SlideShare a Scribd company logo
1 of 3
Download to read offline
IDENTIFYING TOP/BOTTOM FIELD IN INTERLACED VIDEO
Raja Subramanian and Sanjeev Retna
Mistral Solutions Pvt Ltd., India
ABSTRACT
This paper elaborates an approach that can be adopted to
determine top/bottom fields in an interlaced video. Knowing
the top and bottom field is important if the video is de-
interlaced using Field Combination, Weaving + Bob, Discard
and other algorithms based on motion detection.
Determining the field information helps to re-construct the
frame with lesser artifacts. This approach can be used if the
top/bottom field information is not provided by video
decoder chip.
INTRODUCTION
Interlaced video has been in use for more than 50 years.
When dealing with interlaced video, de-interlacing
algorithms are essential to remove any interlacing artifacts.
There are many de-interlacing algorithms available for
NTSC/PAL interlaced video. For low-end systems (system
with less processing capability), following approaches can be
considered:-
Algorithm Description Quality
Stationary Moving
Weaving /
Field
Combine
Combine the
top and bottom
field to form a
single frame
Same as the
source
Artifacts due to
time delay
between top and
bottom field
Discard Discard the top
or bottom field,
resize (double
the height)
other field to
construct a
frame
Stair case
artifacts due to
resize/line copy.
Image quality
degrades.
Stair case
artifacts due to
resize/line copy.
Image quality
degrades.
BOB Interpolate the
top and bottom
field (resize)
and construct
the frame at
doubled FPS
Image quality is
same or better
than the discard
algorithm.
Edges of the
objects in the
video frame
may be
incorrect in size
and colour.
Flickers if the
display refresh
rate is incorrect
or variable.
Image quality is
same or better
than the discard
algorithm.
Edges of the
objects in the
video frame may
be incorrect in
size and colour.
Flickers if the
display refresh
rate is incorrect
or variable.
BOB +
Weaving
Use the
previous top or
bottom field
data to
construct the
frame at
doubled FPS
Same as the
source
Close to the
source. Lesser
artifacts than all
the above
The captured frame shown below depicts the stair-case
artifacts of discard algorithm.
If the field order is unknown (or not provided by the video
decoder chip) then Discard or Bob algorithm is the efficient
method to de-interlace.
In most cases, frames re-constructed by Field Combination
or BOB + Weaving quality will be close to the original input
frame.
Most of the NTSC/PAL video decoder provides the field
order information. However, in some systems, it may not be
possible to get field order due to some operations like
serial/parallel conversion. Also the video can be of various
resolutions and type (NTSC, PAL, etc.). Therefore, the
algorithm needs to use a common mechanism to determine
the field order.
Figure 2: Bob + Weaving
Figure 1: Discard
June 2013
Field Dominance Algorithm
Each digitized input video frame contains two parts which is
received as a top and bottom field with respect to time order.
If the field order information from video decoder chip is not
available and the de-interlacing algorithm receives an top
field first, then re-construction of the frame is illustrated
below.
In this case the video frames are re-constructed correctly.
If the algorithm receives the bottom field as a first field then
the field order is reversed. Re-construction of the video
frames in such a case is illustrated as below:
For Example, YUV 4:2:2 little-endian data with top field
dominance will be stored as shown
Line 1(top1) → U0 Y0 V0 Y1 U2 Y2 V2 Y3...
Line 2(bottom1) → U0 Y0 V0 Y1 U2 Y2 V2 Y3...
Line 3(top2) → U0 Y0 V0 Y1 U2 Y2 V2 Y3...
Line 4(bottom2) → U0 Y0 V0 Y1 U2 Y2 V2 Y3...
.
.
.
Line n → U0 Y0 V0 Y1 U2 Y2 V2 Y3..
In this case the quality of the re-constructed frame is same as
the input source.
If field order is reversed, then progressively weaved data
may look as follows
Line 1(bottom1) → U0 Y0 V0 Y1 U2 Y2 V2 Y3...
Line 2(top2) → U0 Y0 V0 Y1 U2 Y2 V2 Y3...
Line 3(bottom2) → U0 Y0 V0 Y1 U2 Y2 V2 Y3...
Line 4(top3) → U0 Y0 V0 Y1 U2 Y2 V2 Y3...
.
.
.
Line n → U0 Y0 V0 Y1 U2 Y2 V2 Y3..
The re-constructed frame quality is not good due to the
swapped lines.
To determine the field order, the algorithm must assume the
field order is correct and then interpolate the two consecutive
lines of the top fields using the 2-points or 6-points median
algorithm and compare each pixel's luminance against the
bottom line's pixels. This procedure can be repeated again
assuming the field is reversed. Comparing these two helps in
identifying the field order.
Assuming the luminance values of the frame are arranged in
the following order
Line 1, y(1) → x1,x2,x3...
Line 2, y(2) → x1,x2,x3...
Line 3, y(3) → x1,x2,x3...
Line 4, y(4) → x1,x2,x3...
.
Line n-1, y(n-1) → x1,x2,x3...x(m-1),x(m),x(m+1),...
Line n, y(n) → x1,x2,x3...x(m-1),x(m),x(m+1),...
Line n+1, y(n+1) → x1,x2,x3...x(m-1),x(m),x(m+1),...
the top field confidence for any point fi( x(m), y(n) ) point
can be calculated by,
Figure 4: Field-swap artifacts
Figure 3: Proper Field Ordering
In the above formula, luminance value of column x(m) and
row y(n) can compared with the calculated median
value(interpolated).
We can also calculate the bottom Field confidence by,
Sum of all top field confidence gives the total top confidence
value for the frame. Similarly the total
bottom field dominance values can be calculated by adding
all the bottom confidence values .
Following formula helps to determine if the field is swapped
or not.
Once the field information is detected, it is necessary to run
this algorithm at regular interval to reaffirm the file order.
Accuracy
Input Video Data Accuracy (in %)
Moving Picture 99 – 100
Still picture (paused video or
picture)
95 – 100
Still picture (paused video or
frame with same color)
>75
Performance
Event Timings (in millisecond)
Time to determine field order (2
- point mean algorithm running
33 - 99
on 400mhz ARM Cortex-M3 )
Time to detect re-plug and
correct the Field order
(Algorithm runs every 1000 ms)
1000-3000
CONCLUSION
The interpolated video data will approximately be equal to
the original data. This method is more reliable to identify the
field dominance for most video data and hence improves the
picture quality dramatically.
2-point median algorithm can be substituted to reduce the
CPU load further.
Further, this algorithm is influenced by input video data and
the field order may change as a result of the video input
cable being removed/re-plugged. So, it's necessary to run this
algorithm at periodic interval to correct the field dominance
accordingly.
About Authors
Raja Subramanian is a member of the engineering team at
Mistral Solutions Pvt Ltd. He has completed Masters in
Computer Science from Bharathidasan University and has 14
years of experience in designing technical solutions in the
area of audio, video, storage and user interface for consumer
electronic products. His expertise is in the field of
autonomous navigation, video surveillance and data security.
Sanjeev Retna is a member of the engineering team at
Mistral Solutions Pvt Ltd. He has a Bachelor of Engineering
degree from Anna University. His expertise is in the field of
autonomous navigation, camera and video related products.

