Scribd Logo
Upload

Welcome to Scribd!

  • 68 views

    Handwritten Digit Recogntion Using Mnist Dataset

    Uploaded by

    185G1 Pragna Koneru
    This document presents a project on handwritten digit recognition using the MNIST dataset. The proposed system uses convolutional neural networks (CNNs) for digit detection with a graphical user interface (GUI) output. CNNs are able to learn feature extraction, using convolution filters to generate invariant features at each layer. The project architecture includes preprocessing, segmentation, feature extraction using CNN layers, and classification/recognition with classifiers like KNN, random forest, and SVM. The system aims to achieve high accuracy for real-world applications like mail sorting and check processing.

    Copyright:

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

    Handwritten Digit Recogntion Using Mnist Dataset

    Uploaded by

    185G1 Pragna Koneru
    68 views20 pages
    This document presents a project on handwritten digit recognition using the MNIST dataset. The proposed system uses convolutional neural networks (CNNs) for digit detection with a graphical user interface (GUI) output. CNNs are able to learn feature extraction, using convolution filters to generate invariant features at each layer. The project architecture includes preprocessing, segmentation, feature extraction using CNN layers, and classification/recognition with classifiers like KNN, random forest, and SVM. The system aims to achieve high accuracy for real-world applications like mail sorting and check processing.

    Original Title

    MINI PROJECT RE-1 (1)

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document presents a project on handwritten digit recognition using the MNIST dataset. The proposed system uses convolutional neural networks (CNNs) for digit detection with a graphical user interface (GUI) output. CNNs are able to learn feature extraction, using convolution filters to generate invariant features at each layer. The project architecture includes preprocessing, segmentation, feature extraction using CNN layers, and classification/recognition with classifiers like KNN, random forest, and SVM. The system aims to achieve high accuracy for real-world applications like mail sorting and check processing.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    68 views20 pages

    Handwritten Digit Recogntion Using Mnist Dataset

    Uploaded by

    185G1 Pragna Koneru
    This document presents a project on handwritten digit recognition using the MNIST dataset. The proposed system uses convolutional neural networks (CNNs) for digit detection with a graphical user interface (GUI) output. CNNs are able to learn feature extraction, using convolution filters to generate invariant features at each layer. The project architecture includes preprocessing, segmentation, feature extraction using CNN layers, and classification/recognition with classifiers like KNN, random forest, and SVM. The system aims to achieve high accuracy for real-world applications like mail sorting and check processing.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 20

    HANDWRITTEN

    DIGIT RECOGNTION
    USING MNIST
    DATASET
    PRESENTED BY:
    LEKKALA HARIKA CHOWDARY (187R1A05F3)
    KONERU PRAGNA (187R1A05G1)
    NALLA BHAVANI (187R1A05G7)
    • ABSTRACT

    • EXISTING SYSTEM

    • PROPOSED SYSTEM

    CONTENTS • HARDWARE REQUIREMENTS

    • SOFTWARE REQUIREMENTS

    • PROJECT ARCHITECTURE

    • CONCLUSION
    ABSTRACT
    ▪ Handwritten character recognition is one of the practically important issues
    in pattern recognition applications. The applications of digit recognition
    includes postal mail sorting, bank check processing, form data entry, etc.
    The heart of the problem lies within the ability to develop an efficient
    algorithm that can recognize handwritten digits submitted by users by the
    way of a scanner, tablet, and other digital devices. This project presents an
    approach to off-line handwritten digit recognition based on different
    machine learning techniques. The main objective of this project is to
    develop effective and reliable approaches for recognition of handwritten
    digits.
    EXISTING SYSTEM
    As an active research topic, there have been several novel methods for
    handwritten detection which achieved extremely high measure of accuracy. There
    have been researches recently in this work line to improve it, Abhishikth
    Ayyadevara, PNV Sai Ram Teja,“Handwritten Recognition Using Unique
    Feature Extraction Technique”, 2020 IEEE International Conference on Recent
    Trends in Electronics, Information & Communication Technology. In this paper,
    they performed a multilayer perceptron neural network using a backpropagation
    algorithm and they achieved  86.5385%. Rohan Sethi, Ila Kaushik,“HandWritten
    Digit Recognition using Machine Learning” 2020 IEEE 9th International
    Conference on Communication Systems and Network Technologies (CSNT). In
    this paper, they used KNN algorithm to recognize handwritten digits accurately
    Feature Extraction method using PCA is used.
    DISADVANTAGES

    ➢The accuracy of these models is less when compared to


    Convolutional Neural Networks.
    ➢The datasets are now available in abundance to get better working
    model.
    ➢  There are no proper GUI, which can bypass the internal working and

    can give the final outputs directly to users.


