A New Era in Video Surveillance

Before you purchase your next new CCTV surveillance system, we urge you to explore our video surveillance option.  Our design enables the use of fewer cameras to capture even more useful surveillance.   Simply text 520.955.3067 to learn more.

lenses_ricom_q1645_2mp_8-26mm_1837063_front_1805_768px
Video Optics Breakthrough
Optospatial Engineering

Optics breakthrough in the video surveillance industry.  See much more with far less!

Cloud Computing
Hybrid Cloud Computing

Design your surveillance using AWS cloud computing environment to save money.

IoT Edge Computing
IoT Edge Computing

Bring supercomputer performance to the edge of your surveillance system.

artificial intelligence smart surveillance
AI-Powered Video Analytics

Video analytics - Surveillance operators can efficiently monitor 10x more than humanly possible.

Breakthrough Innovation

A discovery in the field of video optics has the potential to accelerate the evolution of surveillance technology for military and commercial applications. This discovery will increase efficiencies, decrease costs, and advance the surveillance industry as we know it.

 

In addition, the optical solution intellectual property is independent of any specific hardware, software, or commercial brand. The power behind this technical concept is in algorithms, geospatial sciences and positioning.

Send a Message

Video Surveillance Design

Optospatial Engineering has proven to save our clients from purchasing fewer cameras.  That translates into less compute requirements, storage, and any software that charge by the camera.  The design below saved the client 75 cameras from the original system containing 450 cameras.

Wach 1 Design Layout

Image courtesy of Wach 1 Design

The Edge of Video Surveillance

Save tens of thousands of dollars by investing in edge computing.  Before you buy another bare metal server for $30k to $50k, consider the edge computing at about 25% of the cost.  Bring supercomputer performance to the edge in a small form factor system.  Run modern neural networks to process data from multiple high-resolution sensors – a requirement for full AI smart surveillance systems.

 

Are you still following the old way?  We can help you realize present-day technology and advance your surveillance while keeping costs down.

Benefits Gained from Choosing Our Design

Longer observation periods: Compared to traditional techniques, our development allows cameras to capture 5-times as many square-feet with useful detail in live and recorded video. This means objects remain clearly in view 5-times longer than they do in traditional designs.

 

Dramatically improved Artificial Intelligence (AI) confidence scores: A surveillance system that is augmented by AI does not return "IS" or "IS NOT" values when evaluating objects within the scene. Instead, they show the percentage of probability that the object is what the assigned label shows it to be. Our designs return 4 to 5 times higher confidence scores than traditional designs.

 

Better results AND lower cost: Our AI-powered application will simplify the design process and reduce overall camera, server, storage, and infrastructure cost by 30% compared to traditional designs. It will also allow end-users to design their own surveillance systems.

Side-by-Side Comparison

VIDEO #1: Traditional Video Surveillance

Video courtesy of Wach 1 Design.

Video Explained: These videos show two different shots being compared at the same time.  In video #1 is a traditional security camera view, video #2 is our applied proprietary design camera view. Keep in mind there are two stops signs in the scene that are approximately 900 ft apart. Try to look for it in video #1, then see the difference is video #2.

VIDEO #2: ENHANCED Video Surveillance

Video courtesy of Wach 1 Design.

Our surveillance design requires a change from the traditional security camera positions.  The enhanced design enables camera placement to be much further from traditional positions.  In the enhanced version, the video analytics software (Amazon) dramatically increases the object and people detection performance.  Our newly improved design advances the performance levels by nearly 5 times.

Objectives

To increase the confidence scores of artificial intelligence algorithms
Our optical design solution increases the usefulness of each of the camera’s pixels. The modified optics enable a higher pixel-per-foot ratio in the video output. Artificial intelligence algorithms will generate higher ‘confidence scores’ in detection and recognition as the pixel-per-foot increases in the video output. This enables higher levels of object detection and recognition capabilities when compared to traditional surveillance operating the same software.

To showcase leap technology in surveillance video optics capabilities
Our solution enables a maximized pixel configuration within the specified surveillance viewshed. Virtually all cameras sold today can achieve maximum efficiency of 18%. Yes, the best-case scenario for nearly all cameras deployed is less than 20% efficiency.Therefore, there is a 9-out-of-10 chance that your cameras are covering less than 1/5 the square-footage they are capable of.

 

To reduce hardware costs by using a 5G backbone to transmit data payloads

Our solution design can produce a 30% hard cost savings in surveillance hardware and software. Expanding the video surveillance depth-of-field and using cloud reach back over 5G enables fewer cameras to cover the same area. In addition to reducing camera hardware requirements, costs relating to AI image and video analysis will incrementally reduce with fewer cameras.

 

Video courtesy of Wach 1 Design.
Notice how much more time the person stays in view after detection.  Faster detection, recognition and longer periods of analysis.

AI-Powered Video Analytic Capabilities

Real-time alerts

  • Motion detection
  • Trip wire
  • Object removal or abandonment
  • Counter-flow detection
  • Tailgate detection
  • License-plate recognition
  • Combination alerts
  • Customized alerts
  • Crowd forming / running
  • Loitering
  • Directional motion

Indexing and search

  • Attribute-based search (size, color, speed)
  • Date and time ranges
  • By location, in field of view
  • License-plate search (partial or full)
  • Across multiple cameras
  • Track objects in view
  • Attributes-based search (add: time, duration)
  • Counting
  • Able to work in crowded scenes and challenging environment conditions

 

Why Use AI-Powered Video Analytics

Humans have performance thresholds that are physically impossible to exceed without the intervention of cognitive computing. Computer vision technology is a software and not a camera hardware solution.  Video Analytics and Visual Recognition software can operate on video and photos taken from body cameras, cell phones, and fixed cameras of many types.  However, quality video output plays a major role in quality computer vision. 

 

Human ability has limits

Studies have proven that a single officer watching two videos over a 22-minute span can lose up to 95% of optimized focus due to fatigue.  It is physically impossible to sustain a high level of attention span for long periods of time. Video analytics provides a technical assistant that dramatically improves the effectiveness of the surveillance operator.

 

Enhance rather than replace

Video Analytics is able to observe, tag, track, and search objects, people, and vehicles.  Each object, person, and vehicle is observed and classified by color, size, speed, features, and trajectory. Video analytics is also able to count people, identify motion trends, anomalies, and facial recognition. More importantly, it is able to create alerts that notify surveillance officers in real-time to deploy solutions in a quicker response time.  

 

Visual learning models improve accuracy

Applying teaching models over a very short amount of time improves the computing classification to optimized levels of required accuracy.  You will not need a Ph.D. in Computer Science Programming to manage your visual learning models. You may be pleasantly surprised to see the low code application is user-friendly.  Visual learning models can be deployed over just a few short days to see noticeable differences in recognition.

 

 

 

 

 

Use cases include, but are not limited to the following:
  • Security surveillance
  • Law enforcement & Military
  • Agriculture & Indoor Grow Facilities
  • Medical
  • Retail
  • Manufacturing
  • Energy & Utilities
  • School Campus & Universities
  • Sporting Arenas
Picture3
Picture1
Picture6