PureTech’s patented location and AI based intrusion detection software automates wide area and perimeter security proteciton for the most critical applications, where high probability of detection and low nuisance alarm rates are mandatory. Market Solutions PureTech’s patented location and AI based intrusion detection software automates wide area and perimeter protection for the most critical applications, where high probability of detection and low nuisance alarm rates are mandatory. PureTech Systems Inc. PureActiv® Software Solution provides a scalable layered automated detection, classification, tracking, and deterrent solution for protecting any critical infrastructures, facilities, and borders. PureActiv can dramatically increase surveillance effectiveness and efficiencies by acting as a force multiplier. PureActiv’s open architecture software solution easily integrates with numerous perimeter and field sensors and systems to enable your organization to quickly leverage as much or as little of PureTech’s capabilities as needed. PureActiv can automate the entire process from intrusion detection, auto-verification, PTZ Auto-follow, and invoking non-lethal acoustic deterrents. PureActiv AI Video analytics are in wide use today automating the protection of borders, numerous types of transportation infrastructures (Airports, Seaports, Transit agencies, Parking management), Man overboard, Military bases, and utilities, among others. Since PureTech’s PureActiv geospatial AI Video Analytics boast the longest detection ranges in the market (over 5 miles) , camera counts are minimized, which dramatically lowers infrastructure costs including fewer cameras, recording, poles, power and network infrastructure, and installation labor. Pure Tech Markets Skeptical? Put us to the test. If your company has an initiative to secure critical infrastructure or country borders, we’d be happy to engage with you to prove our performance including comparison against any competitor in the market. We are that confident in our performance. Learn More Request a Live Web Demo Geospatial AI-Boosted Video Analytics PureActiv’s decades of time-tested video analytics now incorporate automated verification capability using AI Deep Learning Neural nets, which adds a second high-fidelity object classification into an already robust video analytics pipeline. The net effect is even lower nuisance alarm levels with very high probability of detection. Automated Detect, Classify & Deter Auto-Detect and Geolocate Intrusions PureTech Geospatial Video Analytics, Fence Sensors, Radars, Ground, and Other sensors. Auto Classify Intruders Autonomously steer cameras, dispatch drones to gelocation of intrusion. Using ML, classify target as a threat/non-threat. Auto Alert Personnel & Keep Track of Intruders Issue alarms - Mobile devices, Workstations. Autonomously follow intruders. Keep airborne and ground intruders in camera view. Auto Deter & Defeat Blast intruders with deterrent sounds and strobe lights. Jame drone communications and GPS. Long Range Detection (>5 miles) Pure Activ ® AlertView Command and Control Interface PureActiv geospatial AI-boosted video analytics can be integrated into third party NVR and PSIM systems or used with PureActiv AlertView real-time video and sensor management command and control client provide a complete end-to-end software solution. PureActiv AI-boosted video analytics can be deployed on edge-devices and enterprise servers. Testimonials “I have relied heavily on PureActiv and thrived off its benefits. PureActiv is a gamechanger , no doubt about it”. - FEDERAL GOVERNMENT USER PureActiv® AI-Boosted Video Analytics Integrated into 3rd Party VMS and/or PSIM PureActiv® AI-Boosted Video Analytics Integrated into PureActiv® AlertView Command and Control Many times, the right detection solution is not a single sensor technology. In addition to its geospatial AI-boosted video analytics, PureActiv integrates with other perimeter sensors such as radar, buried or on-fence fiber detection systems, and ground sensors, either singularly or combining them, to provide just the right level of protection your perimeter requirements. PureActiv® Open System, Agile Real-time Detection and Response Command and Control

our platforms, along with the expertise PureTech Systems, including Machine/Deep Learning, enables us to readily address not only new and novel safety and security surveillance applications but also many other image-based pattern recognition and computer vision application. Other Market Solutions The flexibility and adaptability of the our platforms, along with the expertise PureTech Systems has in image processing, including Machine/Deep Learning, enables us to readily address not only new and novel safety and security surveillance applications but also many other image-based pattern recognition and computer vision applications. PureTech Systems focuses on pursuing and providing solutions for markets both domestically and internationally. In addition to providing security solutions for Borders , Military Bases , Airports, Seaports , Mass Transit and Utilities , PureTech Sytems can be applied to following markets: Vehicle Counting PureTech Systems Inc. delivers vehicle counting solutions that utilize advanced AI-boosted video analytics to accurately detect and track vehicles in real-time by turning cameras into intelligent counting sensors. Learn More Smart Cities By leveraging our expertise in geospatial video analytics we empower cities with real-time monitoring, intelligent decision-making, and enhanced safety. Learn More Man Overboard Detection Our solution utilizes our award-winning geospatial AI-Boosted video analytics to deliver a proactive detection system that accurately detects and alerts of man overboard events enhancing the safety of passengers and crew members aboard cruise ships. Learn More Intelligent Traffic System(ITS) Apply our advanced technologies and industry expertise to enables cities and transportation agencies to make data-driven decisions, optimize traffic flow, and improve overall transportation efficiency, resulting in smoother commutes and reduced congestion. Learn More

< Back PureTech Announces Two New Faces to the Team Aug 4, 2020 Please join us in giving a warm welcome to the two newest team members to the PureTech team. PureTech is proud to announce and welcome Rick Cottle as the Company's Operations Manager. Rick will provide strong leadership and oversight of the customer support and system deployment teams as PureTech continues to expand into new markets. Prior to joining the PureTech Team, he has worked in the semiconductor industry, space systems avionics, and commercial aerospace software, as an engineer, manufacturing manager and software development manager. He has a BS in Electrical Engineering and MS in Computer Information Systems, and has taught undergraduate and graduate level courses in project management and information systems. PureTech is also excited to welcome Shannon Staples as the Company's Marketing Manager. Shannon will be developing and implementing an overall corporate marketing strategy and translating the company’s business objectives into marketing strategies. Before joining PureTech, Shannon served as the Executive Director of the Border Patrol Foundation, a non-profit that provides support to the families of fallen Border Patrol Agents. Prior to that, she served as a Border Patrol Agent on the Arizona border working collaterally as an agent and a Public Affairs Specialist at the Tucson Sector Headquarters. "We are excited to welcome Rick and Shannon to the PureTech Team" said Larry Bowe, PureTech President. "Both will bring an increased focus to the quality of our customer relationships and help build on the success and strength of an already great team." Previous Next