More Related Content

Similar to Field Dominance Algorithm

Error Resiliency and Concealment in H.264 MPEG-4 Part 10
Error Resiliency and Concealment in H.264 MPEG-4 Part 10Error Resiliency and Concealment in H.264 MPEG-4 Part 10
Error Resiliency and Concealment in H.264 MPEG-4 Part 10coldfire7
 
Performance analysis of NOR CAM cell using CMOS-HP, CMOS-LP and FinFET 16nm t...
Performance analysis of NOR CAM cell using CMOS-HP, CMOS-LP and FinFET 16nm t...Performance analysis of NOR CAM cell using CMOS-HP, CMOS-LP and FinFET 16nm t...
Performance analysis of NOR CAM cell using CMOS-HP, CMOS-LP and FinFET 16nm t...IRJET Journal
 
Atc On An Simd Cots System Wmpp05
Atc On An Simd Cots System   Wmpp05Atc On An Simd Cots System   Wmpp05
Atc On An Simd Cots System Wmpp05Ülger Ahmet
 
Vibration Analysis of a Frame
Vibration Analysis of a FrameVibration Analysis of a Frame
Vibration Analysis of a FrameFaisalManzoor32
 
Adaptive Trilateral Filter for In-Loop Filtering
Adaptive Trilateral Filter for In-Loop FilteringAdaptive Trilateral Filter for In-Loop Filtering
Adaptive Trilateral Filter for In-Loop Filteringcsandit
 
AREA OPTIMIZED FPGA IMPLEMENTATION FOR GENERATION OF RADAR PULSE COM-PRESSION...
AREA OPTIMIZED FPGA IMPLEMENTATION FOR GENERATION OF RADAR PULSE COM-PRESSION...AREA OPTIMIZED FPGA IMPLEMENTATION FOR GENERATION OF RADAR PULSE COM-PRESSION...
AREA OPTIMIZED FPGA IMPLEMENTATION FOR GENERATION OF RADAR PULSE COM-PRESSION...VLSICS Design
 
