A multi-band fusion imaging device for smoke detection search and rescue

Abstract

The utility model discloses a kind of multi-band fusion imaging devices for thick smoke detection search and rescue, it is related to fire emergency rescue equipment technical field, including core mainboard, the side welding of the core mainboard is provided with system switch and function switch, the side of the core mainboard is electrically connected with infrared thermal imaging module, the visible light imaging module and near-infrared active illumination module side by side, the utility model solves the single band imaging technology that existing fire field is applied to the thermal imaging equipment of thick smoke environment, cannot fully meet the urgent needs of complex fire environment, the utility model system innovatively combines visible light, infrared thermal imaging and near-infrared active illumination technology organically, relies on high-performance domestic ARM+NPU computing power platform, provides the fusion image with thermal radiation characteristics and visible light details, and with less than 50ms End-to-end processing delay and 200ms within wireless image transmission delay, guarantee the real-time and accuracy of information, provide panoramic visual perception ability for firemen.

Description

Technical Field

[0001] This utility model relates to the field of fire emergency rescue equipment technology, specifically a multi-band fusion imaging device for dense smoke detection and search and rescue. Background Technology

[0002] Fire rescue, especially fires occurring in complex environments such as high-rise buildings, large commercial complexes, underground spaces, tunnels, and petrochemical facilities, are often accompanied by the generation of large amounts of dense smoke and toxic gases, creating extremely poor visual conditions. The 2023 annual report of the International Association of Fire and Rescue clearly points out that up to 72% of firefighter injuries and fatalities occur in dense smoke environments with visibility of less than 0.3 meters. In such scenarios, the visual perception of rescuers is severely obscured, and traditional search and rescue methods that rely on visual observation are almost ineffective, greatly increasing the blindness and danger of rescue operations.

[0003] Currently, imaging equipment used in firefighting in dense smoke environments mainly relies on thermal imaging technology. However, traditional thermal imagers use single-band imaging technology, which cannot fully meet the urgent need for "clear vision, long-range vision, and accurate vision" in complex fire environments. After entering the scene, firefighters still find it difficult to quickly and accurately obtain comprehensive visual information sufficient to support tactical decisions, such as the precise location and status of trapped personnel, the direction of fire spread, the integrity of building structures, and the location of hazardous chemical leaks. This lack of information perception capability directly leads to low rescue efficiency and an increased risk of casualties.

[0004] To address the aforementioned issues, a multi-band fusion imaging device for dense smoke detection and search and rescue is proposed. Utility Model Content

[0005] The purpose of this invention is to provide a multi-band fusion imaging device for dense smoke detection and search and rescue, which solves the problem that existing thermal imaging devices used in the fire protection field for dense smoke environments use single-band imaging technology and cannot fully meet the urgent need for "clear vision, long-distance vision, and accurate vision" in complex fire environments.

[0006] To achieve the above objectives, this utility model provides the following technical solution: a multi-band fusion imaging device for dense smoke detection and search and rescue, comprising a core motherboard, characterized in that: a system switch and a function switch are welded to one side of the core motherboard, a high-resolution display module for display is provided on one side of the core motherboard, a sensor board is electrically connected to one side of the high-resolution display module, and an infrared thermal imaging module, a visible light imaging module, and a near-infrared active illumination module are electrically connected side-by-side on one side of the core motherboard, wherein the infrared thermal imaging module and the visible light imaging module achieve pixel-level alignment through rigorous optical calibration and AI feature point extraction.

[0007] Preferably, the infrared thermal imaging module uses an uncooled infrared detector.

[0008] Preferably, the near-infrared active illumination module incorporates a high-power near-infrared LED light source in the 850nm band, a driving circuit, and an optical lens group.

[0009] Preferably, the visible light imaging module is a global shutter CMOS sensor.

[0010] Preferably, the core motherboard is an embedded core board that integrates an ARM architecture processor and a neural network processing unit.

[0011] Preferably, the high-resolution display module includes a display adapter board electrically connected to the core motherboard, and a display screen electrically connected to one side of the display adapter board.

[0012] Preferably, the sensor board is an independent PCB board that integrates multiple high-precision environmental sensors.

[0013] Preferably, the system switch and the function switch are fixed in a direction perpendicular to each other.

