A method for identifying and analyzing water-containing coal on a coal mine belt based on AI technology

Abstract

The present invention belongs to the technical field of water-coal identification, and specifically discloses a method for identifying and analyzing water-coal on a coal mine belt based on AI technology, including: collecting image signals of the coal conveying belt in real time to quickly monitor water-coal signals; through interference analysis of dust moisture and geological condition signals, eliminating the influence of interference signals during the coal mining process in the coal mine on the obstruction of the camera's line of sight; by detecting the gravity offset tendency of the coal conveying belt, analyzing the state signals of the coal conveying belt at each image signal acquisition moment, and screening the reference impact area and reference sputtering area of the corresponding coal flow impact at each image signal acquisition moment, which helps to quantitatively analyze the coal flow impact; by identifying the reflective components of each identification point domain to which each image signal belongs, identifying the point domain area of the water-coal component, and further analyzing the dynamic behavior of the coal flow during transportation according to the corresponding point domain areas of the impact state and sputtering state in each image signal.

Description

Technical Field

[0001] The invention belongs to the technical field of water-coal identification and relates to a method for identifying and analyzing water-coal in a coal mine belt based on AI technology. Background Art

[0002] With the increasing depth of coal mining, hydrogeological conditions are becoming more and more complex, and water inrush disasters in coal mines occur from time to time, causing serious casualties and property losses. In order to ensure the safety of coal mining and operation management, it is necessary to use various advanced image detection and recognition technologies to improve the production environment of coal mines and enhance their risk resistance. Among them, water-coal monitoring is an important part of coal mine monitoring. However, water-coal monitoring usually requires manual inspections, mainly relying on underground workers to report to the dispatching room through underground telephones after water inrush, which is inefficient and information is delayed. Therefore, the development of an automatic water-coal monitoring system has important practical significance and application value.

[0003] There are also some solutions related to water-coal identification in the existing technology, but there are still the following limitations: the existing technology does not take interference signals into consideration when identifying water-coal signals. Due to the influence of vibration and dust accumulation during the mining process in coal mines, the water-coal detection signal has a certain degree of camera line of sight obstruction, and the actual water-coal content may be higher than the content detected in the collected image.

[0004] In addition, when identifying the water content of coal on the coal transport belt, the existing technology does not conduct in-depth analysis based on the impact and sputtering state of the coal flow and the reflective state of the coal blocks. Due to the different weight distribution of the coal blocks and the running speed of the belt, different impact and sputtering effects will be produced. These dynamic changes put higher requirements on the capture and processing of the image. If this factor is not fully considered, inaccurate recognition results may result. Summary of the invention

[0005] In view of this, in order to solve the problems raised in the above background technology, a method for identifying and analyzing water coal in coal mine belts based on AI technology is proposed.

[0006] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a method for identifying and analyzing water and coal in a coal mine belt based on AI technology, including: E1, image acquisition unit: using an industrial camera to collect a real-time picture of the water and coal situation on the coal transport belt, and transmit it as an image signal to an image processing unit.

[0007] E2, image processing unit: receiving the coal transport image set sent by the image acquisition unit, and numbering each image signal in the coal transport image set as , and detect the image interference signal and transmit it to the water-coal signal detection unit.

[0008] E3. Determine the alarm trigger mechanism of the terminal device.

[0009] E4. Belt status detection: Detect the operating data of the coal conveyor belt at each image signal acquisition moment, generate the coal conveyor belt status signal at each image signal acquisition moment, and import it into the water-coal signal detection unit.

[0010] E5. Water-coal signal detection unit: Monitor the water-coal signals in each image signal. When continuous water-coal signals are detected in the coal transportation image set, generate an alarm signal.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By collecting the image signals of the coal conveyor belt in real time, the present invention monitors and analyzes the coal conveyor belt in the coal mine underground in real time, can quickly monitor the water-coal signals, and solves the problems existing in manual inspection.

[0012] (2) By performing interference analysis on the dust-wet condition signals and geological condition signals, the present invention can identify and eliminate the signal attenuation and distortion caused by interference factors, exclude the influence of the interference signals on the camera line of sight during the coal mining process underground, improve the accuracy and reliability of the water-coal signal monitoring, thereby improving the accurate grasp of the on-site situation by the production scheduling personnel. At the same time, by determining the alarm trigger mechanism of the terminal device, the invalidity of the device detection of the industrial camera can be found in time.

