Larry Bowe is the Founder and President of PureTech Systems Inc. As an entrepreneur, Larry has been creating shareholder value in the software industry since 1987. Prior to founding PureTech Systems, Inc... < Back Larry Bowe, Jr. President & CEO Larry Bowe is the Founder and President of PureTech Systems Inc. As an entrepreneur, Larry has been creating shareholder value in the software industry since 1987. Prior to founding PureTech Systems, Inc. in 2005, Larry was Vice President of Business Development for Verint Video Systems, a leading provider of digital video recording systems to the physical security market. In this role, Larry had overall responsibility for Product Management, technology licensing, acquisitions, and their mobile digital video business unit. During his 4 year tenure, Larry’s contributions were instrumental in growing revenue 250% both organically and through strategic acquisitions, transforming Verint. Prior to joining Verint, Larry had a 17 year career at Honeywell where he held several technical and leadership positions in both their Homes and Buildings and Avionics SBUs. As the youngest Director in Honeywell, Larry shared global responsibility for technology planning and budgeting for their billion dollar Homes and Buildings Solutions SBU. He was also assigned the technical lead, responsible for due diligence of software business acquisitions. Prior to joining Honeywell Homes and Buildings, Larry held technical leadership positions in Honeywell’s Avionics businesses. Larry was one of a select few to be nominated to participate in Honeywell’s two-year Leadership training program which he successfully completed. Larry also led the software team responsible for the first ever implementation of the Future Air Navigation Communications System (FANS) on business class jets. In his early career at Honeywell, Larry was a successful avionics software engineer. Prior to joining Honeywell, Larry worked for Motorola’s government electronics group whose focus was the manufacturing of satellite electronic systems. Larry holds a B.S. in Computer Science from Arizona State University School of Engineering and an MBA from the University of Phoenix. Larry completed a number of courses in e-commerce from Stanford University. Mail Document

Reduce parking frustration with PureTech's vehicle counting solutions. Our advanced geospatial AI-boosted video analytics transforms standard cameras into smart counting sensors as part of a new or existing parking management system. PureTech delivers real-time parking data and stall occupancy via XML over TCP/IP. Vehicle Counting Schedule a Demo A rising incident of travelers rage, mounting complaints, illegal parking, lost revenue, and high emissions from cars looking for parking spaces have institutions searching for a better way to route cars to available parking spaces. Expansive parking structures and high cost of installing traditional counting systems further complicate the issue. That’s where the PureActiv Vehicle Counting technology can help. PureActiv Vehicle Counting converts cameras into intelligent counting sensors, communicating directional count and stall occupancy data to parking management systems via XML over TCP/IP. This open communication standard allows it to easily integrate into new or existing parking management systems. Features: Significantly more accurate than loop detectors (over 98% accurate). Simple install and less intrusive than loop detectors. Utilizes cameras to accurately count parking traffic by stall, zone, or floor. Ignore non-vehicles (people, lighting, baggage, carts). Count in one or both directions. Reconfigurable (cameras can be easily moved). Wireless capability. Ability to act as full video security system. Benefits Increase Revenue - Utilizes full potential of parking structure with high accuracy rates. Decrease Costs - Save installation dollars over traditional counting technologies. Increase customer satisfaction. Reduce emissions by getting cars to their parking space quicker. Schedule a Demo Pure Tech Products