[0014] Compared with the prior art, the beneficial effects of this utility model are as follows: 1. This utility model provides a multi-band fusion imaging device for dense smoke detection and search and rescue. Through multi-band fusion and intelligent processing, it achieves a fundamental improvement. The system innovatively combines visible light, infrared thermal imaging, and near-infrared active illumination technology. Relying on a high-performance domestic ARM+NPU computing platform, it achieves high-definition smoke-penetrating imaging at a distance of more than 5 meters in dense smoke environments with visibility of less than 0.5 meters, with an output resolution of 1920×1080@30Hz. It not only overcomes the inherent defects of single technologies and provides fused images that combine thermal radiation characteristics and visible light details, but also realizes real-time identification and labeling of targets such as fire sources and human bodies through built-in AI algorithms. It elevates the original perception to intelligent auxiliary decision-making, and initially solves the problem that existing thermal imaging equipment used in the fire protection field for dense smoke environments uses single-band imaging technology and cannot fully meet the urgent need for "clear vision, long vision, and accurate vision" in complex fire environments.

[0015] 2. This utility model provides a multi-band fusion imaging device for dense smoke detection and search and rescue. The system has an end-to-end processing latency of less than 50ms and a wireless image transmission latency of less than 200ms, ensuring the real-time performance and accuracy of information. The integrated portable design and a battery life of more than 3 hours ensure its practicality and reliability in complex fire scene environments. It provides firefighters with unprecedented panoramic visual perception capabilities, significantly improving rescue efficiency and safety. It solves the problem that existing thermal imaging equipment used in dense smoke environments in the fire protection field adopts single-band imaging technology and cannot fully meet the urgent need for "clear vision, long-distance vision, and accurate vision" in complex fire scene environments. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the technical route of this utility model; Figure 2 This is a schematic diagram of the overall structure of this utility model; Figure 3 This is a schematic diagram of the control flow structure of this utility model.

[0017] In the diagram: 1. Infrared thermal imaging module; 2. Near-infrared active illumination module; 3. Visible light imaging module; 4. Core motherboard; 5. High-resolution display module; 51. Display adapter board; 52. Display screen; 6. Sensor board; 7. System switch; 8. Function switch. Detailed Implementation

[0018] The technical solutions of the present utility model will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present utility model, and not all embodiments. Based on the embodiments of the present utility model, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the present utility model.

[0019] To further understand the content of this utility model, a detailed description of this utility model will be provided in conjunction with the accompanying drawings.

[0020] Combination Figure 1 This utility model discloses a multi-band fusion imaging device for dense smoke detection and search and rescue, comprising a core motherboard 4, a system switch 7 and a function switch 8 welded to one side of the core motherboard 4, a high-resolution display module 5 for display on one side of the core motherboard 4, a sensor board 6 electrically connected to one side of the high-resolution display module 5, and an infrared thermal imaging module 1, a visible light imaging module 3 and a near-infrared active illumination module 2 electrically connected side-by-side on one side of the core motherboard 4. The infrared thermal imaging module 1 and the visible light imaging module 3 achieve pixel-level alignment through rigorous optical calibration and AI feature point extraction.

[0021] Specifically, the infrared thermal imaging module 1 of this multi-band fusion imaging device uses an uncooled infrared detector with a response band of 8-14µm to capture thermal radiation distribution images. The near-infrared active illumination module 2 emits 850nm infrared light to enhance smoke penetration. The visible light imaging module 3 is equipped with an f / 1.4 large aperture low-distortion lens to acquire high-resolution detailed images. This band fusion imaging device uses dual-light fusion imaging technology of 8-14µm and 850nm infrared light. The system is based on a VIS+IR dual-channel heterogeneous data processing architecture. The infrared thermal imaging module 1 and the visible light imaging module 3 achieve pixel-level alignment through rigorous optical calibration and AI feature point extraction. It also utilizes a continuously optimized edge AI model of pruning-quantization-distillation to compress the inference task, which originally could only run in the cloud, to the front-end embedded hardware for real-time completion. The smoke penetration imaging results are transmitted back to the command center with one click via a 5G private network, providing continuous, clear, and zero-latency "first-person perspective" for on-site decision-making. The video stream enables precise registration and real-time fusion of multi-band images, achieving a smoke-penetrating imaging distance of over 5 meters even in dense smoke environments with visibility below 0.5 meters. It outputs a high-definition video stream with a resolution of 1920×1080@30Hz, and the device has a continuous working time of at least 3 hours. Overall, it features clear imaging, fast response, long battery life, and strong portability, making it widely applicable in fields such as fire rescue, security monitoring, military reconnaissance, and underground space operations. It significantly improves visual perception and rescue efficiency in complex environments. The core motherboard 4 of this multi-band fusion imaging device runs an image fusion algorithm that is a lightweight deep learning model optimized through pruning and quantization to adapt to the computing resource limitations of edge devices, achieving real-time inference with a latency of no more than 50ms at 1080p resolution. Simultaneously, the system supports low-bitrate wireless image transmission via the RTMP protocol, with end-to-end latency controlled within 200ms. The video stream can be simultaneously pushed to firefighter helmet displays, drone ground stations, or large screens in rear command centers.