< Back PureTech Announces Successful Integration and Multiple Deployments with OptaSense Sep 29, 2020 PHOENIX, Ariz. – PureTech Systems announced today the successful integration and deployments of PureActiv® software with OptaSense Linear Ground Detection Systems (LGDS). The LGDS provides perimeter intrusion detection using advanced distributed acoustic sensing technology capable of detecting, classifying and locating multiple threats in real time. When coupled with PureTech’s AI Deep Learning based software, the PureActiv® system verifies the detected threats as person, vehicle, animal, or other. This enables the integrated system with machine to machine validation to automatically verify real threats and dramatically reduce nuisance alarms and the time spent reacting to them. Working in cooperation, OptaSense and PureTech have deployed perimeter protection at several high security power generation plants in the United States with plans to install additional sites by the end of 2020. OptaSense currently has over 20,000 miles of systems operational around the world, including on the U.S. southern border and international borders. “PureActiv® provides the highest probability of detection with the lowest nuisance alarm rates by forming a nearly impenetrable virtual wall. Successfully integrated into a multitude of camera, radar, and LDGS systems, our software provides cutting-edge perimeter security around the most critical facilities, infrastructure, and country borders ” said PureTech Systems President Larry Bowe. “ PureTech Systems software outperforms other analytics in detection range and accuracy of detection and classification, allowing security personnel to more efficiently and effectively identify threats.” OptaSense’s Linear Ground Detection System is a real-time awareness edge device that detects and assesses activities and behaviors for perimeters, borders and infrastructure monitoring. The solution “listens” and “feels” acoustic and seismic energies near a fiber-optic cable, then, through advanced algorithms, alerts and classifies on human activities through software, via a user interface or integrated C2 platforms. “OptaSense leads the fiber-optic sensing industry in deployed miles, spanning multiple verticals and multiple applications. We have an open architecture that integrates with other layers of security and common operating pictures like PureActiv®, with our system often providing an early warning and tip and cue. We’re proud to integrate with industry leading technologies on critical national infrastructure projects,” said Jeff Williamson, OptaSense Managing Director. About PureTech Systems® PureTech Systems Inc. is a privately owned company established in 2004 that develops, markets, and supports patented location-based AI video analytics software, PureActiv©, for real time safety and security applications. The company’s software improves situational awareness with AI video analytics, sensor integration and information fusing for automated real-time event detection and forensic video content analysis with primary emphasis on automated intrusion detection and camera tracking for country borders and coastlines, facility perimeters and critical infrastructures (pipelines, railroads, dams, bridges, ports, utilities, power plants, military bases, and airports). To find out more about PureTech Systems Inc. visit our website at www.puretechsystems.com or contact at 602-424-9842 or info@PureTechSystems.com . About OptaSense OptaSense, founded in 2007, provides intelligent business value through the deployment of distributed fiber-optic sensing in a wide variety of application areas serving a global market. A wholly-owned subsidiary of the QinetiQ Group, OptaSense is the trusted partner for leading edge Distributed Fiber-Optic Sensing (DFOS) solutions that reduce the cost of asset ownership by optimizing operational efficiency, performance and safety. Our solutions provide real-time, actionable data, dedicated expertise and global experience to multiple industries, including oil and gas, pipeline, security, transport and utilities. Operating in over 50 countries with more than 20,000 miles of assets under contract, we are monitoring and protecting some of the world’s most valuable assets. Learn more at www.OptaSense.com . Previous Next

< Back PureTech Announces Successful Deployments with Siklu Dec 3, 2020 PHOENIX, Ariz. -- PureTech Systems announced today the successful deployments of PureActiv® software with the Siklu gigabit wireless network. “The PureTech real-time AI video processing system relies on receiving low latency, quality, error free video streams. The Siklu mmWave systems are the perfect complement as they provide gigabit capacities and highly flexible, rapid deployment of a network connecting cameras where ever they may be to the PureTech solution”. For security reasons the client cannot be discussed however, the combined solution of the Siklu network with the PureActiv software ensures the highest probability of detection at these U.S. based power generation plants. Additionally, the PureActiv software monitors the quality of the network and if a network link goes down, the PureActiv software immediately notifies the user in the event the loss is due to a perimeter attack. TSG Solutions Inc., a Continental Mapping Company, is the integrator that architected and deployed the perimeter protection solution. “As the wireless network for the security systems at Nuclear Power plants, this means you have to be rock solid and immune to jamming with down times measured in seconds per year,” Said Alex Doorduyn, VP and GM of Americas for Siklu. “ Not only does the network have to be always on and available, it has to have the gigabit capacities new cameras and sensors demand. The images and data that form the input for PureTech advanced algorithms have to be as they come from the camera without network latency or artifacts.” “PureActiv® provides a fully integrated automated perimeter security system for the most critical facilities and infrastructure ” said PureTech Systems President Larry Bowe. “PureTech Systems software outperforms other analytics in detection range and accuracy of detection and classification, allowing security personnel to more efficiently and effectively identify threats.” About PureTech Systems® PureTech Systems Inc. is a privately owned company established in 2004 that develops, markets, and supports patented location-based AI video analytics software, PureActiv©, for real time safety and security applications. The company’s software improves situational awareness with AI video analytics, sensor integration and information fusing for automated real-time event detection and forensic video content analysis with primary emphasis on automated intrusion detection and camera tracking for country borders and coastlines, facility perimeters and critical infrastructures (pipelines, railroads, dams, bridges, ports, utilities, power plants, military bases, and airports). To find out more about PureTech Systems Inc. visit our website at www.puretechsystems.com , call 602-424-9842 or info@PureTechSystems.com . About Siklu Siklu delivers multi-gigabit wireless fiber connectivity in urban, suburban and rural areas. Operating in the millimeter wave bands, Siklu’s wireless solutions are used by leading service providers and system integrators to provide 5G Gigabit Wireless Access services. In addition, Siklu solutions are ideal for Smart City projects requiring extra capacity such as video security, WiFi backhaul and municipal network connectivity all over one network. Thousands of carrier-grade systems are delivering interference-free performance worldwide. Easily installed on street-fixtures or rooftops, these radios have been proven to be the ideal solution for networks requiring fast and simple deployment of secure, wireless fiber. www.siklu.com . Previous Next

