Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 38

WELCOME

VISVESVARAYA TECHNOLOGICAL UNIVERSITY


BELGAUM-590018, KARNATAKA, INDIA

EAST WEST INSTITUE OF TECHNOLOGY


DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

PHASE 2 PRESENTATION
“AI Based Smart Voice Assistance For villages and Rural Places”

PRESENTED BY: UNDER THE GUIDANCE OF :

AARZOO MUSKAN MN (1EW17CS002) Prof. Dr ACHYUTHA PRASAD


BHUVAN KARTHIK S (1EW17CS028) N
SANJANA NS (1EW17CS129) HOD , Dept. of CSE
SOWMYA VS (1EW17CS154) EWIT
LIST OF CONTENT

1. Introduction
2. Literature Survey
3. Problem Statement
4. Goals of Project
5. Proposed System
6. Implementation
7. Conclusion
8. References
INTRODUCTION

 In India there are around 6,00,000 villages out of them 1,25,000 villages are backward, so there is a need
for designing and building the village as a smart village. Village is main criteria for development of
nation.

 The development of the information technology and communication has been complex in implementing of
artificial intelligent systems. The systems are approaching of human activities such as decision support
systems, robotics, natural language processing, expert systems, etc.

 This application work is very simpler because the knowledge already programmed in advance. One of
methods used in this application is to match the pattern (pattern-matching).
 Machine learning (ML) is a category of an algorithm that allows software applications to become more

accurate in predicting outcomes without being explicitly programmed. The basic premise of machine

learning is to build algorithms that can receive input data and use statistical analysis to predict an output

while updating outputs as new data becomes available.

 IoT or the internet of things is the interconnection of devices to collect and transfer the data. It uses

multiple technologies and algorithms to communicate such as Raspberry pi, Camera, Mic , Speaker.

 The interconnection of these multiple embedded devices will result in automation in nearly all fields, also
enabling advanced application like improved accuracy, efficiency and economic benefit with reduces
human intervention.

 In this project we describe about “AI Based Smart Voice Assistance For villages and Rural Places”. The
proposed design is a system which provides answer to queries related to Agriculture, leaf disease
detection, panchayat officer contact details, acts as voice machine.
Literature Survey

S. Tittle of the Author name methodology Advantages Disadvantages


No project
1 ROBOTIC VEHICLE Aamina The proposed system aims to After the images of the Cannot recognize the all
FOR AUTOMATED Nooraiyeen automate the process of image
DETECTION OF capturing of the plant leaves leaves have been type of diseases
LEAF DISEASES using the robotic vehicle and captured using the
later processes the captured camera they are sent to
images on a remote system the remote computer
using MATLAB. An additional
feature of voice control has for further processing.
been included in the robotic The events that occur
vehicle for actuating during the processing of
movements according to the
users voice commands. images

2 RASPBERRY PI Piyush Vashistha This research work aims to build The system experiences Unable to communicate
BASED VOICE- Juginder Pal Singh up a personal assistant by using the downside that just in native language
OPERATED Pranav Jain Raspberry Pi as a processing
PERSONAL Jitendra Kumar chip and underlying predefined voices.These
ASSISTANT(NEOB architecture. It emphasizes the systems are playing out
OT) substitution of screen-based the restricted
interaction by utilizing ambient
technologies, Robotics and IoT, assignment either by
means the user interface is voice controlled
integrated with the physical commands .
gadget.
S. Tittle of the Author name methodology Advantages Disadvantages
No project

3 Smart Sivaraman The summoning of an ambulance Timely prediction of The proposed system can be
summoning of Karthikeyan during an emergency or accident vehicle accident and extended to interact with
ambulance is attained by the arrangement of sending the message and traffic signals in order reach
during a vehicle the following components location to concerned the accident location and
accident microcontroller, a GPS module, a person. hospital in short duration.
gsm module and a crash sensor. This saves many lifes.
This whole setup is installed as a
black box inside the vehicle so as
to instill reliability of appliance
inside the vehicle.
4 Detection Saumitra The method consists of character With the help of OCR and Persons should be educated to
Of Godbole , area detection and character Text Summarization we operate
medicine Dhananjay extraction by image processing. can extract the name of the
information with Joijode , Kshitij and character recognition is medicine and display
optical character Kadam , Sameer realized by the designed slight information like name,
recognition using Karoshi seven-layer network. The usage, dosage, etc. on the
android recognition accuracy of the neu ral application.
network is upped to 98.19%, and
the medicine images are also
recognized correctly. Furthermore,
the recognition time is very short,
which shows that the proposal can
be used for practical application.
S . No Tittle of the project Author name methodology Advantages Disadvantages

5 Grape Leaf Disease S.M. Jaisakthi; We have proposed an Using SVM we Cannot recognize
Identification using P. Mirunalini; D. automatic system for have obtained a the all type of
Machine Learning Thenmozhi; detecting the diseases in better testing diseases
Techniques Vatsala the grape vines using
image processing and
accuracy of 93%.
machine learning
technique. The system
segments the leaf (Region
of Interest) from the
background image using
grab cut segmentation
method.
PROBLEM STATEMENT

 The rural people face a lot of communication gap with the present developing society filled with
the technologies. They are unaware of new methods and technologies being used, which can
make their work easy and more efficient.

