SlideShare a Scribd company logo
1 of 36
Download to read offline
S W I S S G E R M A N U N I V E R S I T Y
DESIGNING A SMART CAR DATA LOGGING AND TRACKING
WITH VISION
A Thesis Presentation
By
Willy Haliim
In Partial Fulfillment of the Requirements for
BACHELOR’S DEGREE
IN
MECHATRONICS ENGINEERING
SWISS GERMAN UNIVERSITY
EduTown BSDCity
Tangerang – 15339
INDONESIA
S W I S S G E R M A N U N I V E R S I T Y
15/07/14 2
PRESENTATION OVERVIEW
• Background Problem
• Thesis Purpose
• Methodology
• Results
• Conclusion
• Discussion / Further Recommendation
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
S W I S S G E R M A N U N I V E R S I T Y
15/07/14 3
JOINT PROJECT OF THREE STUDENTS
1. Building a Data Logger as Part of Intelligent Vehicle
System Using 32-bit ARM Cortex-M3 with Cubieboard and
Android Phone Interfacing. (Richard‘s Part)
2. Designing a Smart Car Data Logging and Tracking
With Vision. (Author‘s Part)
3. Designing an Inter-Cars Safety Communication System
Based on Android Smartphone. (Rendi‘s Part)
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
S W I S S G E R M A N U N I V E R S I T Y
4
BACKGROUND
15/07/14
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
Accident Prevention Accident Handling
S W I S S G E R M A N U N I V E R S I T Y
OVERALL SYSTEM IDEA
15/07/14
Smart Mobility with Connected Cooperative Cars System
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
S W I S S G E R M A N U N I V E R S I T Y
6
THESIS PURPOSE
15/07/14
Author’s Scope:
1. Prevent accident by developing an early warning range detection
system using stereovision system
2. Communicate with other subsystem to create a whole working smart
mobility conected cooperative cars.
3. Use Dual Core ARM Cortex-A7 Mini PC and Linux
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
S W I S S G E R M A N U N I V E R S I T Y
7
PROBLEM DEFINITION
1. How to detect obstacle‘s range in front of a stereo camera using
stereovision system.
2. How to record the car‘s journey using one the stereo camera.
3. Which Communication Method will be used to synchronize the
Cubieboard 2 with the sensor and the server subsystems.
4. How to explore the exposed I/O pin on Cubieboard 2 to minimize
hardware requirements.
5. The processing speed is limited by the specification of
Cubieboard 2. The Focus is on Functionality.
15/07/14
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
S W I S S G E R M A N U N I V E R S I T Y
15/07/14
OVERALL SYSTEM OVERVIEW
(Author’s Scope)
8
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
S W I S S G E R M A N U N I V E R S I T Y
SYSTEM OVERVIEW
STEREOVISION
RANGE CALCULATION
PROCESS
ESTIMATED
OBJECT’S RANGE
INPUT PROCESSING OUTPUT
15/07/14 9
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
Minoru 3D Camera Cubieboard 2
IMAGES
UI PUSHBUTTON
S W I S S G E R M A N U N I V E R S I T Y
15/07/14 10
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
CUBIEBOARD 2. DUAL CORE 1 GHz. 1 GB RAM
S W I S S G E R M A N U N I V E R S I T Y
19/02/10 [Your Thesis Title] – [Your Name] – [Your Faculty]
INSIDE THE ARM CORTEX A7 PROCESSOR
S W I S S G E R M A N U N I V E R S I T Y
ARM PROCESSORS
15/07/14 12
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
OS Application Based.
Programmable Using C++
Low Power Consumption
Low Cost
S W I S S G E R M A N U N I V E R S I T Y
THESIS METHODOLOGY
15/07/14 13
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
S W I S S G E R M A N U N I V E R S I T Y
INSTALLING WORKSPACE ENVIRONMENT
15/07/14
CUBIAN OS REMOTE TO SMARTPHONE
14
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
S W I S S G E R M A N U N I V E R S I T Y
STEREOVISION METHOD 1 - DEPTH IMAGE
15/07/14
DISPARITYRECTIFICATION
CAMERA
CALIBRATION
15
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
STEREO MATCHING
S W I S S G E R M A N U N I V E R S I T Y
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
|l-r|
α
α
β
β
16
STEREOVISION METHOD 2 – RANGE CALCULATION
b
S W I S S G E R M A N U N I V E R S I T Y
17
PROGRAM FLOWCHART - RANGE CALCULATION
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
OBJECT DETECTING EARLY WARNING SYSTEM CAPTURE AND SEND
S W I S S G E R M A N U N I V E R S I T Y
OBJECT DETECTING STEPS – METHOD 2
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
3. Color Filter
Erode Dilate
4. Find & Draw
Contours
5. Calculate
the COG
1. RGB Image
(Original)
2. Convert to
HSV
18