In his current capacity as VP Federal Sector, Ilia Rosenberg manages all US Federal and International Government pursuits in the area of national security. Leveraging his extensive experience, Ilia drives all... < Back Ilia Rosenberg VP Federal Sector In his current capacity as VP Federal Sector, Ilia Rosenberg manages all US Federal and International Government pursuits in the area of national security. Leveraging his extensive experience, Ilia drives all aspects of growth for technology and services for PureTech Systems. In his prior role as Managing Director for Blue Ocean Advisory Group and ISR GeoSensing Mr. Rosenberg was responsible for managing programs related to national defense, security, as well as heading their Artificial Intelligence and Machine Learning practices. In 2016-2019 he was VP of Programs at a Silicon Valley AI start-up where he was leading technical developments and commercial negotiations with a variety of government and private customers and successfully brought AI technology to DoD and the intelligence community. In 2011-2015 he served as Chief Technologist for AGT International Global Delivery Organization in its Headquarters in Zurich, Switzerland leading the worldwide engineering team to assure effective development and transition of new products and creative solutions for Smart and Safe Cities and Smart Borders programs. Before joining AGT International in 2011, Ilia served as the Director of Technology Assessment for the Boeing Company Security Solutions division in Arlington, VA. Before joining Boeing he worked in the field of atmospherical physics and completed his postdoctoral studies at the Department of Chemical Engineering at The University of Texas at Austin. He published extensively in the peer review journals and his projects were covered by the New York Times, Time Magazine, San Francisco Chronicle, New Scientist, The BBC, Discovery Channel, and Washington Technology. Mail Document

< Back PureTech Systems Announces New Design Calculator Tool Mar 16, 2021 PHOENIX, Ariz. – PureTech Systems announced today the successful release of their proprietary Security Design Imaging Calculator. The purpose of the PTS Design Tool is to provide security planning personnel with the ability to ascertain the distance and pixels needed to view, detect, and classify a specified target to properly secure their perimeters. A common obstacle in designing security systems for wide-area perimeters is determining the location and number of cameras, as well as the camera pixels required for accurately detecting threats. A handful of parameters such as those listed below, once confirmed or modified by the user, help lead to the desired result: Width & height of image sensor, Number of lateral & vertical camera pixels, Horizontal Field of View (HFOV) of the camera or alternatively the focal length of its optics, Camera installation height, Parameters through which the target of interest and tallest expected target is specified namely the heights and widths of encompassing bounding boxes and the ranges from the camera to the targets (of interest and tallest). Corresponding focal length to the input HFOV or vice-versa when the input is focal length, Vertical field-of-view (VFOV), Angle subtending the blind zone, and Plus several more. PureTech Systems is positioned as a leader in the location (geospatial) and Machine Learning / Deep Learning —based AI Video Analytics market with a current focus on applications involving security of critical infrastructure and facilities and false alarm reduction for Video Central Monitoring Stations where it offers its AI Video Analytics as a Service. Development and patent protection of technology is a core philosophy and adding value through sensor collaboration has been centric to outperforming competitors. The company strives to provide solutions to problems, which results in long-term relationships from its customer base. To maintain its technological innovation and advancement, it continuously invests in research and development. While its technology is applicable to many markets such as driverless cars, robotic manufacturing and inspection and medical image detection, PureTech Systems focuses on pursuing and providing automated wide-area and perimeter protection. Access to the PureTech Design Tool can be requested through the PureTech Systems' website . Following the use of the calculator, experts in security design are available to assist potential clients with market-leading perimeter security solutions for facilities, infrastructure and borders. About PureTech Systems® PureTech Systems Inc. is a privately owned company established in 2004 that develops, markets, and supports patented location-based AI video analytics software, PureActiv©, for real time safety and security applications. The company’s software improves situational awareness with AI video analytics, sensor integration and information fusing for automated real-time event detection and forensic video content analysis with primary emphasis on automated intrusion detection and camera tracking for borders and coastlines, facility perimeters and critical infrastructures (pipelines, railroads, dams, bridges, ports, utilities, power plants, military bases, and airports). To find out more about PureTech Systems Inc. visit our website at www.puretechsystems.com , call 602-424-9842 or email info@PureTechSystems.com . Previous Next

< Back Surpassing Legacy Standards Jan 12, 2026 1. Introduction Some surveillance vendors claim they can automatically detect and classify humans and vehicles at extreme distances—sometimes 1 to 6 miles —using only a few pixels per target. In some cases, claims as low as 2–100 pixels are made. These claims are often justified by citing DRI (Detection, Recognition, Identification) or DORI tables, which were created decades ago for human observers , not autonomous computer-vision systems. At the same time, some vendors attempt to avoid AI and machine learning altogether, relying instead on simple motion detection, thresholding, or rule-based analytics, while still claiming “automatic detection.” Both approaches— misusing DRI/DORI or avoiding ML entirely —lead to systems that fail in real-world deployments. These claims conflict with the physics of imaging , the limitations of sensors , and the fundamental requirements of modern machine-learning (ML) —especially in environments where atmospheric turbulence, reduced contrast, camera shake, background complexity, partial occlusions, animals, and environmental motion are common. This white paper explains why DRI and DORI apply only to human perception, why they cannot be used to predict autonomous classification performance, why ML systems require substantially more pixels on target , and why systems that do not use AI/ML suffer from unacceptably high false-alarm rates . It also explains why long-range conditions require even more margin, and why no company can bypass physics with software. Any extraordinary claim must be validated through a Proof of Concept (POC) . 2. What DRI and DORI Actually Measure 2.1 DRI (Detection, Recognition, Identification) DRI was developed in 1958 to estimate how far a human observer could visually interpret a target using optical or thermal equipment. It describes whether a person can detect that “something is there,” recognize a general category (such as human versus vehicle), or identify a specific type. Humans can often recognize a person with very limited visual information—on the order of 12–16 vertical pixels , which might correspond to roughly: 12 pixels high × ~4 pixels wide ≈ ~48 total pixels , or 16 pixels high × ~5 pixels wide ≈ ~80 total pixels This is possible because the human brain can infer missing detail, guess intent, and apply context. DRI was never designed to evaluate autonomous systems. 