Overview of signal integrity simulation for sfp+ interface serial links with ...
Overview of signal integrity simulation for sfp+ interface serial links with ...Overview of signal integrity simulation for sfp+ interface serial links with ...
Overview of signal integrity simulation for sfp+ interface serial links with ...Conference Papers
 
Image Storage, Indexing and Recognition
Image Storage, Indexing and RecognitionImage Storage, Indexing and Recognition
Image Storage, Indexing and RecognitionAdeel Riaz
 
Explaining the decisions of image/video classifiers
Explaining the decisions of image/video classifiersExplaining the decisions of image/video classifiers
Explaining the decisions of image/video classifiersVasileiosMezaris
 
igorFreire_UCI_real-time-dsp_reports
igorFreire_UCI_real-time-dsp_reportsigorFreire_UCI_real-time-dsp_reports
igorFreire_UCI_real-time-dsp_reportsIgor Freire
 
Post Production Glossary
Post Production GlossaryPost Production Glossary
Post Production GlossaryJoe Nasr
 
A Wavelet - Based Object Watermarking System for MPEG4 Video
A Wavelet - Based Object Watermarking System for MPEG4 VideoA Wavelet - Based Object Watermarking System for MPEG4 Video
A Wavelet - Based Object Watermarking System for MPEG4 VideoCSCJournals
 
Motion estimation overview
Motion estimation overviewMotion estimation overview
Motion estimation overviewYoss Cohen
 
De Interlacing Techniques
De Interlacing TechniquesDe Interlacing Techniques
De Interlacing TechniquesRamesh Prasad
 

Similar to Field Dominance Algorithm (20)

Mo2521632166
Mo2521632166Mo2521632166
Mo2521632166
 
Error Resiliency and Concealment in H.264 MPEG-4 Part 10
Error Resiliency and Concealment in H.264 MPEG-4 Part 10Error Resiliency and Concealment in H.264 MPEG-4 Part 10
Error Resiliency and Concealment in H.264 MPEG-4 Part 10
 
Performance analysis of NOR CAM cell using CMOS-HP, CMOS-LP and FinFET 16nm t...
Performance analysis of NOR CAM cell using CMOS-HP, CMOS-LP and FinFET 16nm t...Performance analysis of NOR CAM cell using CMOS-HP, CMOS-LP and FinFET 16nm t...
Performance analysis of NOR CAM cell using CMOS-HP, CMOS-LP and FinFET 16nm t...
 
Atc On An Simd Cots System Wmpp05
Atc On An Simd Cots System   Wmpp05Atc On An Simd Cots System   Wmpp05
Atc On An Simd Cots System Wmpp05
 
Vibration Analysis of a Frame
Vibration Analysis of a FrameVibration Analysis of a Frame
Vibration Analysis of a Frame
 
Adaptive Trilateral Filter for In-Loop Filtering
Adaptive Trilateral Filter for In-Loop FilteringAdaptive Trilateral Filter for In-Loop Filtering
Adaptive Trilateral Filter for In-Loop Filtering
 
AREA OPTIMIZED FPGA IMPLEMENTATION FOR GENERATION OF RADAR PULSE COM-PRESSION...
AREA OPTIMIZED FPGA IMPLEMENTATION FOR GENERATION OF RADAR PULSE COM-PRESSION...AREA OPTIMIZED FPGA IMPLEMENTATION FOR GENERATION OF RADAR PULSE COM-PRESSION...
AREA OPTIMIZED FPGA IMPLEMENTATION FOR GENERATION OF RADAR PULSE COM-PRESSION...
 
video comparison
video comparison video comparison
video comparison
 
Overview of signal integrity simulation for sfp+ interface serial links with ...
Overview of signal integrity simulation for sfp+ interface serial links with ...Overview of signal integrity simulation for sfp+ interface serial links with ...
Overview of signal integrity simulation for sfp+ interface serial links with ...
 
Image Storage, Indexing and Recognition
Image Storage, Indexing and RecognitionImage Storage, Indexing and Recognition
Image Storage, Indexing and Recognition
 
Gg3311121115
Gg3311121115Gg3311121115
Gg3311121115
 
Explaining the decisions of image/video classifiers
Explaining the decisions of image/video classifiersExplaining the decisions of image/video classifiers
Explaining the decisions of image/video classifiers
 
igorFreire_UCI_real-time-dsp_reports
igorFreire_UCI_real-time-dsp_reportsigorFreire_UCI_real-time-dsp_reports
igorFreire_UCI_real-time-dsp_reports
 
Downlink speech quality
Downlink speech qualityDownlink speech quality
Downlink speech quality
 