Detects yellow object
Centroid object detected
inRange
S W I S S G E R M A N U N I V E R S I T Y
A B C
EARLY WARNING RANGE DETECTION SYSTEM
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
RANGE LED INDICATOR DISPLAY
100 – 150 CM GREEN SAFE
50 – 99 CM YELLOW -
BELOW 50 CM RED AND BUZZER WARNING!
19
R1 R2 R3
S W I S S G E R M A N U N I V E R S I T Y
SERIAL
DATA COMMUNICATION
RECEIVED AND SENT
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
FROM
FLYMAPLE BOARD
(SENSOR)
TO
ANDROID PHONE
(SERVER)
FTP
20
Location data
Orientation Data
Accident Log
(Trigger)
Location data
Accident Log
(Trigger)
Image Captured
S W I S S G E R M A N U N I V E R S I T Y
21
RESULTS – OVERVIEW DESIGN
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
Stereo Camera
Cubieboard 2
Remote Desktop
Flymaple
(Sensor)
Wifi dongle
Early warning
system
S W I S S G E R M A N U N I V E R S I T Y
22
RESULTS – DEPTH IMAGE
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
CAMERA ON FIXED POSITION.
OBJECT MOVED EVERY 10 CM.
S W I S S G E R M A N U N I V E R S I T Y
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
Distance
(cm)
Average
Grayscale
Value
10 -
20 152
30 111
40 86
50 65
60 52
70 44
80 37
90 31
100 26
110 22
120 19
130 16
140 13
150 11
23
RESULTS – DEPTH IMAGE
S W I S S G E R M A N U N I V E R S I T Y
24
RESULTS – DEPTH IMAGE
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
CUBIAN (320 X 480) WINDOWS 7 (640X480)
S W I S S G E R M A N U N I V E R S I T Y
RESULTS – RANGE CALCULATION
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14 25
Program UI and
Display Frame
S W I S S G E R M A N U N I V E R S I T Y
RESULTS – RANGE CALCULATION
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14 26
Object on fixed position. Camera moved every 10 cm.
S W I S S G E R M A N U N I V E R S I T Y
15/07/14
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology 27
Real
Range(mm)
Avg Disparity
(Pix)
Average(mm) Error(mm) Error(%)
100 56 625 -525 84
200 79 443 -243 55
300 86 404 -104 26
400 75 464 -64 14
500 63 547 -47 9
600 55 631 -31 5
700 49 705 -5 1
800 45 774 26 3
900 41 841 59 7
1000 39 897 103 11
1100 36 957 143 15
1200 35 1001 199 20
1300 34 1031 269 26
1400 32 1095 305 28
1500 30 1195 305 30
1600 29 1201 399 33
= 3.3%
S W I S S G E R M A N U N I V E R S I T Y
15/07/14
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology 28
= 3.3%
S W I S S G E R M A N U N I V E R S I T Y
15/07/14
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology 29
= 3.3%
The closer you look the bigger the disparity. The less you see
S W I S S G E R M A N U N I V E R S I T Y
A B C
EARLY WARNING RANGE DETECTION SYSTEM
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
FAR MEDIUM CLOSE
30
S W I S S G E R M A N U N I V E R S I T Y
DATA COMMUNICATION
RECEIVED AND SENT
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
FROM FLYMAPLE BOARD DATA ON ANDROID SERVER
31
song.mp3
S W I S S G E R M A N U N I V E R S I T Y
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
IMPROVEMENTS
32
Without multithreaded
low FPS
With multithreaded
higher FPS
S W I S S G E R M A N U N I V E R S I T Y
33
CONCLUSION
1. Obstacle in front of camera successfully detected with effective
range 40 – 100 cm and error around 1 – 10 % .
2. Wired and Wireless Communication between Flymaple board
sensor and Android server are successfully attempted.
3. I/O pins explored and it is useful for input output sensors, like
capturing button and early warning range detection system.
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
S W I S S G E R M A N U N I V E R S I T Y
34
DISCUSSION / FURTHER RECOMMENDATION
1. Use ARM Cortex-A7 processor with better
specification to provide faster calculation and gives
better frame rate result.
2. Stereo Camera with higher resolution, bigger baseline
and focal length to acquire further distance detection.
3. Better algorithm on stereovision programming.
4. Improve stereovision range detection by using another
stereovision calculation method.
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
S W I S S G E R M A N U N I V E R S I T Y
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
PRESENTATION RESULTS DEMO
35
EARLY RANGE DETECTION WARNING SYSTEM
S W I S S G E R M A N U N I V E R S I T Y
Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim –
Engineering and Information Technology
15/07/14
PRESENTATION RESULTS DEMO
36
FTP COMMUNICATION

