Intelligent Recognition Method and Device for Video Images of Longwall Mining Face

Abstract

This disclosure proposes an intelligent recognition method and device for video images from fully mechanized mining faces, relating to the field of coal mine safety image recognition technology. The method includes: acquiring video monitoring images of the fully mechanized mining face, including images acquired by an image acquisition device mounted on a hydraulic support; identifying the fully mechanized mining face based on the video monitoring images to obtain successfully identified target objects and assigning them numbers; acquiring feature data of the target objects and updating their status information based on the feature data; and generating a display image based on the updated status information, the target object's number, the video monitoring images, and the physical number of the hydraulic support. Thus, by acquiring video monitoring images, identifying target objects, and obtaining their feature data, the status information of target objects can be accurately monitored and updated in real time, increasing the efficiency and accuracy of video monitoring recognition.

Description

Technical Field

[0001] This disclosure relates to the field of coal mine safety image recognition technology, and in particular to a method and device for intelligent recognition of video images from fully mechanized mining faces. Background Technology

[0002] The fully mechanized longwall face is the main production site in underground coal mines. Hydraulic supports are the support equipment for the longwall face, which support the roof of the working face and ensure the safety of the working space.

[0003] During the support and advancement of the working face, hydraulic supports are monitored by corresponding sensors. Column pressure sensors detect the operating condition of the hydraulic support's supporting columns, while push-stroke sensors detect the operating condition of the hydraulic support's pushing jacks. These sensor devices can only monitor the operating condition of the hydraulic support itself and cannot achieve spatial correspondence between the hydraulic support and the working face floor, the hydraulic support and the scraper conveyor, or the hydraulic support and the coal mining machine. Therefore, a method is needed to achieve comprehensive monitoring of the spatial position of the hydraulic supports in the fully mechanized mining face relative to the mining area and other fully mechanized mining equipment.

[0004] Public content

[0005] This disclosure aims to at least partially address one of the technical problems in the related art.

[0006] Therefore, one objective of this disclosure is to propose an intelligent recognition method for video images of fully mechanized mining faces.

[0007] The second objective of this disclosure is to propose an intelligent recognition device for video images of fully mechanized mining faces.

[0008] The third objective of this disclosure is to propose an electronic device.

[0009] The fourth objective of this disclosure is to provide a non-transitory computer-readable storage medium.

[0010] The fifth objective of this disclosure is to provide a computer program product.

[0011] To achieve the above objectives, the first aspect of this disclosure proposes an intelligent recognition method for video images of a fully mechanized mining face, comprising: acquiring video monitoring images of the fully mechanized mining face, the video monitoring images including images acquired by an image acquisition device mounted on a hydraulic support; identifying the fully mechanized mining face based on the video monitoring images to obtain successfully identified target objects and assigning a number to the target objects; acquiring feature data of the target objects and updating the status information of the target objects based on the feature data; and generating a display image based on the updated status information, the number of the target objects, the video monitoring images, and the physical number of the hydraulic support.

[0012] According to one embodiment of this disclosure, identifying a mining face based on video surveillance images to obtain a successfully identified target object includes: obtaining candidate objects to be identified; matching the candidate objects with video surveillance images, and using the successfully matched candidate objects as target objects.

[0013] According to one embodiment of this disclosure, the candidate object includes a hydraulic support base. Obtaining feature data of the target object includes: when the hydraulic support base is the target object, determining the floor height of the hydraulic support base from the working surface base plate based on a video surveillance image, and using the floor height as feature data of the hydraulic support base.

[0014] According to one embodiment of this disclosure, updating the state information of a target object based on feature data includes: obtaining historical floor height; in response to a floor height greater than the historical floor height and a floor height less than a height threshold, updating the state information of the hydraulic support base to a bottom-scraping and frame-moving state; and in response to a floor height greater than the historical floor height and a floor height greater than or equal to a height threshold, updating the state information of the hydraulic support base to a high-lift bottom-moving and frame-moving state.

[0015] According to one embodiment of this disclosure, the candidate object includes a pedestrian passage space. Acquiring feature data of the target object includes: when the pedestrian passage space is the target object, determining the interval distance between the image acquisition device and the pedestrian passage space based on video surveillance images, and using the interval distance as feature data of the pedestrian passage space.

