graph Signal Processing for Machine Learning A Review and New Perspectives - ...
Graph Signal Processing for Machine Learning A Review and New Perspectives - ...
lauratoni4324 vistas
Graph Signal Processing for Machine Learning A Review and New Perspectives - ... por lauratoni4
Graph Signal Processing for Machine Learning A Review and New Perspectives - ...Graph Signal Processing for Machine Learning A Review and New Perspectives - ...
Graph Signal Processing for Machine Learning A Review and New Perspectives - ...
lauratoni4198 vistas
Laplacian-regularized Graph Bandits por lauratoni4
Laplacian-regularized Graph BanditsLaplacian-regularized Graph Bandits
Laplacian-regularized Graph Bandits
lauratoni4122 vistas
fuzzy LBP for face recognition ppt por Abdullah Gubbi
fuzzy LBP for face recognition pptfuzzy LBP for face recognition ppt
fuzzy LBP for face recognition ppt
Abdullah Gubbi4.1K vistas
Graph Signal Processing for Machine Learning A Review and New Perspectives - ... por lauratoni4
Graph Signal Processing for Machine Learning A Review and New Perspectives - ...Graph Signal Processing for Machine Learning A Review and New Perspectives - ...
Graph Signal Processing for Machine Learning A Review and New Perspectives - ...
lauratoni4247 vistas
Surface reconstruction using point cloud por ishan kossambe
Surface reconstruction using point cloudSurface reconstruction using point cloud
Surface reconstruction using point cloud
ishan kossambe1.9K vistas
Unsupervised/Self-supervvised visual object tracking por Yu Huang
Unsupervised/Self-supervvised visual object trackingUnsupervised/Self-supervvised visual object tracking
Unsupervised/Self-supervvised visual object tracking
Yu Huang129 vistas
3-d interpretation from single 2-d image for autonomous driving por Yu Huang
3-d interpretation from single 2-d image for autonomous driving3-d interpretation from single 2-d image for autonomous driving
3-d interpretation from single 2-d image for autonomous driving
Yu Huang1.6K vistas
Fisheye Omnidirectional View in Autonomous Driving por Yu Huang
Fisheye Omnidirectional View in Autonomous DrivingFisheye Omnidirectional View in Autonomous Driving
Fisheye Omnidirectional View in Autonomous Driving
Yu Huang1.3K vistas
Interactive segmentation for change detection in multispectral remote sensing... por Ecway Technologies
Interactive segmentation for change detection in multispectral remote sensing...Interactive segmentation for change detection in multispectral remote sensing...
Interactive segmentation for change detection in multispectral remote sensing...
Ecway Technologies192 vistas

Similar a An Efficient Arabic Text Spotting from Natural Scenes Images

Road Segmentation from satellites images por
Road Segmentation from satellites imagesRoad Segmentation from satellites images
Road Segmentation from satellites imagesYoussefKitane
66 vistas5 diapositivas
Enhanced characterness for text detection in the wild por
Enhanced characterness for text detection in the wildEnhanced characterness for text detection in the wild
Enhanced characterness for text detection in the wildPrerana Mukherjee
60 vistas16 diapositivas
Overview of Convolutional Neural Networks por
Overview of Convolutional Neural NetworksOverview of Convolutional Neural Networks
Overview of Convolutional Neural Networksananth
7.7K vistas43 diapositivas
Object Elimination and Reconstruction Using an Effective Inpainting Method por
Object Elimination and Reconstruction Using an Effective Inpainting MethodObject Elimination and Reconstruction Using an Effective Inpainting Method
Object Elimination and Reconstruction Using an Effective Inpainting MethodIOSR Journals
282 vistas5 diapositivas
Ocr using tensor flow por
Ocr using tensor flowOcr using tensor flow
Ocr using tensor flowNaresh Kumar
166 vistas13 diapositivas
Wits presentation 6_28072015 por
Wits presentation 6_28072015Wits presentation 6_28072015
Wits presentation 6_28072015Beatrice van Eden
848 vistas33 diapositivas

Similar a An Efficient Arabic Text Spotting from Natural Scenes Images(20)