Link adaptation
Link adaptationLink adaptation
Link adaptation
 
Post Production Glossary
Post Production GlossaryPost Production Glossary
Post Production Glossary
 
5 ofdm
5 ofdm5 ofdm
5 ofdm
 
A Wavelet - Based Object Watermarking System for MPEG4 Video
A Wavelet - Based Object Watermarking System for MPEG4 VideoA Wavelet - Based Object Watermarking System for MPEG4 Video
A Wavelet - Based Object Watermarking System for MPEG4 Video
 
Motion estimation overview
Motion estimation overviewMotion estimation overview
Motion estimation overview
 
De Interlacing Techniques
De Interlacing TechniquesDe Interlacing Techniques
De Interlacing Techniques
 

More from Mistral Solutions

An infographic from Mistral explaining Semiconductors
An infographic from Mistral explaining SemiconductorsAn infographic from Mistral explaining Semiconductors
An infographic from Mistral explaining SemiconductorsMistral Solutions
 
Interfacing a Thermal Printer using an AM437x PoM
Interfacing a Thermal Printer using an AM437x PoMInterfacing a Thermal Printer using an AM437x PoM
Interfacing a Thermal Printer using an AM437x PoMMistral Solutions
 
IoT based Industrial Gateway (IoT-SDK) built around Sitara™ AM437x processors...
IoT based Industrial Gateway (IoT-SDK) built around Sitara™ AM437x processors...IoT based Industrial Gateway (IoT-SDK) built around Sitara™ AM437x processors...
IoT based Industrial Gateway (IoT-SDK) built around Sitara™ AM437x processors...Mistral Solutions
 
Designing a Product on Module for Health and Wellness Applications
Designing a Product on Module for Health and Wellness ApplicationsDesigning a Product on Module for Health and Wellness Applications
Designing a Product on Module for Health and Wellness ApplicationsMistral Solutions
 
iOS Application Development for Drones
iOS Application Development for DronesiOS Application Development for Drones
iOS Application Development for DronesMistral Solutions
 
Hardware Software Co-Design - White Paper
Hardware Software Co-Design - White PaperHardware Software Co-Design - White Paper
Hardware Software Co-Design - White PaperMistral Solutions
 
RTM Break Out Card - Datasheet
RTM Break Out Card - DatasheetRTM Break Out Card - Datasheet
RTM Break Out Card - DatasheetMistral Solutions
 
Power Spikes Isolation to avoid corruption within sensitive ICs Solutions
Power Spikes Isolation to avoid corruption within sensitive ICs SolutionsPower Spikes Isolation to avoid corruption within sensitive ICs Solutions
Power Spikes Isolation to avoid corruption within sensitive ICs SolutionsMistral Solutions
 
Algorithm for Lossy Image Compression using FPGA
Algorithm for Lossy Image Compression using FPGAAlgorithm for Lossy Image Compression using FPGA
Algorithm for Lossy Image Compression using FPGAMistral Solutions
 
High-Definition Rugged DVR - Case Study
High-Definition Rugged DVR - Case StudyHigh-Definition Rugged DVR - Case Study
High-Definition Rugged DVR - Case StudyMistral Solutions
 
Customized SoC for Industrial Automation Markets
Customized SoC for Industrial Automation MarketsCustomized SoC for Industrial Automation Markets
Customized SoC for Industrial Automation MarketsMistral Solutions
 
The Amazing World of Infotainment
The Amazing World of InfotainmentThe Amazing World of Infotainment
The Amazing World of InfotainmentMistral Solutions
 
Infographic - Sonar Subsystems
Infographic - Sonar Subsystems Infographic - Sonar Subsystems
Infographic - Sonar Subsystems Mistral Solutions
 

More from Mistral Solutions (20)

An infographic from Mistral explaining Semiconductors
An infographic from Mistral explaining SemiconductorsAn infographic from Mistral explaining Semiconductors
An infographic from Mistral explaining Semiconductors
 
SMARC Development Kit
SMARC Development KitSMARC Development Kit
SMARC Development Kit
 
Satellite based SCADA systems
Satellite based SCADA systemsSatellite based SCADA systems
Satellite based SCADA systems
 
AM437x Application board
AM437x Application boardAM437x Application board
AM437x Application board
 
SMARC module
SMARC moduleSMARC module
SMARC module
 
Interfacing a Thermal Printer using an AM437x PoM
Interfacing a Thermal Printer using an AM437x PoMInterfacing a Thermal Printer using an AM437x PoM
Interfacing a Thermal Printer using an AM437x PoM
 
IoT based Industrial Gateway (IoT-SDK) built around Sitara™ AM437x processors...
IoT based Industrial Gateway (IoT-SDK) built around Sitara™ AM437x processors...IoT based Industrial Gateway (IoT-SDK) built around Sitara™ AM437x processors...
IoT based Industrial Gateway (IoT-SDK) built around Sitara™ AM437x processors...
 
