A multimodal common optical path gas leakage monitoring and positioning system and method

Abstract

The application relates to the technical field of optical detection and gas leakage monitoring, in particular to a multimodal common optical path gas leakage monitoring positioning system and method, which comprises an active laser emission module and a common optical path optical receiving module, wherein the active laser emission module is used for integrating and emitting infrared wave band laser and visible light wave band laser to an external target as active detection laser beams through a beam combining element; the common optical path optical receiving module comprises a plurality of different types of lens elements, the plurality of lens elements form a plurality of detection channels, the common optical path optical receiving module is used for receiving light signals reflected or scattered by the external target, and the light signals are disassembled into sub-signals of different wave bands through reflection or transmission of the lens elements to guide the sub-signals to corresponding detectors of the detection channels to obtain multimodal detection signals, wherein the detection channels comprise a TDLAS channel, an infrared channel, and at least one spectral imaging channel and / or a ranging channel. The application can improve the accuracy and effect of multispectral synchronous acquisition.

Description

Technical Field

[0001] This application relates to the technical field of optical detection and gas leak monitoring, and in particular to a multimodal common optical path gas leak monitoring and positioning system and method. Background Technology

[0002] Industrial site gas leak detection places extremely high demands on equipment safety, detection accuracy, and location efficiency. Existing technologies have the following limitations:

[0003] Single TDLAS detection technology: It can achieve highly sensitive quantitative detection of specific gases, but it can only provide path average concentration data and cannot directly obtain the three-dimensional location information of the leak point. The location depends on additional equipment, which is inefficient and has large errors.

[0004] Single imaging detection technology: Infrared / visible / ultraviolet imaging can achieve scene visualization, but infrared imaging is easily affected by background temperature, ultraviolet imaging is only suitable for corona discharge or specific fluorescence leakage detection, visible light cannot identify colorless and odorless gases, and none of them can provide accurate quantitative data on gas concentration.

[0005] Split-type multi-device solution: In the existing technology, TDLAS, laser ranging, and multispectral imaging are mostly independent devices, which have problems such as large device size, complex installation, non-coaxial optical paths leading to data asynchrony, low positioning accuracy, and high cost. They cannot simultaneously meet the comprehensive needs of "quantitative detection + three-dimensional positioning + multi-dimensional visualization". Summary of the Invention

[0006] To improve the accuracy and effectiveness of multispectral synchronous acquisition, this application provides a multimodal common optical path gas leak monitoring and location system and method.

[0007] Firstly, this application provides a multimodal common optical path gas leak monitoring and location system, which adopts the following technical solution:

[0008] A multimodal common-path gas leak monitoring and location system includes an active laser emitting module and a common-path optical receiving module, wherein,

[0009] The active laser emission module is used to integrate the emitted infrared and visible lasers through a beam combiner and emit them to an external target as an active detection laser beam.

[0010] The common-path optical receiving module includes several different types of lens elements, which together form several detection channels. The common-path optical receiving module is used to receive light signals reflected or scattered by an external target, and to decompose the light signals into sub-signals of different wavelengths through reflection or transmission by each lens element to guide them to the detectors corresponding to each detection channel to obtain multi-mode detection signals. The detection channels include a TDLAS channel, an infrared channel, and at least one spectral imaging channel and / or a ranging channel.

[0011] In some embodiments, the common-path optical receiving module includes a first dichroic mirror, a second dichroic mirror, a third dichroic mirror, a beam splitter, a reflector, a first focusing mirror, and a second focusing mirror;

[0012] in,

[0013] The infrared channel is composed of the transmission through the first dichroic mirror and the transmission through the third dichroic mirror;

[0014] The TDLAS channel is composed of transmission through the first dichroic mirror, reflection through the third dichroic mirror, reflection through the reflecting mirror, and the first focusing mirror;

[0015] The visible light channel in the spectral imaging channel is composed of reflections from the first dichroic mirror and reflections from the second dichroic mirror or the beam splitter.

[0016] The ultraviolet channel in the spectral imaging channel is composed of reflection from the first dichroic mirror and transmission from the second dichroic mirror;

[0017] The ranging channel is composed of reflection from the first dichroic mirror and transmission from the beam splitter, or reflection from the first dichroic mirror, reflection from the second dichroic mirror, and the second focusing mirror.

[0018] In some of these embodiments, it also includes:

[0019] A multi-channel signal processing unit is used to receive the multi-modal detection signals output by each of the detectors and send them to the corresponding processing circuit for preprocessing to obtain multi-modal parameters.

[0020] The data fusion processing unit, based on the laser emission time corresponding to the active laser emission module as a reference, aligns the timestamps of the multimodal detection signals of each detection channel, maps the multimodal parameters to the same pixel coordinate system to achieve pixel-level alignment, extracts the feature data corresponding to each channel and binds them into the same leakage feature tuple, superimposes the multimodal parameters onto the visible light image to perform image fusion and superimposes annotation information to obtain a fused image, and performs information judgment based on the fused image to generate a judgment result;

[0021] The uploading unit compares the judgment result with the set threshold to generate a leakage alarm result and a detection report, and stores the fused image and the timestamp together in local storage and uploads them to the cloud.

[0022] The alarm unit triggers an abnormal alarm based on the leakage alarm result.

[0023] Secondly, this application provides a multimodal common optical path gas leak detection and location method, which adopts the following technical solution:

[0024] A multimodal common-optical-path gas leak detection and location method includes the following steps:

[0025] An active detection laser beam is emitted, wherein the active detection laser beam comprises at least a coaxial laser obtained by combining infrared laser and visible laser;

[0026] The light signal is received by reflection or scattering, and the light signal is guided and split into different wavelengths by a combination of different types of lens elements to the detectors corresponding to each detection channel to obtain multimodal detection signals. The detection channels include TDLAS channels, infrared channels, and at least one spectral imaging channel and / or ranging channel.

[0027] All multimodal detection signals are synchronously triggered and acquired based on a unified hardware clock, and bound to the same timestamp to obtain a time-synchronized multimodal data set.

[0028] Based on the preset optical parameters of the common optical path, the data in the multimodal data group are mapped to the same pixel coordinate system to obtain spatial pixel-level registered multimodal fusion data;

[0029] Based on the multimodal fusion data, multi-dimensional feature association and leakage determination are performed, and leakage detection and localization results including gas concentration, leakage location, target distance, and visual markers are output.

[0030] In some embodiments, a modal combination consisting of several detection channels is obtained by combining different types of lens elements, the modal combination including:

[0031] Combination 1: TDLAS detection + infrared imaging + visible light imaging + laser ranging;

[0032] Combination 2: TDLAS detection + infrared imaging + visible light imaging + ultraviolet imaging;

[0033] Combination 3: TDLAS detection + infrared imaging + ultraviolet imaging + laser ranging.

[0034] In some embodiments, all multimodal detection signals are synchronously triggered and acquired based on a unified hardware clock, and bound to the same timestamp to obtain a time-synchronized multimodal data set, including the following steps:

[0035] Based on the emission time of the detected laser beam, a synchronous trigger pulse signal is generated and simultaneously sent to all lasers and detectors, synchronously starting the TDLAS wavelength scanning, laser ranging pulse emission, and exposure of all detectors;

[0036] After the data collection is completed, a stop signal is sent out simultaneously, and all output data is acquired and bound to the same timestamp.

[0037] In some embodiments, after synchronously triggering the acquisition of all multimodal detection signals based on a unified hardware clock, the process further includes preprocessing of the multimodal detection signals, specifically:

[0038] The first / second harmonics of the TDLAS gas absorption signal are extracted and phase-locked amplified. The target gas concentration, concentration gradient, and over-threshold marker are obtained based on the absorption peak intensity inversion.

[0039] Laser time-of-flight processing, noise and jitter removal are performed on the ranging pulse signal;

[0040] Non-uniformity correction, dead spot removal, temperature conversion, and gas cloud identification are performed on infrared images.

[0041] Distortion correction, white balance processing, and target recognition are performed on visible light images.

[0042] The ultraviolet images were subjected to solar blindness background suppression, discharge point extraction, and intensity calculation.

[0043] In some embodiments, multi-dimensional feature association and leakage determination are performed based on the multimodal fusion data, including the following steps:

[0044] Using pixel coordinates as keys, the TDLAS concentration value, infrared temperature value, ultraviolet discharge intensity, and laser ranging distance are analyzed based on the feature data corresponding to each detection channel at the corresponding position and bound to the same leakage feature tuple. The feature data includes at least the concentration value and concentration gradient of the TDLAS channel, the temperature anomaly area and gas cloud outline of the infrared channel, the pipe / flange / valve target of the visible light channel, the corona spot characteristics of the ultraviolet channel, and the distance value of the ranging channel.

[0045] The leakage feature tuples are analyzed based on a preset multi-condition weighted logic to determine the leakage level and confidence level.

[0046] In some embodiments, the output of leak detection and location results, including gas concentration, leak location, target distance, and visual markers, includes the following steps:

[0047] Using a visible light image as the base image, an infrared pseudo-color layer, an ultraviolet highlighting layer, and a TDLAS concentration pseudo-color cloud map are overlaid. Leakage boxes, concentration values, distance values, timestamps, and leakage levels are marked at corresponding locations to generate a multi-dimensional fused image output.

[0048] In some embodiments, the leakage level includes normal, minor leakage, general leakage, severe leakage, discharge and leakage combined anomaly, and the leakage detection and location results also include the leakage location coordinates and recommended treatment measures.

[0049] The technical solutions provided by the embodiments of this application have the following technical effects:

[0050] Data time synchronization, no time drift: All modules share the same hardware master clock, and TDLAS transmission, laser ranging, and visible / infrared / ultraviolet exposure all start at the same time, eliminating time deviations between multiple devices at the source. Imaging and point detection data are bound one-to-one in time, avoiding the misalignment problem of "concentration alarm, image mismatch", making positioning more reliable.

[0051] Shared optical path and high integration: All channels share the same incident light from the same receiving lens, acquiring light at the same time, from the same angle, and with the same luminous flux. No "image misalignment or density mismatch" will occur when the device moves. At the same time, the shared optical front end reduces the number of optical components, lowers the size and cost of the device, and facilitates vehicle-mounted, airborne, or portable deployment.

[0052] Image processing registration and fusion is simple: multi-spectral synchronous acquisition, visible light / infrared / ultraviolet pixels are naturally aligned at the same time and from the same viewpoint, without the need for complex registration algorithms, and the fusion speed is fast and the error is small.

[0053] Multimodal complementarity and comprehensive judgment: TDLAS provides accurate quantitative data, laser ranging realizes target distance measurement, infrared / visible light / ultraviolet imaging realizes the visualization and location of leakage points and scene perception, and adopts multi-dimensional decision fusion to avoid false triggering by a single sensor. The system maintains high accuracy under strong light, backlight and complex working conditions. Attached Figure Description

[0054] Figure 1 This is a schematic diagram of the module connection of a multimodal common optical path gas leak monitoring and positioning system provided in this embodiment.

[0055] Figure 2 This is a schematic diagram of the common optical path corresponding to combination 1 in the multimodal common optical path gas leak monitoring and positioning system provided in the embodiments of this application.

[0056] Figure 3This is a schematic diagram of the common optical path corresponding to combination 2 in the multimodal common optical path gas leak monitoring and positioning system provided in the embodiments of this application.

[0057] Figure 4 This is a schematic diagram of the common optical path corresponding to combination 3 in the multimodal common optical path gas leak monitoring and positioning system provided in the embodiments of this application.

[0058] Figure 5 This is a schematic diagram of the steps of a multimodal common optical path gas leak monitoring and location method provided in an embodiment of this application.

[0059] Figure 6 This is a flowchart illustrating a multimodal common optical path gas leak monitoring and location method provided in this embodiment. Detailed Implementation

[0060] To better understand the purpose, technical solutions, and advantages of this application, it has been described and illustrated below with reference to the accompanying drawings and embodiments. However, those skilled in the art should understand that this application can be implemented without these details. In some cases, to avoid obscuring various aspects of this application due to unnecessary description, well-known methods, processes, systems, components, and / or circuits already described at a higher level will not be elaborated upon. It will be apparent to those skilled in the art that various modifications can be made to the embodiments disclosed in this application, and the general principles defined in this application can be applied to other embodiments and application scenarios without departing from the principles and scope of this application. Therefore, this application is not limited to the illustrated embodiments, but conforms to the broadest scope consistent with the scope of protection claimed in this application.

[0061] It should be noted that the descriptions of these embodiments are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0062] In the description of this application, "several" means one or more, "more than" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0063] In the description of this application, the terms "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples.

[0064] like Figure 1 As shown in the figure, this application discloses a multimodal common-path gas leak monitoring and location system, including an active laser emitting module and a common-path optical receiving module, wherein,

[0065] The active laser emission module is used to integrate the emitted infrared and visible laser beams through a beam combiner and emit them to an external target as an active detection laser beam.

[0066] The active laser emission module is used to provide an active light source for the system and consists of a laser drive circuit, a TDLAS tunable laser (infrared band), a laser ranging laser (visible band), a beam combiner, and an emission lens.

[0067] The laser drive circuit provides drive control for the TDLAS laser and the ranging laser, enabling the modulation and emission of the laser.

[0068] TDLAS lasers emit lasers in specific infrared bands (targeting the absorption peaks of the target gas, such as 1650nm for methane and 1570nm for hydrogen sulfide) for gas absorption detection; rangefinder lasers emit lasers in visible light bands for target distance measurement.

[0069] Two lasers are combined into one beam by a beam combiner, which is emitted from the same emitting lens to illuminate the target under test. In this embodiment, the beam combiner is a beam combiner mirror.

[0070] The common-path optical receiving module includes several different types of lens elements, which together form several detection channels. The common-path optical receiving module is used to receive light signals reflected or scattered by external targets, and to decompose the light signals into sub-signals of different wavelengths through reflection or transmission of each lens element to guide them to the detectors corresponding to each detection channel to obtain multi-mode detection signals. The detection channels include TDLAS channels, infrared channels, and at least one spectral imaging channel and / or ranging channel.

[0071] The common-path optical receiving module is used to acquire light signals reflected or scattered by external targets through a single receiving lens, achieving common-path reception. By using multiple optical lenses such as dichroic mirrors, reflectors, beam splitters, and focusing lenses, the light signal is split according to its spectrum and guided to the corresponding detectors in the corresponding detection channels for detection. For example, 200-400nm ultraviolet light is guided to the ultraviolet imaging detector corresponding to the ultraviolet channel; 400-780nm visible light is guided to the visible light imaging detector corresponding to the visible light channel or to the laser ranging detector via the ranging channel; 780-2500nm infrared light is guided to the corresponding TDLAS single-point detector via the TDLAS channel; and 2500-14000nm infrared light is guided to the corresponding infrared imaging detector via the infrared channel. This achieves "same incident light, multi-spectral parallel acquisition" with no crosstalk and no energy loss between channels.

[0072] In other embodiments, the common-path optical receiving module includes a first dichroic mirror, a second dichroic mirror, a third dichroic mirror, a beam splitter, a reflector, a first focusing mirror, and a second focusing mirror.

[0073] The first dichroic mirror reflects light in the 200–780nm band and transmits light in the 780–14000nm band; the second dichroic mirror reflects light in the 400–780nm band and transmits light in the 200–400nm band; the third dichroic mirror reflects light in the 780–2500nm band and transmits light in the 2500–14000nm band; the beam splitter reflects light in the 400–780nm band and transmits light in the 400–780nm band; the reflector is used to change the direction of the light path and guide the TDLAS band light to the focusing lens; the first focusing lens and the second focusing lens are used to focus the specific TDLAS band light and the narrowband visible light band light of the laser, respectively.

[0074] in,

[0075] The infrared channel consists of the transmission through the first dichroic mirror and the transmission through the third dichroic mirror;

[0076] The TDLAS channel consists of transmission through the first dichroic mirror, reflection through the third dichroic mirror, reflection through the reflecting mirror, and the first focusing mirror.

[0077] The visible light channel in the spectral imaging channel is composed of reflections from the first dichroic mirror and the second dichroic mirror or beam splitter.

[0078] The ultraviolet channel in the spectral imaging channel is composed of reflection from the first dichroic mirror and transmission from the second dichroic mirror;

[0079] The ranging channel consists of reflection from the first dichroic mirror and transmission from the beam splitter, or reflection from the first dichroic mirror, reflection from the second dichroic mirror, and a second focusing mirror.

[0080] like Figure 2 As shown, in some embodiments, the multimodal common-path gas leak monitoring and location system uses a combination of lens elements to form a detection channel combination of: TDLAS detection + infrared imaging + visible light imaging + laser ranging. Specifically,

[0081] First, the laser drive circuit controls the TDLAS laser to emit an infrared detection laser, and at the same time controls the laser ranging laser to emit a visible light ranging laser. The two lasers are combined into a single beam by a beam combiner and emitted through the transmitting lens to the scene being measured.

[0082] Scene reflected / scattered light enters the system through the same receiving lens and is split by the first dichroic mirror: light in the 200–780nm band is reflected and enters the corresponding beam splitter at the reflection position; light in the 780–14000nm band is transmitted and enters the third dichroic mirror at the transmission position.

[0083] Infrared channel: Infrared light in the 2500–14000nm band passes through the third dichroic mirror, enters the infrared imaging array detector through the infrared optical channel, and the output signal is sent to the infrared image processing circuit for thermal imaging and temperature anomaly detection.

[0084] TDLAS Channel: Light in the 780–2500nm band is reflected by the third dichroic mirror, then reflected by the mirror in the reflecting position, and the first focusing mirror focuses the light signal to the TDLAS single-point detector. The output signal is sent to the TDLAS signal processing circuit for quantitative detection of gas concentration.

[0085] Visible light channel and ranging channel: The beam splitter splits the 400–780nm visible light beam. Part of the light is reflected into the visible light imaging array detector, and the output signal is sent to the visible light image processing circuit. The other part of the light passes through the beam splitter and is focused by the second focusing lens to the laser ranging single-point detector. The output signal is sent to the ranging signal processing circuit. The visible light detection result is used for scene visualization, and the ranging detection result is used for target distance measurement.

[0086] like Figure 3 As shown, in some embodiments, the multimodal common-path gas leak monitoring and location system uses a combination of lens elements to form a detection channel combination of: TDLAS detection + infrared imaging + visible light imaging + ultraviolet imaging. Specifically,

[0087] First, the laser drive circuit controls the TDLAS laser to emit an infrared detection laser, and at the same time controls the laser ranging laser to emit a visible light ranging laser. The two lasers are combined into a single beam by a beam combiner and emitted through the transmitting lens to the scene being measured.

[0088] Scene reflected / scattered light enters the system through the same receiving lens and is split by the first dichroic mirror: light in the 200–780nm band is reflected and enters the second dichroic mirror at the corresponding reflection position; light in the 780–14000nm band is transmitted and enters the third dichroic mirror at the transmission position.

[0089] Infrared channel: Infrared light in the 2500–14000nm band passes through the third dichroic mirror, enters the infrared imaging array detector through the infrared optical channel, and the output signal is sent to the infrared image processing circuit.

[0090] TDLAS Channel: Light in the 780–2500nm band is reflected by the third dichroic mirror, then reflected by the mirror in the reflecting position, and the first focusing mirror focuses the light signal to the TDLAS single-point detector. The output signal is then sent to the TDLAS signal processing circuit.

[0091] Visible light channel and ultraviolet channel: Visible light in the 200–780nm band is reflected by the second dichroic mirror and enters the visible light imaging array detector through the visible light optical channel. The output signal is sent to the visible light image processing circuit. Ultraviolet light in the 200–400nm band passes through the second dichroic mirror and enters the ultraviolet imaging array detector through the ultraviolet optical channel. The output signal is sent to the ultraviolet imaging processing circuit. The ultraviolet detection results are used for the visual detection of corona discharge or specific leaks.

[0092] like Figure 4 As shown, in some embodiments, the multimodal common-path gas leak monitoring and location system uses a combination of lens elements to form a detection channel combination of: TDLAS detection + infrared imaging + ultraviolet imaging + laser ranging, wherein...

[0093] First, the laser drive circuit controls the TDLAS laser to emit an infrared detection laser, and at the same time controls the laser ranging laser to emit a visible light ranging laser. The two lasers are combined into a single beam by a beam combiner and emitted through the transmitting lens to the scene being measured.

[0094] Scene reflected / scattered light enters the system through the same receiving lens and is split by the first dichroic mirror: light in the 200–780nm band is reflected and enters the second dichroic mirror at the corresponding reflection position; light in the 780–14000nm band is transmitted and enters the third dichroic mirror at the transmission position.

[0095] Infrared channel: Infrared light in the 2500–14000nm band passes through the third dichroic mirror, enters the infrared imaging array detector through the infrared optical channel, and the output signal is sent to the infrared image processing circuit.

[0096] TDLAS Channel: Light in the 780–2500nm band is reflected by the third dichroic mirror, then reflected by the mirror in the reflecting position, and the first focusing mirror focuses the light signal to the TDLAS single-point detector. The output signal is then sent to the TDLAS signal processing circuit.

[0097] Ultraviolet channel and ranging channel: 200–400nm ultraviolet light passes through the second dichroic mirror, enters the ultraviolet imaging array detector through the ultraviolet optical channel, and the output signal is sent to the ultraviolet imaging processing circuit; 400–780nm visible light is reflected by the second dichroic mirror, focused by the focusing lens to the laser ranging single-point detector, and the output signal is sent to the ranging signal processing circuit.

[0098] This embodiment can simultaneously achieve infrared thermal imaging, ultraviolet corona / leakage detection, TDLAS quantitative detection, and laser ranging, and is suitable for power + petrochemical composite scenarios.

[0099] In other embodiments, it also includes:

[0100] The multi-channel signal processing unit is used to receive the multi-mode detection signals output by each detector and send them to the corresponding processing circuit for preprocessing to obtain multi-mode parameters.

[0101] The processing circuit includes:

[0102] Infrared image processing circuit: performs noise reduction, temperature correction, and outlier extraction on infrared imaging data;

[0103] TDLAS signal processing circuit: calculates parameters such as the concentration and leakage rate of the target gas through harmonic detection / lock-in amplification algorithm;

[0104] Visible light image processing circuit: enhances, corrects distortion, and identifies targets in visible light images;

[0105] Ultraviolet imaging processing circuit: performs background suppression and discharge point extraction on ultraviolet images;

[0106] Ranging signal processing circuit: calculates the time of flight of the laser ranging signal and calculates the target distance.

[0107] The above circuit can be achieved using existing circuitry in the technology.

[0108] The data fusion processing unit uses the laser emission time corresponding to the active laser emission module as a benchmark, aligns the timestamps of the multimodal detection signals of each detection channel, maps the multimodal parameters to the same pixel coordinate system to achieve pixel-level alignment, extracts the feature data corresponding to each channel and binds them into the same leakage feature tuple, superimposes the multimodal parameters onto the visible light image to perform image fusion and superimposes annotation information to obtain a fused image, and performs information judgment based on the fused image to generate a judgment result.

[0109] First, time synchronization is performed: using the active laser emission time as a reference, the timestamps of the data from each channel are aligned. Specifically:

[0110] A unified hardware master clock is provided by the data fusion processing unit, ensuring clock synchronization across the entire network;

[0111] Simultaneous activation by the same trigger pulse: TDLAS scanning, ranging emission, visible / infrared / ultraviolet exposure;

[0112] All data must be stamped with the same timestamp T;

[0113] Construct a synchronized data set: {timestamp T, concentration C, distance D, infrared Img_IR, visible light Img_VIS, ultraviolet Img_UV};

[0114] If a frame is lost, the entire group is discarded and immediately retried to ensure that there is no misaligned data.

[0115] Next, spatial registration is performed: based on the common optical path parameters, the TDLAS detection path, ranging data and multispectral images are mapped to coordinates to achieve accurate correlation between three-dimensional position and concentration data.

[0116] All channels share the same receiving lens and optical axis, making them naturally coaxial.

[0117] Using the visible light image as the reference coordinate system, pixel mapping is performed on the infrared / ultraviolet / TDLAS spot / range measuring spot;

[0118] Geometric correction was performed using factory-calibrated internal parameters and distortion coefficients;

[0119] The final implementation includes: pixel-level image alignment, pixel-level localization of TDLAS detection points, and binding of distance values ​​to corresponding pixels.

[0120] Feature extraction is performed: multi-channel features are extracted in parallel, specifically:

[0121] TDLAS: Concentration value, concentration gradient, and threshold marker;

[0122] Infrared: Temperature anomaly region, gas cloud outline, hotspot coordinates;

[0123] Visible light: Pipe / flange / valve target recognition, scene structure;

[0124] Ultraviolet: Corona spot, intensity, number of points, area, gas cloud outline, hotspot coordinates;

[0125] Laser ranging: target distance, distance stability.

[0126] Finally, fusion annotation and visualization are performed, annotating the gas concentration, leakage rate, and distance data at corresponding locations in the visible / infrared / ultraviolet images to generate a multi-dimensional fused image. Specifically:

[0127] Using a visible light image as the base map;

[0128] Superimposed infrared pseudo-color layer (temperature / gas cloud);

[0129] Superimposed ultraviolet high-brightness marking layer (corona discharge point);

[0130] Overlaying a pseudo-color cloud map of TDLAS concentration;

[0131] Overlay annotations: leak box, concentration, distance, time, level;

[0132] Output a fused image to achieve "one-click location and one-click evidence collection".

[0133] The upload unit compares the judgment result with the set threshold to generate a leakage alarm result and a detection report, which are then stored locally along with the fused image and timestamp and uploaded to the cloud.

[0134] The alarm unit triggers abnormal alarms based on the leakage alarm results.

[0135] Based on a set threshold, a leak alarm signal is generated, a detection report is output, and the data is uploaded to a host computer or cloud system. A comprehensive assessment is made to determine the output level: normal / minor leak / general leak / serious leak / complex anomaly of discharge + leak. Output information includes: leak level, location coordinates, distance, and recommended actions.

[0136] Through the above steps, a four-level multimodal data fusion mechanism is adopted. First, data layer time synchronization is achieved through hardware synchronization triggering and unified timestamp. Then, spatial pixel-level registration is completed by utilizing the natural coaxial advantage of the common optical path. Subsequently, at the feature layer, TDLAS concentration, infrared temperature / gas signal, visible light target structure, ultraviolet corona / gas signal, and laser ranging distance are jointly extracted and correlated. At the decision layer, the leakage level is comprehensively determined through multi-condition weighted logic. Finally, at the visualization layer, the multispectral image and detection data are superimposed and fused for output.

[0137] This fusion approach achieves information complementarity and enhancement at multiple levels, including time, space, features, and decision-making, significantly improving the sensitivity, location accuracy, and anti-interference capability of leak detection. It is suitable for high-precision inspection and safety monitoring in complex industrial sites such as petrochemical, gas, and power industries.

[0138] like Figure 5 and 6 As shown, this application also discloses a multimodal common optical path gas leak detection and location method, including the following steps:

[0139] S100 emits an active detection laser beam, which includes at least an infrared laser and a visible laser beam combined to form a coaxial laser beam.

[0140] First, perform system power-on initialization:

[0141] After the system is powered on, the data fusion unit completes hardware self-test, clock synchronization, and parameter loading.

[0142] The data fusion unit, as the system's main controller, initializes the internal clock source and configuration parameters for each channel (such as TDLAS modulation frequency, detector exposure time, and infrared non-uniformity correction coefficient), and establishes a unified time reference to provide a clock basis for subsequent synchronous acquisition.

[0143] Secondly, a "single clock source + star distribution" architecture is adopted to synchronously transmit the same clock signal to: TDLAS laser driver circuit, laser ranging laser driver circuit, visible light imaging detector, infrared imaging detector, and ultraviolet imaging detector. This eliminates clock drift between modules at the hardware level and provides a foundation for microsecond-level synchronous acquisition.

[0144] The synchronization trigger signal is sent out. Based on the master clock, the fusion unit generates a synchronization trigger pulse signal and sends it to all modules at the same time: the TDLAS laser starts wavelength scanning, the laser rangefinder emits a ranging pulse, and the visible light / infrared / ultraviolet detectors simultaneously start exposure, realizing hardware-level synchronization of "same time, same scene, same trigger" without any channel delay deviation.

[0145] S200 receives reflected or scattered light signals and guides the light signals to the detectors corresponding to each detection channel according to the wavelength through a combination of different types of lens elements, thereby obtaining multimodal detection signals. The detection channels include TDLAS channels, infrared channels, and at least one spectral imaging channel and / or ranging channel.

[0146] Among them, a modal combination consisting of several detection channels is obtained by combining different types of lens elements. The modal combination includes:

[0147] Combination 1: TDLAS detection + infrared imaging + visible light imaging + laser ranging;

[0148] Combination 2: TDLAS detection + infrared imaging + visible light imaging + ultraviolet imaging;

[0149] Combination 3: TDLAS detection + infrared imaging + ultraviolet imaging + laser ranging.

[0150] The S300 synchronously triggers and acquires all multimodal detection signals based on a unified hardware clock and binds them to the same timestamp to obtain a time-synchronized multimodal data set.

[0151] All multimodal detection signals are synchronously triggered and acquired based on a unified hardware clock, and bound to the same timestamp to obtain a time-synchronized multimodal data set, including the following steps:

[0152] S310 generates a synchronous trigger pulse signal based on the emission time of the detected laser beam and sends it to all lasers and detectors simultaneously, thus synchronously starting the TDLAS wavelength scanning, laser ranging pulse emission, and exposure of all detectors.

[0153] S311: After the data acquisition is completed, a stop signal is sent simultaneously, all output data are acquired and bound to the same timestamp.

[0154] After setting the acquisition duration, the fusion unit simultaneously sends a stop signal:

[0155] The TDLAS laser stops wavelength scanning, the ranging laser stops emitting, and all detectors simultaneously end exposure. Each module outputs raw data to the corresponding preprocessing circuit, forming a "multi-modal raw data set under the same timestamp".

[0156] Simultaneously, after synchronously triggering and acquiring all multimodal detection signals based on a unified hardware clock, the process also includes preprocessing of the multimodal detection signals, specifically:

[0157] S320 performs first / second harmonic extraction and lock-in amplification on the TDLAS gas absorption signal, and obtains the target gas concentration value, concentration gradient, and over-threshold marker based on the absorption peak intensity inversion.

[0158] The TDLAS signal processing circuit preprocesses the absorption spectrum signal output by the single-point detector: first / second harmonic extraction to remove noise and background interference; phase-locked amplification to improve the signal-to-noise ratio; and inversion of the target gas concentration based on the absorption peak intensity. The output results include: TDLAS detection coordinates, concentration value, concentration gradient, and threshold mark.

[0159] S321 performs laser time-of-flight processing and noise and jitter removal on the ranging pulse signal;

[0160] The ranging signal processing circuit processes the pulse signal output by the laser ranging detector: calculates the laser flight time and converts it into target distance; filters multiple ranging data to remove noise and jitter; and outputs ranging coordinates, distance value, and distance stability judgment result.

[0161] S322 performs non-uniformity correction, dead spot removal, temperature conversion, and gas cloud identification on infrared images.

[0162] The infrared image processing circuit preprocesses the thermal image output by the infrared imaging array detector: Non-uniformity correction (NUC) to eliminate detector response inconsistencies; dead / bad pixel removal; temperature conversion, image enhancement, and gas identification; the output results include: temperature feature distribution, hot spot coordinates, gas cloud identification, and location coordinates.

[0163] S323 performs distortion correction, white balance processing, and target recognition on visible light images;

[0164] The visible light image processing circuit preprocesses the images output by the visible light imaging array detector: lens distortion correction; automatic white balance and automatic exposure control; edge enhancement and noise reduction; the output results include: pipe / flange / valve target recognition, scene structure, and positioning reference point coordinates.

[0165] S324 performs solar blindness background suppression, discharge point extraction, and intensity calculation on ultraviolet images.

[0166] The ultraviolet imaging processing circuit preprocesses the image output by the ultraviolet imaging array detector: suppressing solar blind background; reducing image noise and enhancing corona spot; extracting discharge point / leak point; the output results include: corona features and location coordinates, gas identification and location coordinates, and discharge intensity / area.

[0167] S400, based on preset optical parameters of the common optical path, maps the data in the multimodal data group to the same pixel coordinate system, and obtains multimodal fusion data with spatial pixel-level registration.

[0168] Based on common optical path parameters, and using visible light images as a reference, infrared images, ultraviolet images, TDLAS detection points, and ranging points are mapped to the same pixel coordinate system to achieve pixel-level alignment.

[0169] The S500 performs multi-dimensional feature association and leak determination based on multi-modal fusion data, and outputs leak detection and location results including gas concentration, leak location, target distance, and visual markers.

[0170] Feature association includes the following steps:

[0171] S510 uses pixel coordinates as keys to analyze the TDLAS concentration value, infrared temperature value, ultraviolet discharge intensity, and laser ranging distance based on the feature data corresponding to each detection channel at the corresponding position, and binds them into the same leakage feature tuple. The feature data includes at least the concentration value and concentration gradient of the TDLAS channel, the temperature anomaly area and gas cloud outline of the infrared channel, the pipe / flange / valve target of the visible light channel, the corona spot characteristics of the ultraviolet channel, and the distance value of the ranging channel.

[0172] S520 analyzes leakage feature tuples based on preset multi-condition weighted logic to determine the leakage level and confidence level.

[0173] The S530 uses a visible light image as a base map, overlays an infrared pseudo-color layer, an ultraviolet highlighting layer, and a TDLAS concentration pseudo-color cloud map, and marks the leakage box, concentration value, distance value, timestamp, and leakage level at the corresponding positions to generate a multi-dimensional fused image output.

[0174] The specific steps of the fusion process are as follows:

[0175] Factory calibration mapping relationship: The intrinsic parameters of the receiving system (focal length, principal point coordinates, radial / tangential distortion parameters) are calibrated in advance. At the same time, the image plane offset and pixel size ratio of the infrared detector, ultraviolet detector, TDLAS / range measuring optical axis relative to the visible light detector are calibrated to generate the pixel mapping matrix of each channel to the visible light coordinate system.

[0176] Achieving geometric correction and alignment: For the original infrared and ultraviolet images, lens distortion is first eliminated through distortion coefficients, and then coordinate transformation is performed through mapping matrix to scale, translate, and align them to the pixel coordinate system of the visible light image, ensuring that each pixel in the image corresponds to the same point in space; TDLAS and laser ranging are single-point detection, and their optical axis coincides with the main optical axis of the receiving system, which can directly determine the center pixel coordinates of their spot on the visible light image.

[0177] Four independent images are generated at the same aligned coordinates:

[0178] Visible light bottom layer: Visible light color / grayscale image after distortion correction, white balance, and enhancement processing, serving as the bottom layer background.

[0179] Infrared pseudo-color layer: First, the grayscale signal output by the infrared detector is converted into actual temperature values ​​through the factory temperature calibration curve; then, a preset pseudo-color mapping table (such as iron red or rainbow color palette) is used to map the temperature values ​​one by one to RGB color values, generating a color infrared image. To avoid background occlusion, threshold segmentation can be used to retain only the pseudo-color in areas of temperature anomalies and gas cloud outlines, while normal background areas are set to be completely transparent.

[0180] Ultraviolet Highlight Marking Layer: Perform solar blindness background suppression and threshold segmentation on the ultraviolet image to extract the signal regions of corona spots and leakage points; highlight these regions (such as bright purple and white glowing effects), and set the remaining non-signal areas to be completely transparent, marking only abnormal points.

[0181] TDLAS Concentration Pseudo-Color Cloud Layer: TDLAS itself is a single-point quantitative detection system, obtaining only one concentration value and corresponding to one pixel on the visible light image per detection. When the system moves / scans for inspection, it continuously collects multiple sets of discrete sampling points consisting of "concentration value + corresponding pixel coordinates + distance value"; a continuous concentration distribution plane is generated through spatial interpolation algorithms (inverse distance weighted IDW, Kriging interpolation), and then converted into a color cloud map through pseudo-color mapping (low concentration green, high concentration red), while setting the overall transparency.

[0182] Finally, multiple layers are fused and overlaid:

[0183] The layering is completed using an alpha transparency blending algorithm, following a bottom-to-top hierarchical order:

[0184] The final pixel RGB = bottom layer RGB × (1 - top layer Alpha) + top layer RGB × top layer Alpha.

[0185] The overlay order is as follows: visible light base map → infrared pseudo-color layer (transparency 40%-60%) → TDLAS concentration cloud layer (transparency 30%-50%) → ultraviolet highlight marker layer (opaque) → text annotation layer (concentration, distance, grade and other text information), and finally outputs a fused image that can be directly observed.

[0186] Leakage levels include normal, minor leak, general leak, serious leak, discharge and combined leak anomaly. The leak detection and location results also include the leak location coordinates and recommended handling measures.

[0187] Using pixel coordinates as keys, TDLAS concentration, infrared temperature, ultraviolet corona discharge, and ranging distance are bound to the same leakage feature tuple. The infrared pseudo-color layer, ultraviolet high-brightness marker layer, and TDLAS concentration pseudo-color cloud map are superimposed on the visible light base map. Based on multi-feature weighted logic, the leakage / discharge level is determined, and the confidence level, location coordinates, and distance value are output.

[0188] Based on multi-feature weighted logic, the leakage / discharge level is determined, and the confidence level, location coordinates, and distance value are output, including the following steps:

[0189] First, multi-feature standardization and quantization are performed to normalize the heterogeneous features extracted from each channel into anomaly scores in the 0-1 range. Higher scores indicate stronger anomalies.

[0190] TDLAS gas characteristics: S_tdlas = current concentration / severe leak concentration threshold, and can also be superimposed with concentration change gradient (additional points for rapid concentration increase), capped at 1.

[0191] Infrared features: S_ir = (abnormal zone temperature - background temperature) / severe abnormal temperature difference threshold; if the gas cloud outline is identified, an additional fixed weight score is added.

[0192] Ultraviolet characteristics: S_uv = peak intensity of light spot / severe discharge intensity threshold, and can also be superimposed with auxiliary scores calculated by light spot area and number of points.

[0193] All feature scores range from 0 to 1, with higher values ​​indicating more significant anomalies.

[0194] Multi-feature weighted fusion calculation is performed. Weights are assigned to different features based on the application scenario (petrochemical leak / power line inspection), with a total weight sum of 1. The total anomaly score is calculated using linear weighting: S_total = ×S_tdlas+ ×S_ir+ ×S_uv.

[0195] In pure gas leak scenarios: TDLAS quantitative detection has the highest reliability and the largest weighting (e.g., =0.5, =0.3, =0.2);

[0196] In power-related scenarios: the weight of ultraviolet discharge detection is increased, and the weight of gas features is balanced.

[0197] At the same time, a "weighted + rule" composite logic is adopted: if a single core indicator reaches the severe level (such as TDLAS concentration far exceeding the threshold), the highest level judgment can be directly triggered to avoid missed judgments due to other sensors not responding.

[0198] Based on the total anomaly score S_total, a preset grading threshold is matched, and five corresponding grading levels are output:

[0199] S_total < 0.1: Normal

[0200] 0.1 ≤ S_total < 0.3: Minor leakage

[0201] 0.3 ≤ S_total < 0.6: General leakage

[0202] S_total≥0.6: Severe leakage

[0203] If both the gas leakage score and the ultraviolet emission score exceed the threshold, it is judged as a "combined anomaly of discharge + leakage".

[0204] Confidence level: Calculated by combining "multi-feature consistency" and "signal strength", ranging from 0 to 1 (can be converted to a percentage): the higher the multi-feature consistency (multiple sensors simultaneously detecting anomalies) and the higher the signal-to-noise ratio, the higher the confidence level. For example: Confidence level = 0.6 × S_total + 0.4 × (number of features triggering anomalies / total number of features)

[0205] Location coordinates:

[0206] Image pixel coordinates: Based on the visible light base map, the centroid pixel coordinates of the abnormal area (center of the infrared gas cloud, center of the ultraviolet spot, and TDLAS optical axis pixel) are taken as the coordinates of the leak point;

[0207] Three-dimensional spatial coordinates: By combining the laser ranging value and the field of view deflection angle corresponding to the pixel coordinates (calculated from the lens focal length and pixel size), the three-dimensional spatial position of the leak point relative to the equipment is calculated.

[0208] Distance value: directly taken from the measurement result of the laser ranging channel and output bound to the coordinates of the anomaly point; if it is a planar anomaly area, the distance value corresponding to the center of the area is taken.

[0209] Finally, it also includes:

[0210] Abnormal alarm: If the concentration / temperature / UV intensity exceeds the set threshold, an audible and visual / remote alarm will be triggered immediately;

[0211] Data storage and uploading: The fused image, concentration data, distance data, and timestamp are stored locally and then uploaded to the host computer / cloud platform;

[0212] Report generation: Automatically generates detection reports, including leak location, concentration, distance, level, and historical change curves, for traceability and monitoring.

[0213] The implementation principle is as follows:

[0214] A four-level multimodal data fusion mechanism is adopted. First, data layer time synchronization is achieved through hardware synchronization triggering and unified timestamp. Then, spatial pixel-level registration is completed by leveraging the natural coaxial advantage of the common optical path. Subsequently, TDLAS concentration, infrared temperature / gas signal, visible light target structure, ultraviolet corona / gas signal, and laser ranging distance are jointly extracted and correlated at the feature layer. At the decision layer, the leakage level is comprehensively determined through multi-condition weighted logic. Finally, the multispectral image and detection data are superimposed and fused at the visualization layer for output.

[0215] It should be understood that although the steps in the flowcharts in the accompanying drawings are shown sequentially as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise expressly stated herein, there is no strict order in which these steps are performed, and they may be performed in other orders.

[0216] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A multimodal common optical path gas leak monitoring and location system, characterized in that, It includes an active laser emission module and a common-path optical receiver module, wherein, The active laser emission module is used to integrate the emitted infrared and visible lasers through a beam combiner and emit them to an external target as an active detection laser beam. The common-path optical receiving module includes several different types of lens elements, which together form several detection channels. The common-path optical receiving module is used to receive light signals reflected or scattered by an external target, and to decompose the light signals into sub-signals of different wavelengths through reflection or transmission by each lens element to guide them to the detectors corresponding to each detection channel to obtain multi-mode detection signals. The detection channels include a TDLAS channel, an infrared channel, and at least one spectral imaging channel and / or a ranging channel.