2.2 DORI (IEC 62676-4:2015) DORI extends similar ideas to CCTV system design and again describes what a human operator can interpret when viewing video. Recognition-level DORI values often correspond to 7–12 pixels across the target width , still assuming a human is making the judgment. Neither DRI nor DORI evaluates whether a computer can autonomously classify a target, nor do they account for turbulence, camera shake, background complexity, camouflage, or occlusions. 3. Why DRI/DORI Cannot Be Applied to Machine Learning Machine-learning systems such as Convolutional Neural Networks (CNNs) and Transformers classify objects by extracting visual features from the image, including shape, edges, texture gradients, motion consistency, and frame-to-frame stability. If these features do not physically exist in the pixels, the ML system cannot classify the object. For example, a person appearing 10 pixels high × ~3 pixels wide ≈ ~30 total pixels does not contain enough information to reliably determine head shape, limb movement, torso structure, or vehicle geometry. A human observer might guess; an algorithm cannot. DRI/DORI recognition thresholds describe what humans can guess from incomplete data. ML systems require real, measurable information. 4. What Happens If You Do NOT Use AI / Machine Learning It is equally important to understand the consequences of not using AI/ML at all . Systems that rely solely on traditional video analytics—such as simple motion detection, pixel change thresholds, background subtraction, or rule-based logic—lack the ability to understand what is moving. They can detect motion, but they cannot reliably classify it. As a result, non-AI systems typically suffer from: Extremely high false-alarm rates Inability to distinguish humans from animals Inability to reject nuisance motion Poor scalability to large or complex environments 4.1 Why Non-AI Systems Generate Excessive False Alarms Without ML classification, a system must alarm on any motion that meets basic criteria. This includes: Animals Blowing vegetation Shadows Clouds and moving sun patterns Heat shimmer and atmospheric turbulence Camera shake Insects and birds Rain, snow, and dust Rule-based filters can reduce some noise, but they quickly break down in real environments because natural motion is highly variable. As thresholds are tightened to reduce false alarms, real threats are missed. As thresholds are loosened to avoid misses, false alarms explode. This tradeoff cannot be solved without classification. 4.2 Non-AI Systems Cannot Scale As coverage areas grow larger or more complex, non-AI systems become unmanageable: Operators are overwhelmed by alarms Alarm fatigue sets in Systems are ignored or turned down Real threats are lost in noise In practice, many non-AI deployments are eventually disabled or relegated to “monitoring only” because they generate too many alarms to be useful. 4.3 Detection Without Classification Is Operationally Dangerous A system that “detects motion” but cannot determine whether the object is a human, vehicle, animal, or irrelevant noise is not an autonomous security system. It simply shifts the burden to the operator, increasing workload and increasing the chance of human error. This is why modern perimeter security requires both detection and classification , and why AI/ML—used correctly and within physical limits—is essential. 5. Long-Range Physics Further Increase ML Requirements (1–6 Miles) At long ranges, multiple physical effects degrade imagery beyond what DRI/DORI assume: Reduced contrast Background complexity Atmospheric turbulence Camera shake Loss of gradients PureTech mitigates camera-induced motion by performing its proprietary image stabilization as the first processing step , ensuring downstream analytics operate on a stable image. Even so, long-range ML classification requires more pixels , not fewer. 6. Occlusions: Why Real-World Systems Must Design for More Pixel Margin Real environments include frequent occlusions caused by vegetation, terrain, infrastructure, and partial self-occlusion. When only part of a target is visible, the effective usable pixel count drops sharply. For example: 40 px high × 15 px wide ≈ 600 total pixels may be sufficient for a fully visible person. Seeing only half the body may require significantly more total pixels to maintain classification confidence. Designing only to ideal conditions guarantees failure. 7. Independent Evidence: Pixel Requirements for Reliable Autonomous Classification Independent research and industry experience consistently show that reliable autonomous classification cannot be achieved with only a handful of pixels , regardless of algorithm choice or marketing claims. In practical deployments, autonomous classification systems must achieve high probability of correct classification , low false-alarm rates , and low misclassification rates simultaneously. Achieving all three requires substantial spatial and temporal information about the target. Across a wide range of studies and real-world deployments, several consistent observations emerge: Very small targets (on the order of only tens of total pixels) do not contain sufficient structure for reliable autonomous classification. As pixel counts increase into the hundreds of total pixels , classification accuracy improves substantially, particularly when combined with temporal information such as motion consistency. At long ranges, additional factors—including atmospheric turbulence, reduced contrast, background complexity, and partial occlusions—further reduce usable information, increasing the amount of image data (pixels) required to maintain high accuracy. Importantly, there is no single universal pixel threshold that guarantees reliable classification at long range. The effective pixel requirement depends on multiple factors, including sensor modality, environmental conditions, target contrast, degree of occlusion, and system architecture. Systems that rely primarily on static image appearance and single-frame analysis tend to require significantly larger target images (in the thousands) to achieve acceptable performance under degraded conditions. More advanced systems that exploit stabilized imagery, coherent motion over time, and real-world constraints can extract more information from the same imagery—but no credible system can achieve reliable autonomous classification at DRI/DORI recognition levels or at 2 to 10s of total pixels . For long-range applications, it is realistic to expect that classification accuracy improves as available target information (pixel count) grows from a few tens of pixels into the hundreds or more , depending on conditions. Claims of reliable classification far below this regime are not supported by physics, industry experience, or independent research. 8. Training Data: Why Good ML Requires Large, Clean, Real-World Datasets ML performance depends heavily on training data quality and quantity. Modern vision models typically require hundreds of thousands to millions of representative examples. PureTech has been training visible and thermal ML models for 8 years , using hundreds of thousands of real-world images collected under operational conditions. Garbage In, Garbage Out Poor training data leads directly to poor performance. Garbage data includes: Targets that are too small Unrealistic close-ups never seen in deployment Low-contrast imagery Partial fragments without sufficient structure Severe blur or turbulence distortion Incorrect or inconsistent labeling PureTech applies proprietary preprocessing and quality controls to prevent such data from contaminating training. 9. Thermal vs. Visible Imaging Thermal imaging often outperforms visible cameras at long range and at night because it measures emitted heat rather than reflected light. Advantages include better target-background separation, no need for lighting, reduced impact from shadows, and reduced effectiveness of visual camouflage. Thermal does not eliminate physics limits, but it improves signal quality under difficult conditions. 