Safety evaluation method based on laser ranging and laser measuring instrument used by same

Abstract

The invention relates to a safety evaluation method based on laser ranging and a laser measuring instrument used by the safety evaluation method, and belongs to the field of measurement and safety monitoring. The invention relates to a safety evaluation method based on laser ranging, which comprises a safety standard library and presetting a plurality of rule entries, and comprises the following evaluation steps of: acquiring two views by using two groups of side-by-side cameras, performing block matching on video frames to generate a disparity map and a grey-scale map, and determining gradient amplitude; after noise pixels are filtered out through a threshold value, main edge line segments are fitted in a least square mode, line segments which do not conform to the distance range are removed in combination with depth information, finally starting and ending coordinates, a direction angle and an average depth line segment are output, and the line segments are overlapped to select a target to start laser ranging; carrying out multi-time measurement and calculation; selecting a distance measurement position for laser distance measurement; executing multi-time measurement and obtaining a distance value; carrying out single-dimensional comparison, determining a single-dimensional distance, carrying out multi-dimensional synthetic comparison, determining a safety library type of an evaluation target, carrying out multi-dimensional measurement, carrying out synthetic operation on measurement data, and comparing a calculated value with a numerical value in the safety library; and judging whether compliance is met.

Description

TECHNICAL FIELD

[0001] The present application relates to a safety evaluation method based on laser ranging and a laser measuring instrument used thereby, and belongs to the field of measurement and safety monitoring. BACKGROUND

[0002] The existing safety inspection or building detection mostly uses manual tape or single-point range finder to collect size, and lacks effective guided measurement and subsequent automatic analysis function.

[0003] The operator needs to record the measurement position, value and the corresponding area by himself, which is easy to miss or record inconsistently; if the operation of fire compartment judgment, passage width checking and the like needs to be performed, it is often necessary to manually compare with the regulation manual or safety specification, which is large in workload and easy to make mistakes.

[0004] In addition, the traditional range finder lacks a combination means with image information or voice information, can only output a value, and cannot directly associate the specific measurement object corresponding to the value, so that subsequent data arrangement and tracing are difficult.

[0005] A paperless automatic detection method and device are needed to detect and give compliance prompt or suspicious state in time, reduce manual comparison and recording errors, and improve the efficiency of inspection and detection. SUMMARY

[0006] In view of the above problems, the present application provides a safety evaluation method based on laser ranging and a laser measuring instrument used thereby, which gives compliance prompt or suspicious state in time during detection, reduces manual comparison and recording errors, realizes paperless detection record, and significantly improves the efficiency and accuracy of measurement and safety evaluation. The present application overcomes the defect that manual comparison of measurement data is needed in the traditional method, and ensures the consistency and traceability of measurement record.

[0007] To achieve the above object, the technical scheme adopted by the present application is as follows:

[0008] The safety evaluation method based on laser ranging of the present application comprises a safety standard library, a plurality of rule items are preset, and the evaluation steps are as follows,

[0009] 1) image acquisition and edge detection, two views are obtained by using two groups of cameras arranged side by side, block matching is performed on the video frames to generate a parallax map, and then the parallax value and the baseline distance and the focal length of the camera are substituted into the following formula to calculate the depth:

[0010] ,

[0011] According to the calculated depth, a gray scale image of the depth map is obtained, Sobel gradient operation is performed on each gray scale image to calculate the horizontal direction gradient Gradient in vertical direction , and according to the formula

[0012] ,

[0013] The gradient amplitude is determined, the main edge line segment is fitted by using the least square method after filtering out the noise pixels by a threshold value, and the line segment that does not conform to the distance range is removed in combination with the depth information, and finally the start and end coordinates, the direction angle and the average depth line segment are output, and the selected target is started to superimpose these line segments to start the laser ranging;

[0014] 2) Multiple measurement calculations, laser ranging at selected ranging positions, multiple measurements are performed and distance values are obtained (unit: m), the dispersion is calculated by the following formula ,

[0015] ,

[0016] ,

[0017] Wherein represents the average distance value, if , is the final measurement result of the distance value, if , re-measurement;

[0018] 3) Single dimension comparison, determine single dimension distance, compare the single distance value measured with the corresponding safety library limit value Li as follows: if , then "compliant"; if , then "non-compliant", wherein represents that the measured value is directly taken participate in comparison;

[0019] 4) Multi-dimensional synthetic comparison, determine the safety library type of the evaluation target, perform multi-dimensional measurement, and compare the calculated value with the value in the safety library to determine whether it is compliant.