More Related Content

Viewers also liked

Presentation on Shodhganga - National Repository of Indian ETDs
Presentation on Shodhganga - National Repository of Indian ETDsPresentation on Shodhganga - National Repository of Indian ETDs
Presentation on Shodhganga - National Repository of Indian ETDsManoj Kumar K
 
Chapter 5 : Conclusion & Suggestion
Chapter 5 : Conclusion & SuggestionChapter 5 : Conclusion & Suggestion
Chapter 5 : Conclusion & SuggestionMei Poe
 
Essay Structure (Introduction and Outline)
Essay Structure (Introduction and Outline)Essay Structure (Introduction and Outline)
Essay Structure (Introduction and Outline)theLecturette
 
Hospital Management System
Hospital Management SystemHospital Management System
Hospital Management Systemidowume
 
Baskin Project Report (Brand Analysis or Promotional Scheme)
Baskin Project Report (Brand Analysis or Promotional Scheme)Baskin Project Report (Brand Analysis or Promotional Scheme)
Baskin Project Report (Brand Analysis or Promotional Scheme)Abhishek Keshri
 
Baskin Robbins Global Marketing
Baskin Robbins Global MarketingBaskin Robbins Global Marketing
Baskin Robbins Global Marketingmayankl jain
 
Thesis my documentation
Thesis  my documentationThesis  my documentation
Thesis my documentationcas123
 
Project Proposal Sample: RFID on Warehouse Management System
Project Proposal Sample: RFID on Warehouse Management SystemProject Proposal Sample: RFID on Warehouse Management System
Project Proposal Sample: RFID on Warehouse Management SystemCheri Amour Calicdan
 
Online Shopping Full Project Presentation (20 slides)
Online Shopping Full Project Presentation (20 slides)Online Shopping Full Project Presentation (20 slides)
Online Shopping Full Project Presentation (20 slides)Krishna Mohan Shakya
 
Online shopping ppt by rohit jain
Online shopping ppt by rohit jainOnline shopping ppt by rohit jain
Online shopping ppt by rohit jainRohit Jain
 
Online shopping report-6 month project
Online shopping report-6 month projectOnline shopping report-6 month project
Online shopping report-6 month projectGinne yoffe
 

Viewers also liked (15)

Presentation on Shodhganga - National Repository of Indian ETDs
Presentation on Shodhganga - National Repository of Indian ETDsPresentation on Shodhganga - National Repository of Indian ETDs
Presentation on Shodhganga - National Repository of Indian ETDs
 
Article Links
Article LinksArticle Links
Article Links
 
honda report
honda reporthonda report
honda report
 
Chapter 5 : Conclusion & Suggestion
Chapter 5 : Conclusion & SuggestionChapter 5 : Conclusion & Suggestion
Chapter 5 : Conclusion & Suggestion
 
Essay Structure (Introduction and Outline)
Essay Structure (Introduction and Outline)Essay Structure (Introduction and Outline)
Essay Structure (Introduction and Outline)
 
Hospital Management System
Hospital Management SystemHospital Management System
Hospital Management System
 
