SlideShare a Scribd company logo
1 of 13
Distracted Drivers
Weekday Summer 1
Andrew, Angel, David, Karthik, Sanaa and Minsung
1
The Bigger Picture
2
● 1.6 million crashes happen every year because of
someone using a phone while driving.Around 390,000
injuries happen yearly from accidents caused by
distracted driving.
● Computer vision is used to improve road safety by
detecting distracted drivers, it is being applied in several
areas for security and traffic monitoring.It improves safety
and accuracy and plummets the risk of accidents.
Different distractions
while driving
3
● Can you think of any other ways in
which we can use AI to prevent
distracted driving?
● Could you implement these solutions to
prevent distracted driving in different
parts of the world?
4
Distracted Driving
and Our Mission
5
First let’s briefly talk about AI in general. Artificial
intelligence known as AI is a wide range of a
computer science with machine buildings capable
of performing the tasks that are typically require
human intelligence. For our topic, we focused on
how to detect whether the driver is driving safely
or not. In United States, there are so many cases
of drowsy driving and unrealized accidents, such
caused due to smartphones and etc. This was the
time where the AI came in. By training the
computers and using the Ai technology, we can
detect further number of distracted drivers and
therefore guarantee more safety along with less
percentages car accidents. Furthermore,
development of AI not just protects people from
many accidents, but it also improved every
human’s quality of life.
6
What We Tried:
The Good, the
Bad, and the Meh
7
Basic Models:
● Our task was to build a model
capable of differentiating between
different driver behaviors based on
images (computer vision)
● Nearest Neighbor Model (KNN
Model)
○ Testing Accuracy: about 50%
○ Overpredicted that driver was using
radio
● Logistic Regression Model
○ Accuracy: about 35%
○ Overpredicted that driver was using
mirrors
8
KNN
Predictions
Logistic Regression
Predictions
Neural Networks and CNNs
● Next, we used Neural Networks to see if they
would be any better
● Our basic neural network (top) trained 10
times on the training set
● Unfortunately, it only reached a 40% testing
accuracy
● The Convolutional Neural Network (bottom)
didn’t do much better
9
Transfer Learning Neural Networks
● The next attempt was to take pre-
established machine learning algorithms
such as VGG16, VGG19, ResNet50, and
DenseNet121
● These algorithms have already been
extensively trained, and are very adept at
computer vision tasks.
● When applied to our task of identifying driver
behaviors, these networks performed much
better than our initial attempts.
● In particular, VGG 19 was highly effective
10
VGG 19, 15 training epochs
DenseNet 121, 10 training Epochs
Transfer Learning with Some Customization
● The final thing we tried was combining
specific layers of the VGG neural network
with some new layers of our own creation
● While we were unable to improve the model
any further, we still maintained about an
80% validation accuracy, which is pretty
good.
11
VGG 19 with customized dense layers,
15 epochs of training
DenseNet121 with customized layers,
50 epochs of training
Conclusion
We did a lot of trial and error to make our
data as accurate as possible. We want to
focus on the bigger picture. There are
many different ways to prevent
distracted driving using AI.
12
Acknowledgements
Special thanks to our instructor, Yoichi, for teaching us this week!
We would also like to thank the entire Inspirit AI team for helping us through this
project. Thank you all for teaching us, for all the wonderful talks and it was a great
learning experience.
13

More Related Content

What's hot

Autonomous car Working (Deep learning)
Autonomous car Working (Deep learning)Autonomous car Working (Deep learning)
Autonomous car Working (Deep learning)Bharath Chava
 
Artificial intelligence
Artificial intelligenceArtificial intelligence
Artificial intelligenceAbdullahMajid9
 
Machine Learning Overview
Machine Learning OverviewMachine Learning Overview
Machine Learning OverviewMykhailo Koval
 
Self driving cars -
Self driving cars - Self driving cars -
Self driving cars - Hany G. Amer
 
Artificial Intelligence In The Automotive Industry - M&A Trend Analysis
Artificial Intelligence In The Automotive Industry - M&A Trend AnalysisArtificial Intelligence In The Automotive Industry - M&A Trend Analysis
Artificial Intelligence In The Automotive Industry - M&A Trend AnalysisNetscribes
 
Presentation on driverless cars by shahin hussan
Presentation on driverless cars by shahin hussan Presentation on driverless cars by shahin hussan
Presentation on driverless cars by shahin hussan Shahinhussan
 