[0013] (3) By detecting the gravity offset tendency of the coal conveyor belt at each image signal acquisition moment, and combining the acceleration at each image signal acquisition moment, analyzing the coal conveyor belt status signal at each image signal acquisition moment, the dynamic impact of the coal flow impact on the belt can be accurately captured. Based on this, the reference impact area and reference sputtering area of the corresponding coal flow impact at each image signal acquisition moment can be screened, and the coal flow impact can be quantitatively analyzed to determine the reference impact area and reference sputtering area, providing an accurate basis for subsequent monitoring and early warning.

[0014] (4) By identifying the reflective components in each identification point domain to which each image signal belongs, identifying the point domain area of the water-coal component, and according to the corresponding point domain area of each reflective point domain in the impact state and sputtering state in each image signal, further analyzing the dynamic behavior of the coal flow during transportation, especially the flow characteristics of the water-coal mixture, which helps to more accurately judge the presence and proportion of water-coal. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention.

[0017] Figure 2 This is a physical diagram of the water-coal situation of the present invention.

[0018] Figure 3 This is a schematic diagram of the content of the image interference signal of the present invention. Detailed implementation manners

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] Please refer to Figure 1 , Figure 3 As shown, the present invention provides a method for identifying and analyzing water-coal on a coal mine belt based on AI technology, including: E1. Image acquisition unit: The physical diagram of the water-coal situation on the coal conveying belt is collected in real time through an industrial camera and transmitted to the image processing unit as an image signal.

[0021] Please refer to Figure 2 As shown, in a preferred implementation manner, the image interference signal includes a dust and humidity condition signal and a geological condition signal.

[0022] The dust and humidity condition signal includes a dust concentration signal and a water vapor content signal.

[0023] The geological condition signal includes a transportation distortion signal and a route structure signal.

[0024] The present invention can perform real-time monitoring and analysis of the coal conveying belt in the coal mine by collecting the image signal of the coal conveying belt in real time, quickly monitor the water-coal signal, and solve the problems existing in manual inspection.

[0025] E2. Image processing unit: Receive the set of coal conveying images sent by the image acquisition unit, number each image signal in the set of coal conveying images as , and detect the image interference signal and transmit it to the water-coal signal detection unit.

[0026] In a preferred embodiment, the detection of image interference signals includes: E2-11. Analyzing dust concentration signals: Obtain the image signal acquisition times in the coal transportation image set, monitor the dust distribution spectral images in the area near the coal transportation belt at each image signal acquisition time through a portable spectrometer, and evenly divide the area near the coal transportation belt into sub-areas, denoted as each micro-region, and identify the dust concentration in each corresponding micro-region of the dust distribution spectral image at each image signal acquisition time. , and then take as the dust fuzzy signal in each corresponding micro-region at each image signal acquisition time, where represents the preset reference concentration of dust, represents the micro-region number, .

[0027] The dust concentration in each micro-region is determined by the image chromaticity value of the corresponding micro-region part in the dust distribution spectral image. Specifically, identify the image chromaticity value of each micro-region part from the dust distribution spectral image, and match the image chromaticity value of each micro-region part with the corresponding chromaticity value of each preset dust concentration to obtain the dust concentration in each micro-region.

[0028] E2-12. Analyzing water vapor content signals: Detect the water vapor content in each corresponding micro-region at each image signal acquisition time through infrared spectroscopy , and detect the lens wetting characteristics of the industrial camera at each image signal acquisition time , and evaluate the water vapor fuzzy signal in each corresponding micro-region at each image signal acquisition time , where represents the preset reference content of water vapor.

[0029] Specifically, scan the area within the image acquisition range through a sensor, obtain the radiation data in each band and present it in the form of an image, where each pixel represents the spectral radiation information of a certain area. Water vapor has strong absorption characteristics in certain bands of the infrared spectrum. Therefore, the sensor can identify the water vapor content by measuring the radiation intensity in these bands.