Baskin Project Report (Brand Analysis or Promotional Scheme)
Baskin Project Report (Brand Analysis or Promotional Scheme)Baskin Project Report (Brand Analysis or Promotional Scheme)
Baskin Project Report (Brand Analysis or Promotional Scheme)
 
Baskin Robbins Global Marketing
Baskin Robbins Global MarketingBaskin Robbins Global Marketing
Baskin Robbins Global Marketing
 
Onlineshopping
OnlineshoppingOnlineshopping
Onlineshopping
 
Thesis my documentation
Thesis  my documentationThesis  my documentation
Thesis my documentation
 
Project Proposal Sample: RFID on Warehouse Management System
Project Proposal Sample: RFID on Warehouse Management SystemProject Proposal Sample: RFID on Warehouse Management System
Project Proposal Sample: RFID on Warehouse Management System
 
Online shopping
Online shoppingOnline shopping
Online shopping
 
Online Shopping Full Project Presentation (20 slides)
Online Shopping Full Project Presentation (20 slides)Online Shopping Full Project Presentation (20 slides)
Online Shopping Full Project Presentation (20 slides)
 
Online shopping ppt by rohit jain
Online shopping ppt by rohit jainOnline shopping ppt by rohit jain
Online shopping ppt by rohit jain
 
Online shopping report-6 month project
Online shopping report-6 month projectOnline shopping report-6 month project
Online shopping report-6 month project
 

Similar to Thesis Presentation Willy Haliim

Speed Detecting Camera by Kandarp Tiwari
Speed Detecting Camera by Kandarp TiwariSpeed Detecting Camera by Kandarp Tiwari
Speed Detecting Camera by Kandarp TiwariKandarp Tiwari
 
SMART CAR PARKING SYSTEM USING IR SENSOR
SMART CAR PARKING SYSTEM USING IR SENSORSMART CAR PARKING SYSTEM USING IR SENSOR
SMART CAR PARKING SYSTEM USING IR SENSORIRJET Journal
 
IRJET- Raspberry Pi based Security System with Automatic Alert System in Road...
IRJET- Raspberry Pi based Security System with Automatic Alert System in Road...IRJET- Raspberry Pi based Security System with Automatic Alert System in Road...
IRJET- Raspberry Pi based Security System with Automatic Alert System in Road...IRJET Journal
 
Automotive Cybersecurity Challenges for Automated Vehicles: Jonathan Petit
Automotive Cybersecurity Challenges for Automated Vehicles: Jonathan PetitAutomotive Cybersecurity Challenges for Automated Vehicles: Jonathan Petit
Automotive Cybersecurity Challenges for Automated Vehicles: Jonathan PetitSecurity Innovation
 
Advanced car security system
Advanced car security systemAdvanced car security system
Advanced car security systemAmi Goswami
 
CAR BLACK BOX SYSTEM
CAR BLACK BOX SYSTEMCAR BLACK BOX SYSTEM
CAR BLACK BOX SYSTEMIRJET Journal
 
SMART SOLUTION FOR RESOLVING HEAVY TRAFFIC USING IOT
SMART SOLUTION FOR RESOLVING HEAVY TRAFFIC USING IOTSMART SOLUTION FOR RESOLVING HEAVY TRAFFIC USING IOT
SMART SOLUTION FOR RESOLVING HEAVY TRAFFIC USING IOTIRJET Journal
 
TII_Thierry_LESTABLE_WCNC_2022_v10_Short.pdf
TII_Thierry_LESTABLE_WCNC_2022_v10_Short.pdfTII_Thierry_LESTABLE_WCNC_2022_v10_Short.pdf
TII_Thierry_LESTABLE_WCNC_2022_v10_Short.pdfThierry Lestable
 
IRJET - Ingenious Car Parking System using IoT in Commercial Areas
IRJET - Ingenious Car Parking System using IoT in Commercial AreasIRJET - Ingenious Car Parking System using IoT in Commercial Areas
IRJET - Ingenious Car Parking System using IoT in Commercial AreasIRJET Journal
 
Apresentação smart parking microio
Apresentação smart parking microioApresentação smart parking microio
Apresentação smart parking microioMicroiolda
 