Artificial Intelligence In Automotive Industry: Surprisingly Slow Uptake And ...
Artificial Intelligence In Automotive Industry: Surprisingly Slow Uptake And ...Artificial Intelligence In Automotive Industry: Surprisingly Slow Uptake And ...
Artificial Intelligence In Automotive Industry: Surprisingly Slow Uptake And ...Bernard Marr
 
Introduction to Artificial Intelligence (Std.X)
Introduction to Artificial Intelligence (Std.X)Introduction to Artificial Intelligence (Std.X)
Introduction to Artificial Intelligence (Std.X)Jitendra Kumar Yadav
 
Artificial inteligence
Artificial inteligenceArtificial inteligence
Artificial inteligenceSharath Raj
 
Artificial Intelligence PPT .
Artificial Intelligence PPT .Artificial Intelligence PPT .
Artificial Intelligence PPT .VAIBHAVNAGPURE6
 
Artificial Intelligence (A.I.) || Introduction of A.I. || HELPFUL FOR STUDENT...
Artificial Intelligence (A.I.) || Introduction of A.I. || HELPFUL FOR STUDENT...Artificial Intelligence (A.I.) || Introduction of A.I. || HELPFUL FOR STUDENT...
Artificial Intelligence (A.I.) || Introduction of A.I. || HELPFUL FOR STUDENT...Shivangi Singh
 
ARTIFICIAL INTELLIGENCE
ARTIFICIAL INTELLIGENCEARTIFICIAL INTELLIGENCE
ARTIFICIAL INTELLIGENCEMidhuti
 

What's hot (20)

Ai ppt
Ai pptAi ppt
Ai ppt
 
Autonomous car Working (Deep learning)
Autonomous car Working (Deep learning)Autonomous car Working (Deep learning)
Autonomous car Working (Deep learning)
 
Artificial Intelligence
Artificial IntelligenceArtificial Intelligence
Artificial Intelligence
 
Artificial intelligence
Artificial intelligenceArtificial intelligence
Artificial intelligence
 
Autonomous car
Autonomous carAutonomous car
Autonomous car
 
Machine Learning Overview
Machine Learning OverviewMachine Learning Overview
Machine Learning Overview
 
Self driving car
Self driving carSelf driving car
Self driving car
 
Self driving cars -
Self driving cars - Self driving cars -
Self driving cars -
 
Artificial Intelligence In The Automotive Industry - M&A Trend Analysis
Artificial Intelligence In The Automotive Industry - M&A Trend AnalysisArtificial Intelligence In The Automotive Industry - M&A Trend Analysis
Artificial Intelligence In The Automotive Industry - M&A Trend Analysis
 
Presentation on driverless cars by shahin hussan
Presentation on driverless cars by shahin hussan Presentation on driverless cars by shahin hussan
Presentation on driverless cars by shahin hussan
 
Artificial Intelligence In Automotive Industry: Surprisingly Slow Uptake And ...
Artificial Intelligence In Automotive Industry: Surprisingly Slow Uptake And ...Artificial Intelligence In Automotive Industry: Surprisingly Slow Uptake And ...
Artificial Intelligence In Automotive Industry: Surprisingly Slow Uptake And ...
 
Introduction to Artificial Intelligence (Std.X)
Introduction to Artificial Intelligence (Std.X)Introduction to Artificial Intelligence (Std.X)
Introduction to Artificial Intelligence (Std.X)
 
Driverless car
Driverless carDriverless car
Driverless car
 
Artificial inteligence
Artificial inteligenceArtificial inteligence
Artificial inteligence
 
Artificial Intelligence PPT .
Artificial Intelligence PPT .Artificial Intelligence PPT .
Artificial Intelligence PPT .
 
Artificial Intelligence (A.I.) || Introduction of A.I. || HELPFUL FOR STUDENT...
Artificial Intelligence (A.I.) || Introduction of A.I. || HELPFUL FOR STUDENT...Artificial Intelligence (A.I.) || Introduction of A.I. || HELPFUL FOR STUDENT...
Artificial Intelligence (A.I.) || Introduction of A.I. || HELPFUL FOR STUDENT...
 
