Scribd Logo
Upload

Welcome to Scribd!

  • 107 views

    Traffic Sign Recognition Project

    Uploaded by

    Nikhil Pawar
    This synopsis describes a project to recognize traffic signs using convolutional neural networks. The proposed system will train and classify traffic signs from over 35,000 images of 43 classes using CNNs with TensorFlow and Keras. A sequential model with convolutional, max pooling, flatten, and dense layers will be implemented in PyCharm using OpenCV for real-time prediction from a webcam. The goal is to achieve high test accuracy in identifying traffic signs to help enable autonomous vehicle navigation.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    Traffic Sign Recognition Project

    Uploaded by

    Nikhil Pawar
    107 views9 pages
    This synopsis describes a project to recognize traffic signs using convolutional neural networks. The proposed system will train and classify traffic signs from over 35,000 images of 43 classes using CNNs with TensorFlow and Keras. A sequential model with convolutional, max pooling, flatten, and dense layers will be implemented in PyCharm using OpenCV for real-time prediction from a webcam. The goal is to achieve high test accuracy in identifying traffic signs to help enable autonomous vehicle navigation.

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This synopsis describes a project to recognize traffic signs using convolutional neural networks. The proposed system will train and classify traffic signs from over 35,000 images of 43 classes using CNNs with TensorFlow and Keras. A sequential model with convolutional, max pooling, flatten, and dense layers will be implemented in PyCharm using OpenCV for real-time prediction from a webcam. The goal is to achieve high test accuracy in identifying traffic signs to help enable autonomous vehicle navigation.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    107 views9 pages

    Traffic Sign Recognition Project

    Uploaded by

    Nikhil Pawar
    This synopsis describes a project to recognize traffic signs using convolutional neural networks. The proposed system will train and classify traffic signs from over 35,000 images of 43 classes using CNNs with TensorFlow and Keras. A sequential model with convolutional, max pooling, flatten, and dense layers will be implemented in PyCharm using OpenCV for real-time prediction from a webcam. The goal is to achieve high test accuracy in identifying traffic signs to help enable autonomous vehicle navigation.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    You are on page 1of 9

    Sir Parashurambhau College (Autonomous), Pune - 30

    Department of Computer Science


    M.Sc. Computer Science Part I Semester II 2020-21

    Synopsis Format

    1. Title of the project


    Traffic Sign Recognition

    2. Introduction to the problem / Problem Definition


    The self-driving cars in which the passenger can fully depend on the
    car for traveling. But to achieve best goal, its necessary for vehicles
    to understand and follow all traffic rules.
    In the world of Artificial Intelligence and advancement in
    technologies, many researchers and big companies like Tesla,
    Mahindra, Google, Mercedes-Benz, Toyota, Ford, Audi are working
    on autonomous vehicles and self-driving cars. So, for achieving
    accuracy in this technology, the vehicles should be able to interpret
    traffic signs and make decisions accordingly.
    3. Review of existing System/methods/techniques (if any),
    literature review
    of existing systems (along with pros and cons)
    Previous system has traditional workload type. Traffic signs will
    simply classified by uploading image of it. Traffic sign image must be
    stored already in system, we just have to select the image and
    upload it on the screen then Image recognition will be done
    automatically.
    Advantage - Basic application used in older versions of autonomous
    cars. Traffic signs/rules can recognized and helpful for understanding
    purpose. one of best machine learning application.
    Disadvantage - The system is not fully automated. Its not
    implemented as it designed.

    4. Proposed System / Scope of work


    Train and classify Traffic Signs using Convolutional neural networks
    This will be done using OPENCV in real time using a simple webcam .
    CNNs have been gaining popularity in the past couple of years due to
    their ability to generalize and classify the data with high accuracy. In
    this project we will train traffic signs with over 35000 images of 43
    different classes with the help of tensorflow and keras.
    We will try to achieve maximum Accuracy and Test Score for Traffic
    Signs Recognition and Classification by CNN. It will use OPENCV in
    real time by simple webcam. Based on this cars can accurately
    identify Traffic Signs and work accordingly.
    5. Objective of the project (a clear picture to choose this particular
    topic for
    project), Expected Outcomes.
    AUTONOMOUS CARS is one of the trending topics in today’s. Traffic
    Signs Recognition is one of best feature from it. There are several
    different types of traffic signs like speed limits, no entry, traffic
    signals, turn left or right, children crossing, no passing of heavy
    vehicles, etc. Companies like Tesla, Mahindra, Google, Mercedes-
    Benz, Toyota, Ford, Audi are working on autonomous vehicles and
    self-driving cars with this model. This will able to read and
    understand traffic signs Automatically which are a very important
    task for all autonomous vehicles.
    6. Methodology and techniques that will be used (the summary of
    the
    project must also be incorporated)
    In this project we are using Convolutional Neural Network by
    OPENCV in real time using a simple webcam. Simple methodology
    like Training, Testing and Validation of the dataset is used.
    Image Preprocessing – Grayscale and Equalizing Images with colour
    mapping.
    Image Augmentation – To make Images more generic.
    Plotting Barcharts for sample distribution of Images. Plotting Graphs
    of Loss and Accuracy for Training and Validation of the dataset.
    Prediction of the Images and Classes by model.predict method.
    7. Details about the Hardware & Software’s that may be used
    Operating System – Windows 10
    Software - PyCharm Community Edition 2021.2
    8. Describe the software / algorithm to be implemented (I/O
    screens, any
    statistical / mathematical models that will be implemented)
    We are implementing Sequential model for Convolutional Neural
    Network with layers like Conv2D, MaxPooling2D, Flatten, Dense,
    Dropout.
    Screenshots :-

    1. Class wise Distribution of the Images


    2. Sample Images with Class Labels.
    3. Accuracy of Training and Validation Dataset

    4. Loss of Training and Validation Dataset


    5. Augmented Image Samples
    6. Test Score with Test Accuracy

    9. Describe the testing strategies and methods which will be used.


    After Trained model Run the Test model file. Put the Image of any
    random Traffic Sign in front of camera. Model will recognize the
    Image and classify the traffic sign class automatically. After successful
    loading of frames, system will display Image description and Class
    label on the screen.
    10. Reference or Bibliography
    Dataset -
    https://www.kaggle.com/meowmeowmeowmeowmeow/gtsrb-
    german-traffic-sign
    Python Language -
    https://www.w3schools.com/python/python_ref_functions.asp
    https://stackoverflow.com/questions/19326711/pickle-dump-error-
    for-object-defined-by-class
    OpenCV -
    https://docs.opencv.org/master/d6/d00/tutorial_py_root.html
    Tensorflow Sequential Model -
    https://www.tensorflow.org/guide/keras/sequential_model
    Pycharm Software -
    https://www.jetbrains.com/pycharm/download/#section=windows

    You might also like