PROBE offers thermal deep learning bullet cameras, which will bring enhanced capabilities to perimeter security, including advanced fire detection technology. The new cameras are very cost-effective, with not only deep learning algorithms but also a built-in GPU processors to support updated algorithms in the future.
Based on deep learning algorithms, PROBE’s thermal deep learning bullet cameras deliver powerful and accurate behavior analysis, including detection's such as line crossing, intrusion, region entrance and exit. The intelligent human/vehicle detection feature helps reduce false alarms caused by animals, camera shake, falling leaves, or other irrelevant objects, significantly improving alarm accuracy.
In addition, PROBE’s thermal deep learning bullet cameras are equipped with a built-in GPU chips with advanced imaging processing technology, which can create the best thermal imaging results. The high-performance GPU can support updates with more complex algorithms with larger data samples in the future to further improve the intelligent effect of Video Content Analytics (VCA).
PROBE’s thermal deep learning bullet cameras can be used in a broad range of perimeter security and fire prevention solutions, specifically in industry scenarios like power stations, airports, mines and farms. The new single-screw bracket for the cameras is designed with installers in mind, too. Its small size and neat, stable design makes it convenient to install and adjust the angle freely, either by wall, ceiling or stand mounting
Currently, the PB-T384DV-B and PB-T640DV-B series thermal deep learning bullet cameras are available
The electromagnetic spectrum contains radiation from gamma rays, X-rays, ultraviolet, visible light, infrared, microwaves, and radio waves. Each one has its unique wavelength. Any object with a temperature above absolute zero can emit a detectable amount of infrared radiation. The higher an object’s temperature, the more infrared radiation is emitted.
An infrared camera’s effective range is what is meant by “seeing an object”. Defined thresholds, known as Johnson’s Criteria, refer to the minimum number of line pairs necessary to either detect, recognize, or identify targets captured by scene imagers. The lower limits of detection, recognition, and identification (DRI), according to Johnson criteria are:
• Detection: In order to distinguish an object from the background, the image must be covered by 1.5 or more pixels.
• Recognition: In order to classify the object (animal, human, vehicle, boat, etc.), the image must have at least 6 pixels across its critical dimension.
• Identification: In order to identify the object and describe it in details, the critical dimension must have be least 12 pixels across.
While invisible to human eyes, thermal cameras detect this kind of radiation (from wavelength 8 to 14 μm, or 8,000 – 14,000 nm) and produce images (thermograms) using temperature differences. Thermography makes it possible to see the environment with or without visible light, and thus is widely used in such areas as video surveillance, fire detection, environmental control, building analyses, medical analyses, and others. Wit hthe edition of Artificial Intelligence, many of our thermal cameras now include DEEP LEARNING technology, deep learning technology enables the DeepinView cameras to detect human bodies while filtering out insignificant objects and movements within a scene where conventional VCA systems trigger false alarms. This is particularly useful for perimeter protection, where users often spend too much time and monetary resources locating significant alarms and relevant information.
Thermal cameras can be used to augment dusty or fragile mining environments where visible light or digital cameras are used for monitoring. Thermal are extremely sensitive to temperature changes and see through dust and smoke due to their infrared wavelength, allowing them to detect any heat energy through most environmental conditions.
Thermal cameras work well alongside sensors to maximize intrusion warnings to potential perimeter breaches. As thermal cameras see in total darkness, heat sources are detected . All living or energy emanating Intruders generate a heat source (hot or cold) in relative temperature to the environment. Thermal cameras are capable of setting permission zones and alarms both visible and audible when breached.
Drones equipped with thermal cameras help farmers accurately track and manage livestock, as well as surveying pastures to monitor weather damage and irrigation. Ranchers are adopting aerial thermal cameras to spot livestock under forest canopies and are also effective in detecting nearby predators.
Thermal cameras with video analytics are used to create a virtual fencing to stop and detect intrusions. Smart thermal cameras are perfect for detecting human activity in bright light and complete darkness. On-board image processing can filter irrelevant light sources, and reflections, as well distinguishing movement from animals, trees and blowing trash while accurately detecting the presence of unauthorized persons over great distances.
Oil and gas refineries as well as energy stations are constantly at risk from criminal and terrorist attacks, requiring security personal to quickly identify threats before perimeters are breached. Thermal imaging systems are especially designed to effectively identify any threat assessment in complete darkness or bad weather including light fog, haze, dust or smoke.
Warehouses are stocked with highly valuable goods. Protecting them against intrusion and arson remains the main priority of security professionals. Statistically asset loss due to fires at warehouses well equipped with fire alarms and fire fighting systems is increasing and is not effective in deterring intentional arson. Thermal cameras can be used to set virtual fencing surrounding the target building adding another level or real-time intrusion detection.