Utilizing Cloud Computing for Image Retrieval Optimization: Review

Verified

Added on  2024/06/28

|86
|10423
|268
Literature Review
AI Summary
This literature review examines various techniques for optimizing image retrieval processes using cloud computing, with a focus on mobile devices. It covers approaches such as content-based multi-source encrypted image retrieval, efficient privacy-preserving content-based image retrieval schemes, and distributed image-retrieval methods in multi-camera systems for smart cities. The review highlights the goals, components, processes, advantages, and limitations of each technique, considering factors like privacy preservation, security, processing rate, and feature extraction. The studies also explore the use of algorithms like k-nearest neighbor and fault-tolerant processing to enhance efficiency and security in cloud-based image retrieval systems, addressing challenges related to data leakage, slow processing speeds, and image management in large cities. Desklib provides access to similar solved assignments and resources for students.
Document Page
Literature Review
Student Name & CSU ID

Project Topic Title
Utilising cloud computing for the optimization of image retrieval process used in mobile devices
1
tabler-icon-diamond-filled.svg

Paraphrase This Document

Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
Document Page
Version 1.0 _ Week 1 (5 Journal Papers from CSU Library)
1

Reference in APA format that will be in

'Reference List'

Shen, M., Cheng, G., Zhu, L., Du, X., & Hu, J. (2018). Content-based multi-source encrypted image retrieval

in clouds with privacy preservation.
Future Generation Computer Systems.
Citation that will be in the content
Shen et. al., 2018
URL of the
Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference
https://www-sciencedirect-

com.ezproxy.csu.edu.au/science/article/

pii/S0167739X17321969

Level of Journal: Q1
Privacy-preserving, Secure, image retrieval, Multi-source,
Content-based image retrieval, Searchable encryption, Image

encryption

The Name of the Current Solution

(Technique/ Method/ Scheme/

Algorithm/ Model/ Tool/ Framework/ ...

etc )

The Goal (Objective) of this Solution &

What is the Problem that need to be solved

What are the components of it?

Technique/Algorithm name:

Homomorphic encryption

Tools:

Multiple Image owners with Privacy

Protection

Applied Area:

Medical diagnoses

Art collection

Problem:

The major problem with
the Content-based
image retrieval major issue is that when

images are outsourced to the servers it has

chances of data leakage.

Goal:
The goal is to overcome problems and
make data secure and to allow
Multiple Image
owners with Privacy Protection (
MIPP).
ï‚·
Image owner
ï‚·
Keys
ï‚·
The cloud
ï‚·
The KMC
The Process (Mechanism) of this Work; The process steps of the Technique/system

2
Document Page
Process Steps Advantage (Purpose of this step) Disadvantage (Limitation/Challenge)
1

Generation of key:
A secret image encryption
key is obtained so that authorized users and

image owners can use it.

It will ensure image content privacy and its

features privacy.

N/A

2

Encryption and decryption of image: A

standard
key is used for image encryption
which is secured by
Chosen-Plaintext Attacks
(CPA).

The only authorized user can communicate

with image encryption and decryption key.

N/A

3

Encrypting the features of image
: Used for
preserving the image features and features

should be encrypted before uploading it to

cloud.

Without communicating with others the owner

of image and user can select it. Also,

parameters can be discarded.

With KMC, the algorithm is used to decrypt

images.

Validation Criteria (Measurement Criteria)

Dependent Variable
Independent Variable
Retrieval scheme
Images
Input and Output
Critical Thinking: Feature of this work,
and Why (Justify)

Critical Thinking:
Limitations of the
research current solution, and Why

(Justify)

3
Document Page
Input (Data) Output (View)
ï‚·
Input in the medical sector
is a medical image.

Image retrieval is done

efficiently and images are

gathered with a guarantee of

maintaining the privacy

aspects.

The techniques used in this increases the

privacy on an image and make it secure.

Also, accuracy while retrieval increases.

The third point is efficiency. It shows the

time used for retrieving a secured image.

The major limitations are:

ï‚·
Privacy of images and features of
images that belong to the different

owner their privacy should be

ensured.

