1/14/2016
Active Surveillance as a Service
(ASaaS)
Andrew Fisher (EE), Robert Marshall (EE), Brahm Powell (ME), Titus
Lungu (ME), Mark Heller (CE), Nick White (CE)
Faculty Advisors:
Dr. Pong Chu (ECE) and Dr. Majid Rashidi (ME)
Parker Hannifin
Dr. Joseph Kovach
Background
More
Captured
Suspects
Less
Repeat
Offenders
Less
Campus
Crime
Safer
Campus for
Students
and Faculty
1
1/14/2016
Design Objectives
Maximize
• Accurate Detections
• Usability
• Utility and Relevance
Minimize
• False Readings
• Network Usage
• Interaction and Setup
Technical Approach
Unmet Needs
Design Concepts
Selected Design
Technical Specifications
2
1/14/2016
Unmet Needs
Real-time
analysis
Uninterrupted
Surveillance
Multiplecamera
tracking
Ground-level
Analytics
Technical Approach
Unmet Needs
Design Concepts
Selected Design
Technical Specifications
3
1/14/2016
Measuring Design Feasibility
Effective (Solves Current Issues)
Project Complexity
5
4.5
4
4
3.5
3
3
2.5
2
2
1.5
1
1
0.5
0
0
0.5
1
1.5
2
2.5
Efficiency (Scalable, Performance Driven)
Technical Approach
Unmet Needs
Design Concepts
Selected Design
Technical Specifications
4
1/14/2016
Design Analysis
Design
Design
Scalable
Cost
Effective
Secure
Flexible
Design
Feasible Design
Technical Approach
Unmet Needs
Design Concepts
Selected Design
Technical Specifications
5
1/14/2016
Technical
Specifications
Algorithm
Design
System
Architecture
Distribution &
Maintenance
Detection
Tracking
Layered Design
Scalable
Effective
Software Delivery
Updates, User
Support
Algorithm Design
Computer
Vision
Library
Detect
Track
Make
Decisions
6
1/14/2016
Detect
EmguCV
OpenCV
RGB
Gait
Track
Size
Motion
7
1/14/2016
Make Decisions
ID
Match
Detect
Crime
Active
Suspect
Tracking
System Architecture
Presentation Layer
Business Logic Layer
API
Operational Database
Operational
Database
Data
Warehouse
OLAP System
8
1/14/2016
Presentation Layer
Web Application
• Microsoft ASP.NET MVC 6
Desktop Application
• Microsoft Windows Presentation Foundation (WPF)
Mobile Application
• Native iOS & Android Support
Presentation Layer
Business Logic Layer
API
Operational Database
Operational
Database
Data Warehouse
OLAP System
Business Logic Layer
Independent
• Logic remains unchanged
Controller
• Holds algorithm and data access logic
Presentation Layer
Business Logic Layer
API
Operational Database
Operational
Database
Data Warehouse
OLAP System
9
1/14/2016
Database Layer
Operational EntityRelationship Diagram
• Does not concern itself with logic layer details
Independent
Optimized
• RDMS is efficient and capacity is near limitless
Presentation Layer
Business Logic Layer
API
Operational Database
Operational
Database
Data Warehouse
OLAP System
Big Data Analytics
Presentation Layer
Business Logic Layer
API
Operational Database
Operational
Database
Data Warehouse
OLAP System
10
1/14/2016
Project Management
Deliverables
Budget
Communication
& Planning
Deliverables
WPF Application
• User interaction, presentation layer
Logic Library
• Actual tracking functionality, logic layer
Operational Database
• Ability to persist and make intelligent decisions
Data Warehouse
• Ability to perform exploratory analysis on historical data
11
1/14/2016
Budget
Cloud
Processing
•$0
Camera
Implementation
•$0
Software
Licenses
•$0
Communication & Planning
Development Strategy
• Agile Approach
Version Control
• Visual Studio Online (Git based)
Communication & Task
Management
• Visual Studio Online (Scrum Manager)
12
1/14/2016
Market Potential
In 2013
●Video surveillance - $14.98 billion
●VSaaS - $1.94 billion
In 2020
●Video surveillance - $48.32 billion
●VSaaS - $6.28 billion
Our Estimation
• 10 million cameras
• $10 per camera per month
• $100 million per month
• $1.2 billion per year and growing
13
1/14/2016
Timeline
Design
• Architecture Diagrams
• Process Flow Diagrams
Fall 2015
Spring 2016
Development
• Logic and Algorithm
Development
• System Architecture
Development
Gantt Chart Representation
Questions
Design Objectives
Technical Approach
Project Management
Deliverables
Budget
Communication & Planning
Market Potential
Timeline
14