Tailgating and piggybacking applications False Alarm Elimination Learn More! Perimeter breach nuisance alarms not only hinder the effectiveness of security personnel, but also takes time away from completing other important work. PureTech’s patented DL Auto-Verify (PurifAI) provides relief to users experiencing excessive nuisance alarms caused by radars and fence sensors. Leveraging decades of surveillance video, PureTech’s machine learned analytics applied to video from PTZ cameras nearly eliminates nuisance alarms while still maintaining industry high probability of detection. This plug-and-play system allows you to utilize your existing infrastructure and improve your overall perimeter security and personnel efficiency at a minimal cost. Key Benefits Eliminates nearly all Nuisance Alarms, Dramatically increases operator efficiency and effectiveness by reducing their workload, Low-cost enhancement to your current system, Improved overall situational awareness, and Simple plug and play installation. Available on Servers, Edge Devices or in the Cloud Radars, Fence & Cameras Cloud, Edge Devices, Servers Your VMS Outputs to VMS Options: Alarms via dry contact, Alarms via software integration, Live video with annotations, Recorded alarm video with annotations, and Optionally, dynamic map as a video output. Request PDF PureTech’s patented ML/DL Auto-Verify in tandem with a PTZ camera, rejects the following false alarms caused by fence fiber and radar because they are not seen as objects of interest in the detection area. Vibrations Due to Loud Sounds Fence/Buried Fiber Something Moving Other Than Person or Vehicle Radar Fence Vibrations Not From Person or Vehicle Fence/Buried Fiber

PureTech Systems Inc. has reference installations in markets ranging from government to utilities, nuclear power plants to airports and seaports. The following installations are key installations. Customers PureTech Systems Inc. has successful deployments of our cutting-edge security and surveillance solutions across various markets. Our advanced technologies have been implemented in critical infrastructure, government facilities, commercial buildings, and more, providing our clients with unparalleled protection and situational awareness. Through a combination of state-of-the-art software and hardware, our solutions enable comprehensive monitoring, threat detection, and efficient incident response. U.S. Border Fixed Towers, Relocatable Towers, and Mobile (truck) Systems. Learn more about Border Security Solutions Nuclear Power Plants Multi-sensor integrated Automated Early Warning Perimeter System. See more on Border Security Learn more about Utilities Solutions Rapid Transit Rail Intrusion Detection System. Learn more about Rapid Transit Solutions Electric Utilities Perimeter Intelligent video detection integrated into 3rd Party NVR. Learn more about Utilities Solutions U.S. & International Commercial Seaports Geospatial Intelligent Video Management Solution. Learn more about Seaport Solutions Water Utilities Geospatial Intelligent Video Perimeter protection system. Learn more about Utilities Solutions Foreign Ministry of Defense Airport, high-risk sites, and mobile systems. Learn more about Airport Solutions U.S. Defense Contractor Multiple high security sites around the U.S., centrally monitored. Learn more about Market Solutions U.S. Military Seaports Water-side geospatial perimeter protection solution. Learn more about Military Solutions Testimonials “I have relied heavily on PureActiv and thrived off its benefits. PureActiv is a gamechanger , no doubt about it”. - FEDERAL GOVERNMENT USER

Stay connected with the latest advancements in security and surveillance technology by joining PureTech Systems at upcoming industry events. Explore our time-tested solutions, attend live demonstrations, and engage with our experts to discover how PureTech can enhance your security infrastructure.