 They lack in knowledge regarding basic education, medicine usage, agricultural remedies and
usage of pesticides , hygiene and steps to be taken during emergency .
GOALS OF PROJECT

Educating about schemes for farmers by the government and current affairs , providing right
pesticide information.

People need not to wait for panchayat officer to solve their queries, human dependence will be
reduced

A Smart model for communication between the model and rural people to help and educate them .

To monitor hydrogen levels in the surrounding and alert to higher authorities if detected more.

Our proposed system has emergency dial button for the emergency purpose , it notifies the near
by hospital
PROPOSED SYSTEM

 The aim of this project is to build the communication between the Smartbox and the village
people and make it easy for the people who are not able to read or write.
 It consists of Raspberry pi, camera,Speaker,Mic and Power Supply.
 The kannada language has been trained by using Open Cv and Python.
CONCEPTUAL/ANALYSIS MODELLING

Sequence diagram:
Use case diagram : Voice to Voice
Use case diagram : Prescription Reading
Use case diagram : Leaf disease detection
Activity Diagram:
Data Flow diagram:
COMPONENTS DESIGNED

RASPBERRY PI
A small credit-card sized computer capable of performing various functionalities such as in surveillance
systems, military applications,
SPECIFICATIONS
SD CARD

 Raspberry PI has no storage on board.


 SD/micro SD Card stores the OS. The total memory of the SD card is about
8GB.
 Class 10 is preferred owing to its high speed.
RPI CAMERA
 To enable the camera go to the command window and go to the configuration window.
 sudo raspi-config
ALGORITHM:CNN

 Apply convolution filter in first layer


 The sensitivity of filter is reduced by smoothing the convolution filter (i.e) subsampling
 The signal transfers from one layer to another layer is controlled by activation layer
 Fasten the training period by using rectified linear unit (RELU)
 The neurons in proceeding layer is connected to every neuron in subsequent layer
 During training Loss layer is added at the end to give a feedback to neural network
SOFTWARE REQUIREMENT SPECIFICATION

Functional Requirements

To develop system which ,

 Recognize the voice and process it.

 Identify the captured image and process it.

 Detects the leaf diseases by leaf image using CNN

 Also give remedies to farmer with respect to diseases

 Automatically updates the database .


NON FUNCTIONAL REQUIREMENTS

 Usability

 Reliability

 Performance

 Supportability

 Portability
IMPLEMENTATION

• Modules
Leaf Disease Module
 There are 5 sub modules in the system. They are (i) Image Acquisition (ii) Image Preprocessing (iii)
Image Segmentation (iv) Feature Extraction (v) Classification.
 In Image Acquisition, real-time tomato leaf images from the fields nearby are captured. It consists
of both healthy leaf images as well as diseased leaf images
 Diseases like Brown Spot and Leaf Blast are considered. The captured images are cropped to a
specific size.
 The cropped RGB images are converted to grayscale in Image Preprocessing. Image Segmentation
is the third component.
 It consists of segmenting the converted grayscale images using K-Means Clustering. This helps to
get rid of problems like backgrounds, illumination of light, etc.
 Feature Extraction is extracting or showing the diseased portion of the leaf so that classification
becomes easy.
CONVOLUTION NEURAL NETWORK
Output Classes
Output
Layer
pooling FC
convolutional

Hidden Layer
VOICE TO VOICE APPROACH

Speech recognition is an interdisciplinary subfield of computer science and


computational linguistics that develops methodologies and technologies that
enable the recognition and translation of spoken language into text by computers.

Train voice commands


 Store in database
 Recognize Speech
 Play Replay Sound
IMAGE TO TEXT CONVERSION

The above shows the flow of Text-To-Speech. The first block is the image pre-processing
modules and the OCR. It converts the pre-processed image, which is in .png form, to
a .txt file. We are using the Tesseract OCR
APPLICATIONS

Educates rural people about the medicine and its usage.


Helps the farmers to recognise the diseases of leaves and get remedies for them .
CONCLUSION

 The device is used to convert text/image for the deaf and also used to convert voice for
the blind.
 We have designed the prototype model for human and robot communication into a
single compact device.
 The device can be used as smart assistant for differently abled people to communicate
with others and it is a language independent system.
 Reduce mis guidance \ mis communication
REFERENCES

[1] Jiansheng Liu, Bilan Zhu.“An intelligent personal assistant robot: BoBi secretary”,2nd
International Conference on Advanced Robotics and Mechatronics (ICARM),2017

[2] Jishnu U.K., Indu V., K.J. Ananthakrishnan, Korada Amith, P Sidharth Reddy, Pramod S.“Voice
Controlled Personal Assistant Robot for Elderly People”, 5th International Conference on
Communication and Electronics Systems (ICCES),2020

[3] Nidhal Hadj Abdallah, Eya Affes, Yassine Bouslimani, Mohsen Ghribi, Azeddine Kaddouri, Moez
Ghariani.“Smart Assistant Robot for Smart Home Management” ,1st International Conference on
Communications, Control Systems and Signal Processing (CCSSP),2020

[4]S.M. Jaisakthi, P. Mirunalini; D. Thenmozhi.“Grape Leaf Disease Identification using Machine


Learning Techniques” ,International Conference on Computational Intelligence in Data Science
(ICCIDS), 2019
QUERIES
THANK YOU

You might also like