ï‚·
Eavesdroppers
ï‚·
Cloud
(
Describe the research/current solution) Evaluation Criteria How this research/current solution is
valuable for your project

In this, an image retrieval scheme that is used in clouds with

privacy concerns using content-based multi-source encryption

technique is explained. Also with secure multi-party computation,

we used features of the encrypted image so that owner of that

image can encrypt the features by his own key (
Shen et. al., 2018).
Image owner provides image database.

Then an authorised user demands an image

retrieval. The cloud is responsible for

storing and retrieving an image. And at last

Key management centre is responsible for

managing encryption and decryption key.

This solution helps in CBIR with

multiple image sources. And it helps in

maintaining the image content privacy

plus the features of the image.

Diagram/Flowchart

4
tabler-icon-diamond-filled.svg

Paraphrase This Document

Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
Document Page
Figure: System Model of the Secure Multi-Source CBIR Scheme.
5
Document Page
2
Reference in APA format that will be in

'Reference List'

Xia, Z., Xiong, N. N., Vasilakos, A. V., & Sun, X. (2017). EPCBIR: An efficient and privacy-preserving

content-based image retrieval scheme in cloud computing.
Information Sciences, 387, 195-204.
Citation that will be in the content
Xia, et. Al., 2017
URL of the
Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference
https://www-sciencedirect-

com.ezproxy.csu.edu.au/science/article/

pii/S0020025516321971

Level of Journal: Q1
Locality-sensitive hashing, Content-based image retrieval,
Searchable encryption, Secure k-nearest neighbour algorithm

The Name of the Current Solution

(Technique/ Method/ Scheme/

Algorithm/ Model/ Tool/ Framework/ ...

etc )

The Goal (Objective) of this Solution &

What is the Problem that need to be solved

What are the components of it?

Technique/Algorithm name:

k-nearest neighbour

min-hash algorithm

Tools:

Mobile computing

Cloud computing

Applied Area:

Mobile Commerce

Mobile Healthcare

Mobile Gaming

Problem:

The major problem is to store images on cloud

servers. Also, privacy is not preserved for

sensitive images.

Goal:

The two major goals are, first is to maintain

efficiency and second is to maintain the

security of data.

ï‚·
Keys
ï‚·
Images
ï‚·
Cloud
6
Document Page
Mobile Learning
Mobile social networking

The Process (Mechanism) of this Work; The process steps of the Technique/system

Process Steps
Advantage (Purpose of this step) Disadvantage (Limitation/Challenge)
1
The generation of unencrypted index: By
some methods
image feature vector is extracted.
It helps to rank query image with the help of

one-one map index.

N/A

2
The index encryption:
The pain-text based information is revealed via

image feature vector format regarding the

content of image

It helps in saving the time.
N/A
Validation Criteria (Measurement Criteria)

Dependent Variable
Independent Variable
Private content-based image retrieval
Privacy of unencrypted image.
Homomorphic-encryption
High complexities
Input and Output
Critical Thinking: Feature of this work, and
Why (Justify)

Critical Thinking:
Limitations of the research
current solution, and Why (Justify)

Input (Data)
Output (View)
Data: text and images
These image features
These encryption techniques and algorithm help to

maintain the privacy of an image. Also in index

feature vectors in trapdoors are encrypted and are

The security of image features should be

improved. And the extraction features of

encrypted image help the future work for securing

7
tabler-icon-diamond-filled.svg

Paraphrase This Document

Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
Document Page
The data in these two
format are considered

to be the input for this

module.

are secured against

ciphertext and

efficiency is also

improved.

secured with the k-NN algorithm. These trapdoors

protect information from leakage and maintain

privacy.

CBIR outsourcing.

(
Describe the research/current solution) Evaluation Criteria How this research/current solution is valuable
for your project

In this, a proposal is given that supports CBIR,

and the encoded imaginings are revealed and

information which is sensitive, is forwarded to

cloud. Also, feature vectors are used for

representing the images. With locality sensitive

hashing searching efficiency increases. And with

a k-nearest neighbour (kNN) algorithm features

are secured (Xia, et. Al., 2017).

There is one image owner and one user. They need

encryption and decryption technique to be used.

