A laser cleaning system and method capable of safe beam interruption

Abstract

The application discloses a laser cleaning system and method capable of safely cutting off light, which takes a first cleaning distance as a basis for judging whether laser cleaning is in a normal operation range, and then effectively controls the operation of a laser emission module, especially can realize non-starting of the laser emission module when the laser cleaning range is exceeded, so that the safety of laser cleaning is effectively ensured; and takes a second cleaning distance as a basis for judging whether a human body signal in a laser cleaning environment needs to be identified, and then controls the operation of an AI vision sensor, especially can timely stop laser cleaning when the human body signal is identified, so that the safety of laser cleaning is effectively ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of laser cleaning, in particular to a laser cleaning system and method capable of safely interrupting laser emission. BACKGROUND

[0002] Laser cleaning technology is an efficient, pollution-free, precise, non-destructive and environmentally friendly cleaning method. With the continuous innovation and application expansion of laser technology, laser cleaning technology is increasingly diversified in various industries and has become one of the preferred methods for cleaning in many fields. Laser cleaning equipment includes high-energy lasers, optical systems and laser beam control systems. Its internal control system needs to identify the cleaning target in real time through visual recognition technology or other sensors to ensure that the laser beam can accurately irradiate the location of the contaminants.

[0003] At present, although the laser cleaning equipment in the market can effectively remove surface particles, the safety problem of laser emission during the cleaning process has not been effectively solved, that is, due to the dangerous nature of laser, the laser beam generated during the laser cleaning process has a certain energy, which may cause skin burns and eye damage if directly irradiated on the human body, so safety control of laser cleaning emission is needed. However, there is no special method for laser safety protection in the prior art, resulting in safety hazards during laser cleaning.

[0004] Therefore, in the prior art, there is a problem of safety hazards due to the lack of special protection measures during laser cleaning. SUMMARY

[0005] Therefore, it is necessary to provide a laser cleaning system and method capable of safely interrupting laser emission to solve the problem of safety hazards due to the lack of special protection measures during laser cleaning in the prior art.

[0006] To solve the above problems, the present application provides a laser cleaning system capable of safely interrupting laser emission, comprising:

[0007] A laser emission module for emitting laser to a target to be cleaned according to a cleaning instruction;

[0008] A distance measuring module for obtaining a first cleaning distance and a second cleaning distance between the laser emission module and the target to be cleaned;

[0009] An AI vision sensor for obtaining an environmental image of an environment in which the target to be cleaned is located;

[0010] The data processing module is connected with the laser emitting module, the distance measuring module and the AI vision sensor respectively, and is configured to determine whether to activate the laser emitting module according to the first cleaning distance and a first preset cleaning distance threshold; determine whether to activate the AI vision sensor according to the second cleaning distance and a second preset cleaning distance threshold; and send a light interruption instruction to the laser emitting module when it is determined that the environment image contains a portrait signal.

[0011] Further, the AI vision sensor includes a thermal infrared camera and a laser radar sensor; and the environment image includes a thermal infrared image and a laser point cloud image.

[0012] The thermal infrared camera is configured to acquire the thermal infrared image, and the laser radar sensor is configured to acquire the laser point cloud image.

[0013] Further, the laser cleaning system capable of safely interrupting light further includes an alarm module connected with the data processing module, and configured to send an alarm signal according to the light interruption instruction.

[0014] Further, the alarm signal includes at least one of a sound signal, a light signal, a digital signal and a vibration signal.

[0015] Further, the distance measuring module includes at least one of a laser distance measuring sensor, an ultrasonic distance measuring sensor and an infrared distance measuring sensor.

[0016] To solve the above problems, the application further provides a laser cleaning method capable of safely interrupting light, comprising:

[0017] acquiring a first cleaning distance and a second cleaning distance between the laser emitting module and the target to be cleaned based on the distance measuring module;

[0018] determining whether to activate the laser emitting module according to the first cleaning distance and a first preset cleaning distance threshold, and determining whether to activate the AI vision sensor according to the second cleaning distance and a second preset cleaning distance threshold, based on the data processing module;

[0019] after activating the AI vision sensor, acquiring an environment image of an environment where the target to be cleaned is located based on the AI vision sensor;

[0020] determining whether there is a portrait signal in the environment where the target to be cleaned is located based on the data processing module according to the environment image;

[0021] when there is a portrait signal in the environment where the target to be cleaned is located, sending a light interruption instruction to the laser emitting module based on the data processing module.