< Back Successful Integration with OWL Mar 15, 2022 PHOENIX, Ariz. – PureTech Systems announced today the successful integration of the OWL GroundAware GA1360 2D digital multibeam forming radar system with the PureActiv Rapid-Deploy Autonomous Perimeter Surveillance System (R-DAPSS). R-DAPSS is equipped with PureTech's patented geospatial and AI boosted video analytics, PureActiv, which provides among the highest probability of detection while maintaining near-zero nuisance alarms for perimeter and wide-area protection. The easily relocatable trailer is equipped with radar, PTZ Cameras and PureActiv Auto-verification to eliminate nuisance alarms and perform auto-tracking, and deterrents such as loud hailers and strobe lights. In addition, the R-DAPSS is solar and backup generator powered, needing little to no fuel, depending on installation location, thus providing an environmentally friendly protection system that does not require ground penetration and has minimal impact on local wildlife. The GA1360 system from Observation Without Limits (OWL) combines advanced digital beamforming radar technology, classification intelligence, reconfigurability, and easy integration with other security systems (such as video management and access control systems) to bring 360 ̊ of real-time, all-weather situational awareness for the physical security of perimeters and other sensitive areas of critical sites. As the foundation for event-based layered security, GA1360 enables automated detection, tracking, deterrence, and response to intrusions within a 1 km radius of critical sites -- that’s up to 750 acres or 300 hectares covered by a single radar. Electric utility transmission substations, power generation plants, manufacturing facilities and airports are among the sites where GA1360 units are commonly deployed. Next week at the ISC West Conference in Las Vegas, PureTech and OWL will be showcasing their cooperation by displaying the newest R-DAPSS model, equipped with OWL’s GroundAware Radar. Other OWL radar models are also available on the R-DAPSS platform. Attendees of the ISC West conference that are interested in a cost-effective, rapid-deploy solution for perimeter security can visit PureTech at booth 13121 and OWL at booth 11099. “OWL’s customers use radar systems in all types of environments, including areas where mounting, power and network connectivity are unavailable,” says Mike Stokes, OWL president. “We are delighted to partner with PureTech to make OWL radars available on R-DAPSS. This versatile platform provides the relocatable infrastructure needed for easy deployment of OWL radars and other surveillance assets already integrated by PureTech.” “PureTech is pleased to be adding this impressive radar to the family of detection and tracking sensors supported by the R-DAPSS and PureActiv. Our objective is to always offer the best-fit perimeter protection solution to address our customer’s needs. Adding the GroundAware Radar family to the PureActiv R-DAPSS will enable us to address an ever-expanding set of rapid-deploy protection scenarios” said Larry Bowe, Jr., President of PureTech Systems. About Pure Tech 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 email info@PureTechSystems.com . About OWL Backed by 48 years of advanced radar expertise, Observation Without Limits (OWL) develops, manufactures, and supports the OWL GroundAware® family of 2D and 3D digital beamforming radar solutions. Critical infrastructure and government sectors around the world use GroundAware solutions for ground and low-altitude airspace surveillance as an important element of their event-based layered security plans. OWL's R&D, support, and manufacturing facilities are located in Huntsville, Ala. Visit  www.owlknows.com to learn more. Previous Next

PureTech Applications PureTech Applications Plug-in play automated outdoor wide-area and perimeter protection can be utilized in a number of different applications. PureActiv is among the highest in probability of detection and lowest in nuisance alarm rates, providing a superior level of security for your facility, infrastructure, and border. Detect, Track, Classify, and Deter with one system. Your safety is our priority. Counter Drone PureActiv autonomously detects, classifies, and tracks aerial threats, delivering increased situational awareness. Learn More Access Control False Alarm Elimination PureTech validates access control events using its patented geospatial AI-boosted video analytics to verify the legitimacy of PACS Alarms. Learn More False Alarm Elimination PureTech’s patented DL Auto-Verify (PurifAI) provides relief to users experiencing excessive nuisance alarms. Learn More Rail Intrusion Detection PureActiv delivers advanced rail intrusion detection, real-time alerts, and proactive deterrence of potential safety and security threats, safeguarding lives and critical assets. Learn More Mobile Delivers the same best-in-class perimeter intrusion detection, auto-verification, and automated deterrence as a traditional hard-wired solution. Learn More Perimeter Intrusion Detection Offers industry-leading perimeter intrusion detection with seamless accuracy and reliability. Learn More Tailgating This specialized perimeter intrusion detection capability ensures your site is protected from those violating security policy, and potentially intending harm to critical assets and personnel. Learn More Testimonials “I have relied heavily on PureActiv and thrived off its benefits. PureActiv is a gamechanger , no doubt about it”. - FEDERAL GOVERNMENT USER

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

< Back PurifAI for TMA's Dispatch Spring Edition May 28, 2025 We’re excited to announce that PureTech Systems’ PurifAI® solution is featured in the Spring 2025 edition of TMA Dispatch , the official publication of The Monitoring Association. The ad highlights how PurifAI® empowers central stations with AI-boosted geospatial video analytics — delivering smarter, faster alarm verification from any video alarm source. Fully integrated with platforms like Immix and more, PurifAI® helps reduce nuisance alarms, improve operator efficiency, and elevate monitoring service value. Check out the full issue here: TMA Dispatch Spring 2025 Previous Next