[0016] According to one embodiment of this disclosure, updating the state information of a target object based on feature data includes: acquiring historical interval distances; updating the state information of the pedestrian passage space to a hydraulic support forward movement state in response to a historical interval distance being greater than the interval distance; and updating the state information of the pedestrian passage space to a hydraulic support forward movement position error state in response to a historical interval distance being less than or equal to the interval distance and the interval distance not being within a preset set interval distance.

[0017] According to one embodiment of this disclosure, the candidate object includes a coal mining machine drum. Obtaining feature data of the target object includes: when the coal mining machine drum is the target object, determining the drum distance between the coal mining machine drum and the hydraulic support, and using the drum distance as feature data of the coal mining machine drum.

[0018] According to one embodiment of this disclosure, updating the state information of a target object based on feature data includes: comparing the drum distance with a set support distance; in response to the drum distance being less than the set support distance, updating the state information of the coal mining machine drum to a state of collision tendency; and in response to the drum distance being greater than or equal to the set support distance, updating the state information of the coal mining machine drum to a normal state.

[0019] According to one embodiment of this disclosure, the intelligent recognition method for video images of fully mechanized mining faces further includes: generating reminder information based on status information.

[0020] To achieve the above objectives, a second aspect of this disclosure provides an intelligent recognition device for video images of a fully mechanized mining face, comprising: a data acquisition module for acquiring video surveillance images of the fully mechanized mining face, the video surveillance images including images acquired by an image acquisition device mounted on a hydraulic support; a recognition module for recognizing the fully mechanized mining face based on the video surveillance images to obtain successfully recognized target objects and assigning a number to the target objects; an update module for acquiring feature data of the target objects and updating the status information of the target objects based on the feature data; and a generation module for generating a display image based on the updated status information, the number of the target objects, the video surveillance images, and the physical number of the hydraulic support.

[0021] To achieve the above objectives, a third aspect of this disclosure provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to implement the intelligent recognition method for video images of fully mechanized mining faces as described in the first aspect of this disclosure.

[0022] To achieve the above objectives, a fourth aspect of this disclosure provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to implement the intelligent recognition method for video images of fully mechanized mining faces as described in the first aspect of this disclosure.

[0023] To achieve the above objectives, a fifth aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, is used to implement the intelligent recognition method for video images of fully mechanized mining faces as described in the first aspect of this disclosure.

[0024] Therefore, by acquiring video surveillance images, identifying target objects, and obtaining their feature data, the status information of target objects can be accurately monitored and updated in real time, increasing the efficiency and accuracy of video surveillance recognition. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of an intelligent recognition method for video images of a fully mechanized mining face according to one embodiment of the present disclosure;

[0026] Figure 2 This is a schematic diagram of another intelligent recognition method for video images of a fully mechanized mining face according to one embodiment of this disclosure;

[0027] Figure 3 This is a schematic diagram of the overall process of an intelligent recognition method for video images of a fully mechanized mining face according to one embodiment of the present disclosure;

[0028] Figure 4 This is a schematic diagram of an intelligent video image recognition device for a fully mechanized mining face according to one embodiment of the present disclosure;

[0029] Figure 5 This is a schematic diagram of an electronic device according to one embodiment of the present disclosure. Detailed Implementation

[0030] Embodiments of this disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting this disclosure.

[0031] Figure 1 This is a schematic diagram illustrating an exemplary implementation of a method for intelligent recognition of video images from a fully mechanized mining face, as proposed in this disclosure. Figure 1 As shown, the intelligent recognition method for video images of the fully mechanized mining face includes the following steps:

[0032] S101, collects video monitoring images of the fully mechanized mining face, including images collected by an image acquisition device mounted on a hydraulic support.

[0033] It should be noted that the video surveillance images of the fully mechanized mining face can be real-time or historically collected; no restrictions are imposed here.

[0034] It should be noted that the image acquisition device can be set up in multiple locations on the fully mechanized mining face, without any limitation, but it must include the image acquisition device set up on the hydraulic support, which is used to acquire two types of video monitoring images: those facing the support arrangement direction and those facing the coal wall. Image acquisition devices at other locations can assist in acquiring the two types of video monitoring images: those facing the support arrangement direction and those facing the coal wall.

[0035] The image acquisition device can be of various types, such as explosion-proof cameras, underground cameras, etc., without any limitation.

[0036] S102, based on video surveillance images, identifies the integrated mining face to obtain successfully identified target objects and assigns them numbers.