[0022] Further, determining whether to activate the laser emitting module according to the first cleaning distance and a first preset cleaning distance threshold, and determining whether to activate the AI vision sensor according to the second cleaning distance and a second preset cleaning distance threshold, comprises:

[0023] determining whether the first cleaning distance is within a range of a first preset cleaning distance threshold value;

[0024] If not, the laser emission module remains closed;

[0025] If yes, the laser emission module is activated, and the AI vision sensor is activated when the second cleaning distance exceeds a second preset cleaning distance threshold value.

[0026] Further, based on the data processing module determining whether there is a human image signal in the environment where the target to be cleaned is located according to the environment image, comprising:

[0027] Based on the AI vision sensor, a thermal infrared image and a laser point cloud image in the environment where the target to be cleaned is located are obtained;

[0028] According to the trained neural network model, the thermal infrared image and the laser point cloud image are fused and recognized to determine whether there is a human image signal in the environment.

[0029] Further, when there is a human image signal in the environment where the target to be cleaned is located, the data processing module sends a light interruption instruction to the laser emission module, and then further comprising:

[0030] Based on the alarm module, an alarm signal is sent according to the light interruption instruction.

[0031] Further, the first preset cleaning distance threshold value is set to 10-30 cm, and the second preset cleaning distance threshold value is set to 50 cm.

[0032] The beneficial effects of the present application are: the present application provides a laser cleaning system and method capable of safely interrupting light, which takes the first cleaning distance as the basis for judging whether the laser cleaning is within the normal operation range, thereby effectively controlling the operation of the laser emission module, especially enabling the laser emission module to be turned off when the laser cleaning range is exceeded, thereby effectively ensuring the safety of laser cleaning; by taking the second cleaning distance as the basis for judging whether the human signal in the laser cleaning environment needs to be identified, thereby controlling the operation of the AI vision sensor, especially enabling the laser cleaning to be stopped in time when the human signal is identified, thereby effectively ensuring the safety of laser cleaning. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 Structure diagram of an embodiment of the laser cleaning system capable of safely interrupting light provided by the present application;

[0034] Figure 2 Three-dimensional effect diagram of the whole machine assembly of an embodiment of the laser cleaning system capable of safely interrupting light provided by the present application;

[0035] Figure 3 A three-dimensional structure sectional view of the whole machine assembly of an embodiment of the laser cleaning system capable of safe light interruption provided in the present application;

[0036] Figure 4 A result schematic diagram of an embodiment of the mathematical model of the direct laser triangulation method provided in the present application;

[0037] Figure 5 A flowchart of an embodiment of the laser cleaning method capable of safe light interruption provided in the present application;

[0038] Figure 6 A flowchart of an embodiment of the laser emission module and AI vision sensor provided in the present application;

[0039] Figure 7 A flowchart of an embodiment of determining whether there is a human image signal in the environment where the target to be cleaned is located provided in the present application. DETAILED DESCRIPTION

[0040] The preferred embodiments of the present application will be specifically described below in conjunction with the accompanying drawings, wherein the drawings form a part of the present application and are used to illustrate the principles of the present application together with the embodiments of the present application, but are not used to limit the scope of the present application.

[0041] Before the embodiments are stated, the AI vision sensor is described:

[0042] The AI vision sensor is an artificial intelligence vision sensor combining a thermal infrared camera and a laser radar sensor, which can reliably detect and identify objects even in poor visibility environments such as daytime or night through capturing thousands of pixels of light from a whole image.

[0043] Laser cleaning technology is a high-efficiency, pollution-free, precise, non-destructive, and environmentally friendly cleaning method. With the continuous innovation and application expansion of laser technology, laser cleaning technology is becoming more and more diversified in various industries and has become one of the preferred methods for cleaning in many fields. Laser cleaning equipment includes a high-energy laser, an optical system, and a laser beam control system. The internal control system needs to identify the cleaning target in real time through vision recognition technology or other sensors to ensure that the laser beam can accurately irradiate the position of the contaminant.

[0044] Currently, while laser cleaning equipment on the market can effectively remove surface particles, the safety issues related to laser emission during the cleaning process have not been effectively resolved. This means that due to the inherent danger of lasers, the laser beams generated during laser cleaning possess a certain amount of energy, which, if directly irradiated onto the human body, could cause skin burns and eye damage. Therefore, safety control of laser cleaning emissions is necessary. However, existing technologies lack specific methods for laser safety protection, resulting in safety hazards during laser cleaning.

[0045] Therefore, in the existing technology, there are safety hazards in the laser cleaning process due to the lack of special protective measures.

[0046] To address the aforementioned issues, this application provides a laser cleaning system and method capable of safely cutting off the laser beam, which will be described in detail below.

[0047] like Figure 1 As shown, Figure 1 This is a schematic diagram of an embodiment of the laser cleaning system capable of safely cutting off laser light provided in this application. The laser cleaning system 10 capable of safely cutting off laser light includes:

[0048] Laser emitting module 11 is used to emit laser to the target to be cleaned according to the cleaning command;

[0049] The ranging module 12 is used to obtain the first cleaning distance and the second cleaning distance between the laser emitting module 11 and the target to be cleaned;

[0050] AI vision sensor 13 is used to acquire environmental images of the environment in which the target to be cleaned is located.

[0051] The data processing module 14 is connected to the laser emission module 11, the ranging module 12 and the AI ​​vision sensor 13 respectively. It is used to determine whether to activate the laser emission module 11 based on the first cleaning distance and the first preset cleaning distance threshold; to determine whether to activate the AI ​​vision sensor 13 based on the second cleaning distance and the second preset cleaning distance threshold; and to send a light cut-off command to the laser emission module 11 when it is determined that there is a human image signal in the environmental image.

[0052] In this embodiment, the first cleaning distance and the second cleaning distance between the laser emitting module 11 and the target to be cleaned are measured by setting the distance measuring module 12, and then the laser emitting module 11 is controlled by the data processing module 14 according to the first cleaning distance and the first preset cleaning distance threshold, which realizes the effectiveness of the laser emitting module 11 controlled by the first cleaning distance as the judgment condition before the laser cleaning action starts, avoiding the safety hazards caused by misoperation; further, the AI vision sensor 13 is also controlled according to the second cleaning distance and the second preset cleaning distance threshold, which realizes the timely discovery of the portrait signal in the laser cleaning environment during the laser cleaning process, and then sends the light-off instruction to the laser emitting module 11 in time, thereby avoiding the safety hazards caused by the laser emission to the human body.

[0053] As a preferred embodiment, the distance measuring module 12 includes at least one of a laser distance measuring sensor, an ultrasonic distance measuring sensor, and an infrared distance measuring sensor.

[0054] As a preferred embodiment, the AI vision sensor 13 includes a thermal infrared camera and a laser radar sensor; the environment image includes a thermal infrared image and a laser point cloud image.

[0055] The thermal infrared camera is used to acquire the thermal infrared image, and the laser radar sensor is used to acquire the laser point cloud image.

[0056] In this embodiment, during the laser cleaning process, the thermal infrared image and the laser point cloud image in the environment are acquired by the AI vision sensor 13, and then the thermal infrared image and the laser point cloud image are fused and recognized by the data processing module 14, which can realize the acquisition of dynamic changes in the environment, such as personnel movement and obstacle appearance, and then determine whether there is a portrait signal after recognizing the dynamic change, and after recognizing the portrait signal, the cleaning action is closed by sending the light-off instruction to the laser emitting module 11, thereby effectively improving the safety of laser cleaning.

[0057] In this embodiment, a multi-level safety perception method is adopted, which realizes real-time monitoring and automatic light-off during the laser cleaning process by combining laser distance measuring sensor technology, AI vision sensor technology and intelligent judgment, to ensure that the laser beam only works in the specified area.

[0058] In a specific embodiment, as shown in Figures 2-3 , the three-dimensional effect diagram of the whole machine assembly of the laser cleaning system capable of safe light-off provided by the present application is shown in Figure 2 , and the three-dimensional structure section view of the whole machine assembly of the laser cleaning system capable of safe light-off provided by the present application is shown in Figure 3

[0059] ​Wherein, the distance measuring module 12 is composed of a laser distance measuring sensor and a distance measuring sensor fixing plate 121, the AI vision sensor 13 is fixedly installed on the laser cleaning system 10 capable of safely breaking light through a vision sensor fixing plate 131, and A is a laser emission path.

[0060] In a specific embodiment, the measurement principle and process of the distance measuring module 12 are briefly introduced. The laser distance measuring sensor is mainly used for measuring the distance of the laser beam in space. Precise distance measurement is realized through a high-precision photodiode or laser radar. If it is detected that the laser is within the emission range, the laser is turned on. This module is responsible for real-time measurement of the distance between the laser beam and the target surface, uses laser triangulation to process and analyze data, and transmits distance data to the data processing module 14 of the whole machine for analysis and judgment.

[0061] Laser ranging technology is a distance measurement technology that measures the distance of a target by using a laser pulse or continuous wave laser beam emitted towards the target. Essentially, it calculates the distance between the emission point and the target to be measured by using the information difference between the emitted laser and the received laser. Generally, a laser ranging system mainly includes laser emission components and receiving components, power supply, and control system components. Compared with other ranging methods, the main advantages of laser ranging are:

[0062] 1. High precision. The measurement result error is mainly related to the precision of the instrument and the natural environment during measurement, and is not affected by the measured distance and the operator's operation behavior;

[0063] 2. The distance measuring module is light and easy to carry and operate. Because of its excellent directivity, the optical transmitting antenna has a very small volume, and the emitted light beam is very narrow, which can meet the measurement requirements;

[0064] 3. Strong anti-interference ability. The circuit and ground wave will not interfere with it;

[0065] 4. High resolution. The light beam emitted by the distance measuring machine is very narrow, and the pulse width is very short, so it has very high target resolution in both horizontal and vertical directions.

[0066] Laser triangulation can be divided into direct laser triangulation and oblique laser triangulation according to the different angles between the incident light emitted by the laser and the normal line of the measured object surface. In the direct laser triangulation, the incident light and the normal line of the measured object surface coincide, and in the oblique laser triangulation, the incident light and the normal line of the measured object surface have a certain angle, and the incident light and the photoelectric sensor are located on the two sides of the normal line of the measured object surface.

[0067] Most of the object surface measured in daily life is rough, and the light scattering on the surface is good, so the direct method can avoid the limitation of the measuring environment. In addition, the direct method is easy to locate the center of the light spot, the data processing is convenient, and the precision of the direct method is high. Therefore, the direct laser triangulation method is adopted.

[0068] Specifically, according to the analysis of the actual situation, a mathematical model is established for the direct laser triangulation method. As shown in Figure 4 Figure 4 The result schematic diagram of the mathematical model of the direct laser triangulation method provided in the application is shown in the figure. In the figure, O is the origin of the rectangular two-dimensional coordinate system; D is the reference position of the light spot on the measured object irradiated by the laser; M is the center of the cylindrical plano-convex lens; P is the imaging center position of the light spot on the linear array CCD sensor; N is the focal point position of the cylindrical plano-convex lens; s is the distance from the center of the cylindrical plano-convex lens to the laser optical axis; f is the focal length of the cylindrical plano-convex lens; d is the distance between the surface of the target to be measured and the front panel of the testing device; and L is the position of the spot center position to the origin.

[0069] In the mathematical model, the extensions of the laser axis beam, the imaging lens main plane and the photoelectric sensor surface intersect at a point, which satisfies the condition of the Scheimpflug law. The Scheimpflug law states that the focusing plane (the plane on which the clear imaging point is located, or the object plane), the lens plane and the imaging plane all intersect at a straight line. The mathematical model constrains the position information of the laser, the cylindrical plano-convex lens and the photoelectric sensor in the system, and connects the moving track of the light spot center with the distance of the target to be measured through the establishment of a two-dimensional coordinate system. Thus, the distance between the front panel of the measuring device and the surface of the target to be measured can be calculated through the track coordinates of the light spot.

[0070] Combined with the Gaussian imaging law, the lens coordinates are (-s, 0), the corresponding focal point coordinates of the lens are (-s, -f), and the light spot coordinates are (0, d).

[0071] Through mathematical operation, the formulas (1) and (2) of the straight lines PD and PO can be obtained. The track coordinates of the image point P of the target to be measured distance d can be obtained through the formulas (3) and (4) by combining the formulas (1) and (2). At this time, the distance L of the image point P to the origin O can be calculated through the formula (5), and then the distance d of the target to be measured and the distance L of the image point P to the origin can be obtained through the formula (6).

[0072] Formula (1):

[0073] Formula (2):

[0074] Formula (3):​

[0075] Equation (4):

[0076] Equation (5):

[0077] Equation (6):

[0078] In the above manner, the laser cleaning system 10 capable of safe light interruption in the present application can effectively obtain the cleaning distance between the laser emitting module 11 and the target to be cleaned.

[0079] In a specific embodiment, the AI vision sensor 13 uses advanced visual recognition technology to monitor and analyze images in real time in the cleaning area, identify potential safety hazards, and provide important data support for the data processing module 14. Advanced AI visual recognition algorithms are used for real-time image processing of the cleaning area to identify the cleaning target and the surrounding environment. Image data is transmitted to the data processing module 14 of the whole machine for comprehensive analysis with laser ranging sensor data, mainly for detecting dynamic changes in the environment, including personnel movement, obstacle appearance, etc. When factors that may affect the laser cleaning process are detected, the safety light interruption system can be immediately started.

[0080] General vision sensors may have difficulty accurately identifying objects. An AI vision sensor 13 fusion system is proposed in this paper, which combines a thermal infrared camera and a laser radar sensor to reliably detect and identify objects even in poor visibility environments such as daytime or night. This method calibrates the external parameters of the infrared thermal imager and the laser radar sensor by designing and manufacturing a three-dimensional calibration target, thereby obtaining the external parameters of the two sensors.

[0081] The process of extracting key points using the characteristics of the thermal infrared camera and the laser radar sensor and the calibration of the external parameters, and the transformation matrix from the laser radar to the thermal infrared camera is t1, which is expressed as equation (7), where r is the rotation matrix and t is the translation matrix. The present application uses a direct external parameter calibration algorithm for the thermal infrared camera and the laser radar sensor. First, the brightness distribution of the entire thermal image is analyzed using a histogram, which is a technique for analyzing the frequency of image brightness values in a thermal image. The three-dimensional calibration target is separated by the value corresponding to the upper n of the histogram brightness value. Where n is the distribution ratio of the upper brightness value. Apply the image processing technique of binarization to the thermal image t(x, y), and separate the three-dimensional calibration target using the n value on the histogram to obtain equation (8). A corresponding point is extracted from the extracted three-dimensional calibration target, and the distance between the pixel point t(x, y) in the thermal image and the corresponding point t max (t x ,t y ), tmin (t x ,t y ) of the coordinate values, a total of 4 corresponding points are extracted, which are formula (9) and formula (10).

[0082] Formula (7):

[0083]

[0084] Formula (8):

[0085]

[0086] Formula (9):

[0087]

[0088] Formula (10):

[0089]

[0090] In order to fit the field of view (FOV) of the thermal infrared camera without using the complete 3D point cloud, only the front 180 ° distance value, extract the three-dimensional point cloud data corresponding to the three-dimensional calibration target. Then, the RANSAC algorithm is used to extract the three-dimensional calibration target. The optimal plane is calculated using the internal value determined by the RANSAC algorithm. By comparing the three-dimensional point cloud coordinate values l(x,y,z) corresponding to the three-dimensional calibration target in the three-dimensional point cloud, a total of 4 corresponding points l max (l x ,l y ,l z ), l min (l x ,l y ,l z ), the 4 corresponding points are extracted as formula (11) and formula (12):

[0091] Formula (11):

[0092]

[0093] Formula (12):

[0094]

[0095] In the formula, l(x,y,z) is the three-dimensional coordinate system axis of the laser radar sensor.

[0096] Based on the inherent parameters of the thermal imager, if the unique parameters of the thermal infrared camera are known, the PnP algorithm implemented in the OpenCV library is used to estimate the relationship R|t between the world coordinate system of the LiDAR sensor and the image coordinate system of the thermal infrared camera. Here, R is the matrix that transforms the three-dimensional world coordinate system of the LiDAR sensor into the two-dimensional image coordinate system of the thermal infrared camera, R is the rotation matrix, and t is the translation matrix. The estimation relationship between the thermal infrared camera and the LiDAR sensor is given by formula (13):

[0097]

[0098] In the formula, (T) u ,T v ) represents the image pixel coordinates, (f x ,f y (c) is the focal length. x ,c y Let ) be the optical center, r be the rotation matrix, t be the transformation matrix, and (L) be the transformation matrix. x ,L y ,L z ) represents the coordinate system of the lidar sensor.

[0099] The AI ​​vision sensor 13 plays a crucial role in the safe laser shut-off process during laser cleaning. By collecting image data of the surrounding environment and using advanced artificial intelligence technology for real-time analysis and judgment, it identifies potential safety hazards and provides important data support for safe laser shut-off.

[0100] It should be noted that although the AI ​​vision sensor 13 can detect information in the image well, it is generally suitable for light sources and good environmental conditions. Therefore, in order to effectively cope with the harsh environment of laser cleaning, a passive infrared sensor is specially set up to supplement the functions of the AI ​​vision sensor 13.

[0101] In other words, the laser cleaning system 10 that can safely cut off the light also includes a passive infrared sensor. The passive infrared sensor is connected to the data processing module 14 and is used to acquire the motion infrared rays between the target to be cleaned and the passive infrared sensor.

[0102] The data processing module 14 is also used to send a light-off command to the laser emitting module 11 when moving infrared light is detected.

[0103] In this embodiment, since the target to be cleaned is generally in a static state, by setting a passive infrared sensor to monitor the environmental area during the laser cleaning process, it is possible to identify moving elements in the environmental area - moving infrared rays. Then, it is possible to determine the interfering elements outside the target to be cleaned based on the moving infrared rays, and identify the interfering elements as abnormal situations, thereby sending a light cut-off command to the laser emitting module 11, which can effectively cope with harsh cleaning environments.

[0104] Furthermore, in order to effectively alert staff and respond promptly to interruptions in laser cleaning operations, the laser cleaning system 10, which can safely cut off the light, also includes an alarm module. The alarm module is connected to the data processing module and is used to issue an alarm signal according to the light-cutting command.

[0105] In this embodiment, by setting up an alarm module connected to the AI ​​vision sensor 13, an alarm signal is issued at the same time as the light-off command is generated, so that staff can promptly detect the abnormality when the cleaning is interrupted due to abnormal light loss.

[0106] In a preferred embodiment, the alarm signal includes at least one of sound signal, light signal, digital signal and vibration signal.

[0107] To address the aforementioned problems, this application also provides a laser cleaning method capable of safely cutting off the laser beam, such as... Figure 5 As shown, Figure 5 A schematic flowchart of an embodiment of the laser cleaning method capable of safely cutting off light provided in this application includes:

[0108] Step S101: Obtain the first cleaning distance and the second cleaning distance between the laser emitting module and the target to be cleaned based on the ranging module;

[0109] Step S102: Based on the data processing module, determine whether to activate the laser emission module according to the first cleaning distance and the first preset cleaning distance threshold; determine whether to activate the AI ​​vision sensor according to the second cleaning distance and the second preset cleaning distance threshold.

[0110] Step S103: After activating the AI ​​vision sensor, acquire an environmental image of the environment where the target to be cleaned is located based on the AI ​​vision sensor;

[0111] Step S104: Based on the data processing module, determine whether there is a human image signal in the environment where the target to be cleaned is located according to the environmental image;

[0112] Step S105: When there is a human image signal in the environment where the target to be cleaned is located, the data processing module sends a light cut-off command to the laser emission module.

[0113] In this embodiment, firstly, a first cleaning distance and a second cleaning distance between the laser emitting module and the target to be cleaned are obtained based on the ranging module; then, based on the data processing module, it is determined whether to activate the laser emitting module according to the first cleaning distance and a first preset cleaning distance threshold; based on the second cleaning distance and the second preset cleaning distance threshold, it is determined whether to activate the AI ​​vision sensor; further, after activating the AI ​​vision sensor, an environmental image of the environment where the target to be cleaned is located is obtained based on the AI ​​vision sensor; and based on the data processing module, it is determined whether there is a human image signal in the environment where the target to be cleaned is located; finally, when there is a human image signal in the environment where the target to be cleaned is located, a light-cutting command is sent to the laser emitting module based on the data processing module.

[0114] In this embodiment, a first cleaning distance and a second cleaning distance between the laser emitting module and the target to be cleaned are measured by a ranging module. Then, a data processing module controls the laser emitting module based on the first cleaning distance and a first preset cleaning distance threshold. This allows the effectiveness of the laser emitting module to be controlled before the laser cleaning operation begins, using the first cleaning distance as a judgment condition, thus avoiding safety hazards caused by misoperation. Furthermore, the AI ​​vision sensor is controlled based on the second cleaning distance and the second preset cleaning distance threshold. This enables the timely detection of human images in the laser cleaning environment during the laser cleaning process, and prompts a light-cutting command to the laser emitting module, thereby avoiding safety hazards caused by laser emission onto the human body.

[0115] In a preferred embodiment, in step S102, after obtaining the first cleaning distance and the second cleaning distance, it is also necessary to determine whether the cleaning distance meets the requirements according to the data processing module, so as to activate the corresponding device, such as... Figure 6 As shown, Figure 6 A flowchart illustrating an embodiment of activating the laser emitting module and the AI ​​vision sensor provided in this application includes:

[0116] Step S121: Determine whether the first cleaning distance is within the range of the first preset cleaning distance threshold;

[0117] Step S122: If not, keep the laser emission module off;

[0118] Step S123: If yes, then activate the laser emission module and activate the AI ​​vision sensor when the second cleaning distance exceeds the second preset cleaning distance threshold.

[0119] In this embodiment, firstly, it is determined whether the first cleaning distance is within the range of the first preset cleaning distance threshold; if not, the laser emission module is kept off; if yes, the laser emission module is activated, and when the second cleaning distance exceeds the second preset cleaning distance threshold, the AI ​​vision sensor is activated.

[0120] In this embodiment, the operation of the laser emitting module is effectively controlled by using the first cleaning distance as the basis for determining whether the laser cleaning is within the normal operating range. In particular, the laser emitting module is not activated when the laser cleaning range is exceeded, thus effectively ensuring the safety of laser cleaning. The operation of the AI ​​vision sensor is controlled in a targeted manner by using the second cleaning distance as the basis for determining whether it is necessary to identify human signals in the laser cleaning environment. In particular, the laser cleaning is stopped in time when human signals are identified, thus effectively ensuring the safety of laser cleaning.

[0121] In one specific embodiment, the first preset cleaning distance threshold is set to 10-30 cm, and the second preset cleaning distance threshold is set to 50 cm.

[0122] It should be noted that in other embodiments, the ranges of the first preset cleaning distance threshold and the second preset cleaning distance threshold can be adaptively adjusted according to actual needs to meet the needs of different situations, and no limitation is made here.

[0123] In a preferred embodiment, in step S104, in order to determine whether there is a human image signal in the environment where the target to be cleaned is located, such as... Figure 7 As shown, Figure 7 A flowchart illustrating an embodiment of this application for determining whether a human facial signal exists in the environment of a target to be cleaned includes:

[0124] Step S141: Acquire thermal infrared images and laser point cloud images of the environment in which the target to be cleaned is located based on AI visual sensors;

[0125] Step S142: Based on the fully trained neural network model, perform fusion recognition of thermal infrared images and laser point cloud images to determine whether there are human facial signals in the environment.

[0126] In this embodiment, firstly, thermal infrared images and laser point cloud images of the environment in which the target to be cleaned is located are acquired based on an AI visual sensor; then, the thermal infrared images and laser point cloud images are fused and identified according to a fully trained neural network model to determine whether there are human facial signals in the environment.

[0127] In this embodiment, an AI visual sensor acquires real-time images of the area to be cleaned by laser. Then, a well-trained neural network model fuses and identifies the thermal infrared image and the laser point cloud image to determine whether there is a human face signal in the environment. When a human face signal is captured, a light-cutting command is sent to the laser emission module in a timely manner to avoid safety accidents.

[0128] Furthermore, in step S105, when the light-off command is sent to the laser emitting module, since the light-off command is an abnormal interruption of the laser cleaning operation, it is necessary to issue an alarm signal based on the light-off command by the alarm module to effectively remind the staff to check the abnormal situation in time.

[0129] In one specific embodiment, when the laser cleaning equipment is started, the laser ranging module first accurately measures the working area to ensure that laser cleaning is carried out only within the predetermined range, thus guaranteeing the foundation for safe operation. The accurate measurement by the laser ranging module is crucial for safe operation. The cleaning process will only officially begin once the laser ranging module confirms that the working area meets the preset conditions.

[0130] Once the working area meets the preset conditions, the laser begins to emit a highly focused laser beam, precisely aiming it at the target surface to ensure efficient cleaning. Subsequently, the AI ​​vision sensor module becomes a crucial component, constantly monitoring the laser's irradiation range. It captures images of the cleaning area at high speed and analyzes them in real time using advanced image processing algorithms to guarantee the accuracy and efficiency of the cleaning process.

[0131] When the sensor detects that the laser has exceeded the predetermined range or detects any movement of human features, it reacts immediately. This reaction is highly timely and effective. First, it triggers an early warning system, alerting operators through sound, light, or other warning signals to ensure they can take necessary action promptly. Second, and more importantly, it automatically shuts off the laser, stopping its operation. This measure prevents the laser from accidentally illuminating areas that should not be cleaned or inadvertently hitting people, ensuring operator safety. Simultaneously, the AI ​​vision sensor module can also identify any moving objects within the work area, especially people. If it detects someone entering the work area, it reacts immediately, issuing a warning and shutting off the laser to ensure no potential dangers occur. After the laser is shut off, the AI ​​vision sensor module also promptly determines whether the work area has been cleaned. If contaminants remain, the laser emission system will re-enter the laser emission distance determination process and begin a new round of laser cleaning.