10. MWIR, LWIR, and SWIR Overview LWIR (8–14 µm): uncooled, durable, good short- to medium-range performance MWIR (3–5 µm): superior long-range performance, higher contrast, requires cooling SWIR (~1–2 µm): reflected-light imaging, good detail in low light, poor in fog or total darkness Each has tradeoffs; none can violate physics. 11. PureTech’s Physics-Aligned Multi-Cue Approach PureTech Systems combines: Image stabilization (first step) Terrain-mapped object tracking for real-world size, speed, and direction Motion consistency filtering Shape plausibility checks Speed profiling Contextual and trajectory filtering ML classification applied PureTech holds 16 issued patents covering image processing, stabilization, and computer vision. 12. Why This Matters: Missed Detections, False Alarms, and ROI A missed detection can mean loss of life, loss of critical infrastructure, regulatory penalties, lawsuits, and reputational damage. False alarms waste time, consume resources, cause alarm fatigue, and obscure real threats. Excessive false alarms are functionally equivalent to missed detections because operators stop responding appropriately. 1. Introduction Some surveillance vendors claim they can automatically detect and classify humans and vehicles at extreme distances—sometimes 1 to 6 miles —using only a few pixels per target. In some cases, claims as low as 2–100 pixels are made. These claims are often justified by citing DRI (Detection, Recognition, Identification) or DORI tables, which were created decades ago for human observers , not autonomous computer-vision systems. At the same time, some vendors attempt to avoid AI and machine learning altogether, relying instead on simple motion detection, thresholding, or rule-based analytics, while still claiming “automatic detection.” Both approaches— misusing DRI/DORI or avoiding ML entirely —lead to systems that fail in real-world deployments. These claims conflict with the physics of imaging , the limitations of sensors , and the fundamental requirements of modern machine-learning (ML) —especially in environments where atmospheric turbulence, reduced contrast, camera shake, background complexity, partial occlusions, animals, and environmental motion are common. This white paper explains why DRI and DORI apply only to human perception, why they cannot be used to predict autonomous classification performance, why ML systems require substantially more pixels on target , and why systems that do not use AI/ML suffer from unacceptably high false-alarm rates . It also explains why long-range conditions require even more margin, and why no company can bypass physics with software. Any extraordinary claim must be validated through a Proof of Concept (POC) . 2. What DRI and DORI Actually Measure 2.1 DRI (Detection, Recognition, Identification) DRI was developed in 1958 to estimate how far a human observer could visually interpret a target using optical or thermal equipment. It describes whether a person can detect that “something is there,” recognize a general category (such as human versus vehicle), or identify a specific type. Humans can often recognize an object is a person with very limited visual information—on the order of 12–16 vertical pixels , which might correspond to roughly: 12 pixels high × ~4 pixels wide ≈ ~48 total pixels , or 16 pixels high × ~5 pixels wide ≈ ~80 total pixels This is possible because the human brain can infer missing detail, guess intent, and apply context. DRI was never designed to evaluate autonomous systems. 2.2 DORI (IEC 62676-4:2015) DORI extends similar ideas to CCTV system design and again describes what a human operator can interpret when viewing video. Recognition-level DORI values often correspond to 7–12 pixels across the target width , still assuming a human is making the judgment. Neither DRI nor DORI evaluates whether a computer can autonomously classify a target, nor do they account for turbulence, camera shake, background complexity, camouflage, or occlusions. 3. Why DRI/DORI Cannot Be Applied to Machine Learning Machine-learning systems such as Convolutional Neural Networks (CNNs) and Transformers classify objects by extracting visual features from the image, including shape, edges, texture gradients, motion consistency, and frame-to-frame stability. If these features do not physically exist in the pixels, the ML system cannot classify the object. For example, a person appearing 10 pixels high × ~3 pixels wide ≈ ~30 total pixels does not contain enough information to reliably determine head shape, limb movement, torso structure, or vehicle geometry. A human observer might guess; an algorithm cannot. DRI/DORI recognition thresholds describe what humans can guess from incomplete data. ML systems require real, measurable information. 4. What Happens If You Do NOT Use AI / Machine Learning It is equally important to understand the consequences of not using AI/ML at all . Systems that rely solely on traditional video analytics—such as simple motion detection, pixel change thresholds, background subtraction, or rule-based logic—lack the ability to understand what is moving. They can detect motion, but they cannot reliably classify it. As a result, non-AI systems typically suffer from: Extremely high false-alarm rates Inability to distinguish humans from animals Inability to reject nuisance motion Poor scalability to large or complex environments 4.1 Why Non-AI Systems Generate Excessive False Alarms Without ML classification, a system must alarm on any motion that meets basic criteria. This includes: Animals Blowing vegetation Shadows Clouds and moving sun patterns Heat shimmer and atmospheric turbulence Camera shake Insects and birds Rain, snow, and dust Rule-based filters can reduce some noise, but they quickly break down in real environments because natural motion is highly variable. As thresholds are tightened to reduce false alarms, real threats are missed. As thresholds are loosened to avoid misses, false alarms explode. This tradeoff cannot be solved without classification. 4.2 Non-AI Systems Cannot Scale As coverage areas grow larger or more complex, non-AI systems become unmanageable: Operators are overwhelmed by alarms Alarm fatigue sets in Systems are ignored or turned down Real threats are lost in noise In practice, many non-AI deployments are eventually disabled or relegated to “monitoring only” because they generate too many alarms to be useful. 4.3 Detection Without Classification Is Operationally Dangerous A system that “detects motion” but cannot determine whether the object is a human, vehicle, animal, or irrelevant noise is not an autonomous security system. It simply shifts the burden to the operator, increasing workload and increasing the chance of human error. This is why modern perimeter security requires both detection and classification , and why AI/ML—used correctly and within physical limits—is essential. 5. Long-Range Physics Further Increase ML Requirements (1–6 Miles) At long ranges, multiple physical effects degrade imagery beyond what DRI/DORI assume: Reduced contrast Background complexity Atmospheric turbulence Camera shake Loss of gradients PureTech mitigates camera-induced motion by performing its proprietary image stabilization as the first processing step , ensuring downstream analytics operate on a stable image. Even so, long-range ML classification requires more pixels , not fewer. 6. Occlusions: Why Real-World Systems Must Design for More Pixel Margin Real environments include frequent occlusions caused by vegetation, terrain, infrastructure, and partial self-occlusion. When only part of a target is visible, the effective usable pixel count drops sharply. For example: 40 px high × 15 px wide ≈ 600 total pixels may be sufficient for a fully visible person. Seeing only half the body may require significantly more total pixels to maintain classification confidence. Designing only to ideal conditions guarantees failure. 