Road Segmentation from satellites images por YoussefKitane
Road Segmentation from satellites imagesRoad Segmentation from satellites images
Road Segmentation from satellites images
YoussefKitane66 vistas
Enhanced characterness for text detection in the wild por Prerana Mukherjee
Enhanced characterness for text detection in the wildEnhanced characterness for text detection in the wild
Enhanced characterness for text detection in the wild
Prerana Mukherjee60 vistas
Overview of Convolutional Neural Networks por ananth
Overview of Convolutional Neural NetworksOverview of Convolutional Neural Networks
Overview of Convolutional Neural Networks
ananth7.7K vistas
Object Elimination and Reconstruction Using an Effective Inpainting Method por IOSR Journals
Object Elimination and Reconstruction Using an Effective Inpainting MethodObject Elimination and Reconstruction Using an Effective Inpainting Method
Object Elimination and Reconstruction Using an Effective Inpainting Method
IOSR Journals282 vistas
Ocr using tensor flow por Naresh Kumar
Ocr using tensor flowOcr using tensor flow
Ocr using tensor flow
Naresh Kumar166 vistas
Deep learning for 3-D Scene Reconstruction and Modeling por Yu Huang
Deep learning for 3-D Scene Reconstruction and Modeling Deep learning for 3-D Scene Reconstruction and Modeling
Deep learning for 3-D Scene Reconstruction and Modeling
Yu Huang9.6K vistas
YARCA (Yet Another Raycasting Application) Project por graphitech
YARCA (Yet Another Raycasting Application) ProjectYARCA (Yet Another Raycasting Application) Project
YARCA (Yet Another Raycasting Application) Project
graphitech370 vistas
201907 AutoML and Neural Architecture Search por DaeJin Kim
201907 AutoML and Neural Architecture Search201907 AutoML and Neural Architecture Search
201907 AutoML and Neural Architecture Search
DaeJin Kim950 vistas
intro-to-cnn-April_2020.pptx por ssuser3aa461
intro-to-cnn-April_2020.pptxintro-to-cnn-April_2020.pptx
intro-to-cnn-April_2020.pptx
ssuser3aa46115 vistas
Image segmentation for high resolution images por jeet patalia
Image segmentation for high resolution imagesImage segmentation for high resolution images
Image segmentation for high resolution images
jeet patalia858 vistas
A STUDY OF METHODS FOR TRAINING WITH DIFFERENT DATASETS IN IMAGE CLASSIFICATION por ADEIJ Journal
A STUDY OF METHODS FOR TRAINING WITH DIFFERENT DATASETS IN IMAGE CLASSIFICATIONA STUDY OF METHODS FOR TRAINING WITH DIFFERENT DATASETS IN IMAGE CLASSIFICATION
A STUDY OF METHODS FOR TRAINING WITH DIFFERENT DATASETS IN IMAGE CLASSIFICATION
ADEIJ Journal37 vistas
Error Rate Analysis of MIMO System Using V Blast Detection Technique in Fadin... por IJERA Editor
Error Rate Analysis of MIMO System Using V Blast Detection Technique in Fadin...Error Rate Analysis of MIMO System Using V Blast Detection Technique in Fadin...
Error Rate Analysis of MIMO System Using V Blast Detection Technique in Fadin...
IJERA Editor315 vistas
MDCT audio coding with pulse vector quantizers por Ericsson
MDCT audio coding with pulse vector quantizersMDCT audio coding with pulse vector quantizers
MDCT audio coding with pulse vector quantizers
Ericsson682 vistas
C04741319 por IOSR-JEN
C04741319C04741319
C04741319
IOSR-JEN266 vistas

Último

GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N... por
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...James Anderson
142 vistas32 diapositivas
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ... por
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...ShapeBlue
52 vistas10 diapositivas
Developments to CloudStack’s SDN ecosystem: Integration with VMWare NSX 4 - P... por
Developments to CloudStack’s SDN ecosystem: Integration with VMWare NSX 4 - P...Developments to CloudStack’s SDN ecosystem: Integration with VMWare NSX 4 - P...
Developments to CloudStack’s SDN ecosystem: Integration with VMWare NSX 4 - P...ShapeBlue
120 vistas62 diapositivas
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&T por
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&TCloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&T
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&TShapeBlue
81 vistas34 diapositivas
CloudStack Object Storage - An Introduction - Vladimir Petrov - ShapeBlue por
CloudStack Object Storage - An Introduction - Vladimir Petrov - ShapeBlueCloudStack Object Storage - An Introduction - Vladimir Petrov - ShapeBlue
CloudStack Object Storage - An Introduction - Vladimir Petrov - ShapeBlueShapeBlue
63 vistas15 diapositivas
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue por
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlueElevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlueShapeBlue
149 vistas7 diapositivas

Último(20)

GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N... por James Anderson
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
James Anderson142 vistas
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ... por ShapeBlue
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...
ShapeBlue52 vistas
Developments to CloudStack’s SDN ecosystem: Integration with VMWare NSX 4 - P... por ShapeBlue
Developments to CloudStack’s SDN ecosystem: Integration with VMWare NSX 4 - P...Developments to CloudStack’s SDN ecosystem: Integration with VMWare NSX 4 - P...
Developments to CloudStack’s SDN ecosystem: Integration with VMWare NSX 4 - P...
ShapeBlue120 vistas
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&T por ShapeBlue
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&TCloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&T
CloudStack and GitOps at Enterprise Scale - Alex Dometrius, Rene Glover - AT&T
ShapeBlue81 vistas
CloudStack Object Storage - An Introduction - Vladimir Petrov - ShapeBlue por ShapeBlue
CloudStack Object Storage - An Introduction - Vladimir Petrov - ShapeBlueCloudStack Object Storage - An Introduction - Vladimir Petrov - ShapeBlue
CloudStack Object Storage - An Introduction - Vladimir Petrov - ShapeBlue
ShapeBlue63 vistas
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue por ShapeBlue
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlueElevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue
ShapeBlue149 vistas
Backup and Disaster Recovery with CloudStack and StorPool - Workshop - Venko ... por ShapeBlue
Backup and Disaster Recovery with CloudStack and StorPool - Workshop - Venko ...Backup and Disaster Recovery with CloudStack and StorPool - Workshop - Venko ...
Backup and Disaster Recovery with CloudStack and StorPool - Workshop - Venko ...
ShapeBlue114 vistas
Digital Personal Data Protection (DPDP) Practical Approach For CISOs por Priyanka Aash
Digital Personal Data Protection (DPDP) Practical Approach For CISOsDigital Personal Data Protection (DPDP) Practical Approach For CISOs
Digital Personal Data Protection (DPDP) Practical Approach For CISOs
Priyanka Aash103 vistas
Igniting Next Level Productivity with AI-Infused Data Integration Workflows por Safe Software
Igniting Next Level Productivity with AI-Infused Data Integration Workflows Igniting Next Level Productivity with AI-Infused Data Integration Workflows
Igniting Next Level Productivity with AI-Infused Data Integration Workflows
Safe Software373 vistas
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue por ShapeBlue
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlueCloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue
ShapeBlue68 vistas
State of the Union - Rohit Yadav - Apache CloudStack por ShapeBlue
State of the Union - Rohit Yadav - Apache CloudStackState of the Union - Rohit Yadav - Apache CloudStack
State of the Union - Rohit Yadav - Apache CloudStack
ShapeBlue218 vistas
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit... por ShapeBlue
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...
Transitioning from VMware vCloud to Apache CloudStack: A Path to Profitabilit...
ShapeBlue86 vistas
The Power of Heat Decarbonisation Plans in the Built Environment por IES VE
The Power of Heat Decarbonisation Plans in the Built EnvironmentThe Power of Heat Decarbonisation Plans in the Built Environment
The Power of Heat Decarbonisation Plans in the Built Environment
IES VE67 vistas
Business Analyst Series 2023 - Week 4 Session 7 por DianaGray10
Business Analyst Series 2023 -  Week 4 Session 7Business Analyst Series 2023 -  Week 4 Session 7
Business Analyst Series 2023 - Week 4 Session 7
DianaGray10110 vistas
Extending KVM Host HA for Non-NFS Storage - Alex Ivanov - StorPool por ShapeBlue
Extending KVM Host HA for Non-NFS Storage -  Alex Ivanov - StorPoolExtending KVM Host HA for Non-NFS Storage -  Alex Ivanov - StorPool
Extending KVM Host HA for Non-NFS Storage - Alex Ivanov - StorPool
ShapeBlue56 vistas
DRaaS using Snapshot copy and destination selection (DRaaS) - Alexandre Matti... por ShapeBlue
DRaaS using Snapshot copy and destination selection (DRaaS) - Alexandre Matti...DRaaS using Snapshot copy and destination selection (DRaaS) - Alexandre Matti...
DRaaS using Snapshot copy and destination selection (DRaaS) - Alexandre Matti...
ShapeBlue69 vistas
"Surviving highload with Node.js", Andrii Shumada por Fwdays
"Surviving highload with Node.js", Andrii Shumada "Surviving highload with Node.js", Andrii Shumada
"Surviving highload with Node.js", Andrii Shumada
Fwdays49 vistas
Data Integrity for Banking and Financial Services por Precisely
Data Integrity for Banking and Financial ServicesData Integrity for Banking and Financial Services
Data Integrity for Banking and Financial Services
Precisely76 vistas