[0132] Based on the above technical solution, by using the first cleaning distance as the basis for determining whether laser cleaning is within the normal operating range, the operation of the laser emitting module can be effectively controlled. In particular, when the laser cleaning range is exceeded, the laser emitting module can be kept off, thus effectively ensuring the safety of laser cleaning. By using the second cleaning distance as the basis for determining whether it is necessary to identify human signals in the laser cleaning environment, the operation of the AI ​​vision sensor can be controlled in a targeted manner. In particular, when human signals are identified, laser cleaning can be stopped in time, thus effectively ensuring the safety of laser cleaning.

[0133] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A laser cleaning system capable of safely cutting off light, characterized in that, include: The laser emitting module is used to emit a laser to the target to be cleaned according to the cleaning command; The ranging module is used to obtain a first cleaning distance and a second cleaning distance between the laser emitting module and the target to be cleaned; An AI vision sensor is used to acquire environmental images of the environment in which the target to be cleaned is located. The AI ​​vision sensor includes a thermal infrared camera and a lidar sensor. The environmental images include thermal infrared images and lidar point cloud images. The thermal infrared camera is used to acquire the thermal infrared images, and the lidar sensor is used to acquire the lidar point cloud images. The data processing module is connected to the laser emitting module, the ranging module, and the AI ​​vision sensor, respectively. It is used to determine whether to activate the laser emitting module based on the first cleaning distance and the first preset cleaning distance threshold; to determine whether to activate the AI ​​vision sensor based on the second cleaning distance and the second preset cleaning distance threshold; and to send a light-off command to the laser emitting module when it is determined that there is a human image signal in the environmental image.

2. The laser cleaning system capable of safely cutting off light according to claim 1, characterized in that, The laser cleaning system capable of safely cutting off light also includes an alarm module, which is connected to the data processing module and is used to issue an alarm signal according to the light-cutting command.

3. The laser cleaning system capable of safely cutting off light according to claim 2, characterized in that, The alarm signal includes at least one of sound signal, light signal, digital signal and vibration signal.

4. The laser cleaning system capable of safely cutting off light according to claim 1, characterized in that, The ranging module includes at least one of a laser ranging sensor, an ultrasonic ranging sensor, and an infrared ranging sensor.

5. A laser cleaning method capable of safely cutting off laser light, applied to the laser cleaning system capable of safely cutting off laser light as described in any one of claims 1-4, characterized in that, include: The first and second cleaning distances between the laser emission module and the target to be cleaned are obtained based on the ranging module. The data processing module determines whether to activate the laser emission module based on the first cleaning distance and the first preset cleaning distance threshold. Based on the second cleaning distance and the second preset cleaning distance threshold, determine whether to activate the AI ​​vision sensor; After activating the AI ​​vision sensor, an environmental image of the environment in which the target to be cleaned is located is obtained based on the AI ​​vision sensor; Based on the data processing module, it is determined whether there is a human image signal in the environment where the target to be cleaned is located, according to the environmental image. When the human image signal is present in the environment where the target to be cleaned is located, the data processing module sends a light-cutting command to the laser emission module.

6. The laser cleaning method capable of safely cutting off light according to claim 5, characterized in that, The laser emitting module is activated based on the first cleaning distance and the first preset cleaning distance threshold. Based on the second cleaning distance and the second preset cleaning distance threshold, determine whether to activate the AI ​​vision sensor, including: Determine whether the first cleaning distance is within the range of the first preset cleaning distance threshold; If not, then keep the laser emitting module off; If so, the laser emitting module is activated, and the AI ​​vision sensor is activated when the second cleaning distance exceeds the second preset cleaning distance threshold.

7. The laser cleaning method capable of safely cutting off light according to claim 6, characterized in that, The step of determining whether there is a human image signal in the environment of the target to be cleaned based on the environmental image by the data processing module includes: The AI ​​vision sensor is used to acquire thermal infrared images and laser point cloud images of the environment in which the target to be cleaned is located; The thermal infrared image and the laser point cloud image are fused and identified using a well-trained neural network model to determine whether there are human facial signals in the environment.

8. The laser cleaning method capable of safely cutting off light according to claim 7, characterized in that, When the human image signal is present in the environment where the target to be cleaned is located, the data processing module sends a light-cutting command to the laser emitting module, and then the process further includes: The alarm module issues an alarm signal based on the light-off command.

9. The laser cleaning method capable of safely cutting off light according to claim 5, characterized in that, The first preset cleaning distance threshold is set to 10-30 cm, and the second preset cleaning distance threshold is set to 50 cm.

Images