7. Independent Evidence: Pixel Requirements for Reliable Autonomous Classification Independent research and industry experience consistently show that reliable autonomous classification cannot be achieved with only a handful of pixels , regardless of algorithm choice or marketing claims. In practical deployments, autonomous classification systems must achieve high probability of correct classification , low false-alarm rates , and low misclassification rates simultaneously. Achieving all three requires substantial spatial and temporal information about the target. Across a wide range of studies and real-world deployments, several consistent observations emerge: Very small targets (on the order of only tens of total pixels) do not contain sufficient structure for reliable autonomous classification. As pixel counts increase into the hundreds of total pixels , classification accuracy improves substantially, particularly when combined with temporal information such as motion consistency. At long ranges, additional factors—including atmospheric turbulence, reduced contrast, background complexity, and partial occlusions—further reduce usable information, increasing the amount of image data (pixels) required to maintain high accuracy. Importantly, there is no single universal pixel threshold that guarantees reliable classification at long range. The effective pixel requirement depends on multiple factors, including sensor modality, environmental conditions, target contrast, degree of occlusion, and system architecture. Systems that rely primarily on static image appearance and single-frame analysis tend to require significantly larger target images (in the thousands) to achieve acceptable performance under degraded conditions. More advanced systems that exploit stabilized imagery, coherent motion over time, and real-world constraints can extract more information from the same imagery—but no credible system can achieve reliable autonomous classification at DRI/DORI recognition levels or at 2 to 10s of total pixels . For long-range applications, it is realistic to expect that classification accuracy improves as available target information (pixel count) grows from a few tens of pixels into the hundreds or more , depending on conditions. Claims of reliable classification far below this regime are not supported by physics, industry experience, or independent research. 8. Training Data: Why Good ML Requires Large, Clean, Real-World Datasets ML performance depends heavily on training data quality and quantity. Modern vision models typically require hundreds of thousands to millions of representative examples. PureTech has been training visible and thermal ML models for 8 years , using hundreds of thousands of real-world images collected under operational conditions. Garbage In, Garbage Out Poor training data leads directly to poor performance. Garbage data includes: Targets that are too small Unrealistic close-ups never seen in deployment Low-contrast imagery Partial fragments without sufficient structure Severe blur or turbulence distortion Incorrect or inconsistent labeling PureTech applies proprietary preprocessing and quality controls to prevent such data from contaminating training. 9. Thermal vs. Visible Imaging Thermal imaging often outperforms visible cameras at long range and at night because it measures emitted heat rather than reflected light. Advantages include better target-background separation, no need for lighting, reduced impact from shadows, and reduced effectiveness of visual camouflage. Thermal does not eliminate physics limits, but it improves signal quality under difficult conditions. 10. MWIR, LWIR, and SWIR Overview LWIR (8–14 µm): uncooled, durable, good short- to medium-range performance MWIR (3–5 µm): superior long-range performance, higher contrast, requires cooling SWIR (~1–2 µm): reflected-light imaging, good detail in low light, poor in fog or total darkness Each has tradeoffs; none can violate physics. 11. PureTech’s Physics-Aligned Multi-Cue Approach PureTech Systems combines: Image stabilization (first step) Terrain-mapped object tracking for real-world size, speed, and direction Motion consistency filtering Shape plausibility checks Speed profiling Contextual and trajectory filtering ML classification applied PureTech holds 16 issued patents covering image processing, stabilization, and computer vision. 12. Why This Matters: Missed Detections, False Alarms, and ROI A missed detection can mean loss of life, loss of critical infrastructure, regulatory penalties, lawsuits, and reputational damage. False alarms waste time, consume resources, cause alarm fatigue, and obscure real threats. Excessive false alarms are functionally equivalent to missed detections because operators stop responding appropriately. Organizations that choose systems based solely on lowest acquisition cost often incur far higher total cost of ownership and risk exposure. Investing upfront in systems designed around physics, robust ML, stabilization, terrain mapping, and multi-cue validation delivers far better ROI by avoiding catastrophic failures and operational collapse. 13. Proof of Concept: The Only Valid Verification Any vendor claiming autonomous classification at DRI/DORI pixel levels, at less than several hundred total pixels , especially under extreme long-range and occluded conditions must demonstrate the claim in a Proof of Concept . Physics always wins. 14. Conclusion DRI and DORI describe what humans can infer. They do not describe what autonomous systems require. Systems that ignore ML generate unacceptable false alarms. Systems that misuse ML or ignore physics miss real threats. PureTech Systems delivers reliable autonomous detection and classification by respecting physical reality, using stabilized imagery, terrain-mapped measurements, multi-cue analytics, disciplined ML training, and patented computer-vision technology—producing operational security systems that actually work as demonstrated by its real-world deployments in the most challenging environments such as country borders. Organizations that choose systems based solely on lowest acquisition cost often incur far higher total cost of ownership and risk exposure. Investing upfront in systems designed around physics, robust ML, stabilization, terrain mapping, and multi-cue validation delivers far better ROI by avoiding catastrophic failures and operational collapse. 13. Proof of Concept: The Only Valid Verification Any vendor claiming autonomous classification at DRI/DORI pixel levels, at less than several hundred total pixels , especially under extreme long-range and occluded conditions must demonstrate the claim in a Proof of Concept . Physics always wins. 14. Conclusion DRI and DORI describe what humans can infer. They do not describe what autonomous systems require. Systems that ignore ML generate unacceptable false alarms. Systems that misuse ML or ignore physics miss real threats. PureTech Systems delivers reliable autonomous detection and classification by respecting physical reality, using stabilized imagery, terrain-mapped measurements, multi-cue analytics, disciplined ML training, and patented computer-vision technology—producing operational security systems that actually work as demonstrated by its real-world deployments in the most challenging environments such as country borders. Previous Next