Designing a Product on Module for Health and Wellness Applications
Designing a Product on Module for Health and Wellness ApplicationsDesigning a Product on Module for Health and Wellness Applications
Designing a Product on Module for Health and Wellness Applications
 
iOS Application Development for Drones
iOS Application Development for DronesiOS Application Development for Drones
iOS Application Development for Drones
 
Hardware Software Co-Design - White Paper
Hardware Software Co-Design - White PaperHardware Software Co-Design - White Paper
Hardware Software Co-Design - White Paper
 
India Needs an OEM Evolution
India Needs an OEM EvolutionIndia Needs an OEM Evolution
India Needs an OEM Evolution
 
RTM Break Out Card - Datasheet
RTM Break Out Card - DatasheetRTM Break Out Card - Datasheet
RTM Break Out Card - Datasheet
 
Power Spikes Isolation to avoid corruption within sensitive ICs Solutions
Power Spikes Isolation to avoid corruption within sensitive ICs SolutionsPower Spikes Isolation to avoid corruption within sensitive ICs Solutions
Power Spikes Isolation to avoid corruption within sensitive ICs Solutions
 
Algorithm for Lossy Image Compression using FPGA
Algorithm for Lossy Image Compression using FPGAAlgorithm for Lossy Image Compression using FPGA
Algorithm for Lossy Image Compression using FPGA
 
High-Definition Rugged DVR - Case Study
High-Definition Rugged DVR - Case StudyHigh-Definition Rugged DVR - Case Study
High-Definition Rugged DVR - Case Study
 
Customized SoC for Industrial Automation Markets
Customized SoC for Industrial Automation MarketsCustomized SoC for Industrial Automation Markets
Customized SoC for Industrial Automation Markets
 
The Amazing World of Infotainment
The Amazing World of InfotainmentThe Amazing World of Infotainment
The Amazing World of Infotainment
 
Assistive technology
Assistive technologyAssistive technology
Assistive technology
 
Infographic - Sonar Subsystems
Infographic - Sonar Subsystems Infographic - Sonar Subsystems
Infographic - Sonar Subsystems
 
Infographic - Drone
Infographic - DroneInfographic - Drone
Infographic - Drone
 

Recently uploaded

Novel 3D-Printed Soft Linear and Bending Actuators
Novel 3D-Printed Soft Linear and Bending ActuatorsNovel 3D-Printed Soft Linear and Bending Actuators
Novel 3D-Printed Soft Linear and Bending ActuatorsResearcher Researcher
 
Road Development plans of India-Transportation Engineering
Road Development plans of India-Transportation EngineeringRoad Development plans of India-Transportation Engineering
Road Development plans of India-Transportation EngineeringSayli Madhale
 
Pyrolysis process control: temperature control design and application for opt...
Pyrolysis process control: temperature control design and application for opt...Pyrolysis process control: temperature control design and application for opt...
Pyrolysis process control: temperature control design and application for opt...IJECEIAES
 
Secure Key Crypto - Tech Paper JET Tech Labs
Secure Key Crypto - Tech Paper JET Tech LabsSecure Key Crypto - Tech Paper JET Tech Labs
Secure Key Crypto - Tech Paper JET Tech Labsamber724300
 
priority interrupt computer organization
priority interrupt computer organizationpriority interrupt computer organization
priority interrupt computer organizationchnrketan
 
Chapter 9 Mechanical Injection Systems.pdf
Chapter 9 Mechanical Injection Systems.pdfChapter 9 Mechanical Injection Systems.pdf
Chapter 9 Mechanical Injection Systems.pdfFaizanAhmed396943
 
22CYT12 & Chemistry for Computer Systems_Unit-II-Corrosion & its Control Meth...
22CYT12 & Chemistry for Computer Systems_Unit-II-Corrosion & its Control Meth...22CYT12 & Chemistry for Computer Systems_Unit-II-Corrosion & its Control Meth...
22CYT12 & Chemistry for Computer Systems_Unit-II-Corrosion & its Control Meth...KrishnaveniKrishnara1
 
Introduction of Object Oriented Programming Language using Java. .pptx
Introduction of Object Oriented Programming Language using Java. .pptxIntroduction of Object Oriented Programming Language using Java. .pptx
Introduction of Object Oriented Programming Language using Java. .pptxPoonam60376
 