[0022] Workflow: After the system powers on, it first executes a hardware self-test program to check the status and initialize parameters of each imaging module and processing unit. The near-infrared active illumination module 2 automatically adjusts the supplementary lighting intensity based on ambient brightness data collected by the built-in ambient light sensor. The visible light imaging module 3 and the infrared thermal imaging module 1 simultaneously perform white balance correction and non-uniformity compensation to ensure the basic quality of the acquired images. After initialization, the system enters the real-time acquisition and processing stage. The near-infrared active illumination module 2 emits 850nm infrared light to penetrate smoke, the visible light imaging module 3 captures scene details at 8-megapixel resolution, and the infrared thermal imaging module 1 simultaneously acquires a 640×512 resolution thermal distribution image. The three image data streams are transmitted to the core processing board via the MIPI interface, where they are first processed based on improvements. Image registration using the ORB feature extraction and mismatch removal algorithm achieves sub-pixel alignment of dual-band images. The registered image data is then fed into a lightweight U-Net fusion network with pruned quantization, where feature-level fusion processing is accelerated by the NPU to generate a 1920×1080 resolution fused image that simultaneously contains thermal radiation features and visible light details. The processed video stream is compressed using an H.265 encoder and output in two paths: one path is output to a display screen 52 for real-time display, and the other path is transmitted to the command center via a 5G wireless module using the RTMP protocol. The entire processing latency is controlled within 200ms to ensure the real-time nature and accuracy of the information. Throughout the entire process, the system continuously monitors the operating status, intelligently adjusts power consumption, and automatically activates protection mechanisms when an anomaly is detected.

[0023] The present invention will be further described below with reference to the embodiments.

[0024] Example 1: Combination Figures 1-3 The infrared thermal imaging module 1 uses an uncooled infrared detector and operates in the 8-14µm band. It is used to collect infrared radiation information of the target and generate a thermal distribution image. The module has the ability to image in completely dark and dense smoke environments. The temperature measurement accuracy is ±0.5℃, the resolution is 640×512@25Hz, the noise equivalent temperature difference is better than 50mK, and it integrates temperature correction and non-uniformity compensation algorithms to accurately generate temperature distribution images.

[0025] The near-infrared active lighting module 2 incorporates a high-power near-infrared LED light source, driving circuit, and optical lens group in the 850nm band. The near-infrared LED array uses 24 high-power chips arranged in a ring and is equipped with a parabolic aluminum reflector to improve illumination efficiency. The driving circuit supports PWM stepless dimming and can automatically adjust the lighting intensity according to the ambient smoke concentration. The near-infrared active lighting module 2 can actively emit near-infrared light in low visibility environments, using its low scattering characteristics to provide illumination and significantly enhance smoke penetration ability.

[0026] The visible light imaging module 3 features a global shutter CMOS sensor with 8 megapixels. It is paired with a lens with a large f / 1.4 aperture and low optical distortion, and supports HDR mode and automatic exposure control to ensure that high-resolution, high-detail environmental images can still be captured in low-light environments.

[0027] The core motherboard 4 is an embedded core board that integrates an ARM architecture processor and a neural network processing unit. The core motherboard 4 adopts an embedded processing platform, integrates an ARM architecture processor and a neural network processing unit, and is electrically connected to the infrared thermal imaging module 1, the near-infrared active illumination module 2 and the visible light imaging module 3 through the MIPI interface, respectively, and is responsible for synchronously acquiring the raw image data of each module.