[0020] The safety evaluation method based on laser ranging according to the present application is characterized in that the multi-dimensional types in the safety library include the following types,

[0021] Fire compartment area, F(S)=S, Li=5000m², wherein S represents the measured area, F(S) indicates that the area value is directly taken according to the calculation formula, and the limit value 5000m² is compared;

[0022] Width of the evacuation passage, F(R)=R, Li=1.2m, wherein R represents the measured width, F(R) indicates that the distance value is directly taken according to the calculation formula, and the limit value 1.2m is compared;

[0023] Container volume limit, F(LxWxH)=LxWxH, Li=100m³, wherein L, W, H represent length, width, height respectively, F(LxWxH) represents the product of the three, and the comparison limit value is 100m³.

[0024] According to the safety evaluation method based on laser ranging, the feature is that the fire compartment area is judged: if multiple edge detection results are defined in the same "compartment", the system divides the coordinate points into multiple polygons according to the polygon division algorithm Input the following formula:

[0025] ,

[0026] Wherein represents the area of the compartment, if it is "suspicious", if it is "compliant", and the limit value is

[0027] According to the safety evaluation method based on laser ranging, the feature is that the fire project in the safety library type further includes the following types,

[0028] High-altitude operation safety distance, measure the edge height of high-altitude operation, and the laser measurement data and the personnel coordinates measured by the image exist at the same time, the system derives the vertical distance between the personnel and the dangerous edge When , the safety threshold value is 2m, and it is automatically marked as "suspicious".

[0029] According to the safety evaluation method based on laser ranging, the feature is that the fire project in the safety library type further includes the following types, entrance arrangement and evacuation efficiency comprehensive evaluation: multiple "evacuation doors" or "walkways" are measured, the system combines the door width and the walkway length to execute evacuation flow estimation ,

[0030] Wherein and are flow coefficients, if , it is determined to be non-compliant.

[0031] According to the safety evaluation method based on laser ranging, the feature is that the fire project in the safety library type further includes the following types, section connectivity compliance, in the connection of workshop passages, multiple passage widths and connection angles are measured, if a corner and exist, it is marked that "there is a safety hazard at this corner", that is, marked as "suspicious".

[0032] The laser measuring instrument used in the safety evaluation method based on laser ranging according to the present invention is characterized in that the instrument itself is in the shape of a triangular prism, the front of the instrument has a "touch screen", above the touch screen is a long strip "status display", and two omnidirectional microphones are hollowed out on the edges of the prism on both sides of the status display.

[0033] Below the touchscreen display are the instrument's function buttons: power on, measurement, and return / cancel.

[0034] The top surface of the instrument has a triangular structure and is equipped with measurement modules, namely a binocular camera module and a laser ranging module. The binocular camera module is located on both sides of the base angle of the isosceles triangle, and the laser emission port and receiving port of the laser ranging module are located at the central axis of the isosceles triangle, with the laser emission port close to the apex angle of the isosceles triangle. A Type-C charging / data port is located at the bottom of the instrument.

[0035] According to the laser measuring instrument of the present invention, the instrument is characterized by having an internal control module, which includes a main control microprocessor and a control module ROM. The voice recognition and semantic standardization module is built-in, and the data analysis and standard comparison program is in the control module ROM. The control module communicates with the binocular camera module and the laser measuring module via a high-speed SPI bus.

[0036] The laser measuring instrument according to the present invention is characterized by the following steps of use:

[0037] After powering on, the control module first initializes the binocular camera module, laser measurement module, voice recognition and semantic standardization module, data analysis and standard comparison program and communication module in sequence, completing the self-test and readiness of the internal hardware and software environment;

[0038] Subsequently, the binocular camera module begins to acquire on-site image data in real time, and the control module performs filtering, gradient calculation, and line segment fitting preprocessing steps. Based on parameters such as line segment length and direction clustering, the main edges that need to be measured are selected.

[0039] The main control microprocessor generates guide marks on the device display interface based on the selected edge position, prompting the operator to align with the measurable target and manually press the measurement button to start laser ranging.