[0037] In this embodiment of the disclosure, there are various methods for identifying the longwall mining face based on video surveillance images. Optionally, the acquired video images can be manually marked, and the hydraulic support base, pedestrian passage area, coal mining machine travel area, and coal wall space area directly opposite the hydraulic support can be manually selected and marked. For example, scenes in video images where displacement occurs between the hydraulic support base and the working face floor can be manually annotated, and image recognition algorithms can be used to extract feature points to identify the positional relationship between the support base and the working face floor. Scenes in video images where pedestrian walkway space changes during the forward movement of hydraulic supports can be manually annotated, and image recognition algorithms can be used to extract spatial feature points of the pedestrian walkway to identify the change process of the pedestrian walkway corresponding to the hydraulic supports. Scenes in video images where the distance between the drum and the support top beam during coal mining machine cutting can be manually annotated, and image recognition algorithms can be used to extract the correlation feature points between the coal mining machine drum and the support top beam to identify the passage process of the coal mining machine drum. Scenes in video images where workers appear in pedestrian walkways can be manually marked, and image recognition algorithms can be used to extract feature points of the workers in the pedestrian walkways to identify the workers' position information, etc.

[0038] Optionally, multiple candidate objects to be identified can be set, and the target object can be located from the video surveillance image based on image processing technology. Compared with manual labeling, image processing technology can greatly save labor costs and improve the speed of localization, although its robustness is relatively poor. Therefore, a suitable identification method can be selected based on actual needs.

[0039] It should be noted that successfully identified target objects can be of the same type or include multiple types. For example, a video surveillance image may contain multiple pedestrian walkway areas. When numbering, different target objects can follow the same numbering rule or different numbering rules; target objects of the same type can follow the same numbering rule. For example, hydraulic support bases can be numbered with numbers, such as 001, 002, 003, etc., while pedestrian walkway areas can be numbered with letters and numbers, such as a1, a2, a3, etc. No restrictions are placed here. This facilitates subsequent data processing and differentiation.

[0040] S103, acquire the feature data of the target object, and update the state information of the target object based on the feature data.

[0041] In this embodiment, the feature data corresponding to different target objects may be different, and the specific details need to be defined according to the actual design requirements. For example, the feature data corresponding to the hydraulic support base may be the height of the hydraulic support base from the floor of the working face; the feature data corresponding to the pedestrian passage space may be the distance between the image acquisition device and the pedestrian passage space; and the feature data corresponding to the coal mining machine drum may be the distance between the coal mining machine drum and the hydraulic support drum, etc.

[0042] After acquiring the feature data, it can be processed to determine the operational status of the target object, and the status information of the target object can be updated based on the operational status. Optionally, the feature data can be compared with normal data to determine the operational status.

[0043] Optionally, the job status can be determined by looking up the feature data in a table. It should be noted that the job status table can be pre-set and contains the mapping relationship between feature data and job status.

[0044] S104 generates a display image based on the updated status information, the target object's number, the video surveillance image, and the physical number of the hydraulic support.

[0045] In this embodiment of the disclosure, after obtaining the updated status information, the updated status information, the target object's number, and the physical number of the hydraulic support can be bound together and marked to the corresponding position in the corresponding video surveillance image to generate a display image. It should be noted that the display images corresponding to different hydraulic supports can be different. They can be displayed separately based on the physical number of the hydraulic supports, or they can be displayed in different areas on the same screen. No limitation is made here.

[0046] In this embodiment, video surveillance images of the fully mechanized mining face are first acquired. These images include those captured by an image acquisition device mounted on a hydraulic support. Then, the fully mechanized mining face is identified based on the video surveillance images to identify successfully identified target objects, which are then assigned a number. Next, the feature data of the target objects is acquired, and their status information is updated based on this feature data. Finally, a display image is generated based on the updated status information, the target object's number, the video surveillance images, and the physical number of the hydraulic support. Thus, by acquiring video surveillance images, identifying target objects, and obtaining their feature data, the status information of target objects can be accurately monitored and updated in real time, increasing the efficiency and accuracy of video surveillance identification.

[0047] In this embodiment of the disclosure, the scene of workers appearing in the pedestrian passage in the video image can also be manually calibrated based on the displayed image. Image recognition algorithms can be used to extract feature points of the workers in the pedestrian passage in the video image, thereby identifying and determining the location information of the workers. If an abnormal state of the target object occurs, alarm information can be sent to the corresponding workers to improve the safety of workers working underground and the efficiency of information acquisition.

