Upload
others
View
7
Download
0
Embed Size (px)
Citation preview
Copyright ©2014 Cyclops Technologies, Inc. This document is available to the public but is considered to be the property of Cyclops Technologies, Inc. 640 Brooker Creek Blvd., Ste. 465, Oldsmar, Florida 34677.
ARES Cloud
The Case for Cloud Based Video Analytics With the increasing number of international conflicts and domestic threats, public safety and threat prevention capabilities are steadily growing in demand by both law enforcement and corporations. By providing real-‐time video analytic and digital forensic capabilities to these entities, threat prevention capabilities are greatly increased in both an efficient and cost-‐effective manner as threats are able to be identified in real-‐time and have the appropriate people notified instantly. With more comprehensive and featured tools, these solutions now have the capability to be completely hosted offsite in a cloud solution. This type of hosting solves many issues that are encountered by traditional onsite solutions.
One such issue is that onsite systems necessitate a large initial investment into processing hardware and leaves the maintenance of the hardware in the hands of the customer. This presents an immense problem for entities which lack either the available funds for the initial hardware or dedicated technical staff to ensure their processing hardware is healthy and fully operational. In addition, this barrier of entry may prevent entities from even pursuing simplistic demo scenarios, as the deployment complexity is greatly increased in comparison to a cloud solution.
With a cloud-‐based solution, both of these issues are overcome as hardware, which can scale to any incoming demand, is available in minutes and can scale to any incoming demand. In addition, both maintenance and software updates can be handled with no interruption to service, providing 24/7 operational uptime. This not only reduces the overhead of a new installation, but it vastly lowers the speed to deployment and the barrier of entry for the customer.
Another complication of classical onsite systems is networking and communication. Businesses that wish to deploy a solution to multiple sites incur an even larger overhead in deployment and maintenance in addition to the overhead on networking systems in keeping real-‐time data ready and available across the company wide-‐area network. These complexities are not limited to multi-‐site solutions as companies that wish to monitor and access real-‐time data from outside of their network are to pursue advanced network utilities such as VPNs in order to access their data. These tools are quite expensive and necessitate a technical staff to perform network maintenance and upkeep.
This is solved in a cloud-‐based solution where the network infrastructure is completely offloaded to the hosting provider and with industry-‐leading security and best practices, real-‐time monitoring is just a click away from anywhere in the world, while maintaining system and data integrity. In addition, a cloud based solution provides a much easier means to seamlessly integrate different devices into the ARES umbrella. Mobile applications, desktop applications and websites can all be powered by a common system with a simple configuration to the end-‐user.
ARES Cloud
Overview The enhanced ARES Cloud solution will serve as the highest-‐level SaaS product offered by PlateSmart. ARES Cloud will utilize multiple backend video analytic clouds which will be responsible for their respective analytics (LPR, Facial, Thermal, Camera Association) Through the utilization of these low-‐level analytic clouds, this information will then be processed against customer alerts/rules and tracked through the ARES system with the capability to report to a multitude of video management systems. The main driving force behind this product is ease of use and flexibility allowing this product to solve extremely specialized issues.
Customer Interface Capabilities The user interface will be provided via a website client, with possible thin-‐client utilities if needed.
• Monitoring o Alerts o Plates o Statistics
• Customized reports • Administration
o User administration § Create/Delete/Update authorized users § Restrict access to specific roles
o Input sources § Cameras § Batch image processing § VMS
o VMS Integrations
o Alert Rule Configurations § Policy/Hotlist Synchronization Configuration
o Billing
Revenue Model
• Subscription based o Ideally, the revenue model needs to include a fee based on the FPS as it will be the main
cost-‐factor. o Discount for reserving camera slots in 1 year increments
§ By predicting our processing needs, we are able to save up to 60% of our server hosting fees.
o Other items could additionally be monetized § # of Users § # of Alert Policies § # of VMS Integrations
o Price to be determined based on market pricing assessment
• Question: How can we monetize the aggregate data across all customers while maintaining privacy?