< Back Transitioning Command and Control platforms to C4ISR Systems. Jun 14, 2022 An article written by Ilia Rosenberg, VP of the Federal Sector was picked up by GIT-Security EMEA , on the topic of transitioning command and control platforms to C4ISR systems. Ilia goes into detail on the lessons learned, as well as a general view and key features of an autonomous C4ISR. Until recently creating a new enterprise C4ISR system, be it Army, Navy, Air Force, or Department of Homeland Security, utilized the same deterministic systems engineering approach. In the last several years the use of AI and Machine Learning (ML) in surveillance systems has become widespread, often with good success, but sometimes with mixed results. As PureTech have observed in recorded surveillance photos and videos, adversarial methods and techniques are sometimes able to defeat the standard AI/ML trained neural networks for example by altering target shapes or adding layers of clothing to humans to become low observables. It is no surprise that to address the standard “Detect-Track-Classify” problem a more sophisticated approach extending beyond AI/ML is imperative. The current DoD Digital Strategyguidelines are embedded in PureTech Systems PureActiv software based architecture as well as in individual security designs that were have used successfully in the US and internationally. Working with DOD JAIC (Joint Artificial Intelligence Center [ https://www.ai.mil/about.html ]) since 2016 provided an early understanding of many of these guidelines and techniques. One of JAIC early recommendations was to not only put bounding boxes around detected objects, but also record relevant metadata to deliver more information for the mission and later trend analysis. PureTech were able to implement and test these recommendations at field installations, continually refining the software algorithms including taking advantage of sensor correlation. Lessons Learned By leveraging such sensor and data correlation, PureActiv C4ISR hosted on edge devices and/or servers can receive detections and tracks from multiple sensors such as motion analytics, radars, and fence, ground and subterranean sensors and then autonomously command PTZ cameras and/or dispatch UASs, and UGVs to go “inspect” the detected targets to automatically confirm target types before issuing alarms, i.e. provide auto-verification. This robust approach also combines our geospatial video analytics with AI/DL and radar tracks to disambiguate the objects of interest. This allows the system to eliminate a substantial, up to 95%, amount of nuisance alarms while maintaining a very high detection rate. Another important lesson that PureTech takes to heart is the AI-assurance. It’s important to remind ourselves that AI can do wrong things and it can learn wrong things. That is why PureTech Systems continuously validates the confidence level of the system through robust and continuous ground truth regression testing. The General View and Key Features of An Autonomous C4ISR In this example screenshot, objects of interest are automatically detected, tracked (upper left and lower left side of the screen) and classified by AI/DL as people, conveyance, and UAV and displayed on the digital terrain map with the correct GIS coordinates in real time (upper right side). Alarms are displayed in the lower right with a still image and looping video of the alarm and the pertinent live camera. The decision of target classification is made autonomously without human interaction. In addition to gradual improvement in sensors’ performances, such as cameras and radars ranges and sensitivities, PureTech recently integrated three relatively new capabilities and made them available to DHS and CBP in particular. These included Android Team Awareness Kit (ATAK), small drones (sUAS) detection, and counter-UAS systems. The same C4ISR package shown above fully supports autonomous counter UAS operations after a combination of the air surveillance radars and high-speed, wideband RF scanner for detection and classification of signals in congested spectrum environments were integrated. Since adversaries use sUAS for smuggling illegal drugs over the border, an integrated counter-UAS capability available where such activity is occurring is badly needed. In addition, C4ISR systems that provide ground and UAS tracking including recording track metadata can server as a tool to learn about smuggling corridors and patterns of drug cartels’ operations. According to Tim Bennett the program manager for air domain awareness at the Department of Homeland Security’s Science and Technology Directorate whose job is devoted to detecting, tracking, and countering drones, “Narcodrones a new big problem that we all have to address. It isn’t just DHS. It’s across all agencies in our government and all governments in the world ( https://www.smithsonianmag.com/air-space-magazine/narcodrones-180974934/).” ATAK is quickly becoming a force multiplier for many agencies by enabling cross collaboration between individual agents, teams, and central command. ATAK becomes an extension of the autonomous command and control by communicating data from and to the field including geolocations, media, and tactical information that is immediately actionable and disseminated across one or more areas of responsibility. All information that is available at the C4ISR level (left picture above) can be displayed on ATAK devices (right picture), whether it is installed on a Windows laptops or Android devices. In conclusion, it is worth noting that in every C4ISR autonomous subsystem and operation is designed to aide, not replace human capabilities. Whether used as a battlefield command and control or border security common operational picture, one of the key goals is to augment the warfighter’s or law enforcement agent’s reasoning skills with an autonomous decision-making systems. Ultimately autonomy reduces cognitive load by filtering nuisance alarms and prioritizing information flow for action. This enables agencies to dramatically reduce operational resources while maintaining mission effectiveness. As former Secretary of the Air Force Heather Wilson once said, “The advantage will go to those who create the best technologies and who integrate and field them in creative operational ways that provide military advantages. ( www.acc.af.mil/News/Article-Display/Article/1448216/secaf-this-is-about-lethality-and-mission-effectiveness/)”The Author: Ilia Rosenberg, PureTech Systems, Vice President, Federal Sector This article can also be read at https://www.git-security.com/news/transitioning-command-and-control-platforms-c4isr-systems . Previous Next

< Back New PureTech Patent for Man Overboard Detections Mar 22, 2021 PHOENIX, Ariz. – PureTech Systems today announced the issuance of a patent by the United States Patent Office - " US Patent 10,922,552 , System and Method for Man Overboard Incident Detection". Detecting man overboard events is a common problem for naval ships, cruise ships, and other large vessels. The newly awarded patent for PureTech Systems addresses some of these issues. The timely detection of man overboard events in the cruise industry and other related markets is difficult. The problem is compounded by the fact that when these events occur the timely availability of important data is missing. Accurate confirmation of the event including time of occurrence, location on the ship and location in the sea is critical, but often unavailable for hours following an occurrence, if at all. Fortunately, proactive detection systems such as the Man Overboard Detection System from PureTech Systems, can accurately detect man overboard events and provide immediate, actionable data to response personnel. In addition to providing necessary and life-saving analytics to sea vessels, the newly patented system can also be applied to bridges and other over-water structures. Key features of the system include instant video validation of the event, latitude and longitude capture and reporting, drone integration capabilities, as well as several other components. "PureTech Systems continuously invests in research and development of systems that can protect against outside threats to infrastructures, utilities, and borders", said Larry Bowe, Jr., President of PureTech Systems. "Preserving life is a driving force behind our security solutions and I'm proud of our team and the advancements that they have made in this field." PureTech Systems owns a total of 13 patents in a variety of solutions that include award-winning AI Video Analytics for electronically securing perimeters around facilities, infrastructures, and borders. Previous Next