[0048] In this embodiment of the disclosure, after obtaining the display image, the display image can also be displayed on the terminal, such as through a relevant website or through a relevant mobile application, so that operators or dispatchers can conveniently obtain underground operation information.

[0049] In the above embodiments, the integrated mining face is identified based on video surveillance images to obtain successfully identified target objects. It can also be achieved through... Figure 2 To further explain, the method includes:

[0050] S201, Obtain the candidate object to be identified.

[0051] It should be noted that the candidate objects in this embodiment are pre-set and can be set according to actual identification needs. No limitations are made here. For example, the candidate objects may include the hydraulic support base, pedestrian passage area, coal mining machine walking area, and coal wall space directly opposite the hydraulic support in the image.

[0052] S202, Match the candidate object with the video surveillance image, and take the successfully matched candidate object as the target object.

[0053] In this embodiment of the disclosure, candidate objects can be matched with video surveillance images based on image recognition technology, and the successfully matched candidate objects can be used as target objects.

[0054] It should be noted that image recognition, also known as pattern recognition, involves extracting features from an image, classifying the image based on its geometric and textural features, and performing structural analysis on the entire image. Typically, image preprocessing is performed before recognition, including filtering noise and interference, improving contrast, enhancing edges, and geometric correction.

[0055] In this embodiment, candidate objects to be identified are first obtained, and then matched with video surveillance images. Successfully matched candidate objects are selected as target objects. Therefore, determining the target object based on matching candidate objects with video surveillance images allows for automatic tracking of the target object, providing a basis for judging the state information of the target object and improving the efficiency of video recognition on the work surface.

[0056] In this embodiment of the disclosure, when the hydraulic support base is the target object, the height of the hydraulic support base from the floor of the working surface is determined based on the video monitoring image, and the floor height is used as the feature data of the hydraulic support base.

[0057] Furthermore, the status information of the hydraulic support base is updated based on feature data. This can be achieved by obtaining historical floor heights. If the floor height is greater than the historical floor height and less than the height threshold, the status information of the hydraulic support base is updated to the bottom-scraping and support-moving state. If the floor height is greater than the historical floor height and greater than or equal to the height threshold, the status information of the hydraulic support base is updated to the high-lift bottom-moving and support-moving state.

[0058] In this embodiment of the disclosure, when the pedestrian passage space is the target object, the distance between the image acquisition device and the pedestrian passage space is determined based on the video surveillance image, and the distance is used as the feature data of the pedestrian passage space.

[0059] Furthermore, by updating the status information of the pedestrian passage space based on feature data, historical interval distances can be obtained. If the historical interval distance is greater than the interval distance, the status information of the pedestrian passage space is updated to the state of hydraulic support moving forward. If the historical interval distance is less than or equal to the interval distance and the interval distance is not within the preset interval distance, the status information of the pedestrian passage space is updated to the state of hydraulic support moving forward position error.

[0060] In this embodiment of the disclosure, when the target object is the coal mining machine drum, the distance between the coal mining machine drum and the hydraulic support drum is determined, and the drum distance is used as the characteristic data of the coal mining machine drum.

[0061] Furthermore, the status information of the coal mining machine drum is updated based on the feature data. The drum distance can be compared with the set support distance. If the drum distance is less than the set support distance, the status information of the coal mining machine drum is updated to a state with a collision tendency. If the drum distance is greater than or equal to the set support distance, the status information of the coal mining machine drum is updated to a normal state.

[0062] In this embodiment of the disclosure, after obtaining the status information of the target object, it is possible to analyze whether the target object is in an abnormal state based on the status information. When the target object is in an abnormal state, a corresponding reminder message is generated and displayed on the display image, or a reminder is given to personnel through other means. For example, an alarm message can be generated and sent to relevant operators, and an alarm voice can also be generated.