IRJET- Number Plate Extraction from Vehicle Front View Image using Image ...
IRJET-  	  Number Plate Extraction from Vehicle Front View Image using Image ...IRJET-  	  Number Plate Extraction from Vehicle Front View Image using Image ...
IRJET- Number Plate Extraction from Vehicle Front View Image using Image ...IRJET Journal
 
IRJET - A Design Thinking based Smart Parking System for Vehicle Parking ...
IRJET -  	  A Design Thinking based Smart Parking System for Vehicle Parking ...IRJET -  	  A Design Thinking based Smart Parking System for Vehicle Parking ...
IRJET - A Design Thinking based Smart Parking System for Vehicle Parking ...IRJET Journal
 
SC4 Workshop 1: Seán Gaines: Vehicle sensors
SC4 Workshop 1: Seán Gaines: Vehicle sensorsSC4 Workshop 1: Seán Gaines: Vehicle sensors
SC4 Workshop 1: Seán Gaines: Vehicle sensorsBigData_Europe
 
Smart vehicle tracking using GPS
Smart vehicle tracking using GPSSmart vehicle tracking using GPS
Smart vehicle tracking using GPSIRJET Journal
 
A Survey on Vehicle Tracking System using IoT
A Survey on Vehicle Tracking System using IoTA Survey on Vehicle Tracking System using IoT
A Survey on Vehicle Tracking System using IoTIRJET Journal
 
Vendeka corporate presentation
Vendeka corporate presentationVendeka corporate presentation
Vendeka corporate presentationMustafa Kuğu
 
한국 건설 기술 전망과 건설 테크 스타트업 소개
한국 건설 기술 전망과 건설 테크 스타트업 소개한국 건설 기술 전망과 건설 테크 스타트업 소개
한국 건설 기술 전망과 건설 테크 스타트업 소개Tae wook kang
 
IRJET- The Smart in-Car Technology
IRJET-  	  The Smart in-Car TechnologyIRJET-  	  The Smart in-Car Technology
IRJET- The Smart in-Car TechnologyIRJET Journal
 
Overcoming the AIoT Obstacles through Smart Component Integration
Overcoming the AIoT Obstacles through Smart Component IntegrationOvercoming the AIoT Obstacles through Smart Component Integration
Overcoming the AIoT Obstacles through Smart Component IntegrationInnodisk Corporation
 

Similar to Thesis Presentation Willy Haliim (20)

Speed Detecting Camera by Kandarp Tiwari
Speed Detecting Camera by Kandarp TiwariSpeed Detecting Camera by Kandarp Tiwari
Speed Detecting Camera by Kandarp Tiwari
 
Telraam
TelraamTelraam
Telraam
 
SMART CAR PARKING SYSTEM USING IR SENSOR
SMART CAR PARKING SYSTEM USING IR SENSORSMART CAR PARKING SYSTEM USING IR SENSOR
SMART CAR PARKING SYSTEM USING IR SENSOR
 
IRJET- Raspberry Pi based Security System with Automatic Alert System in Road...
IRJET- Raspberry Pi based Security System with Automatic Alert System in Road...IRJET- Raspberry Pi based Security System with Automatic Alert System in Road...
IRJET- Raspberry Pi based Security System with Automatic Alert System in Road...
 
Automotive Cybersecurity Challenges for Automated Vehicles: Jonathan Petit
Automotive Cybersecurity Challenges for Automated Vehicles: Jonathan PetitAutomotive Cybersecurity Challenges for Automated Vehicles: Jonathan Petit
Automotive Cybersecurity Challenges for Automated Vehicles: Jonathan Petit
 
Advanced car security system
Advanced car security systemAdvanced car security system
Advanced car security system
 
CAR BLACK BOX SYSTEM
CAR BLACK BOX SYSTEMCAR BLACK BOX SYSTEM
CAR BLACK BOX SYSTEM
 
SMART SOLUTION FOR RESOLVING HEAVY TRAFFIC USING IOT
SMART SOLUTION FOR RESOLVING HEAVY TRAFFIC USING IOTSMART SOLUTION FOR RESOLVING HEAVY TRAFFIC USING IOT
SMART SOLUTION FOR RESOLVING HEAVY TRAFFIC USING IOT
 