• Question: What payment methods need to be supported? Advantages over Onsite Deployments
• Initial software fee is not needed, but is recouped over the long term due to reoccurring camera fees.
• Data is stored in the cloud o Customers are compelled to continue their subscriptions to continue to access data
• Servicing overhead is minimized, reducing hours spent deploying the product o Initial Installation o Subsequent updates
§ Updates are applied universally, with no downtime • Intellectual Property and Licensing
o All hardware is controlled by PlateSmart in the cloud, hence we greatly reduce any means of obtaining our engine(s) and their components.
o Customer licensing can be easily administered from a central location, ensuring no unauthorized use
Low-‐Level Analytic Clouds ARES will be powered by individual clouds of analytics such as License Plate, Facial Detection or Thermal Recognition, but their use can be expanded beyond ARES by being utilized by 3rd Party Applications. They have the ability to be directly monetized outside of the ARES architecture on a per image basis, and provide the means to be the basis of other products.
Use Cases • Batch processing not requiring alerts, or other complex ARES functionality
• PlateSmart Products o Support for onsite ARES deployments o Support for Analytic Mobile Applications
• Third-‐Party integrations and products
Revenue A tiered-‐based pricing model should be used. Provide tiered discounts based on volume of images processed. Exact prices will be determined based on market pricing assessment.
Development Roadmap
Phase One – License Plate Recognition Cloud (Currently Capable): Phase one will consist of the on-‐demand License Plate Recognition Cloud. This analytic cloud will serve as the initial basis for the ARES system and will be completely auto-‐scaling and autonomous in its regular operations. By utilizing traffic prediction and taking advantage of the reserved contracts with customers, the size of this cloud can be painlessly scaled to ensure speedy results while maintaining low costs.
Phase Two – ARES Cloud Alpha (End of Q4 2014/Early Q1 2015): This phase will consist of the initial build out of the ARES Cloud system utilizing the LPR Cloud from Phase One as its initial analytic. This phase aims to implement the following major core elements of the ARES Cloud:
• License Plate Recognition analytics o Processed against custom Alerts or Rules using
customer provided Hotlists (Ex: BOLO List) o Utilizing License Plate Recognition Cloud
• Alert/Rule Notifications via Email/Text or VMS • Complex Reporting Capabilities • Automatic Hotlist/Policy Synchronization • Basic Plate and Alert monitoring • Camera Association • Generic Metadata streams for basic VMS integration
Phase Three – ARES Cloud (Q2 2015): In addition to the above ARES Cloud Alpha, this phase aims to implement and add the following major features into ARES Cloud:
• Automated Payment and Billing Systems • User Interface Advancements
o Customer Dashboard § Advanced Plate and Alert monitoring § System statistics
o Account user administration • Analytic Processing Inputs
o VMS integration o Low-‐priority batch processing
• External Event Inputs – Used to aid in analytic processing o Examples
§ Gunshot locations § Physical Alarm locations
Ongoing – Analytic Clouds Phase four will begin to supplement the ARES system with more complex analytic clouds such as Facial Detection and Thermal Recognition. These clouds will be structured very similarly to the License Plate Recognition Cloud and utilized in the same fashion. ARES Cloud will use the raw information form the various analytics utilized to process this data into intelligent actionable analytic information.
Ongoing – VMS Integrations One of the central components to the ARES Cloud vision is that of a unified video analytic system, therefore it is our goal to provide a wide selection of integration options for our customers. By providing these integrations, we allow our customers to view and monitor their analytics from one central component rather than bleeding man hours into needing to watch multiple various systems. Currently we support the following VMS systems:
• Pelco VMS • OnSSI • OnSSI C2P • ExacqVision
We will also be integrating with Milestone in the near future and VMS Integrations will continue to be performed, expanding the umbrella of the ARES system. By integrating with more systems we allow ourselves to serve as a central leader in the video analytics industry, serving as a central hub for accurate and complex analytics.
ARES Visualization
Alert on VMS Interface
Text Message