You are on page 1of 19

Project Review-1 on

“Privacy preserving with image Segmentation content


based image retrieval in cloud”

Presented by

1
Under the Guidance of

8/2/2019
OUTLINE

Problem Statement
Abstract
Introduction
Objectives
Literature Survey
Methodology
Algorithm
Modules split-up
2

Proposed System
Software Tools and Technology
Proposed Outcomes
References 8/2/2019
PROBLEM STATEMENT

• As on-board cameras get more and more popular, numerous high-resolution


photos are generated every day, which makes storing, sharing image data
become challenging. In fact most of those images contain sensitive secure
data and individuals' protection concerns ruin their investment into un-
trusted services. So these all data have privacy protection which is major
task.

8/2/2019
ABSTRACT

• Many cloud stages rise to meet earnest requirements for extensive volume of
individual image store, sharing and search.
• The present cloud stages give little support to image security assurance that
is privacy protection.
• In this work, present : a Privacy-protecting Image Search framework on
Cloud, or, in other words towards possible cloud services which give secure
content based extensive scale image search with fine-grained access control
using Shape-Based Invariant Texture Index (SITI).
• Users can search on others' images on the off chance that they are approved
by the image owners.
• Our security investigation and model framework evaluation results will
demonstrate that this work effectively secures the image protection requiring
little to no effort of calculation and correspondence

8/2/2019
INTRODUCTION

• What is privacy protection?


• Importance of image security and image segmentation
• What is content based image retrieval?
• Use of this system

8/2/2019
OBJECTIVES

• To support image data privacy protection


• To implement Shape-Based Invariant Texture Index (SITI) for global
feature’s extraction.
• To prevent data from unauthorized users.
• To prevent leakage on the query contents or results

8/2/2019
LITERATURE SURVEY(1/3)

Sr. Author/ Paper Name Proposed System Advantages Drawbacks


No
1 L. Zhang, T. Jung, C. Liu, In this work, they introduce This framework This system not
X. Ding, X.-Y. Li, and Y. a framework POP, which allows users to used OTP for
Liu, “Pop: Privacy- enables privacy-seeking define user
preserving outsourced mobile device users to personalized authentication.
photo sharing and outsource burdensome private content by
searching for mobile photo sharing and searching a simple check-
devices,” in ICDCS. IEEE, safely to untrusted servers. box configuration
2015. Unauthorized parties, and then enjoy the
including the server, learn sharing and
nothing about photos or searching services
search queries. This is as usual.
achieved by carefully
7 designed architecture and
novel non-interactive
privacy-preserving
protocols for image
similarity computation

8/2/2019
LITERATURE SURVEY (2/3)

Sr. Author/ Paper Name Proposed System Advantages Drawbacks


No
2 Y. Hua, H. Jiang, and D. They propose a near-real- FAST is designed 1. This system
Feng, “Real-time semantic time and cost-effective to exploit the only worked on
search using approximate semantic queries based correlation data storage not
methodology for large- methodology, called FAST. property of data content based
scale storage systems,” The idea behind FAST is to by using image retrieval.
IEEE Transactions on explore and exploit the correlation-aware 2. This system
Parallel and Distributed semantic correlation within hashing and not used OTP
Systems, vol. 27, no. 4, pp. and among datasets via manageable flat- for user
1212–1225, 2016. correlation-aware hashing structured authentication.
and manageable flat- addressing.
structured addressing to This enables
significantly reduce the FAST to
8 processing latency, while significantly
incurring acceptably small reduce processing
loss of data-search accuracy latency of
correlated file
detection with
acceptably small
loss of accuracy
8/2/2019
LITERATURE SURVEY(3/3)

Sr. Author/ Paper Name Proposed System Advantages Drawbacks


No
3 Z. Xia, X. Wang, L. They proposed a scheme The feature Low
Zhang, Z. Qin, X. Sun, and that supports CBIR over vectors are embedding
K. Ren, “A privacy- encrypted images without protected by the capacity.
preserving and copy- leaking the sensitive secure KNN
deterrence content-based information to the cloud algorithm, and
image retrieval scheme in server. Firstly, feature image pixels are
cloud computing,” IEEE vectors are extracted to encrypted by a
Transactions on represent the corresponding standard stream
Information Forensics and images. After that, the pre- cipher.
Security, vol. 11, no. 11, filter tables are constructed
pp. 2594–2608, Nov 2016. by locality-sensitive hashing
to increase search efficiency
9