[0040] The laser measurement module emits laser pulses and captures echo signals after being triggered by the operator, while performing temperature compensation to ensure accuracy.

[0041] After the measurement is completed, the speech recognition and semantic standardization module starts listening to the operator's naming and description of the edge or measurement point through two omnidirectional microphones, converts the obtained speech information into text through a language model, and performs standardization processing with reference to the built-in dictionary, and stores it uniformly as the "measurement point ID" field corresponding to the measurement value;

[0042] The data analysis and standard comparison program analyzes the output volume based on semantics. For fire compartments, it compares the output volume with the standard value of the fire compartment according to the selected standard and sends the result back to the main control microprocessor.

[0043] The main control microprocessor generates a visual list on the display interface, indicating whether each measurement record is compliant or non-compliant and suspicious. Batch selection is made manually, and the data is synchronized to a remote server or other devices through the communication module to complete the entire safe measurement process based on laser ranging and automatic identification.

[0044] This invention can automatically identify, accurately measure, and evaluate against standards, significantly reducing human intervention and recording errors, and improving the overall efficiency and accuracy of safety testing and evaluation.

[0045] This invention, based on the comparison of measured values ​​with safety standards, supports multi-dimensional calculations (single distance, area, volume, connectivity, flow estimation, etc.) to form a multi-level evaluation system. Compared with traditional single-shot ranging devices, this invention can perform calculations on synthetic spatial quantities (such as... and It performs automatic judgment to quickly determine whether the limit value has been exceeded. Multiple measurement data aggregation calculations can be completed on-site without relying on complex external software, thus forming a comprehensive three-dimensional judgment of a space or area. This reduces errors from manual annotation. Attached Figure Description

[0046] Figure 1 The structure of the laser measuring instrument of the present invention Figure One ,

[0047] Figure 2 The structure of the laser measuring instrument of the present invention Figure Two ,

[0048] Figure 3 The structure of the laser measuring instrument of the present invention Figure Three ,

[0050] Reference numerals: 1. Instrument, 2. Touch screen, 3. Omnidirectional microphone, 4. Power button, 5. Measurement button, 6. Return / Cancel button, 7. Binocular camera module, 8. Laser receiver port, 9. Laser emitter port, 10. Status display screen, 11. Type-C charging / data port. Detailed Implementation

[0051] The specific structure and content of the present invention will be described in detail below: The present invention includes a hardware laser measuring instrument and a safety evaluation method based on laser ranging using the hardware.

[0052] The following describes the hardware laser measuring instrument. The instrument itself is shaped like a triangular prism with chamfered edges. The front of the instrument has a length-to-width ratio of 35:20. It features a touchscreen display for interactive guidance such as edge selection during measurement. Above the touchscreen is a long strip-shaped status display showing measurement distance, instrument battery level, and error information. Two omnidirectional microphones are located on either side of the prism's edge, positioned through cutouts. Below the touchscreen are the instrument's function buttons: power button, measurement button, and return / cancel button. The top of the instrument has a triangular structure, specifically an isosceles triangle with a base-to-height ratio of 11:20. The top of the instrument houses the measurement modules: a binocular camera module and a laser ranging module. The binocular camera module is located on either side of the base angle of the isosceles triangle, while the laser ranging module's laser emitter and receiver are located along the central axis of the isosceles triangle, with the laser emitter near the apex. A Type-C charging / data port is located at the bottom of the instrument.

[0053] This invention has a built-in control module, including a main control microprocessor. After being powered on, the main control microprocessor completes hardware initialization according to the fixed program, communicates with the binocular camera module, laser measurement module, and voice recognition and semantic standardization module through a high-speed SPI bus, and exchanges bidirectional data with the data analysis and standard comparison module through I2C or UART.

[0054] The main microprocessor defines a "measurement record table" in internal SRAM or external Flash, which includes the following:

[0055] Record Identifier: Record index ID, used for fast querying and association.

[0056] Measurement field: Contains distance values ​​or other geometric quantities.

[0057] Semantic description field: A textual description generated by the speech recognition and semantic standardization module.

[0058] Analysis Identifier: Records the analysis status, including "Not Analyzed", "Compliant", "Non-Compliant", "Suspicious", "Pending Manual Confirmation", etc.

[0059] Timestamp: Used for source tracking and version comparison.

[0060] The external Flash memory stores a security standard library, with rule entries in the form of (keyword K, judgment formula F(), restriction value Li). The basic rules are as follows:

[0061] K = "fire compartment area", F(S) = S, Li = 5000m², where S represents the measured area (unit: square meters), and F(S) indicates that the calculation formula directly takes the area value and compares it with the limit value of 5000m².

[0062] K = "Evacuation route width", F(R) = R, Li = 1.2m, where R represents the measured width (unit: meters), and F(R) indicates that the calculation formula directly takes the distance value and compares it with the limit value of 1.2m.

[0063] K = "Container volume limit", F(L×W×H) = L×W×H, Li = 100m³, where L, W, and H represent length, width, and height (unit: meters) respectively, and F(L×W×H) represents the product of the three, with a limit value of 100m³.

[0064] The system can update the standard library through the communication module, making it easy to load supplementary content for different industries or scenarios.

[0065] The operating steps of the laser measuring instrument using this invention are described below:

[0066] Upon power-on, the control module first initializes the binocular camera module, laser measurement module, speech recognition and semantic standardization module, data analysis and standard comparison program, and communication module sequentially, completing self-checks and readiness checks of the internal hardware and software environment. The speech recognition and semantic standardization module consists of two omnidirectional microphones and a control program built into the control module's ROM. Subsequently, the binocular camera module begins acquiring real-time image data from the site. The control module performs preprocessing steps such as filtering, gradient calculation, and line segment fitting, selecting the main edges to be measured based on parameters such as line segment length and direction clustering. The main control microprocessor generates guide markers on the device display interface based on the selected edge positions, prompting the operator to align with the measurable target. The operator then manually presses the measurement button to initiate laser ranging. The laser measurement module emits laser pulses and captures echo signals after operator triggering, while simultaneously performing temperature compensation to ensure accuracy. After measurement, the speech recognition and semantic standardization module uses the two omnidirectional microphones to listen to the operator's naming and description of the edge or measurement point. The obtained speech information is converted into text using a language model and standardized according to the built-in dictionary, uniformly stored as a "Measurement Point ID" field corresponding to the measured value. Subsequently, the data analysis module analyzes the output volume based on semantics. For volumes involving fire compartments, it compares the output volume with the standard values ​​for the selected fire compartments and transmits the results back to the main control microprocessor. The main control microprocessor generates a visual list on the display interface, indicating whether each measurement record is compliant or suspicious. Batch selection allows for manual judgment, and the data can be synchronized to a remote server or other devices via the communication module, thus completing the entire safety measurement process based on laser ranging and automatic identification.

[0067] This device is used for a laser ranging-based safety evaluation method.

[0068] 1) Image Acquisition and Edge Detection: In practical applications, this invention uses a pair of cameras to acquire left and right views. After performing synchronization and distortion correction using known camera intrinsic and extrinsic parameters, block matching is performed on the video frames to generate a disparity map. Then, the disparity values ​​are... Distance from baseline and camera focal length Substitute into the following formula to calculate the depth:

[0069] ,

[0070] After obtaining the depth map, the system performs Sobel gradient calculations on each grayscale image to calculate the horizontal gradient. gradient in the vertical direction And according to the formula

[0071] ,

[0072] Determine the gradient magnitude. After filtering out noisy pixels using a threshold, fit the main edge segments using a least-squares method. Combine this with depth information to remove segments that do not conform to the distance range. Finally, output the start and end coordinates, orientation angle, and average depth. The system overlays these segments onto the video display for the operator to select the target and initiate laser ranging. If only low precision is required, the depth value calculated by the binoculars is used directly. If millimeter-level precision is required, the binocular output and the laser measurement value are fused and corrected to generate the final measurement record, which is then sent to the subsequent safety analysis module for comparison and judgment.

[0073] 2) Laser Ranging Procedure: After the operator selects a primary edge on the touch interface or in the button menu, the laser measurement module receives the "Start Measurement" command and performs pulse emission and temperature compensation calibration. Single measurement and multiple measurement options can be selected in the system settings. If the multiple measurement function is enabled, the system will emit pulses at 5ms intervals according to the set number of times. Multiple Dispersion Calculation: The module performs multiple measurements and obtains the distance value. (Unit: meters), the dispersion is calculated using the following formula. (Unit: meters):

[0074] ,

[0075] ,

[0076] in This represents the average distance value. If (Precision judgment value), then output As the final measurement result; if The module returns a "noise too high" status, and the system interface prompts for a remeasurement. The final measurement result, along with the measurement point ID, is written to the "Measurement Record Table" for subsequent analysis.

[0077] 3) After the speech recognition and semantic binding measurement is completed, the system enters the speech input stage. The speech acquisition unit records the operator's spoken information, and the speech recognition and semantic standardization module generates a text description, mapping synonymous or similar expressions to a unified field and storing it in the database. Recognition example: If "This is the width of the upper edge of the aisle" is recognized, the module internally maps it and records it as "aisle width" as the standard, binding this standardized name to the corresponding measurement point ID. A one-to-one correspondence is established between the "measurement value field" and the "semantic description field" in the "measurement record table".

[0078] 4) Calculation, Analysis, and Judgment: The data analysis and standard comparison module of this invention includes multiple judgment formulas for performing comprehensive calculations on single measurements, multiple measurements, and complex scenarios. After parsing the semantic description, the system determines which one or more judgment paths to use, thereby generating labels such as "compliant," "non-compliant," "suspicious," and "pending manual confirmation."

[0079] The detection and judgment are achieved through the following:

[0080] 4.1) Single-dimensional comparison: When keywords such as "walkway width," "doorway height," and "fence spacing" appear in the semantic description, the system only needs to compare the measured single distance value. (Unit: meters) and corresponding limit values (Unit: meters) The following comparisons are made:

[0081] like If so, mark it as "compliant";

[0082] like If it does, then mark it as "non-compliant". Here... This indicates that the measured value is taken directly. The comparison is suitable for a single-dimensional scenario.

[0083] 4.2) Multi-dimensional synthesis: When the semantic description contains content such as "container volume" or "fire compartment area", the system searches for multiple corresponding records in the "measurement record table" and performs synthesis calculation.

[0084] Container volume calculation: If three measurement dimensions of the same container are identified. (Unit: meter), system execution:

[0085] ,

[0086] in (Unit: cubic meters) represents the volume result. Then mark it as "non-compliant". Then mark it as "compliant". These are the limit values ​​(unit: cubic meters).

[0087] Fire compartment area determination: If multiple edge detection results are defined under the same "compartment", the system will divide the coordinate points according to the polygon subdivision algorithm. Enter the following formula:

[0088] ,

[0089] in (Unit: square meters) represents the area of ​​this zone. Then mark it as "suspicious". Then mark it as "compliant". These are the limit values ​​(unit: square meters).

[0090] Safety distance for working at heights: If the semantic display shows "height of the edge of the working at heights" and both laser measurement data and personnel coordinates measured from images exist, the system infers the vertical distance between the personnel and the danger edge. (Unit: meter). When (2m is the safety threshold here) When it is automatically marked as "suspicious", the human-computer interaction module will be triggered to issue an alarm.

[0091] Comprehensive evaluation of entrance layout and evacuation efficiency: If the operator measures multiple "evacuation doors" or "corridors," the system will calculate the door width. (Unit: meters) and walkway length (Unit: meters) Combined with the evacuation flow estimation.

[0092] ,

[0093] in and This is the flow coefficient (dimensionless, specific values ​​are available in the pre-set safety standard library). If... If so, it is judged as "non-compliant".

[0094] Section connectivity compliance: If the keyword "workshop passageway connection" appears in the record, the system will search for multiple passageway widths. (Unit: meters) and connection angle (Unit: degrees). If a certain angle and The system marks "There is a safety hazard at this corner" as "suspicious," going beyond the traditional single-check model and forming a comprehensive analysis and judgment.

[0095] This invention, based on the comparison of measured values ​​with safety standards, supports multi-dimensional calculations (single distance, area, volume, connectivity, flow estimation, etc.) to form a multi-level evaluation system. Compared with traditional single-shot ranging devices, this invention can perform calculations on synthetic spatial quantities (such as... and It performs automatic judgment to quickly determine whether the limit value has been exceeded.

[0096] The system of this invention triggers an alarm based on "suspicious" items, which are then displayed in a scrolling manner on the status display screen. When "suspicious" or "non-compliant" items occur multiple times, the human-machine interaction module can prompt the operator or responsible personnel to conduct a focused investigation, enabling real-time intervention in potential safety hazards. The communication module sends the measurement record table (measurement values, analysis status, semantic description, timestamps, etc.) to a remote server in JSON format.

[0097] After completing multi-scenario calculations, the data analysis and standard comparison module, based on human voice commands, writes "compliant," "non-compliant," "suspicious," or "pending human confirmation" in the "Analysis Identifier" field for each measurement record. If marked as "non-compliant" or "suspicious," the main control microprocessor highlights a warning on the display interface, and the operator can choose "verify" or "measure again."

[0098] Multiple measurement data can be aggregated and processed on-site without relying on complex external software, thus forming a comprehensive three-dimensional assessment of a space or area. The automatic detection of hidden scenarios such as high-altitude operation distance, connectivity, and flow estimation overcomes the limitations of single-measurement devices, enabling the system to make global judgments when dealing with complex safety hazards.

[0099] This invention enables immediate on-site calculation of multi-point measurement data, providing a comprehensive three-dimensional assessment advantage. Its immediate determination of scenarios such as distances to high-altitude operations and workshop connectivity effectively identifies potential hazards, maintaining the ability to proactively prevent safety accidents.

[0100] Through the above technical details and processes, this invention comprehensively expands the application of laser rangefinders in safety assessment. While solving the limitations of single-point ranging, it achieves efficient automated safety detection through voice recording and multi-scene judgment algorithms, demonstrating outstanding creativity and practicality in this field.

[0101] In summary, this invention, through detailed calculation analysis and judgment principles in its specific embodiments, elucidates how to combine measured values ​​with multi-scenario calculations and standard comparisons to efficiently obtain safety evaluation conclusions. This multi-level comprehensive judgment and real-time linkage mechanism possesses outstanding inventiveness and practicality in the field, and is of great significance for improving the efficiency and accuracy of on-site safety detection.

Claims

1. A safety evaluation method based on laser ranging, characterized in that, Includes a security standards library with several pre-defined rule entries. The evaluation steps are as follows: 1) Image acquisition and edge detection: Two sets of cameras side-by-side acquire two views. Block matching is performed on the video frames to generate a disparity map, and then the disparity values ​​are... Distance from baseline and camera focal length Substitute into the following formula to calculate the depth: ， A grayscale image of the depth map is obtained based on the calculated depth. Sobel gradient calculation is performed on each grayscale image to calculate the horizontal gradient. gradient in the vertical direction And according to the formula ， After determining the gradient magnitude and filtering out noisy pixels through a threshold, the main edge segments are fitted using the least squares method. The segments that do not conform to the distance range are removed by combining the depth information. Finally, the start and end coordinates, direction angle and average depth segments are output. These segments are superimposed to select the target and start laser ranging. 2) Perform multiple measurements and calculations, select the distance measurement location, perform laser ranging, execute multiple measurements, and obtain the distance value. (Unit: meters), the dispersion is calculated using the following formula. , , , in This represents the average distance value. but As the final measurement result of the distance value, if Measure again; 3) Single-dimensional comparison: Determine the single-dimensional distance by comparing the measured single distance value. The following comparison is made with the corresponding security library limit value Li: If Then "compliant"; if If so, then it is "non-compliant". This indicates that the measured value is taken directly. Participate in the comparison; 4) Multi-dimensional synthesis and comparison: Determine the security library type of the evaluation target, perform multi-dimensional measurements, synthesize the measurement data, compare the calculated value with the value in the security library, and determine whether it is compliant.

2. The safety evaluation method based on laser ranging according to claim 1, characterized in that, The security library includes the following multi-dimensional types: Fire compartment area, F(S)=S, Li=5000m², where S represents the measured area, and F(S) indicates that the calculation formula directly takes the area value and compares it with the limit value of 5000m². The width of the evacuation route is F(R)=R, Li=1.2m, where R represents the measured width, and F(R) indicates that the calculation formula directly takes the distance value, compared with the limit value of 1.2m; The container volume is limited by F(L×W×H)=L×W×H, Li=100m³, where L, W, and H represent the length, width, and height, respectively, and F(L×W×H) represents the product of the three, compared to the limit value of 100m³.

3. The safety evaluation method based on laser ranging according to claim 1, characterized in that, Fire compartment area determination: If multiple edge detection results are defined under the same "compartment", the system will divide the coordinate points according to the polygon subdivision algorithm. Enter the following formula: ， in Indicates the area of ​​the partition, if Then it is "suspicious". Then "compliant", This is a limit value.

4. The safety evaluation method based on laser ranging according to claim 3, characterized in that, The fire protection items in the safe deposit box category also include the following types: The system determines the safe distance for working at heights by measuring the height of the work edge, and simultaneously using laser measurement data and personnel coordinates obtained from image analysis. The system then derives the vertical distance between the personnel and the danger edge. ;when Here, 2m is the safety threshold, and it will be automatically marked as "suspicious".

5. The safety evaluation method based on laser ranging according to claim 3, characterized in that, Fire safety features in safe deposit box types also include the following: comprehensive assessment of entrance layout and evacuation efficiency: measurement of multiple "evacuation doors" or "corridors," and the system will measure door widths. With corridor length Combined, perform evacuation flow estimation , in and Let be the flow coefficient, if If so, it is deemed non-compliant.

6. The safety evaluation method based on laser ranging according to claim 3, characterized in that, safety... Fire safety projects in warehouse types also include the following: compliance with section connectivity requirements, measurement of the width of multiple passageways in workshops. and connection angle If a certain corner and When marking "There is a safety hazard at this corner", it is marked as "suspicious".

7. A laser measuring instrument used in the laser ranging-based safety evaluation method according to claim 1, characterized in that, The instrument itself is triangular prism-shaped, with a "touch screen" on the front and a long "status display" above it. Two omnidirectional microphones are set in the hollowed-out edges of the prism on both sides of the status display. Below the touchscreen display are the instrument's function buttons: power on, measurement, and return / cancel. The top surface of the instrument has a triangular structure and is equipped with measurement modules, namely a binocular camera module and a laser ranging module. The binocular camera module is located on both sides of the base angle of the isosceles triangle, and the laser emission port and receiving port of the laser ranging module are located at the central axis of the isosceles triangle, with the laser emission port close to the apex angle of the isosceles triangle. A Type-C charging / data port is located at the bottom of the instrument.

8. The laser measuring instrument according to claim 7, characterized in that: The instrument has an internal control module, which includes a main control microprocessor and a control module ROM. The speech recognition and semantic standardization module is built-in, and the data analysis and standard comparison program is in the control module ROM. It communicates with the binocular camera module and the laser measurement module via a high-speed SPI bus.

9. The laser measuring instrument according to claim 7, characterized in that: The usage steps are as follows: After powering on, the control module first initializes the binocular camera module, laser measurement module, voice recognition and semantic standardization module, data analysis and standard comparison program and communication module in sequence, completing the self-test and readiness of the internal hardware and software environment; Subsequently, the binocular camera module begins to acquire on-site image data in real time, and the control module performs filtering, gradient calculation, and line segment fitting preprocessing steps. Based on parameters such as line segment length and direction clustering, the main edges that need to be measured are selected. The main control microprocessor generates guide marks on the device display interface based on the selected edge position, prompting the operator to align with the measurable target and manually press the measurement button to start laser ranging. The laser measurement module emits laser pulses and captures echo signals after being triggered by the operator, while performing temperature compensation to ensure accuracy. After the measurement is completed, the speech recognition and semantic standardization module starts listening to the operator's naming and description of the edge or measurement point through two omnidirectional microphones, converts the obtained speech information into text through a language model, and performs standardization processing with reference to the built-in lexicon, and stores it uniformly as the "measurement point ID" field corresponding to the measurement value; The data analysis and standard comparison program analyzes the output volume based on semantics. For fire compartments, it compares the output volume with the standard value of the fire compartment according to the selected standard and sends the result back to the main control microprocessor. The main control microprocessor generates a visual list on the display interface, indicating whether each measurement record is compliant or non-compliant and suspicious. Batch selection is made manually, and the data is synchronized to a remote server or other devices through the communication module to complete the entire safe measurement process based on laser ranging and automatic identification.

Images