Introduction to Machine Learning Part1.pptx
Introduction to Machine Learning Part1.pptxIntroduction to Machine Learning Part1.pptx
Introduction to Machine Learning Part1.pptxPavan Mohan Neelamraju
 
Cost estimation approach: FP to COCOMO scenario based question
Cost estimation approach: FP to COCOMO scenario based questionCost estimation approach: FP to COCOMO scenario based question
Cost estimation approach: FP to COCOMO scenario based questionSneha Padhiar
 
Network Enhancements on BitVisor for BitVisor Summit 12
Network Enhancements on BitVisor for BitVisor Summit 12Network Enhancements on BitVisor for BitVisor Summit 12
Network Enhancements on BitVisor for BitVisor Summit 12cjchen22
 
SOFTWARE ESTIMATION COCOMO AND FP CALCULATION
SOFTWARE ESTIMATION COCOMO AND FP CALCULATIONSOFTWARE ESTIMATION COCOMO AND FP CALCULATION
SOFTWARE ESTIMATION COCOMO AND FP CALCULATIONSneha Padhiar
 
Design and Analysis of Algorithms Lecture Notes
Design and Analysis of Algorithms Lecture NotesDesign and Analysis of Algorithms Lecture Notes
Design and Analysis of Algorithms Lecture NotesSreedhar Chowdam
 
A brief look at visionOS - How to develop app on Apple's Vision Pro
A brief look at visionOS - How to develop app on Apple's Vision ProA brief look at visionOS - How to develop app on Apple's Vision Pro
A brief look at visionOS - How to develop app on Apple's Vision ProRay Yuan Liu
 
Defining the Clouds for entriprises.pptx
Defining the Clouds for entriprises.pptxDefining the Clouds for entriprises.pptx
Defining the Clouds for entriprises.pptxAshwiniTodkar4
 
Detection&Tracking - Thermal imaging object detection and tracking
Detection&Tracking - Thermal imaging object detection and trackingDetection&Tracking - Thermal imaging object detection and tracking
Detection&Tracking - Thermal imaging object detection and trackinghadarpinhas1
 
ADM100 Running Book for sap basis domain study
ADM100 Running Book for sap basis domain studyADM100 Running Book for sap basis domain study
ADM100 Running Book for sap basis domain studydhruvamdhruvil123
 
Madani.store - Planning - Interview Questions
Madani.store - Planning - Interview QuestionsMadani.store - Planning - Interview Questions
Madani.store - Planning - Interview QuestionsKarim Gaber
 

Recently uploaded (20)

Novel 3D-Printed Soft Linear and Bending Actuators
Novel 3D-Printed Soft Linear and Bending ActuatorsNovel 3D-Printed Soft Linear and Bending Actuators
Novel 3D-Printed Soft Linear and Bending Actuators
 
Road Development plans of India-Transportation Engineering
Road Development plans of India-Transportation EngineeringRoad Development plans of India-Transportation Engineering
Road Development plans of India-Transportation Engineering
 
Pyrolysis process control: temperature control design and application for opt...
Pyrolysis process control: temperature control design and application for opt...Pyrolysis process control: temperature control design and application for opt...
Pyrolysis process control: temperature control design and application for opt...
 
Secure Key Crypto - Tech Paper JET Tech Labs
Secure Key Crypto - Tech Paper JET Tech LabsSecure Key Crypto - Tech Paper JET Tech Labs
Secure Key Crypto - Tech Paper JET Tech Labs
 
priority interrupt computer organization
priority interrupt computer organizationpriority interrupt computer organization
priority interrupt computer organization
 
Chapter 9 Mechanical Injection Systems.pdf
Chapter 9 Mechanical Injection Systems.pdfChapter 9 Mechanical Injection Systems.pdf
Chapter 9 Mechanical Injection Systems.pdf
 
22CYT12 & Chemistry for Computer Systems_Unit-II-Corrosion & its Control Meth...
22CYT12 & Chemistry for Computer Systems_Unit-II-Corrosion & its Control Meth...22CYT12 & Chemistry for Computer Systems_Unit-II-Corrosion & its Control Meth...
22CYT12 & Chemistry for Computer Systems_Unit-II-Corrosion & its Control Meth...
 