[0030] The method for obtaining the lens wetting characteristics of the industrial camera at each image signal acquisition time is specifically as follows: Scan the industrial camera lens through sensor technology, identify the water vapor content on the surface of the industrial camera lens at each image signal acquisition time, and then take the ratio of the water vapor content on the surface of the industrial camera lens at each image signal acquisition time to the preset reference water vapor content to obtain the lens wetting characteristics of the industrial camera at each image signal acquisition time.

[0031] E2-13. Analyze the dust and humidity condition interference rate of each image signal through the relational formula , where , where respectively represent the corresponding interference ratios of the preset dust fuzzy signal and water vapor fuzzy signal.

[0032] In a further preferred embodiment, the detection of image interference signals further includes: E2-21. Analyze the transportation distortion signal: Obtain the engineering load at each image signal acquisition moment in the coal mine belt transportation area , and generate the transportation distortion signal at each image signal acquisition moment , where represents the preset reference load of the project.

[0033] The engineering load is specifically the unit construction volume of the coal mine belt transportation project per unit time.

[0034] E2-22. Analyze the route structure signal: Install a laser rangefinder and an angle sensor on the coal transportation belt to monitor the changes in the corresponding slope and curvature of the coal transportation belt in real time, and extract the route slope of the coal transportation belt at each image signal acquisition moment and the route curvature , and generate the corresponding route distortion signal at each image signal acquisition moment , where respectively represent the preset reference route slope and reference route curvature.

[0035] E2-23. Analyze the geological condition interference rate of each image signal through the relational formula , where respectively represent the corresponding interference ratios of the preset transportation distortion signal and route distortion signal. respectively represent the corresponding interference ratios of the preset transportation distortion signal and route distortion signal.

[0036] E3. Discriminate the alarm trigger mechanism of the terminal device.

[0037] In a preferred embodiment, the specific content of the alarm trigger mechanism for discriminating the terminal device is: Compare the dust and humidity condition interference rate of each image signal with the preset dust and humidity condition interference rate threshold. When the dust and humidity condition interference rate of each image signal exceeds the preset dust and humidity condition interference rate threshold, trigger the alarm mechanism at the installation position of the industrial camera.

[0038] By analyzing the interference of the dust and humidity condition signals and the geological condition signals, the present invention can identify and eliminate the signal attenuation and distortion caused by interference factors, exclude the influence of the interference signal on the camera line of sight during the coal mine underground mining process, improve the accuracy and reliability of the water-coal signal monitoring, thereby improving the accurate grasp of the on-site situation by production scheduling personnel. At the same time, by discriminating the alarm trigger mechanism of the terminal device, it can timely detect the invalidity of the device detection of the industrial camera.

[0039] E4. Belt status detection: Install a sensor device on the coal conveyor belt to detect the operating data of the coal conveyor belt at each image signal acquisition moment, generate the coal conveyor belt status signal at each image signal acquisition moment, and import it into the water-coal signal detection unit.

[0040] The water-coal signal refers to the proportion of the water content in the coal on the coal conveyor belt.

[0041] In a preferred embodiment, the operating data of the coal conveyor belt at each image signal acquisition moment includes the acceleration and actual offset amplitude of the coal conveyor belt at each image signal acquisition moment.

[0042] In a further preferred embodiment, the specific generation method of the coal conveyor belt status signal at each image signal acquisition moment is as follows: Identify the position, brightness component, and saturation component of each micro-region corresponding to each image signal through image recognition technology, obtain each reflective point region in each image signal from them, integrate adjacent reflective point regions, and obtain the corresponding point region area and point region position of each reflective point region in each image signal.

[0043] The reflective point region is specifically a micro-region where the brightness component is significantly higher than the surrounding region and the saturation component is lower than the preset saturation threshold.

[0044] Obtain the belt surface monitoring area of the coal conveyor belt in the image signal, divide the area belonging to the gravity offset direction, for example, taking the midline of the belt surface monitoring area as the dividing line, divide the belt surface monitoring area into the area belonging to the left offset direction and the area belonging to the right offset direction, mark each identification point region where the reflective component is a coal block within the corresponding monitoring area of each image signal, and based on the volume analysis method, detect the gravity offset trend of the coal conveyor belt at each image signal acquisition moment, and obtain the gravity offset difference volume at each image signal acquisition moment. 。

[0045] Specifically, the volume analysis method refers to statistically calculating the total volume of coal blocks in the area belonging to the left offset direction and the area belonging to the right offset direction within the monitoring area, comparing and screening out the maximum value, and then obtaining the gravity offset direction to which the maximum value belongs, which is the gravity offset trend of the coal conveyor belt at the image signal acquisition moment.

