Best Practices in Video Analytics Deployments in Retail - WordPress

Document Sample
Best Practices in Video Analytics Deployments in Retail - WordPress Powered By Docstoc
					                            Best Practices
                   Intelligent Video Management

                           National Institute of Standards and Technology
                    Video Surveillance Technologies for Retail Security Symposium
            Session: Current / Emerging Technologies to Detect Shoplifting Behavior (Part I)
                                               June 2009 

The contents of this presentation are considered to be property of Aimetis Corp. The contents herein may not be reproduced
                             or distributed without the specific written consent of Aimetis Corp.
Purpose & Agenda

Share best practices for successful Intelligent Video deployments in enterprise
environments, with focus on loss prevention applications.

  1      Retail & Intelligent Video
  2       Planning for Success
  3       Applications – Where Are the Wins Today
  4       Evaluation, Deployment & Management
  5       Summary

 Estimated time: 20 minutes
Retail & Intelligent Video
What is driving retail to adopt Intelligent Video?

1. Business Opportunity
 Intense competition within the retail segment to increase
 In developed countries, increase in profit will not come from macro
   growth of consumer spending -- rather through better operational
   efficiency and competitiveness.
 Shrinkage in the retail industry is affecting the bottom line
    > 46% by Staff
    > 34% by Customers
    > 16% by Internal / Admin
    > 4% by Suppliers
 Continuous increase of crime toward retailers and their employees

2. Technology Adoption
 Video Surveillance is already a prevalent application in the
 Acceptance of new technologies – so long that it improves the
   bottom line
Retail & Intelligent Video
Definition of Intelligent Video

Two Equally Important Elements of Intelligent Video:
        1.        Image Understanding (video analytics):
                  Transforms raw video into meaningful and organized data that can be leveraged for
                  informational purposes.

        2.        Information Presentation (video management):
                  Presents, manages and distributes analyzed video data in format that optimizes value
                  and usefulness.
Intelligent Video Management
Value Contribution to Video Surveillance Ecosystem

                                                     Information Creation
                                                       (Intelligent Video)
Data Capture:
 Network cameras / encoders

 Server hardware

 Storage devices

 Network routers / switches

 Cable / power

 Services
Retail & Intelligent Video
It’s not just for security anymore

Common Retail Business Requirements for Video Surveillance:
 Prevent Losses / Reduce Shrink

 Protect People & Assets

 Improve Customer Service

 Improve Inventory Management

 Enhance Business Intelligence

 Reduce Time Reviewing Video
Planning for Success
Enterprise Intelligent Video Best Practices

1. Assemble the Team
Internal Team
 Asset Protection
 Security
 Operations
 Merchandising
 IT

Vendor Partnering
 Open and frequent communication video manufacturer
  & system integrator
 Knowledge sharing: business requirements,
  measurement criteria, timelines, budgets, etc.
 Regularly scheduled calls & meetings for
  brainstorming, problem solving; planning & status

Project Management Lead (both Internal & Vendor)
 Manage process and holds continuants accountable
Planning for Success
Enterprise Intelligent Video Best Practices

2. Clear Requirements Statement
 Articulate the business problems company is facing:
      > e.g. smash & grab is up 20% resulting in $10 million
        increase in annual loss
 Prioritize the issues
      > Identify quick wins with good ROI, as well as „game
 Define the desired or acceptable state
      > Will flush out early in the process what is possible
        today, in the future, or never.

3. Site Assessment & Performance Expectation
 Identify any physical limitations or barriers to success
 Discuss what performance is possible and decide if this level
  is acceptable.
Planning for Success
Enterprise Intelligent Video Best Practices

4. System Design, Scope & Support
 Understand the solution design
      > Is it software / COTS based or hardware /
        proprietary design.
 Articulate the desired deployment scope
      > Number of stores, areas within store, etc.
 Evaluate ability to scale across the enterprise
 Identify desired support strategy
      > Does the system fit into my existing video
        surveillance architecture and can our IT / Security
        department support it.
Applications: Where Are the Wins Today
Prevention & Recovery

Identifying Events Before Loss Occurs (Prevention)
 After Hours
      >   Distant perimeter breach
      >   Loitering by entrance, loading dock, glass, etc.
      >   Break & entering / interior activity
      >   Others
 During Store Hours
      >   Excessive loitering in high risk areas
      >   Item removal items in high risk areas
      >   EAS tower alert
      >   Others

Research Events After Loss Occurs (Recovery)
 Intelligent Search
      > Items removed from high risk area
      > Items in aisle (e.g. shrink wrap)
      > Staging
 Information Sharing / Distribution
      > Export / tag / encrypt video footage
      > Controlled access to event / video
Applications: Where Are the Wins Today
Event Driven Response & Mobility

Shift From Reactive to Proactive
 Dynamically update control room „video wall‟ with intelligent qualified
  events driven alerts
 “Push” intelligent driven events to appropriate mobile responders

 Automated and proactive identification of interesting events
  Increases agent productivity / scalability and item recovery.
 Improved responsiveness by enabling on-the-floor coverage & alert
 Less hardware and staff
Evaluation, Deployment & Management

Create Proof of Concept (POC) Charter
 Objective (purpose of POC)

 Scope (duration, data size, funding)

 Pre-Work (set-up, testing/optimization, configuration)

 Measurement Method

 Define Success Criteria

 Test and Data Collection

 Next Steps if Success Achieved (ensure common
 Project Lead Sign-Off
Evaluation, Deployment & Management

 Set deployment team (internal & external)
 Establish deadlines
 Ensure accountability

Phased Deployment
 Phase I (Crawl)
       > Limit the deployment scope (e.g. 5 stores)
       > Flag pitfalls and problem areas
       > Identify what works
       > Modify as needed to enable scale
 Phase II (Walk)
       > Expand the deployment scope (e.g. 40 stores)
       > Tweak any remaining issues that emerge
       > Look to automate installation / set-up (bulk configuration /
         installation capabilities)
 Phase III (Run)
       > Full scale roll out (hundreds of stores)
Evaluation, Deployment & Management

 Success of Intelligent Video dependant on
   management & usability of solution

 Both Administrator & End User

 Initial training / hands-on

 Frequent & repetitive ongoing distant learning using

 Knowledge Base / Flash Tutorials via either customer
   intranet or vendor extranet.
Evaluation, Deployment & Management

Managed Intelligent Video Services
 Simple procurement / licensing models that align vendor incentives with post deployment support

 All inclusive - software licenses, maintenance, bug-fixes, version upgrades

 Dedicated on-site technical support staff

 Key Benefits
      >   Simplified budgeting - makes the expense planning more stable and predictable
      >   Provides access to the latest technology with limited risk;
      >   Reduce costs, including traditional service fees, hardware, and IT operations
      >   Accounting flexibility to treat as Operating Expense or Capital Expense

Intelligent Video can help address challenges we face today
Keys to successful Intelligent Video solutions
1. Advanced Planning
     > Cross functional team
     > Clear requirements statement & performance expectation
     > System design, scope & support
2. Applications that deliver „wins‟ today
     > Prevention
     > Recovery
3. Proper Evaluation, Deployment & Management
                                  Thank You

         Tony Hackett
         Regional Sales Manager
         Aimetis Corporation

         P: 585-547-9675

Shared By: