U18Ini5600 - Engineering Cilincs - V Project Report
U18Ini5600 - Engineering Cilincs - V Project Report
COIMBATORE-641049
(An Autonomous Institution Affiliated to Anna University)
Project Report
Submitted by
December 2020
Faculty Coordinator
1
ACKNOWLEDGEMENT
First, we would like to express our praise and gratitude to the Lord, who
has showered his grace and blessing enabling us to complete this project in an
excellent manner. He has made all things in beautiful in his time. We express
our sincere thanks to our beloved Joint Correspondent, Shri. Shankar
Vanavarayar for his kind support and for providing necessary facilities to carry
out the project work.
We would like to express our sincere thanks to our beloved Principal
Dr.J.Srinivasan, Ph.D who encouraged us with his valuable thoughts. We
would like to express our sincere thanks and deep sense of gratitude to
Dr.S.A.Pasupathy, Ph.D (UK)., Professor and Head, Department of Electronics
and Communication Engineering for his valuable suggestions and
encouragement which paved way for the successful completion of the project.
In particular, we wish to thank and express our everlasting gratitude to
Dr.B.Gopinath, Ph.D, Associate Professor, Department of Electronics and
Communication Engineering for the expert counselling in each and every steps
of project work and we wish to convey our deep sense of gratitude to all
teaching and non-teaching staff members of Electronics and Communication
Engineering Department for their help and cooperation.
2
ABSTRACT
3
TABLE OF CONTENTS
ABSTRACT 3
1 INTRODUCTION 5
2 LITERATURE SURVEY 6
3 PROPOSED BLOCK DIAGRAM 7
4 CODING PART 8
LIST OF TOOLS / COMPONENTS USED
5 9
AND THEIR DESCRIPTION
GESTURE RECOGNITION USING CNN
6 9
DEEP LEARNING
7 NETWORK ARCHITECTURE 10
8 RESULTS AND DISCUSSION 11
9 APPLICATIONS 12
10 CONCLUSION 13
11 REFERENCES 13
4
1. INTRODUCTION
Here the ASL dataset which contains the hand gesture containing (0-9) is
used. Generally an image is preprocessed where it plays a vital role for extracting
the gesture in a static image ( i.e ) background subtraction , image binarization . the
feature is extracted from all the images after binarization .
5
2. LITERATURE SURVEY
Author proposed a real time vision based system for hand gesture
recognition for human computer interaction in many applications. The system can
recognize any different hand gestures given by Indian and American Sign
Language or ISL and ASL at faster rate with virtuous accuracy. RGB-to-GRAY
segmentation technique was used to minimize the chances of false detection.
Authors proposed a method of improvised Scale Invariant Feature Transform
(SIFT) and same was used to extract features. The system is model using
MATLAB. To design and efficient user friendly hand gesture recognition system, a
GUI model has been implemented.
6
3. PROPOSED BLOCK DIAGRAM
4. CODING PART
TESTING
clc;
close all;
clear all;
warning off;
c=webcam;
load myNet;
x=0;
y=0;
height=200;
7
width=200;
bboxes=[x y height width];
while true
e=c.snapshot;
IFaces=insertObjectAnnotation(e,'rectangle',bboxes,'Processing
Area');es=imcrop(e,bboxes);
es=imresize(es,[227,227]);
label=classify(myNet,es);
imshow(IFaces);
title(char(label));
drawnow;
end
TRAINING
clc;
clear all;
close all;
warning off;
g=alexnet;
layers=g.Layers;
layers(23)=fullyConnectedLayer(7);
layers(25)=classificationLayer;
allImages=imageDatastore('Hand
Dataset','IncludeSubfolders,true','LabelSource','foldername');
opts=trainOptions('sgdm','InitialLearnRate',0.001,'MaxEpochs',20,'MinBatchSize',
64);
myNet=trainNetwork(allImages,layers,opts);
save myNet
8
5. LIST OF TOOLS / COMPONENTS USED AND THEIR
DESCRIPTION
a) MATLAB R2020b
b) AI,Deep learning toolbox
c) WEB camera
Initially the dataset is loaded into the network for training . Preprocessing
is done before the feature is extracted. The training is done in Convolutional
neural network . After training an input image is given by capturing from a
webcam. The given input image is tested for recognizing the gesture . A confusion
matrix is produced accordingly to the produced output with its mean accuracy.
A ConvNet is a popular machine learning algorithm. Which is one of the
techniques of Deep learning and is a learning model used to execute classification
tasks through images, video, text, and sound .CNNs specifically give better results
for identifying patterns in an image, which leads to recognizing of hand gesture,
face, and any object. The advantage of CNN is it don’t require any feature
extraction to train the model. CNN is invariant to the scaling and rotation.
9
7. NETWORK ARCHITECTURE
10
Classification Layer: Classification layer is the final layer of the
architecture. This layer classifies the classes based on probabilities
obtained from softmax layer and also calculate cost function.
Training Options
The maximum number of epochs set to 15 and initial learning rate is 0.001.
11
Fig no 3: Real time hand gesture recognition
9. APPLICATIONS
12
10. CONCLUSION
Fully connected layers Scaling Human hand gestures provide the most
important means for non-verbal interaction among people.
11. REFERENCES
• Sebastian Marcel, Oliver Bernier, Jean Emmanuel Viallet and Daniel Collobert.
(2000). “HandGesture Recognition using Input –Output HiddenMarkov Models”,
Proc. of the Fourth IEEE International Conference on Automatic Face and Gesture
Recognition, pp.456 - 461.
• Xia Liu and Kikuo Fujimura. (2004). “HandGesture Recognition using Depth
Data”, Proc. of the Sixth IEEE International conference on automatic Face and
Gesture Recognition,pp. 529-534.
13
14