[0046] The gravity offset difference volume is the difference between the total volume of coal blocks in the area belonging to the left offset direction and the area belonging to the right offset direction.

[0047] Analyze the expected offset amplitude of the coal conveyor belt at the corresponding acceleration at each image signal acquisition moment. Furthermore, determine the coal conveyor belt status signal at each image signal acquisition moment. where represents the preset reference difference volume. represents the The acceleration at each image signal acquisition moment, represents a preset reference acceleration, represents the preset offset amplitude corresponding to the unit difference volume of the coal conveyor belt, represents the actual offset amplitude of the coal conveyor belt at each image signal acquisition moment.

[0048] By detecting the gravity offset trend of the coal conveyor belt at each image signal acquisition moment, combining the acceleration at each image signal acquisition moment, and analyzing the state signal of the coal conveyor belt at each image signal acquisition moment, the present invention can accurately capture the dynamic impact of coal flow impact on the belt. Based on this, the reference impact area and reference sputtering area of the corresponding coal flow impact at each image signal acquisition moment are screened, and the coal flow impact can be quantitatively analyzed to determine the reference impact area and reference sputtering area, providing an accurate basis for subsequent monitoring and early warning.

[0049] E5. Water coal signal detection unit: Monitor the water coal signal in each image signal, and generate an alarm signal when continuous water coal signals are detected in the coal transportation image set.

[0050] In a preferred embodiment, the monitoring of the water coal signal in each image signal includes: comparing the area of each light reflection point domain in each image signal with a preset reference area, screening out each light reflection point domain in each image signal whose point domain area exceeds the preset reference area, and recording them as each identification point domain belonging to each image signal. Then, the light reflection components of each identification point domain belonging to each image signal are identified. The light reflection components are divided into coal blocks and water coal. If the boundary texture characteristics preset for coal blocks exist in the boundary area of a certain identification point domain belonging to a certain image signal, the light reflection component of this identification point domain belonging to this image signal is a coal block; otherwise, the light reflection component of this identification point domain belonging to this image signal is water coal. Based on this, each identification point domain in each image signal whose light reflection component is water coal is obtained, and the corresponding point domain areas of each identification point domain in each image signal whose light reflection component is water coal are statistically analyzed. 。

[0051] Compare the corresponding point domain positions of each identification point domain in each image signal whose light reflection component is water coal with each other, and identify the average distance between the corresponding point domain positions of all identification point domains in each image signal. 。

[0052] Due to the bumps of the conveyor belt caused by uneven distribution of coal blocks, there is a situation of coal flow impact sputtering in coal mines with a relatively large moisture content. Therefore, based on each light reflection point domain in each image signal, the coal flow impact rate in each image signal is detected. , which is one of the evaluation methods for inferring the moisture content in coal mines, and helps to increase the stability and uniformity of coal flow detection.

[0053] Determine the water coal signal in each image signal , where represents a preset reference area represents a preset reference distance, and e is the natural constant

[0054] In a further preferred embodiment, detecting the coal flow impact rate in each image signal includes: identifying the relative states of the corresponding point areas of each reflective point area in each image signal and the position of the coal conveyor belt, including the impact state and the sputtering state, and counting the corresponding point areas of each reflective point area in the impact state and the sputtering state in each image signal, respectively denoted as , represents the number of the reflective point area .

[0055] The sputtering state refers to the disconnection of the corresponding point area of the reflective point area in the image signal and the position of the coal conveyor belt, and the impact state refers to the connection of the corresponding point area of the reflective point area in the image signal and the position of the coal conveyor belt

[0056] Match the coal conveyor belt state signal at the acquisition time of each image signal with the corresponding reference impact area and reference sputtering area of each preset coal conveyor belt state signal, and screen out the reference impact area and the reference sputtering area .