8/2/2019
METHODOLOGY

Image Feature Cloud Server


Extraction
Transfered/ stored
Data
Original
Image Encryption
Image

Access Specifier /
Secret key

User
Database
Query Image
Feature Extraction

10 Search by
Image/ Tag Searched Result
Extract/ Download
Data
Decrpyted Result
using User Image
Secret Key

8/2/2019
ALGORITHM

1. Blowfish Algorithm for image Encryption

2. K-means Clustering for image Segmentation

3. SITI for image feature selection

11

8/2/2019
MODULES SPLIT-UP(1/2)
User Module
1. User Registration
2. User Login
3. Create Image Repository
4. Select Repository and Upload image
4.1 First Apply Encryption algorithm
4.2 Then K means segmentation algorithm
4.3 Then SIFT descriptor feature extraction algorithm
4.4 Store all image
5. Search image
12
5.1 Select input image
5.2 First Apply Encryption algorithm
5.3 Then K means segmentation algorithm
5.4 Then SIFT descriptor feature extraction algorithm
5.5 Match Features. 8/2/2019

5.6 View Relevant Images5


MODULES SPLIT-UP(1/2)

Admin Module

1. Admin Login
2. View all users
3. Authorize Users
4. View Analysis Graph

13

8/2/2019
PROPOSED SYSTEM

• In this system, the authenticated user’s can store the images


on cloud with image features, Segment image, encrypted
image copy (using encryption), secret key and particular
image id.
• Users can search on others' images and firstly get image
encrypted copy in search result.
• Users can get original image result only if they are approved
by the image owners using particular image secret key.
• Image download history will be also stored in database which
14 is extra advantage of this propose work.

8/2/2019
ADVANTAGES OF PROPOSED SYSTEM

• This proposed approach support image data privacy


protection
• Using Shape-Based Invariant Texture Index (SITI) and
Segmentation using K-Means Clustering, proposed work will
get more accurate result.
• In proposed approach, data stored in encrypted format so
more security will be provided.
• Security will be maintained using access control.
• Also security will be maintained using users image secret key.
15

8/2/2019
SOFTWARE TOOLS AND TECHNOLOGY

Hardware Requirements
• Processor - Intel i3 core
• Speed - 1.1 GHz
• RAM - 2GB (min)
• Hard Disk - 20 GB
• Key Board - Standard Windows Keyboard
• Mouse - Two or Three Button Mouse
• Monitor - SVGA
Software Requirements
16 • Operating System - Windows7
• Application Server - Apache Tomcat7.0
• Coding language - JAVA (JDK 1.8)
• Database - MySQL 5.0
• IDE - Eclipse 8/2/2019
PROPOSED OUTCOMES

• The framework enables both encrypted storage and searching using


Content-Based Image Retrieval queries while preserving privacy against
honest-but-curious cloud administrators.
• Proposed work will show that this solution is secure and privacy
preserving, while correctly realizing the goal of content-based image
search.

17

8/2/2019
PROJECT PLAN

Schedule Date Project Activity


July 1stWeek 01/04/2019 Project Topic Searching
2ndWeek 09/04/2019 Project Topic Selection
3rd Week 15/04/2019 Synopsis Submission
August 1stWeek 05/09/2019 Presentation On Project Ideas
2ndWeek 12/09/2019 Submission Of Literature Survey
3rd Week 19/09/2019 Feasibility Assessment
September 1stWeek 02/09/2019 Documentation for paper publishing.
3rdWeek 16/09/2019 Design Of Mathematical Model
4thWeek 23/09/2019 Paper is publish.
October 1stWeek 08/10/2019 Report Preparation And Submission
December 3rdWeek 19/12/2019 1stmodule presentation
4thWeek 26/12/2019 Discussion and implementation of
2ndmodule
18 January 1stWeek 02/01/2020 Preparation for conference
2ndWeek 09/01/2020 Study of algorithm.
3rdWeek 16/01/2020 Discussion about modification.
4thWeek 23/01/2020 1stand 2ndmodule presentation
5thWeek 30/01/2020 Discussion on flow of project and
designing new module

UCOER COMP 8/2/2019


Thank You!!!
19

8/2/2019

You might also like