< Back Unlocking Central Monitoring's Potential Nov 14, 2023 Central Monitoring Stations have entered a new era of efficiency, driven by the power of Artificial Intelligence/Machine Learning (AI/ML)- boosted video analytics. ML involves using advanced algorithms designed to improve performance and accuracy through experience and training data. Video analytics systems start with an initial dataset comprising of examples with defined objects, actions, or events to recognize. These labeled examples encompass a wide range of scenarios, from people walking and animals to intruders entering secured areas. This initial training phase serves as the foundation upon which a system's profound understanding of video content takes shape. Once the system is equipped with this foundational knowledge, it utilizes various techniques for feature extraction from video frames, such as object shapes and motion patterns. These features serve as the building blocks for making predictions and identifying objects or actions within the video feed. The system also selects appropriate ML algorithms, such as convolutional neural networks (CNNs) tailored to specific use cases like object detection, tracking, or activity recognition. But here's where the magic truly unfolds: a feedback loop, comparing these predictions with the desired to measure the accuracy of its predictions. In times where discrepancies arise, it learns, adapts, and evolves. This might look like adjusting algorithm parameters, refining feature extraction methods, or incorporating new data. This constant journey of adaptation and perpetual learning leads to improved accuracy and reliability. How Do Video Analytics Really Work? Now, picture a bustling Central Monitoring Station, where video analytics systems take in and process a steady stream of un-labelled video data from diverse sources. Drawing from learned knowledge, they effortlessly provide predictions on the presence of people, vehicles, and specific behaviors within the video feed. By harnessing the power of advanced computer vision algorithms and ML, video analytics distinguish between true security threats and nuisance/false alarms with precision. Once an object of interest is autonomously detected, the auto-verification process of said object begins. Using ML as explained above, the object will be classified and tracked, while simultaneously issuing an alarm for operators to take swift action. With PureTech Systems’ PurifAI, PureTech’s patented and award-winning AI/ML-boosted video analytics seamlessly integrates with your existing infrastructure, eliminating the need for an extensive overhaul. It's as simple as adding PurifAI to the loop to instantly transform your Central Monitoring Station. Previous Next

< Back PureTech Systems Inc. Introduces PurifAI Dec 12, 2024 Phoenix, AZ – PureTech Systems Inc ., a leader in geospatial AI-boosted video analytics for wide-area perimeter and border security, is excited to announce the launch of PurifAI , an award-winning and Patented SaaS solution designed to filter false and nuisance alarms for self and central monitoring operations . The patented cloud-based service receives video motion alarms from any source (new or existing cameras, DVRs, NVRs, other video analytics, motion sensors), and reassess them using PureTech’s cutting-edge AI-boosted video analytics to filter out false/nuisance alarms. Only auto-verified alarms that pass the AI filtering are forwarded to your central monitoring software, and/or SMS, and/or email. False/nuisance alarms are logged, but not forwarded so you don’t have to waste time looking at them. Key Benefits of PurifAI: · Eliminates nearly all false/nuisance alarms, · Exponential Growth - Enables you to monitor dramatically more sites without adding staff. · Reduced Monitoring Costs - With less time spent dealing with false/nuisance alarms, you can simultaneously improve your overall monitoring efficiency and effectiveness. · No Additional Infrastructure Required - PurifAI works seamlessly with your existing video surveillance infrastructure, such as NVRs, DVRs, and cameras; requiring no additional on-premises equipment. · Happy Employees Equals Happy Customers - Operators are more effective, which contributes to improved job satisfaction, resulting in greater employee and customer retention. "We are thrilled to incorporate our award-winning, patented AI-boosted video analytics in our PurifAI SaaS offering. " said Larry Bowe, CEO of PureTech Systems Inc. " Our AI-powered SaaS solution delivers exceptional filtering accuracy, effectively eliminating nuisance alarms. This empowers central monitoring organizations to enhance service reliability, seamlessly scale their operations, and significantly reduce operating costs." PurifAI is now available for subscription. To learn more about this innovative solution and how it can enhance your security operations, www.verifyalarms.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