(Affiliated to Visvesvaraya Technological University, Belagavi & Approved by AICTE, New Delhi.
Department of Computer Science & Engineering
Presentation On
“Urban Street Cleanliness Assessment Using Mobile Edge
Computing and Deep Learning”
Presented By:
Under the Guidance of :
05/02/2020 1
CONTENTS
Abstract
Introduction
LiteratureSurvey
Block Diagram
Methodology
Hardware and Software Requirements
Applications
Advantages
Conclusion
05/02/2020 2
ABSTRACT
During the process of smart city construction, city managers always
spend a lot of energy and money for cleaning street garbage due to
the random appearances of street garbage. Consequently, visual
street cleanliness assessment is particularly important.
We proposes a novel urban street cleanliness assessment approach
using mobile edge computing and deep learning.
First, the high resolution cameras installed on vehicles collect the
street images. Mobile edge servers are used to store and extract
street image information temporarily.
Second, these processed street data is transmitted to the cloud
datacenter for analysis through city networks .
At the same time, Faster Region-Convolutional Neural Network
(Faster R-CNN) is used to identify the street garbage categories and
count the number of garbage.
Finally, the results are incorporated into the street cleanliness
calculation framework to ultimately visualize the street cleanliness
levels, which provides convenience for city managers to arrange
clean-up personnel effectively
05/02/2020 3
INTRODUCTION
A smart city is an urban area that uses state-of-
the art technologies such as the Internet of Things
(IoT) , Cloud computing and other information
technologies to manage and assess there sources
and environment of a city in an efficient way.
The smart city concept integrates information and
communication technology, and various physical
devices connected to the network to optimize the
efficiency of city operations and services.
05/02/2020 4
INTRODUCTION
However, due to the rapid development of a smart
city, city managers are facing huge challenges in
how to develop and maintain urban infrastructure.
Street cleanliness represents the spiritual outlook
and humanistic atmosphere of a city.
Keeping the streets clean is good for the
development of modern cities. Currently, many
major cities regard urban street cleanliness as one
of the primary tasks of urban civilization
05/02/2020 5
PROBLEM STATEMENT
In street garbage project there are many
errors in segmentation judgment and they do
not provide details of the garbage type.
In Existing approach they have not carried out
an urban street cleanliness assessment.
In existing approach is that the monitoring is
limited to garbage bins and cannot be
extended to monitor the streets.
05/02/2020 6
Objectives
To classify different types garbage type.
To build an efficient urban street cleanliness
assessment system.
To build system to monitor the streets .
05/02/2020 7
LITERATURE SURVEY
S. TITLE AUTHOR DESCRIPTION LIMITATION
NO
1. Smart city reference S. Zygiaris The framework We have
model: Assisting provides an idea studied
planners to for realizing about
conceptualize the sustainable practical
building of smart city development of a usage of
innovation smart city. The smart cities.
ecosystems recent practical
application is to
analyze smart
city planning in
big cities such as
Barcelona,
Amsterdam, and
05/02/2020 8
LITERATURE SURVEY
S. TITLE AUTHOR DESCRIPTION LIMITATION
NO
2. Citizen U. Aguilera, O. The proposed vision is It is suitable
centric Peña, O. achieved by providing a for the text
data Belmonte, and common access based data
services D. López-de- mechanism to the collection
for smarter Ipiña heterogeneous data
cities sources offered by the
city, which reduces the
complexity of accessing
the city’s data whilst
bringing citizens closely
to a prosumer (double
consumer and producer)
role and allowing to
integrate legacy data
05/02/2020 9
into the cities’ data
LITERATURE SURVEY
S. TITLE AUTHOR DESCRIPTION LIMITATION
NO
3. A deep C. Balchandani, The proposed It is not
learning R. K. Hatwar, P. framework for carried out
framework for Makkar, Y. Shah, street cleaning an urban
smart street P. Yelure, and M. leverages the street
cleaning Eirinaki deep learning cleanliness
algorithm pipeline assessment
to analyze the
street
photographs and
determines if the
streets are dirty
by detecting litter
objects.
05/02/2020 10
LITERATURE SURVEY
S. TITLE AUTHOR DESCRIPTION LIMITATION
NO
4. Automatic L. J. C. Brinez, A. The proposed a It is limited
waste Rengifo, and M. system to classify to the school
classification Escobar waste in an environment.
using automatic way as
computer an application of Implementati
vision as an computer vision in on cost is
application in Colombian high high
colombian schools.
high schools
05/02/2020 11
LITERATURE SURVEY
S. TITLE AUTHOR DESCRIPTION LIMITATION
NO
5. Competitive Taylor Buck, N. The proposed a It doesn’t
urbanism and and White, A. system examines support for
the limits to initiatives by the image data.
smart city UK national
innovation: government to It consumes
The UK future facilitate urban more time
cities initiative technological
innovation
through a range
of strategies,
particularly the
TSB Future Cities
Demonstrator
05/02/2020 12
SYSTEM DESIGN
05/02/2020 13
Modules
Data Collection
Data Preprocessing
Garbage Detection
Cleanliness Assessment
05/02/2020 14
METHODOLOGY
DEEP NETWORK
Deep learning originates in artificial neural
networks. By establishing multiple hidden layers
and training large amounts of data, useful features
can be learned to achieve the expected
classification effect.
Working process of RNN:
The image is input to the Convolutional neural
network, and spread to the shared Convolutional
layer to get the feature map;
The feature map extracted by the shared
Convolutional layer generates a suggestion window
through RPN network, and gives region
suggestions and region scores;
05/02/2020 15
METHODOLOGY
The feature map of the first step is input to the
pooling layer in Fast R-CNN to extract area
features. Combined with region suggestions and
region scores, classification probabilities and
bounding box regression are trained, the
classification scores of the region are output, and
the results are finally tested.
Faster R-CNN is considered as one of the most
precise image detection approaches. It has high
detection accuracy and speed.
05/02/2020 16
HARDWARE AND SOFTWARE
REQUIREMENTS
Software Requirements:
Operating system : Windows 7
Coding Language : Python3.6
opencv
05/02/2020 17
HARDWARE AND SOFTWARE
REQUIREMENTS
Hardware Requirements:
System : Pentium i3
Hard Disk : 120GB
Monitor : 15’’LED
Input Device : Keyboard , Mouse
Ram : 4GB
05/02/2020 18
ADVANTAGES
‣ Improves accuracy of detecting the waste.
‣ Reduces the Time complexity by using the efficient feature
extraction method.
‣ It is scalable for the large amount of data.
‣ Project is simple and easy to use and understand.
05/02/2020 19
APPLICATIONS
‣ Swatch Bharath
‣ Smart cities
05/02/2020 20
Possible outcomes
The development of novel technologies has driven
a number of cities into the way to smart cities.
Street cleanliness is one of the concerns for smart
cities. Consequently, this paper proposes a novel
urban street cleanliness assessment approach
using mobile edge computing and deep learning.
A visual street cleanliness road diagram is
presented, such an automated system can help city
administrators to know the cleaning state of the
street easily.
05/02/2020 21
05/02/2020 22