[0028] Example 2: Combination Figures 1-3 The high-resolution display module 5 includes a display adapter board 51 electrically connected to the core motherboard 4, and a display screen 52 electrically connected to one side of the display adapter board 51. The high-resolution display module 5 can display the visual information output by the local real-time display system. The display screen 52 adopts a high-brightness, high-resolution OLED display with a resolution of 1920×1080 and a brightness of 1000 nits. The surface is treated with anti-glare. It is connected to the core motherboard through the MIPI-DSI interface to ensure that the image can still be clearly displayed in a strong light environment. It is used to present the fused image and analysis results processed by the core motherboard. The display adapter board 51 is responsible for protocol conversion and electrical matching of the video signal output by the core motherboard 4, and provides a stable power drive for the display screen 52 to realize reliable signal transmission and physical connection.

[0029] Sensor board 6 is an independent PCB board that integrates multiple high-precision environmental sensors. Sensor board 6 provides environmental perception data for the system to optimize imaging effects and ensure system stability. It provides intelligent exposure control support for visible light imaging module 3 and high-precision temperature compensation and image stabilization support for infrared thermal imaging module 1.

[0030] System switch 7 and function switch 8 are fixed in a perpendicular direction to each other. This perpendicularity reduces the probability of switching errors. System switch 7 is a waterproof self-locking switch with a mechanical life of over 100,000 cycles. System switch 7 is directly connected in series in the main power input path of the device to control the connection and disconnection between the battery pack and the power management module of the core motherboard 4. The switch interface is isolated by an optocoupler to effectively prevent power surges and electromagnetic interference from affecting the core circuit, enabling the device to power on, power off, or wake up from sleep mode. Function switch 8 is a micro switch with a life of over 1 million presses. It can be switched between visible light, thermal imaging, and fusion display modes via software definition. Function switch 8 is connected to the GPIO pin of the core motherboard and configured as a pull-up input mode. Its function is to allow users to quickly switch between different operating modes or start / stop specific functions, such as selecting between pure visible light mode, pure thermal imaging mode, and fusion mode, to adapt to the operational needs of different scenarios.

[0031] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0032] Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-band fusion imaging device for dense smoke detection and search and rescue, comprising a core motherboard (4), characterized in that: A system switch (7) and a function switch (8) are soldered on one side of the core motherboard (4). A high-resolution display module (5) for display is provided on one side of the core motherboard (4). A sensor board (6) is electrically connected to one side of the high-resolution display module (5). An infrared thermal imaging module (1), a visible light imaging module (3), and a near-infrared active illumination module (2) are electrically connected side by side on one side of the core motherboard (4). The infrared thermal imaging module (1) and the visible light imaging module (3) are aligned at the pixel level through strict optical calibration and AI feature point extraction.

2. The multi-band fusion imaging device for dense smoke detection and search and rescue according to claim 1, characterized in that: The infrared thermal imaging module (1) uses an uncooled infrared detector.

3. The multi-band fusion imaging device for dense smoke detection and search and rescue according to claim 1, characterized in that: The near-infrared active illumination module (2) has a built-in high-power near-infrared LED light source, driving circuit and optical lens group in the 850nm band.

4. The multi-band fusion imaging device for dense smoke detection and search and rescue according to claim 1, characterized in that: The visible light imaging module (3) is a global shutter CMOS sensor.

5. The multi-band fusion imaging device for dense smoke detection and search and rescue according to claim 1, characterized in that: The core motherboard (4) is an embedded core board that integrates an ARM architecture processor and a neural network processing unit.

6. The multi-band fusion imaging device for dense smoke detection and search and rescue according to claim 1, characterized in that: The high-resolution display module (5) includes a display adapter board (51) electrically connected to the core motherboard (4) and a display screen (52) electrically connected to one side of the display adapter board (51).

7. The multi-band fusion imaging device for dense smoke detection and search and rescue according to claim 1, characterized in that: The sensor board (6) is an independent PCB board that integrates multiple high-precision environmental sensors.

8. The multi-band fusion imaging device for dense smoke detection and search and rescue according to claim 1, characterized in that: The system switch (7) and the function switch (8) are fixed in a direction perpendicular to each other.

Images