The security of information is an important factor.

And also the data should be efficient enough.

kNN algorithm helps in protecting feature vectors,

as these help in enabling cloud server for ranking

the search results without any additional

communication burdens in an efficient way. With

local sensitive hashing, pre-filter tables are

created. Based on Euclidian distance methods are

purpose that can be compliantly comprehensive to

the CBIR.

Diagram/Flowchart

8
Document Page
Figure: A framework with CBIR scheme for preserving privacy.
9
Document Page
3
Reference in APA format that will be in

'Reference List'

Yang, J., Jiang, B., & Song, H. (2018). A distributed image-retrieval method in the multi-camera system of the

smart city based on cloud computing.
Future Generation Computer Systems, 81, 244-251.
Citation that will be in the content
Yang. et. al., 2018
URL of the
Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference
https://www-sciencedirect-

com.ezproxy.csu.edu.au/science/article/pii/

S0167739X17321362#sec3

Level of Journal: Q1
Cloud computing, Smart city, Multi-camera system,
Distributed image-retrieval method

The Name of the Current Solution

(Technique/ Method/ Scheme/

Algorithm/ Model/ Tool/ Framework/ ...

etc )

The Goal (Objective) of this Solution &

What is the Problem that need to be solved

What are the components of it?

Technique/Algorithm name:

Distributed fault-tolerant processing (DFP)

method
.
Tools:

Multi camera system

Applied Area:

Development of smart cities

Problem:
The major problem is performing
video sensing for the whole city. Also to

correctly retrieve image in
the multi-camera
system is difficult.

Goal:

The goal is to use
cloud computing and
improve processing rate.

ï‚·
Cameras
ï‚·
Monitor
ï‚·
Coaxial cable
ï‚·
Converter
ï‚·
Connector
The Process (Mechanism) of this Work; The process steps of the Technique/system

Process Steps
Advantage (Purpose of this step) Disadvantage (Limitation/Challenge)
1
Client Requests: For a particular purpose a It will improve the processing rate. N/A
10
tabler-icon-diamond-filled.svg

Paraphrase This Document

Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
Document Page
request is send to cloud computation source.
2
Cloud computation: In this, it considers the
chore and makes a decision.

Fast image retrieval is done.
N/A
3
Reply to the client: Final calculated information
and analysis are sent back.

Client get back the helpful information
N/A
Validation Criteria (Measurement Criteria)

Dependent Variable
Independent Variable
Processing rate
Cloud computing
Multi-camera system
Regenerative code
Input and Output
Critical Thinking: Feature of this work, and
Why (Justify)

Critical Thinking:
Limitations of the research
current solution, and Why (Justify)

Input (Data)
Output (View)
Large information set

is in need to be feed

as input in this journal

as illuminated by the

author.

These are sent to

clients and then the

computation is

performed. And then

again it is sent back to

the client.

In this, it is
grounded on a distributed fault-tolerant
processing (DFP) method. The DFP improves

dispensation rate within cloud computing, hence

will overcome slow speed issue of processing rate.

Also dispersed image-retrieval method is

envisioned using cloud-based on a multi-camera

system by using data encryption, data retrieval and

cloud storage.

Dispensation rate based on the cloud has a very

slow speed. In this it mainly used fault tolerant

processing.

(
Describe the research/current solution) Evaluation Criteria How this research/current solution is valuable
for your project

In research distributed fault-tolerant processing

(DFP) method is used. It will improve the rate of

processing. Also, management of image in big

The client is the end user. They need to retrieve the

data from large blocks and in less processing time.

The
distributed fault-tolerant processing (DFP)
method will improve the processing rate of cloud

computing. In smart cities, image management is a

11
Document Page
cities is difficult. But with processing image on
cloud fast image retrieval issue is solved. In this

fault tolerance methods are adopted (
Yang. et. al.,
2018
).
difficult task. But that has also been solved here.

Image retrieval has been done fastly by using

cloud computing.

Diagram/Flowchart

Figure: a Multi-camera system of smart city

12
chevron_up_icon
1 out of 86
circle_padding
hide_on_mobile
zoom_out_icon
[object Object]