[0063] Figure 3 This is a schematic diagram of the overall process of the intelligent recognition method for video images of a fully mechanized mining face according to an embodiment of this disclosure, as follows: Figure 3 As shown, firstly, video surveillance images of the fully mechanized mining face are acquired. Then, based on the video surveillance images, the fully mechanized mining face is identified to obtain successfully identified target objects, which are then numbered. When the hydraulic support base is the target object, the height of the hydraulic support base from the working face floor is determined based on the video surveillance images. This floor height is used as the characteristic data of the hydraulic support base. Historical floor heights are then acquired. If the floor height is greater than the historical floor height and less than a height threshold, the status information of the hydraulic support base is updated to "bottom-scraping and support-moving state." If the floor height is greater than the historical floor height and greater than or equal to the height threshold, the status information of the hydraulic support base is updated to "high-lift bottom-scraping and support-moving state." When the pedestrian passageway is the target object, the distance between the image acquisition device and the pedestrian passageway is determined based on the video surveillance images. This distance is used as the characteristic data of the pedestrian passageway. The system then obtains historical interval distances. If the historical interval distance is greater than the current interval distance, the status information of the pedestrian passage space is updated to the state of hydraulic support forward movement. If the historical interval distance is less than or equal to the current interval distance, and the interval distance is not within the preset interval distance, the status information of the pedestrian passage space is updated to the state of hydraulic support forward movement position error. When the coal mining machine drum is the target object, the drum distance between the coal mining machine drum and the hydraulic support is determined, and the drum distance is used as the characteristic data of the coal mining machine drum. Then, the drum distance is compared with the set support distance. If the drum distance is less than the set support distance, the status information of the coal mining machine drum is updated to the state of collision tendency. If the drum distance is greater than or equal to the set support distance, the status information of the coal mining machine drum is updated to the normal state. Based on the updated status information, the target object number, the video monitoring image, and the physical number of the hydraulic support, a display image is generated.

[0064] Corresponding to the intelligent recognition method for video images of fully mechanized mining faces provided in the above embodiments, one embodiment of this disclosure also provides an intelligent recognition device for video images of fully mechanized mining faces. Since the intelligent recognition device for video images of fully mechanized mining faces provided in this disclosure corresponds to the intelligent recognition method for video images of fully mechanized mining faces provided in the above embodiments, the implementation methods of the above intelligent recognition method for video images of fully mechanized mining faces are also applicable to the intelligent recognition device for video images of fully mechanized mining faces provided in this disclosure, and will not be described in detail in the following embodiments.

[0065] Figure 4 This is a schematic diagram of an intelligent recognition device for video images of a fully mechanized mining face, as disclosed in this disclosure. Figure 4 As shown, the intelligent video image recognition device 400 for the fully mechanized mining face includes: a data acquisition module 410, a recognition module 420, an update module 430, and a generation module 440.

[0066] The acquisition module 410 is used to acquire video monitoring images of the fully mechanized mining face, including images acquired by an image acquisition device installed on the hydraulic support.

[0067] The identification module 420 is used to identify the mining face based on video surveillance images, so as to obtain the successfully identified target objects and assign numbers to the target objects.

[0068] The update module 430 is used to obtain the feature data of the target object and update the state information of the target object based on the feature data.

[0069] The generation module 440 is used to generate a display image based on the updated status information, the target object number, the video surveillance image, and the physical number of the hydraulic support.

[0070] In one embodiment of this disclosure, the identification module 420 is further configured to: obtain candidate objects to be identified; match the candidate objects with video surveillance images, and use the successfully matched candidate objects as target objects.

[0071] In one embodiment of this disclosure, the update module 430 is further configured to: determine the floor height of the hydraulic support base from the working surface base plate based on the video surveillance image when the hydraulic support base is the target object, and use the floor height as the feature data of the hydraulic support base.

[0072] In one embodiment of this disclosure, the update module 430 is further configured to: obtain historical floor height; in response to a floor height greater than historical floor height and a floor height less than a height threshold, update the status information of the hydraulic support base to the bottom-scraping and frame-moving state; in response to a floor height greater than historical floor height and a floor height greater than or equal to a height threshold, update the status information of the hydraulic support base to the high-lift bottom-moving and frame-moving state.

[0073] In one embodiment of this disclosure, the update module 430 is further configured to: when the pedestrian passage space is the target object, determine the interval distance between the image acquisition device and the pedestrian passage space based on the video surveillance image, and use the interval distance as feature data of the pedestrian passage space.

[0074] In one embodiment of this disclosure, the update module 430 is further configured to: obtain historical interval distance; in response to the historical interval distance being greater than the interval distance, update the state information of the pedestrian passage space to the state of hydraulic support moving forward; in response to the historical interval distance being equal to the interval distance and the interval distance not being within the preset set interval distance, update the state information of the pedestrian passage space to the state of hydraulic support moving forward position error.