    MNIST DATASET
    PROPOSED SYSTEM
    The proposed system is set to work with Convolution Neural Networks
    for detecting handwritten digits with final output as hands on GUI. The
    CNN itself learns the feature extraction. CNN uses convolution of
    image and filters to generate invariant features which are passed onto
    the next layer. The features in the next layer are convoluted with
    different filters to generate more invariant and abstract features and the
    process continues till one gets the final feature.
    PROPOSED SYSTEM

    In the feature extraction layers, each layer of the network receives the
    output from its immediate previous layer as its input, and passes the
    current output as input to the next layer. The CNN architecture is
    composed with the combination of three types of layers:
    1. Convolution,
    2. Max-pooling
    3. Classification.
    ADVANTAGES

    ➢ The accuracy of the proposed model would be preferably good as it is


    constructed on Convolutional Neural Networks (CNN).

    ➢ The final GUI can be handy for many users to use the model who
    doesn’t have any idea of internal working.
    HARDWARE REQUIREMENTS

    Following are the hardware requirements used for developing the application

    ●     System                                                 : INTEL CORE i5 Processor


    ●     Hard Disk                                           :20 GB
    ●     Input Devices                                      :Keyboard, Mouse
    ●     RAM                                                   : 8 GB
    SOFTWARE REQUIREMENTS
    ❑Following are the system requirements used for developing the application.

    ❑●     Operating System                               : WINDOWS


    ❑●     Programming Language                     : PYTHON
    ❑●     Tools                                                   : Anaconda, Jupyter Notebook,
    ❑ Google Colab.
    PROJECT ARCHITECTURE
    PROJECT ARCHITECTURE

    The Figure illustrates the architecture diagram of the proposed system. The
    proposed model contains the four stages in order to classify and detect the
    digits:
    A. Pre-processing
    B. Segmentation
    C. Feature Extraction
    D. Classification and Recognition
    A. PRE-PROCESSING

    The role of the pre-processing step is it performs various tasks on the


    input image. It basically upgrades the image by making it reasonable for
    segmentation. The fundamental motivation behind pre-processing is to
    take off a fascinating example from the background. For the most part,
    noise filtering, smoothing and standardization are to be done in this
    stage.
    B. SEGMENTATION

    Once the pre-processing of the input images is completed, sub-images of


    individual digits are formed from the sequence of images. Pre-processed digit
    images are segmented into a sub-image of individual digits, which are
    assigned a number to each digit. Each individual digit is resized into pixels.
    In this step an edge detection technique is being used for segmentation of
    dataset images
    C. FEATURE EXTRACTION

    ▪ After the completion of pre-processing stage and pre-processed images are


    represented in the form of a matrix which contains pixels of the images that are
    of very large size. In this way it will be valuable to represent the digits in the
    images which contain the necessary information. This activity is called feature
    extraction. In the feature extraction stage redundancy from the data is removed.
    D. CLASSIFICATION AND RECOGNITION

    ▪ In the classification and recognition step the extracted feature vectors are
    taken as an individual input to each of the following classifiers. In order to
    showcase the working system model extracted features are combined and
    defined using following three classifiers:
    K-Nearest Neighbor
     Random Forest Classifier
     Support Vector Machine
    CONCLUSION
    ➢ Our work can potentially facilitate research in related domains. The model
    working with CNN can achieve great results. In fact can be used in real life
    situations like online handwriting recognition on computer tablets, recognize
    zip codes on mail for postal mail sorting, processing bank check amounts,
    numeric entries in forms filled up by hand.
    ➢ Human attempts can be diminished in perceiving, learning, recognizing and
    in a lot more regions. Utilising these deep learning techniques.
    ANY QUERIES?
    THANK YOU

    You might also like