Artificial inteligence
Artificial inteligenceArtificial inteligence
Artificial inteligence
 
ARTIFICIAL INTELLIGENCE
ARTIFICIAL INTELLIGENCEARTIFICIAL INTELLIGENCE
ARTIFICIAL INTELLIGENCE
 
Autonomous cars
Autonomous carsAutonomous cars
Autonomous cars
 
Introduction of ai
Introduction of aiIntroduction of ai
Introduction of ai
 

Similar to Distracted Drivers- Inspirit AI

Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Suraj Aavula
 
Sign Language Recognition using Machine Learning
Sign Language Recognition using Machine LearningSign Language Recognition using Machine Learning
Sign Language Recognition using Machine LearningIRJET Journal
 
英語プレゼン.pdf
英語プレゼン.pdf英語プレゼン.pdf
英語プレゼン.pdfssuserab84471
 
Presentation.pdf
Presentation.pdfPresentation.pdf
Presentation.pdfRfujii
 
deep-reinforcement-learning-framework.pdf
deep-reinforcement-learning-framework.pdfdeep-reinforcement-learning-framework.pdf
deep-reinforcement-learning-framework.pdfYugank Aman
 
Introduction to Computer Vision.pdf
Introduction to Computer Vision.pdfIntroduction to Computer Vision.pdf
Introduction to Computer Vision.pdfKnoldus Inc.
 
Self-Navigation Car using Reinforcement Learning
Self-Navigation Car using Reinforcement LearningSelf-Navigation Car using Reinforcement Learning
Self-Navigation Car using Reinforcement LearningDr. Amarjeet Singh
 
Application Oriented Computer Vision Pipeline for Automotive Industry
Application Oriented Computer Vision Pipeline for Automotive IndustryApplication Oriented Computer Vision Pipeline for Automotive Industry
Application Oriented Computer Vision Pipeline for Automotive IndustryOytun Eren Sengul
 
Machine learning for 5G
Machine learning for 5GMachine learning for 5G
Machine learning for 5GTrinh Phuc
 
Deep Learning With Neural Networks
Deep Learning With Neural NetworksDeep Learning With Neural Networks
Deep Learning With Neural NetworksAniket Maurya
 
IRJET - Steering Wheel Angle Prediction for Self-Driving Cars
IRJET - Steering Wheel Angle Prediction for Self-Driving CarsIRJET - Steering Wheel Angle Prediction for Self-Driving Cars
IRJET - Steering Wheel Angle Prediction for Self-Driving CarsIRJET Journal
 
Autonomous driving system using proximal policy optimization in deep reinforc...
Autonomous driving system using proximal policy optimization in deep reinforc...Autonomous driving system using proximal policy optimization in deep reinforc...
Autonomous driving system using proximal policy optimization in deep reinforc...IAESIJAI
 
AI Vs ML Vs DL PowerPoint Presentation Slide Templates Complete Deck
AI Vs ML Vs DL PowerPoint Presentation Slide Templates Complete DeckAI Vs ML Vs DL PowerPoint Presentation Slide Templates Complete Deck
AI Vs ML Vs DL PowerPoint Presentation Slide Templates Complete DeckSlideTeam
 
Reinforcement Learning for Self Driving Cars
Reinforcement Learning for Self Driving CarsReinforcement Learning for Self Driving Cars
Reinforcement Learning for Self Driving CarsSneha Ravikumar
 
Real-Time Driver Drowsiness Detection System
Real-Time Driver Drowsiness Detection SystemReal-Time Driver Drowsiness Detection System
Real-Time Driver Drowsiness Detection SystemIRJET Journal
 
IoT and Computer Vision Based Driver Safety Monitoring System with Risk Predi...
IoT and Computer Vision Based Driver Safety Monitoring System with Risk Predi...IoT and Computer Vision Based Driver Safety Monitoring System with Risk Predi...
IoT and Computer Vision Based Driver Safety Monitoring System with Risk Predi...Sama -E-Shan
 
Ad Click Prediction - Paper review
Ad Click Prediction - Paper reviewAd Click Prediction - Paper review
Ad Click Prediction - Paper reviewMazen Aly
 

Similar to Distracted Drivers- Inspirit AI (20)

Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Convolution Neural Network (CNN)
Convolution Neural Network (CNN)
 
Ibr project report
Ibr project reportIbr project report
Ibr project report
 
Sign Language Recognition using Machine Learning
Sign Language Recognition using Machine LearningSign Language Recognition using Machine Learning
Sign Language Recognition using Machine Learning
 