Introduction of Object Oriented Programming Language using Java. .pptx
Introduction of Object Oriented Programming Language using Java. .pptxIntroduction of Object Oriented Programming Language using Java. .pptx
Introduction of Object Oriented Programming Language using Java. .pptx
 
Introduction to Machine Learning Part1.pptx
Introduction to Machine Learning Part1.pptxIntroduction to Machine Learning Part1.pptx
Introduction to Machine Learning Part1.pptx
 
Cost estimation approach: FP to COCOMO scenario based question
Cost estimation approach: FP to COCOMO scenario based questionCost estimation approach: FP to COCOMO scenario based question
Cost estimation approach: FP to COCOMO scenario based question
 
Network Enhancements on BitVisor for BitVisor Summit 12
Network Enhancements on BitVisor for BitVisor Summit 12Network Enhancements on BitVisor for BitVisor Summit 12
Network Enhancements on BitVisor for BitVisor Summit 12
 
SOFTWARE ESTIMATION COCOMO AND FP CALCULATION
SOFTWARE ESTIMATION COCOMO AND FP CALCULATIONSOFTWARE ESTIMATION COCOMO AND FP CALCULATION
SOFTWARE ESTIMATION COCOMO AND FP CALCULATION
 
Design and Analysis of Algorithms Lecture Notes
Design and Analysis of Algorithms Lecture NotesDesign and Analysis of Algorithms Lecture Notes
Design and Analysis of Algorithms Lecture Notes
 
A brief look at visionOS - How to develop app on Apple's Vision Pro
A brief look at visionOS - How to develop app on Apple's Vision ProA brief look at visionOS - How to develop app on Apple's Vision Pro
A brief look at visionOS - How to develop app on Apple's Vision Pro
 
Defining the Clouds for entriprises.pptx
Defining the Clouds for entriprises.pptxDefining the Clouds for entriprises.pptx
Defining the Clouds for entriprises.pptx
 
Versatile Engineering Construction Firms
Versatile Engineering Construction FirmsVersatile Engineering Construction Firms
Versatile Engineering Construction Firms
 
ASME-B31.4-2019-estandar para diseño de ductos
ASME-B31.4-2019-estandar para diseño de ductosASME-B31.4-2019-estandar para diseño de ductos
ASME-B31.4-2019-estandar para diseño de ductos
 
Detection&Tracking - Thermal imaging object detection and tracking
Detection&Tracking - Thermal imaging object detection and trackingDetection&Tracking - Thermal imaging object detection and tracking
Detection&Tracking - Thermal imaging object detection and tracking
 
ADM100 Running Book for sap basis domain study
ADM100 Running Book for sap basis domain studyADM100 Running Book for sap basis domain study
ADM100 Running Book for sap basis domain study
 
Madani.store - Planning - Interview Questions
Madani.store - Planning - Interview QuestionsMadani.store - Planning - Interview Questions
Madani.store - Planning - Interview Questions
 