TII_Thierry_LESTABLE_WCNC_2022_v10_Short.pdf
TII_Thierry_LESTABLE_WCNC_2022_v10_Short.pdfTII_Thierry_LESTABLE_WCNC_2022_v10_Short.pdf
TII_Thierry_LESTABLE_WCNC_2022_v10_Short.pdf
 
IRJET - Ingenious Car Parking System using IoT in Commercial Areas
IRJET - Ingenious Car Parking System using IoT in Commercial AreasIRJET - Ingenious Car Parking System using IoT in Commercial Areas
IRJET - Ingenious Car Parking System using IoT in Commercial Areas
 
Apresentação smart parking microio
Apresentação smart parking microioApresentação smart parking microio
Apresentação smart parking microio
 
IRJET- Number Plate Extraction from Vehicle Front View Image using Image ...
IRJET-  	  Number Plate Extraction from Vehicle Front View Image using Image ...IRJET-  	  Number Plate Extraction from Vehicle Front View Image using Image ...
IRJET- Number Plate Extraction from Vehicle Front View Image using Image ...
 
IRJET - A Design Thinking based Smart Parking System for Vehicle Parking ...
IRJET -  	  A Design Thinking based Smart Parking System for Vehicle Parking ...IRJET -  	  A Design Thinking based Smart Parking System for Vehicle Parking ...
IRJET - A Design Thinking based Smart Parking System for Vehicle Parking ...
 
SC4 Workshop 1: Seán Gaines: Vehicle sensors
SC4 Workshop 1: Seán Gaines: Vehicle sensorsSC4 Workshop 1: Seán Gaines: Vehicle sensors
SC4 Workshop 1: Seán Gaines: Vehicle sensors
 
Smart vehicle tracking using GPS
Smart vehicle tracking using GPSSmart vehicle tracking using GPS
Smart vehicle tracking using GPS
 
A Survey on Vehicle Tracking System using IoT
A Survey on Vehicle Tracking System using IoTA Survey on Vehicle Tracking System using IoT
A Survey on Vehicle Tracking System using IoT
 
Vendeka corporate presentation
Vendeka corporate presentationVendeka corporate presentation
Vendeka corporate presentation
 
한국 건설 기술 전망과 건설 테크 스타트업 소개
한국 건설 기술 전망과 건설 테크 스타트업 소개한국 건설 기술 전망과 건설 테크 스타트업 소개
한국 건설 기술 전망과 건설 테크 스타트업 소개
 
IRJET- The Smart in-Car Technology
IRJET-  	  The Smart in-Car TechnologyIRJET-  	  The Smart in-Car Technology
IRJET- The Smart in-Car Technology
 
Overcoming the AIoT Obstacles through Smart Component Integration
Overcoming the AIoT Obstacles through Smart Component IntegrationOvercoming the AIoT Obstacles through Smart Component Integration
Overcoming the AIoT Obstacles through Smart Component Integration
 