英語プレゼン.pdf
英語プレゼン.pdf英語プレゼン.pdf
英語プレゼン.pdf
 
Presentation.pdf
Presentation.pdfPresentation.pdf
Presentation.pdf
 
Deep presentation.pptx
Deep presentation.pptxDeep presentation.pptx
Deep presentation.pptx
 
deep-reinforcement-learning-framework.pdf
deep-reinforcement-learning-framework.pdfdeep-reinforcement-learning-framework.pdf
deep-reinforcement-learning-framework.pdf
 
Introduction to Computer Vision.pdf
Introduction to Computer Vision.pdfIntroduction to Computer Vision.pdf
Introduction to Computer Vision.pdf
 
Self-Navigation Car using Reinforcement Learning
Self-Navigation Car using Reinforcement LearningSelf-Navigation Car using Reinforcement Learning
Self-Navigation Car using Reinforcement Learning
 
Application Oriented Computer Vision Pipeline for Automotive Industry
Application Oriented Computer Vision Pipeline for Automotive IndustryApplication Oriented Computer Vision Pipeline for Automotive Industry
Application Oriented Computer Vision Pipeline for Automotive Industry
 
Machine learning for 5G
Machine learning for 5GMachine learning for 5G
Machine learning for 5G
 
Deep Learning With Neural Networks
Deep Learning With Neural NetworksDeep Learning With Neural Networks
Deep Learning With Neural Networks
 
IRJET - Steering Wheel Angle Prediction for Self-Driving Cars
IRJET - Steering Wheel Angle Prediction for Self-Driving CarsIRJET - Steering Wheel Angle Prediction for Self-Driving Cars
IRJET - Steering Wheel Angle Prediction for Self-Driving Cars
 
Autonomous driving system using proximal policy optimization in deep reinforc...
Autonomous driving system using proximal policy optimization in deep reinforc...Autonomous driving system using proximal policy optimization in deep reinforc...
Autonomous driving system using proximal policy optimization in deep reinforc...
 
AI Vs ML Vs DL PowerPoint Presentation Slide Templates Complete Deck
AI Vs ML Vs DL PowerPoint Presentation Slide Templates Complete DeckAI Vs ML Vs DL PowerPoint Presentation Slide Templates Complete Deck
AI Vs ML Vs DL PowerPoint Presentation Slide Templates Complete Deck
 
Reinforcement Learning for Self Driving Cars
Reinforcement Learning for Self Driving CarsReinforcement Learning for Self Driving Cars
Reinforcement Learning for Self Driving Cars
 
Real-Time Driver Drowsiness Detection System
Real-Time Driver Drowsiness Detection SystemReal-Time Driver Drowsiness Detection System
Real-Time Driver Drowsiness Detection System
 
Vehicle Tracking System (VTS)
Vehicle Tracking System (VTS)Vehicle Tracking System (VTS)
Vehicle Tracking System (VTS)
 
IoT and Computer Vision Based Driver Safety Monitoring System with Risk Predi...
IoT and Computer Vision Based Driver Safety Monitoring System with Risk Predi...IoT and Computer Vision Based Driver Safety Monitoring System with Risk Predi...
IoT and Computer Vision Based Driver Safety Monitoring System with Risk Predi...
 
Ad Click Prediction - Paper review
Ad Click Prediction - Paper reviewAd Click Prediction - Paper review
Ad Click Prediction - Paper review
 

Recently uploaded

SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphNeo4j
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Wonjun Hwang
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024Scott Keck-Warren
 
Artificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraArtificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraDeakin University
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Mattias Andersson
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersThousandEyes
 
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsSnow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsHyundai Motor Group
 
Unlocking the Potential of the Cloud for IBM Power Systems
Unlocking the Potential of the Cloud for IBM Power SystemsUnlocking the Potential of the Cloud for IBM Power Systems
Unlocking the Potential of the Cloud for IBM Power SystemsPrecisely
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Alan Dix
 
costume and set research powerpoint presentation
costume and set research powerpoint presentationcostume and set research powerpoint presentation
costume and set research powerpoint presentationphoebematthew05
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptxLBM Solutions
 
APIForce Zurich 5 April Automation LPDG
APIForce Zurich 5 April  Automation LPDGAPIForce Zurich 5 April  Automation LPDG
APIForce Zurich 5 April Automation LPDGMarianaLemus7
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Enterprise Knowledge
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationRidwan Fadjar
 

Recently uploaded (20)

SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
 
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
Bun (KitWorks Team Study 노별마루 발표 2024.4.22)
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
 
Artificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraArtificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning era
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
 
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsSnow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
 
Unlocking the Potential of the Cloud for IBM Power Systems
Unlocking the Potential of the Cloud for IBM Power SystemsUnlocking the Potential of the Cloud for IBM Power Systems
Unlocking the Potential of the Cloud for IBM Power Systems
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
 
costume and set research powerpoint presentation
costume and set research powerpoint presentationcostume and set research powerpoint presentation
costume and set research powerpoint presentation
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food Manufacturing
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptx
 
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptxVulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
 
APIForce Zurich 5 April Automation LPDG
APIForce Zurich 5 April  Automation LPDGAPIForce Zurich 5 April  Automation LPDG
APIForce Zurich 5 April Automation LPDG
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024
 
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 PresentationMy Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
 

Distracted Drivers- Inspirit AI

  • 1. Distracted Drivers Weekday Summer 1 Andrew, Angel, David, Karthik, Sanaa and Minsung 1
  • 3. ● 1.6 million crashes happen every year because of someone using a phone while driving.Around 390,000 injuries happen yearly from accidents caused by distracted driving. ● Computer vision is used to improve road safety by detecting distracted drivers, it is being applied in several areas for security and traffic monitoring.It improves safety and accuracy and plummets the risk of accidents. Different distractions while driving 3
  • 4. ● Can you think of any other ways in which we can use AI to prevent distracted driving? ● Could you implement these solutions to prevent distracted driving in different parts of the world? 4
  • 6. First let’s briefly talk about AI in general. Artificial intelligence known as AI is a wide range of a computer science with machine buildings capable of performing the tasks that are typically require human intelligence. For our topic, we focused on how to detect whether the driver is driving safely or not. In United States, there are so many cases of drowsy driving and unrealized accidents, such caused due to smartphones and etc. This was the time where the AI came in. By training the computers and using the Ai technology, we can detect further number of distracted drivers and therefore guarantee more safety along with less percentages car accidents. Furthermore, development of AI not just protects people from many accidents, but it also improved every human’s quality of life. 6
  • 7. What We Tried: The Good, the Bad, and the Meh 7
  • 8. Basic Models: ● Our task was to build a model capable of differentiating between different driver behaviors based on images (computer vision) ● Nearest Neighbor Model (KNN Model) ○ Testing Accuracy: about 50% ○ Overpredicted that driver was using radio ● Logistic Regression Model ○ Accuracy: about 35% ○ Overpredicted that driver was using mirrors 8 KNN Predictions Logistic Regression Predictions
  • 9. Neural Networks and CNNs ● Next, we used Neural Networks to see if they would be any better ● Our basic neural network (top) trained 10 times on the training set ● Unfortunately, it only reached a 40% testing accuracy ● The Convolutional Neural Network (bottom) didn’t do much better 9
  • 10. Transfer Learning Neural Networks ● The next attempt was to take pre- established machine learning algorithms such as VGG16, VGG19, ResNet50, and DenseNet121 ● These algorithms have already been extensively trained, and are very adept at computer vision tasks. ● When applied to our task of identifying driver behaviors, these networks performed much better than our initial attempts. ● In particular, VGG 19 was highly effective 10 VGG 19, 15 training epochs DenseNet 121, 10 training Epochs
  • 11. Transfer Learning with Some Customization ● The final thing we tried was combining specific layers of the VGG neural network with some new layers of our own creation ● While we were unable to improve the model any further, we still maintained about an 80% validation accuracy, which is pretty good. 11 VGG 19 with customized dense layers, 15 epochs of training DenseNet121 with customized layers, 50 epochs of training
  • 12. Conclusion We did a lot of trial and error to make our data as accurate as possible. We want to focus on the bigger picture. There are many different ways to prevent distracted driving using AI. 12
  • 13. Acknowledgements Special thanks to our instructor, Yoichi, for teaching us this week! We would also like to thank the entire Inspirit AI team for helping us through this project. Thank you all for teaching us, for all the wonderful talks and it was a great learning experience. 13

Editor's Notes

  1. Hello, this is our project and its called distracted drivers. It is by Andrew, Me, David, Karthik, Sanaa and Minsung. Now I have one question, what is Distracted driving? Well Distracted driving is any activity that diverts attention from driving, including talking or texting on your phone, eating or drinking, using the radio —anything that takes your attention away from the task of safe driving.
  2. Go through this section quickly, and just briefly explain each model and how well they did