Field Dominance Algorithm

  • 1. IDENTIFYING TOP/BOTTOM FIELD IN INTERLACED VIDEO Raja Subramanian and Sanjeev Retna Mistral Solutions Pvt Ltd., India ABSTRACT This paper elaborates an approach that can be adopted to determine top/bottom fields in an interlaced video. Knowing the top and bottom field is important if the video is de- interlaced using Field Combination, Weaving + Bob, Discard and other algorithms based on motion detection. Determining the field information helps to re-construct the frame with lesser artifacts. This approach can be used if the top/bottom field information is not provided by video decoder chip. INTRODUCTION Interlaced video has been in use for more than 50 years. When dealing with interlaced video, de-interlacing algorithms are essential to remove any interlacing artifacts. There are many de-interlacing algorithms available for NTSC/PAL interlaced video. For low-end systems (system with less processing capability), following approaches can be considered:- Algorithm Description Quality Stationary Moving Weaving / Field Combine Combine the top and bottom field to form a single frame Same as the source Artifacts due to time delay between top and bottom field Discard Discard the top or bottom field, resize (double the height) other field to construct a frame Stair case artifacts due to resize/line copy. Image quality degrades. Stair case artifacts due to resize/line copy. Image quality degrades. BOB Interpolate the top and bottom field (resize) and construct the frame at doubled FPS Image quality is same or better than the discard algorithm. Edges of the objects in the video frame may be incorrect in size and colour. Flickers if the display refresh rate is incorrect or variable. Image quality is same or better than the discard algorithm. Edges of the objects in the video frame may be incorrect in size and colour. Flickers if the display refresh rate is incorrect or variable. BOB + Weaving Use the previous top or bottom field data to construct the frame at doubled FPS Same as the source Close to the source. Lesser artifacts than all the above The captured frame shown below depicts the stair-case artifacts of discard algorithm. If the field order is unknown (or not provided by the video decoder chip) then Discard or Bob algorithm is the efficient method to de-interlace. In most cases, frames re-constructed by Field Combination or BOB + Weaving quality will be close to the original input frame. Most of the NTSC/PAL video decoder provides the field order information. However, in some systems, it may not be possible to get field order due to some operations like serial/parallel conversion. Also the video can be of various resolutions and type (NTSC, PAL, etc.). Therefore, the algorithm needs to use a common mechanism to determine the field order. Figure 2: Bob + Weaving Figure 1: Discard June 2013
  • 2. Field Dominance Algorithm Each digitized input video frame contains two parts which is received as a top and bottom field with respect to time order. If the field order information from video decoder chip is not available and the de-interlacing algorithm receives an top field first, then re-construction of the frame is illustrated below. In this case the video frames are re-constructed correctly. If the algorithm receives the bottom field as a first field then the field order is reversed. Re-construction of the video frames in such a case is illustrated as below: For Example, YUV 4:2:2 little-endian data with top field dominance will be stored as shown Line 1(top1) → U0 Y0 V0 Y1 U2 Y2 V2 Y3... Line 2(bottom1) → U0 Y0 V0 Y1 U2 Y2 V2 Y3... Line 3(top2) → U0 Y0 V0 Y1 U2 Y2 V2 Y3... Line 4(bottom2) → U0 Y0 V0 Y1 U2 Y2 V2 Y3... . . . Line n → U0 Y0 V0 Y1 U2 Y2 V2 Y3.. In this case the quality of the re-constructed frame is same as the input source. If field order is reversed, then progressively weaved data may look as follows Line 1(bottom1) → U0 Y0 V0 Y1 U2 Y2 V2 Y3... Line 2(top2) → U0 Y0 V0 Y1 U2 Y2 V2 Y3... Line 3(bottom2) → U0 Y0 V0 Y1 U2 Y2 V2 Y3... Line 4(top3) → U0 Y0 V0 Y1 U2 Y2 V2 Y3... . . . Line n → U0 Y0 V0 Y1 U2 Y2 V2 Y3.. The re-constructed frame quality is not good due to the swapped lines. To determine the field order, the algorithm must assume the field order is correct and then interpolate the two consecutive lines of the top fields using the 2-points or 6-points median algorithm and compare each pixel's luminance against the bottom line's pixels. This procedure can be repeated again assuming the field is reversed. Comparing these two helps in identifying the field order. Assuming the luminance values of the frame are arranged in the following order Line 1, y(1) → x1,x2,x3... Line 2, y(2) → x1,x2,x3... Line 3, y(3) → x1,x2,x3... Line 4, y(4) → x1,x2,x3... . Line n-1, y(n-1) → x1,x2,x3...x(m-1),x(m),x(m+1),... Line n, y(n) → x1,x2,x3...x(m-1),x(m),x(m+1),... Line n+1, y(n+1) → x1,x2,x3...x(m-1),x(m),x(m+1),... the top field confidence for any point fi( x(m), y(n) ) point can be calculated by, Figure 4: Field-swap artifacts Figure 3: Proper Field Ordering
  • 3. In the above formula, luminance value of column x(m) and row y(n) can compared with the calculated median value(interpolated). We can also calculate the bottom Field confidence by, Sum of all top field confidence gives the total top confidence value for the frame. Similarly the total bottom field dominance values can be calculated by adding all the bottom confidence values . Following formula helps to determine if the field is swapped or not. Once the field information is detected, it is necessary to run this algorithm at regular interval to reaffirm the file order. Accuracy Input Video Data Accuracy (in %) Moving Picture 99 – 100 Still picture (paused video or picture) 95 – 100 Still picture (paused video or frame with same color) >75 Performance Event Timings (in millisecond) Time to determine field order (2 - point mean algorithm running 33 - 99 on 400mhz ARM Cortex-M3 ) Time to detect re-plug and correct the Field order (Algorithm runs every 1000 ms) 1000-3000 CONCLUSION The interpolated video data will approximately be equal to the original data. This method is more reliable to identify the field dominance for most video data and hence improves the picture quality dramatically. 2-point median algorithm can be substituted to reduce the CPU load further. Further, this algorithm is influenced by input video data and the field order may change as a result of the video input cable being removed/re-plugged. So, it's necessary to run this algorithm at periodic interval to correct the field dominance accordingly. About Authors Raja Subramanian is a member of the engineering team at Mistral Solutions Pvt Ltd. He has completed Masters in Computer Science from Bharathidasan University and has 14 years of experience in designing technical solutions in the area of audio, video, storage and user interface for consumer electronic products. His expertise is in the field of autonomous navigation, video surveillance and data security. Sanjeev Retna is a member of the engineering team at Mistral Solutions Pvt Ltd. He has a Bachelor of Engineering degree from Anna University. His expertise is in the field of autonomous navigation, camera and video related products.