Top Banner

of 25

IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

Apr 14, 2018

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    1/25

    IMAGE COMPRESSIONAND DECOMPRESSION

    USING ANN

    By

    Mr.Mahantesh Paramashetti Anusha.GParveen.A.G

    Pallavi.S.Yadav

    Christeena.S

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    2/25

    CONTENTS

    Introduction

    Biologically Inspired Neuron

    Artificial Neural Networks

    Back Propagation AlgorithmCompression Techniques

    Implementations

    Advantages

    Disadvantages

    Applications

    Conclusion

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    3/25

    INTRODUCTION

    Uncompressed multimedia data requiresconsiderable storage capacity and transmissionbandwidth.

    Apart from the existing technology like JPEGand MPEG standards, new technology such asneural networks are used for imagecompression.

    Natural images are captured using imagesensors and stored in memory banks. Largestorage space is required.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    4/25

    eg: A color image of size 256x256 requires

    a storage space of 1.5 Mega bits.

    Storage cost for 1 GB is approximately Rs.200.

    With the available bandwidth of 64kbpsand 54mbps transmitting a three hour movierequires in uncompressed format takes 2917

    years and 19 days respectively.

    Transmission of huge image data is timeconsuming.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    5/25

    Artificial neural networks has beenchosen for image compression due totheir massively parallel and distributedarchitecture.

    The idea behind this Trainingcommands is the Back propagationalgorithm.

    The focus of this project is to implementthe Neural Architecture Digitally.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    6/25

    Biological Neurons

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    7/25

    The Analogy to the Brain

    Neurons are basic signaling units of the

    nervous system of a living being in which

    each neuron is a discrete cell whose severalprocesses are from its cell body.

    The basic element of human brain has

    abilities to remember, think and apply

    previous experiences to our every action.

    Neural networks process information in a

    similar way the human brain does.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    8/25

    Biologically Inspired Neuron

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    9/25

    Artificial Neural Networks

    Artificial Neural Networks are used toprocess the information the way

    biological systems process analog signals

    like image and sound.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    10/25

    Types of ANN

    Feed forward networks

    Information only flows one way

    One input pattern produces one output

    No sense of time (or memory of previous

    state)

    Recurrency

    Nodes connect back to other nodes orthemselves

    Information flow is multidirectional

    Sense of time and memory of previous state(s)

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    11/25

    Artificial Neuron System

    Input layer

    Hidden layer Output layer

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    12/25

    Block Diagram of Neural Architecture

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    13/25

    Back propagation algorithm

    Information about errors is filtered back

    through the system and it is used to adjust the

    connections between the layers, thusimproving performance.

    The Feed-Forward Neural Network

    architecture is capable of approximating most

    problems with high accuracy and

    generalization ability.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    14/25

    The Back propagation algorithm is used to

    update weights and bias of the neural

    networks.

    Weight and bias elements of the neurondecides the functionality of the network.

    Value of these weight and bias elements arecalculated during training phase.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    15/25

    Image compression refers to the task of

    reducing the amount of data required to store

    or transmit an image.

    The compressed image is then subjected to

    further digital processing such as error control

    coding, encryption or multiplexing with other

    data sources, before being used to modulatethe analog signal that is actually transmitted

    through the channel or stored in a storage

    medium.

    COMPRESSION

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    16/25

    COMPRESSION TECHNIQUES

    JPEG

    Wavelet

    GIF

    M-JPEG

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    17/25

    Original imagesOriginal images

    Image scaling(256x256)

    Adding bias & weights a

    Vector values of the scaled Images (16x4096)

    Combining these images to increase the resolution (16x32768)

    Normalizing the combined image

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    18/25

    normalizing

    a

    Training the network

    testing each image

    Comparing scaled &decompressed image

    by finding their PSNR & MSE

    values

    Each image is converted

    to vector form

    Passing the image

    through the network

    denormalizing

    Each image is converted

    to vector form

    Passing the image

    through the network

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    19/25

    IMPLEMENTATION

    MATLAB version R2007b.

    The Maximum error, MSE andPSNR values are calculated.

    Hardware implementation is

    done using FPGA board (Spatan

    3 ).

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    20/25

    Neural network training & performance

    plots: The neural network is trainedusing the nntraintool, available

    in MATLAB.

    The plot of MSE wrt epochs fordifferent iterations are as shown:

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    21/25

    Advantages

    A neural network can perform tasks that a

    linear program cannot.

    When an element of the neural network fails, itcan continue without any problem by their

    parallel nature.

    A neural network learns and does not need to

    be reprogrammed.

    It works even in the presence of noise with

    good quality output.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    22/25

    Disadvantages

    The neural network needs training to

    operate.

    The architecture of a neural network is

    different from the architecture ofmicroprocessors therefore needs to be

    emulated.

    Requires high processing time for largeneural networks.

    As the number of neurons increases the

    network becomes complex.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    23/25

    Applications

    Pattern Matching

    Pattern RecognitionOptimizationVector Quantization

    Data Clustering

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    24/25

    CONCLUSIONChipscope Pro Analyzer can easily

    implement the design on FPGA kit.The analysis showed that comparision

    between input and output values was

    proved to be similar.

    Using Chipscope Pro Analyzer smaller

    architectures can be easily built.

  • 7/27/2019 IMAGE COMPRESSION AND DECOMPRESSION USING ANN.pptx

    25/25

    THANK YOU