[0057] Combined with the geological condition interference rate of each image signal, analyze the coal flow impact rate in each image signal, where represents that there is no sputtering state in the relative state of the corresponding point area of each reflective point area in each image signal and the position of the coal conveyor belt represents that there is a sputtering state in the relative state of the corresponding point area of each reflective point area in each image signal and the position of the coal conveyor belt

[0058] In a further preferred embodiment, the specific identification method of the continuous water coal signal is: compare the water coal signal in each image signal with a preset water coal signal threshold. When the water coal signals in each image signal all exceed the preset water coal signal threshold, it indicates that there is a continuous water coal signal in the coal transport image set

[0059] The present invention identifies the reflective components of each identification point area of each image signal, determines the point area situation of the water coal components, and further analyzes the dynamic behavior of the coal flow during transportation, especially the flow characteristics of the water coal mixture, based on the corresponding point area of each reflective point area in the impact state and the sputtering state in each image signal. This helps to more accurately judge the presence and proportion of the water coal

[0060] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of the technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A method for identifying and analyzing water-containing coal on a coal mine belt based on AI technology, characterized in that, The method includes the following steps: E1. Image acquisition unit: A physical image of the water - coal situation on the coal - conveying belt is acquired in real - time by an industrial camera and transmitted as an image signal to the image - processing unit. E2. Image processing unit: Receives the coal transportation image set sent by the image acquisition unit, numbers each image signal in the coal transportation image set as , and detects the image interference signal and transmits it to the water-coal signal detection unit; E3. Discriminate the alarm - triggering mechanism of the terminal device. E4. Belt status detection: Detect the operating data of the coal - conveying belt at each image - signal acquisition moment and generate the coal - conveying belt status signal at each image - signal acquisition moment, and import it into the water - coal signal detection unit. E5. Water - coal signal detection unit: Monitor the water - coal signal in each image signal. When continuous water - coal signals are detected in the coal - conveying image set, generate an alarm signal. Monitoring the water-coal signals in each image signal, the content includes: comparing the area of each light-reflecting point domain in each image signal with a preset reference area, screening out the light-reflecting point domains in each image signal whose point domain area exceeds the preset reference area, and recording them as the respective identification point domains belonging to each image signal. Then, identify the light-reflecting components of the respective identification point domains belonging to each image signal. The light-reflecting components are divided into coal blocks and water-coal. If there are boundary texture features preset for coal blocks in the boundary area of a certain identification point domain belonging to a certain image signal, the light-reflecting component of this identification point domain belonging to this image signal is a coal block; otherwise, it is water-coal. Statistically calculate the corresponding point domain areas of the respective identification point domains with light-reflecting components of water-coal in each image signal ; Compare the corresponding point domain positions of the identification point domains where the specular components in each image signal are water coal, and identify the mean distance between the corresponding point domain positions of all the identification point domains in each image signal ; Detect the coal flow impact rate in each image signal based on each specular point region in each image signal ; Determine the water-coal signal in each image signal , where represents a preset reference area, represents a preset reference distance, and e is the natural constant.

2. The coal and water identification and analysis method for coal mine belt based on AI technology according to claim 1, characterized in that The image interference signals include dust - wet condition signals and geological condition signals. The dust - wet condition signals include dust - concentration signals and water - vapor content signals. The geological condition signals include transportation distortion signals and route - structure signals.

3. The coal and water identification and analysis method for coal mine belt based on AI technology according to claim 2, characterized in that, The detection of the image interference signals includes: E2-11. Analyze the dust concentration signal: Obtain the image signal acquisition moments in the coal transportation image set, monitor the dust distribution spectral images in the area near the coal transportation belt at each image signal acquisition moment through a portable spectrometer, divide the area near the coal transportation belt into each micro-region, and identify the dust concentration in each micro-region corresponding to the dust distribution spectral image at each image signal acquisition moment. , and then is used as the dust fuzzy signal in each micro-region corresponding to each image signal acquisition moment, where represents the preset reference concentration of dust, represents the number of the micro-region, ; E2-12. Analyze the water vapor content signal: Detect the water vapor content of each micro-region corresponding to the acquisition time of each image signal through infrared spectroscopy , and detect the lens wetting characteristics of the industrial camera at the acquisition time of each image signal , evaluate the water vapor blur signal of each micro-region corresponding to the acquisition time of each image signal , where represents the preset reference content of water vapor; E2-13. Analyze the dust and humidity condition interference rate of each image signal through the relational formula , where respectively represent the corresponding interference proportions of the preset dust fuzzy signal and water vapor fuzzy signal .