< Back PureTech Delivers Numerous Innovations to Increase ROI May 10, 2023 PURETECH DELIVERS NUMEROUS INNOVATIONS TO INCREASE ROI OF ITS LEADING WIDE AREA VIDEO SURVEILLANCE SOLUTION INCLUDING FALSE ALARM ELIMINATION Customers Benefit from New Capabilities that Further Automate Security Operations and Situational Awareness PHOENIX, Ariz. – PureTech is committed to continuous innovation and customer satisfaction by providing the most comprehensive, easy to integrate, and reliable AI-boosted, geospatial video analytics and sensor fusion software solutions for borders and critical infrastructure. Larry Bowe Jr., CEO, said, “We are pleased to share just some of our progress bringing new capabilities to market in our last few major software releases. This is one of the many reasons why PureTech was named in a recent industry report and received two Govies Awards .” Core Software Developments: PTZ auto follow – the ability to track a specific target, with a PTZ camera leveraging our AI, that violates access control or breaches the perimeter. New cyber security enhancements. Improved throughput for processing various system operations. Enhancements to object tracking. Enhanced server failover automation. Expanded system health monitoring. Support for autonomy zones (geographic zones for generating alarms and other automated actions based on tracks entering, exiting or moving within zones). Novel UDP implementation to support streaming video across latent networks such as cellular and Starlink. Enhanced track management features including recording, searching, and filtering of radar and other tracks. Added support for additional neural nets including, Efficient Detection, Yolo V5 and V7, and ONNX ML runtime. Require (or not) Auto-identification using Machine Learning/Deep Learning (ML/DL) to Alarm. “Resolution boost” (Res Boost™) feature, which enables use of ML/DL on high resolution imagery. Auto-verify of targets using ML/DL on PTZ camera feed when initial detection was by radar, fence, video analytics (i.e., any sensor with geo-location). Support for running video analytics including ML/DL on iGPU inside Intel processors. Enables deployment on lower cost edge computers that do not have a discrete Nvidia GPU. Geographic area alarm shunt to enable temporary gate access, Important Partner Technology Software Integrations: Support for bidirectional TAK communications (Team Awareness Kit). Support for new lenses on ClearAlign VZ-500 PTZ cameras. Added NATO 2525B map icons. Added integration and support for: Elta 2112 radar; SRC R1410 Radar for drone detection; SRC WhisperHunter RF Signal finder for drone classification; DallmeierPss3 panoramic color camera; Magos Radars; Genetec integration to support tracks and alarms on their GIS map; PTZ driver for additional PVP NightHawk cameras; Hensoldt Radar; Health monitoring of Siklu radios, and solar power systems; Rail Intrusion Detection System integration; Echodyne Echoguard Radar; SilentSentinel PTZ Driver; AMAG Access Control enhancements. Previous Next

PureActiv wide area and security perimeter protection solutions helps utilities and critical infrastructure improve the safety and security of generation and distribution facilities and assets with a tangible Return on Investment. Utilities Solutions Learn More Concerns over keeping our nation's utilities protected from terrorist attacks have heightened in today's world. Our societies are entirely dependent on the delivery of power and water. Given the importance of this infrastructure to communities worldwide, it is imperative that necessary proactive measures be taken to secure these critical generation and distribution infrastructures from terrorists intending to do harm. PureTech’s PureActiv wide area and perimeter protection solutions help utilities improve the safety and security of power and water generation and distribution facilities, including their assets with a tangible Return on Investment. The PureActiv system secures perimeters by autonomously detecting, alerting, and deterring ground and air intrusion events at the perimeter, before they result in more catastrophic events, loss of life, and operational downtime. PureActiv provides actionable intelligence and situational awareness enabling utility authorities and law enforcement to intervene to deescalate situations and quickly return to normal operations. With decades of successful fielded deployments and continuous innovation, PureTech’s patented geospatial AI-boosted video analytics provides very high probability of detection with a very low nuisance alarm rate only sending near real-time auto-verified alarms to security personnel. Key Benefits: Increased operator productivity by reducing workload assessing nuisance alarms, Lower total acquisition and support costs, Improved site security by automating perimeter detection, deterrence and response, and providing enhanced situational awareness, and Plug and play integration into your existing system. Request PDF Active Threat Detection "We selected PureTech Systems for our perimeter surveillance software solution due to their ability to integrate best-of-breed perimeter sensors including their award-winning video analytics. After working with them for over 7 years, we are extremely pleased with not only the capabilities of their solution, but also the effort they bring to the table to ensure our needs are met. We consider them a valuable partner." - MANAGER OF NUCLEAR SECURITY PLANNING AND PROJECTS, UTILITY IN NUCLEAR INDUSTRY Key Features: Utilizes existing cameras, radars, fence sensors, access control, NVR, network, Eliminates nearly all nuisance alarms using ML, Gives your NVR live geospatial map capability with no integration required, Autonomous PTZ camera lock-on-target tracking keeps intruder in camera view, Location of detected targets show on GIS Map, No operational changes, Easily scaled to fit your budget and needs today and tomorrow, and Longer detection and classification (PurifAI) ranges reduces infrastructure costs by as much as 30%. Low Signal Detection PureTech’s solutions include and combine integrations with other sensor systems to provide the most effective and affordable safety and security solutions. The solutions scale incrementally so that you only pay for the capabilities you need when you need them. Object Left Behind Behavior Recognition: Zone entry/exit, Stopping, Loiter ing, Crowding, Travel path, Object left behind, Crawling, walking, running, Wrong direction of travel, Thrown object, Removed object, Person & vehicle tailgating, and Counting. Learn More Need a Rapid Deploy Solution? (R-DAPSS) enables airports, borders, military bases, seaports, transit agencies, and utilities to quickly deploy a temporary or permanent high fidelity virtual perimeter system at substantially less cost and time than a hard-wired solution. The PureActiv R-DAPSS provides the same best-in-class level of perimeter intrusion detection, auto-verification, and automated deterrence as a hard wired-solution. Security and facility managers can now secure their perimeters quickly without planning and executing a large construction project. Pure Activ ® Rapid-Deploy Autonomous Perimeter Surveillance System (R-DAPSS) Learn More Pure Tech Products PureActiv AI-Boosted Video Analytics Cameras PurifAI PureActiv Sensor Integration Platform PureActiv AlertView C2 System Rapid Deploy Solutions Radars