Thesis Presentation Willy Haliim

  • 1. S W I S S G E R M A N U N I V E R S I T Y DESIGNING A SMART CAR DATA LOGGING AND TRACKING WITH VISION A Thesis Presentation By Willy Haliim In Partial Fulfillment of the Requirements for BACHELOR’S DEGREE IN MECHATRONICS ENGINEERING SWISS GERMAN UNIVERSITY EduTown BSDCity Tangerang – 15339 INDONESIA
  • 2. S W I S S G E R M A N U N I V E R S I T Y 15/07/14 2 PRESENTATION OVERVIEW • Background Problem • Thesis Purpose • Methodology • Results • Conclusion • Discussion / Further Recommendation Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology
  • 3. S W I S S G E R M A N U N I V E R S I T Y 15/07/14 3 JOINT PROJECT OF THREE STUDENTS 1. Building a Data Logger as Part of Intelligent Vehicle System Using 32-bit ARM Cortex-M3 with Cubieboard and Android Phone Interfacing. (Richard‘s Part) 2. Designing a Smart Car Data Logging and Tracking With Vision. (Author‘s Part) 3. Designing an Inter-Cars Safety Communication System Based on Android Smartphone. (Rendi‘s Part) Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology
  • 4. S W I S S G E R M A N U N I V E R S I T Y 4 BACKGROUND 15/07/14 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology Accident Prevention Accident Handling
  • 5. S W I S S G E R M A N U N I V E R S I T Y OVERALL SYSTEM IDEA 15/07/14 Smart Mobility with Connected Cooperative Cars System Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology
  • 6. S W I S S G E R M A N U N I V E R S I T Y 6 THESIS PURPOSE 15/07/14 Author’s Scope: 1. Prevent accident by developing an early warning range detection system using stereovision system 2. Communicate with other subsystem to create a whole working smart mobility conected cooperative cars. 3. Use Dual Core ARM Cortex-A7 Mini PC and Linux Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology
  • 7. S W I S S G E R M A N U N I V E R S I T Y 7 PROBLEM DEFINITION 1. How to detect obstacle‘s range in front of a stereo camera using stereovision system. 2. How to record the car‘s journey using one the stereo camera. 3. Which Communication Method will be used to synchronize the Cubieboard 2 with the sensor and the server subsystems. 4. How to explore the exposed I/O pin on Cubieboard 2 to minimize hardware requirements. 5. The processing speed is limited by the specification of Cubieboard 2. The Focus is on Functionality. 15/07/14 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology
  • 8. S W I S S G E R M A N U N I V E R S I T Y 15/07/14 OVERALL SYSTEM OVERVIEW (Author’s Scope) 8 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology
  • 9. S W I S S G E R M A N U N I V E R S I T Y SYSTEM OVERVIEW STEREOVISION RANGE CALCULATION PROCESS ESTIMATED OBJECT’S RANGE INPUT PROCESSING OUTPUT 15/07/14 9 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology Minoru 3D Camera Cubieboard 2 IMAGES UI PUSHBUTTON
  • 10. S W I S S G E R M A N U N I V E R S I T Y 15/07/14 10 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology CUBIEBOARD 2. DUAL CORE 1 GHz. 1 GB RAM
  • 11. S W I S S G E R M A N U N I V E R S I T Y 19/02/10 [Your Thesis Title] – [Your Name] – [Your Faculty] INSIDE THE ARM CORTEX A7 PROCESSOR
  • 12. S W I S S G E R M A N U N I V E R S I T Y ARM PROCESSORS 15/07/14 12 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology OS Application Based. Programmable Using C++ Low Power Consumption Low Cost
  • 13. S W I S S G E R M A N U N I V E R S I T Y THESIS METHODOLOGY 15/07/14 13 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology
  • 14. S W I S S G E R M A N U N I V E R S I T Y INSTALLING WORKSPACE ENVIRONMENT 15/07/14 CUBIAN OS REMOTE TO SMARTPHONE 14 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology
  • 15. S W I S S G E R M A N U N I V E R S I T Y STEREOVISION METHOD 1 - DEPTH IMAGE 15/07/14 DISPARITYRECTIFICATION CAMERA CALIBRATION 15 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology STEREO MATCHING
  • 16. S W I S S G E R M A N U N I V E R S I T Y Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 |l-r| α α β β 16 STEREOVISION METHOD 2 – RANGE CALCULATION b
  • 17. S W I S S G E R M A N U N I V E R S I T Y 17 PROGRAM FLOWCHART - RANGE CALCULATION Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 OBJECT DETECTING EARLY WARNING SYSTEM CAPTURE AND SEND
  • 18. S W I S S G E R M A N U N I V E R S I T Y OBJECT DETECTING STEPS – METHOD 2 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 3. Color Filter Erode Dilate 4. Find & Draw Contours 5. Calculate the COG 1. RGB Image (Original) 2. Convert to HSV 18 Detects yellow object Centroid object detected inRange