4. The coal and water identification and analysis method for coal mine belt based on AI technology according to claim 3, characterized in that The detection of the image interference signals also includes: E2-21. Analyze the transportation distortion signal: Obtain the engineering load of the coal mine belt transportation area at each image signal acquisition moment, and generate the transportation distortion signal at each image signal acquisition moment ; E2-22. Analyze the route structure signal: Install a laser rangefinder and an angle sensor on the coal conveyor belt to monitor the changes in the corresponding slope and curvature of the coal conveyor belt in real time, extract the route slope and route curvature at each image signal acquisition moment of the coal conveyor belt, and generate the corresponding route distortion signal at each image signal acquisition moment ; E2-23. Geological condition interference rate of each image signal is analyzed through the relational formula , where respectively represent the corresponding interference ratios of the preset transportation distortion signal and the route distortion signal .

5. The coal and water identification and analysis method for coal mine belt based on AI technology according to claim 4, characterized in that The specific content of the alarm - triggering mechanism of the terminal device is as follows: Compare the dust - wet condition interference rate of each image signal with the preset dust - wet condition interference - rate threshold. When the dust - wet condition interference rate of each image signal exceeds the preset dust - wet condition interference - rate threshold, trigger the alarm mechanism at the installation position of the industrial camera.

6. The coal and water identification and analysis method for coal mine belt based on AI technology according to claim 1, characterized in that, The operating data of the coal - conveying belt at each image - signal acquisition moment includes the acceleration and the actual offset amplitude of the coal - conveying belt at each image - signal acquisition moment.

7. The method for identifying and analyzing water-containing coal on a coal mine belt based on AI technology according to claim 6, wherein The specific generation method of the coal - conveying belt status signal at each image - signal acquisition moment is as follows: Obtain each reflective - point domain in each image signal, integrate adjacent reflective - point domains, and obtain the corresponding point - domain area and point - domain position of each reflective - point domain in each image signal. Taking the transverse area of the belt as the monitoring area, dividing the area belonging to the gravity offset direction, marking each identification point area where the reflective component within the corresponding monitoring area of each image signal is a coal block, based on the volume analysis method, detecting the gravity offset tendency of the coal conveyor belt at each image signal acquisition moment, and obtaining the gravity offset differential volume at each image signal acquisition moment ; Analyze the expected offset amplitude of the coal conveyor belt at the corresponding acceleration at each image signal acquisition moment , and then determine the state signal of the coal conveyor belt at each image signal acquisition moment , where represents the preset reference difference volume, represents the th acceleration at the image signal acquisition moment, represents the preset reference acceleration, represents the preset offset amplitude corresponding to the unit difference volume of the coal conveyor belt, represents the actual offset amplitude of the coal conveyor belt at each image signal acquisition moment.

8. A method for identifying and analyzing water coal on a coal mine belt based on AI technology according to claim 1, characterized in that, Detecting the coal flow impact rate in each image signal includes: identifying the relative states of the corresponding point regions of each reflective point region in each image signal with respect to the position of the coal conveyor belt, including the impact state and the sputtering state, and counting the corresponding point region areas of each reflective point region in the impact state and the sputtering state in each image signal, respectively denoted as , indicating the number of the reflective point region, ; Match the coal conveying belt status signals at each image signal acquisition moment with the corresponding reference impact area and reference sputtering area of each preset coal conveying belt status signal, and screen out the reference impact area of the corresponding coal flow impact at each image signal acquisition moment and the reference sputtering area ; Interference rate of geological conditions combining with each image signal , analyze the coal flow impact rate in each image signal , where indicates that the relative state of the corresponding point area of each reflective point area in each image signal and the position of the coal conveying belt does not have a sputtering state, indicates that the relative state of the corresponding point area of each reflective point area in each image signal and the position of the coal conveying belt has a sputtering state.

9. The coal and water identification and analysis method for coal mine belt based on AI technology according to claim 1, characterized in that, The specific recognition method of the continuous water - coal signals is as follows: Compare the water - coal signals in each image signal with the preset water - coal signal threshold. When the water - coal signals in each image signal exceed the preset water - coal signal threshold, it indicates that continuous water - coal signals exist in the coal - conveying image set.

Images