An Efficient Arabic Text Spotting from Natural Scenes Images

  • 1. AN EFFICIENT ARABIC TEXT SPOTTING FROM NATURAL SCENES IMAGES
  • 2. CONTENTS • EAST Text detection • Pipeline of text detection from natural scene • Results of text detection on our data set • The Optical character Recognition using Tesseract • Results of text recognition our data set • Comparison between Arabic & English recognition • Character segmentation using MSER Using EASTR DATA • Training TESSERACT with Arabic Data
  • 3. EAST TEXT DETECTOR • An Efficient and Accurate Scene Text Detector by Zhou (2017) The core of text detection is the design of features to distinguish text from backgrounds Fully Convolutional Network FCN Non Maximum Suppression • Method consists of two stages: a Fully Convolutional Network and an NMS merging stage. • The pipeline is flexible to produce either word level or line level predictions.
  • 4. COMPARISON OF PIPELINES OF SEVERAL MODELS
  • 5. STRUCTURE The model is a fully-convolutional neural network adapted for text detection that outputs dense per-pixel predictions of words or text lines. This eliminates intermediate steps such as candidate proposal, text region formation and word partition. The post-processing steps only include thresholding and NMS on predicted geometric shapes Adopted DenseBox Method for Object detection
  • 6. DENSEBOX The single convolutional network simultaneously output multiple predicted bounding boxes and class confidence. All components of object detection in DenseBox are modeled as a Fully Convolutional Network FCN except the non-maximum suppression step, so region proposal generation is unnecessary. The system takes an image (at the size of m × n) as input, and output a m/ 4 × n/ 4 feature map with 5 channels. Finally every pixel in the output map is converted to bounding box with score, and non-maximum suppression is applied to those boxes whose scores pass the threshold
  • 8. PIPELINE The model can be decomposed in to three parts: 1. Feature extractor stem (convolutional network pre-trained on ImageNet ) Four levels of feature maps, denoted as fi, are extracted from stem, whose sizes are 1/ 32, 1 /16, 1 /8 and 1 /4 of the input image, respectively 2. Feature-merging branch Fed to an unpooling layer to double its size concatenated with the current feature map conv1×1 bottleneck cuts down the number of channels and reduces computation conv3×3 that fuses the information to finally produce the of this merging stage. 3. Output layer From 32 channel to one channel of score map U- Shape
  • 9. NON MAXIMUME SUPPRESSION  Each output prediction have certain probability Pc  Disregard All boxes whose Pc < 0.6  For the remaining box  Pick the box with the largest Pc  Disregard all the remaining boxes IoU (Intersection over Union) > 0.5 with the output box in the previous step
  • 14. INPUT IMAGE TESSERACT TEXT RECOGNITION EAST DETECTION EXTRACT TEXT ROI OUTPUT TEXT RECOGNITION
  • 15. TESSERACT OCR Tesseract is an open source OCR engine originally developed by Hewlett-Packard Laboratories, Bristol and Hewlett-Packard Co. It is considered one of the most accurate OCR engines that is available. It can read a wide variety of image formats and can convert text written in more than 60 languages.
  • 16. CODE determine the ratio of the original image dimensions to new image dimensions we resize the image The first layer is our output sigmoid activation which gives us the probability of a region containing text or not The second layer is the output feature map that represents the “geometry” of the image
  • 17. Extract Region of interest ROI Text recognition using Tesseract text_recognition.py --east frozen_east_text_detection.pb --image images/offer.jpg text_detection.py --image images/offer.jpg --east frozen_east_text_detection.pb
  • 18. CONVERTING Image into blob passing the path to the EAST detector •The output geometry map used to derive the bounding box coordinates of text in our input images •And similarly, the scores map, containing the probability of a given region containing text
  • 19. •rects : Stores the bounding box (x, y)-coordinates for text regions •confidences : Stores the probability associated with each of the bounding boxes in rects Measure geometric score and probabilities and add bounding box
  • 21. EXTERMAL REGIONS Extremal regions are connected areas that are characterized by uniform intensity and surrounded by a contrast background. The stability of a region can be measured by calculating how resistant the region is to the thresholding variance. This variance can be measured with a simple algorithm: 1. Applying the threshold generates an image A. Detect its connected pixel regions (extremal regions). 2. Increasing the threshold by a delta amount generates an image B. Detect its connected pixel regions (extremal regions). 3. Compare image B with A. If a region in image A is similar to the same region in image B, then add it to the same branch in the tree. The criteria of similarity may vary from implementation to implementation, but it's usually related to the image area or general shape. If a region in image A appears to be split in image B, create two new branches in the tree for the new regions, and associate them with the previous branch
  • 22. MSER The probability of each ER being a character is estimated using novel features calculated with complexity O and only ERs with locally maximal probability are selected for the second stage, ETERMAL REGION DETECTOR
  • 24. CHARACTER RECOGNITION OF EASTR USING MSER