  • 19. S W I S S G E R M A N U N I V E R S I T Y A B C EARLY WARNING RANGE DETECTION SYSTEM Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 RANGE LED INDICATOR DISPLAY 100 – 150 CM GREEN SAFE 50 – 99 CM YELLOW - BELOW 50 CM RED AND BUZZER WARNING! 19 R1 R2 R3
  • 20. S W I S S G E R M A N U N I V E R S I T Y SERIAL DATA COMMUNICATION RECEIVED AND SENT Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 FROM FLYMAPLE BOARD (SENSOR) TO ANDROID PHONE (SERVER) FTP 20 Location data Orientation Data Accident Log (Trigger) Location data Accident Log (Trigger) Image Captured
  • 21. S W I S S G E R M A N U N I V E R S I T Y 21 RESULTS – OVERVIEW DESIGN Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 Stereo Camera Cubieboard 2 Remote Desktop Flymaple (Sensor) Wifi dongle Early warning system
  • 22. S W I S S G E R M A N U N I V E R S I T Y 22 RESULTS – DEPTH IMAGE Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 CAMERA ON FIXED POSITION. OBJECT MOVED EVERY 10 CM.
  • 23. S W I S S G E R M A N U N I V E R S I T Y Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 Distance (cm) Average Grayscale Value 10 - 20 152 30 111 40 86 50 65 60 52 70 44 80 37 90 31 100 26 110 22 120 19 130 16 140 13 150 11 23 RESULTS – DEPTH IMAGE
  • 24. S W I S S G E R M A N U N I V E R S I T Y 24 RESULTS – DEPTH IMAGE Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 CUBIAN (320 X 480) WINDOWS 7 (640X480)
  • 25. S W I S S G E R M A N U N I V E R S I T Y RESULTS – RANGE CALCULATION Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 25 Program UI and Display Frame
  • 26. S W I S S G E R M A N U N I V E R S I T Y RESULTS – RANGE CALCULATION Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 26 Object on fixed position. Camera moved every 10 cm.
  • 27. S W I S S G E R M A N U N I V E R S I T Y 15/07/14 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 27 Real Range(mm) Avg Disparity (Pix) Average(mm) Error(mm) Error(%) 100 56 625 -525 84 200 79 443 -243 55 300 86 404 -104 26 400 75 464 -64 14 500 63 547 -47 9 600 55 631 -31 5 700 49 705 -5 1 800 45 774 26 3 900 41 841 59 7 1000 39 897 103 11 1100 36 957 143 15 1200 35 1001 199 20 1300 34 1031 269 26 1400 32 1095 305 28 1500 30 1195 305 30 1600 29 1201 399 33 = 3.3%
  • 28. S W I S S G E R M A N U N I V E R S I T Y 15/07/14 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 28 = 3.3%
  • 29. S W I S S G E R M A N U N I V E R S I T Y 15/07/14 Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 29 = 3.3% The closer you look the bigger the disparity. The less you see
  • 30. S W I S S G E R M A N U N I V E R S I T Y A B C EARLY WARNING RANGE DETECTION SYSTEM Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 FAR MEDIUM CLOSE 30
  • 31. S W I S S G E R M A N U N I V E R S I T Y DATA COMMUNICATION RECEIVED AND SENT Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 FROM FLYMAPLE BOARD DATA ON ANDROID SERVER 31 song.mp3
  • 32. S W I S S G E R M A N U N I V E R S I T Y Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 IMPROVEMENTS 32 Without multithreaded low FPS With multithreaded higher FPS
  • 33. S W I S S G E R M A N U N I V E R S I T Y 33 CONCLUSION 1. Obstacle in front of camera successfully detected with effective range 40 – 100 cm and error around 1 – 10 % . 2. Wired and Wireless Communication between Flymaple board sensor and Android server are successfully attempted. 3. I/O pins explored and it is useful for input output sensors, like capturing button and early warning range detection system. Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14
  • 34. S W I S S G E R M A N U N I V E R S I T Y 34 DISCUSSION / FURTHER RECOMMENDATION 1. Use ARM Cortex-A7 processor with better specification to provide faster calculation and gives better frame rate result. 2. Stereo Camera with higher resolution, bigger baseline and focal length to acquire further distance detection. 3. Better algorithm on stereovision programming. 4. Improve stereovision range detection by using another stereovision calculation method. Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14
  • 35. S W I S S G E R M A N U N I V E R S I T Y Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 PRESENTATION RESULTS DEMO 35 EARLY RANGE DETECTION WARNING SYSTEM
  • 36. S W I S S G E R M A N U N I V E R S I T Y Designing a Smart Car Data Logging and Tracking with Vision – Willy Haliim – Engineering and Information Technology 15/07/14 PRESENTATION RESULTS DEMO 36 FTP COMMUNICATION