2. The multimodal common optical path gas leak monitoring and location system according to claim 1, characterized in that, The common-path optical receiving module includes a first dichroic mirror, a second dichroic mirror, a third dichroic mirror, a beam splitter, a reflector, a first focusing mirror, and a second focusing mirror; in, The infrared channel is composed of the transmission through the first dichroic mirror and the transmission through the third dichroic mirror; The TDLAS channel is composed of transmission through the first dichroic mirror, reflection through the third dichroic mirror, reflection through the reflecting mirror, and the first focusing mirror; The visible light channel in the spectral imaging channel is composed of reflections from the first dichroic mirror and reflections from the second dichroic mirror or the beam splitter. The ultraviolet channel in the spectral imaging channel is composed of reflection from the first dichroic mirror and transmission from the second dichroic mirror; The ranging channel is composed of reflection from the first dichroic mirror and transmission from the beam splitter, or reflection from the first dichroic mirror, reflection from the second dichroic mirror, and the second focusing mirror.

3. The multimodal common optical path gas leak monitoring and location system according to claim 1, characterized in that, Also includes: A multi-channel signal processing unit is used to receive the multi-modal detection signals output by each of the detectors and send them to the corresponding processing circuit for preprocessing to obtain multi-modal parameters. The data fusion processing unit, based on the laser emission time corresponding to the active laser emission module as a reference, aligns the timestamps of the multimodal detection signals of each detection channel, maps the multimodal parameters to the same pixel coordinate system to achieve pixel-level alignment, extracts the feature data corresponding to each channel and binds them into the same leakage feature tuple, superimposes the multimodal parameters onto the visible light image to perform image fusion and superimposes annotation information to obtain a fused image, and performs information judgment based on the fused image to generate a judgment result; The uploading unit compares the judgment result with the set threshold to generate a leakage alarm result and a detection report, and stores the fused image and the timestamp together in local storage and uploads them to the cloud. The alarm unit triggers an abnormal alarm based on the leakage alarm result.

4. A multimodal common optical path gas leak detection and location method, characterized in that, Includes the following steps: An active detection laser beam is emitted, wherein the active detection laser beam comprises at least a coaxial laser obtained by combining infrared laser and visible laser; The light signal is received by reflection or scattering, and the light signal is guided and split into different wavelengths by a combination of different types of lens elements to the detectors corresponding to each detection channel to obtain multimodal detection signals. The detection channels include TDLAS channels, infrared channels, and at least one spectral imaging channel and / or ranging channel. All multimodal detection signals are synchronously triggered and acquired based on a unified hardware clock, and bound to the same timestamp to obtain a time-synchronized multimodal data set. Based on the preset optical parameters of the common optical path, the data in the multimodal data group are mapped to the same pixel coordinate system to obtain spatial pixel-level registered multimodal fusion data; Based on the multimodal fusion data, multi-dimensional feature association and leakage determination are performed, and leakage detection and localization results including gas concentration, leakage location, target distance, and visual markers are output.

5. The multimodal common optical path gas leak monitoring and location method according to claim 4, characterized in that, A modal combination consisting of several detection channels is obtained by combining different types of lens elements, wherein the modal combination includes: Combination 1: TDLAS detection + infrared imaging + visible light imaging + laser ranging; Combination 2: TDLAS detection + infrared imaging + visible light imaging + ultraviolet imaging; Combination 3: TDLAS detection + infrared imaging + ultraviolet imaging + laser ranging.

6. The multimodal common optical path gas leak monitoring and location method according to claim 4, characterized in that, All multimodal detection signals are synchronously triggered and acquired based on a unified hardware clock, and bound to the same timestamp to obtain a time-synchronized multimodal data set, including the following steps: Based on the emission time of the detected laser beam, a synchronous trigger pulse signal is generated and simultaneously sent to all lasers and detectors, synchronously starting the TDLAS wavelength scanning, laser ranging pulse emission, and exposure of all detectors; After the data collection is completed, a stop signal is sent out simultaneously, and all output data is acquired and bound to the same timestamp.

7. The multimodal common optical path gas leak monitoring and location method according to claim 4, characterized in that, After synchronously triggering and acquiring all multimodal detection signals based on a unified hardware clock, the process also includes preprocessing of the multimodal detection signals, specifically: The first / second harmonics of the TDLAS gas absorption signal are extracted and phase-locked amplified. The target gas concentration, concentration gradient, and over-threshold marker are obtained based on the absorption peak intensity inversion. Laser time-of-flight processing, noise and jitter removal are performed on the ranging pulse signal; Non-uniformity correction, dead spot removal, temperature conversion, and gas cloud identification are performed on infrared images. Distortion correction, white balance processing, and target recognition are performed on visible light images. The ultraviolet images were subjected to solar blindness background suppression, discharge point extraction, and intensity calculation.

8. The multimodal common optical path gas leak monitoring and location method according to claim 4, characterized in that, Based on the multimodal fusion data, multi-dimensional feature association and leakage determination are performed, including the following steps: Using pixel coordinates as keys, the TDLAS concentration value, infrared temperature value, ultraviolet discharge intensity, and laser ranging distance are analyzed based on the feature data corresponding to each detection channel at the corresponding position and bound to the same leakage feature tuple. The feature data includes at least the concentration value and concentration gradient of the TDLAS channel, the temperature anomaly area and gas cloud outline of the infrared channel, the pipe / flange / valve target of the visible light channel, the corona spot characteristics of the ultraviolet channel, and the distance value of the ranging channel. The leakage feature tuples are analyzed based on a preset multi-condition weighted logic to determine the leakage level and confidence level.

9. The multimodal common optical path gas leak monitoring and location method according to claim 8, characterized in that, The output includes leak detection and location results containing gas concentration, leak location, target distance, and visual markers, and includes the following steps: Using a visible light image as the base image, an infrared pseudo-color layer, an ultraviolet highlighting layer, and a TDLAS concentration pseudo-color cloud map are overlaid. Leakage boxes, concentration values, distance values, timestamps, and leakage levels are marked at corresponding locations to generate a multi-dimensional fused image output.

10. The multimodal common optical path gas leak monitoring and location method according to claim 9, characterized in that, The leakage levels include normal, minor leakage, general leakage, severe leakage, discharge and combined leakage anomalies. The leakage detection and location results also include the leakage location coordinates and recommended handling measures.

Images