[0075] In one embodiment of this disclosure, the update module 430 is further configured to: determine the distance between the coal mining machine drum and the hydraulic support drum when the coal mining machine drum is the target object, and use the drum distance as characteristic data of the coal mining machine drum.

[0076] In one embodiment of this disclosure, the update module 430 is further configured to: update the state information of the target object based on feature data, including: comparing the drum distance with the set support distance; in response to the drum distance being less than the set support distance, updating the state information of the coal mining machine drum to a state of collision tendency; in response to the drum distance being greater than or equal to the set support distance, updating the state information of the coal mining machine drum to a normal state.

[0077] In one embodiment of this disclosure, the update module 430 is further configured to: generate reminder information based on status information.

[0078] To implement the above embodiments, this disclosure also proposes an electronic device 500, such as... Figure 5 As shown, the electronic device 500 includes: a processor 501 and a memory 502 communicatively connected to the processor. The memory 502 stores instructions that can be executed by at least one processor. The instructions are executed by at least one processor 501 to implement the intelligent recognition method for video images of fully mechanized mining faces as described in the first aspect of this disclosure.

[0079] To implement the above embodiments, this disclosure also proposes a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to implement the intelligent recognition method for video images of fully mechanized mining faces as described in the first aspect of this disclosure.

[0080] To implement the above embodiments, this disclosure also proposes a computer program product, including a computer program that, when executed by a processor, implements the intelligent recognition method for video images of fully mechanized mining faces as described in the first aspect of this disclosure.

[0081] In the description of this disclosure, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this disclosure and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this disclosure.

[0082] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this disclosure, "a plurality of" means two or more, unless otherwise expressly specified.

[0083] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0084] Although embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present disclosure.

Claims

1. A method for intelligent recognition of video images from a fully mechanized mining face, characterized in that, include: Collect video surveillance images of the fully mechanized mining face, including images collected by an image acquisition device mounted on a hydraulic support; Candidate objects to be identified are obtained based on the video surveillance images; Based on the matching of the candidate objects with the video surveillance images, the successfully matched candidate objects are taken as target objects and numbered. The candidate objects include hydraulic support bases or pedestrian walkway spaces. The feature data of the target object is obtained. In the case that the hydraulic support base is the target object, the height of the hydraulic support base from the floor of the working surface is determined based on the video surveillance image, and the floor height is used as the feature data of the hydraulic support base. Alternatively, in the case that the pedestrian passage space is the target object, the interval distance between the image acquisition device and the pedestrian passage space is determined based on the video surveillance image, and the interval distance is used as the feature data of the pedestrian passage space. The state information of the target object is updated based on the feature data; Based on the updated status information, the target object's number, the video surveillance image, and the hydraulic support's physical number, a display image is generated.

2. The method according to claim 1, characterized in that, The step of updating the state information of the target object based on the feature data includes: When the hydraulic support base is the target object, obtain the historical floor height; In response to the floor height being greater than the historical floor height and the floor height being less than the height threshold, the status information of the hydraulic support base is updated to the bottom-scraping and support-moving state. In response to the floor height being greater than the historical floor height, and the floor height being greater than or equal to the height threshold, the status information of the hydraulic support base is updated to the high-lift and shift status.

3. The method according to claim 1, characterized in that, The step of updating the state information of the target object based on the feature data includes: When the pedestrian walkway space is the target object, obtain the historical interval distance; If the historical interval distance is greater than the interval distance, the state information of the pedestrian passage space is updated to the state of hydraulic support moving forward. If the historical interval distance is less than or equal to the interval distance, and the interval distance is not within the preset set interval distance, then the status information of the pedestrian passage space is updated to the hydraulic support forward movement position error status.

4. The method according to any one of claims 1-3, characterized in that, The method further includes: A reminder message is generated based on the status information.

5. A smart recognition device for video images of a fully mechanized mining face, characterized in that, include: The acquisition module is used to acquire video monitoring images of the fully mechanized mining face, including images acquired by an image acquisition device mounted on a hydraulic support. The identification module is used to obtain candidate objects to be identified based on the video surveillance images; Based on the matching of the candidate objects with the video surveillance images, the successfully matched candidate objects are taken as target objects and numbered. The candidate objects include hydraulic support bases or pedestrian walkway spaces. The update module is used to determine the height of the hydraulic support base from the floor of the working surface plate based on the video monitoring image when the hydraulic support base is the target object, and to use the floor height as the feature data of the hydraulic support base. or, When the pedestrian walkway space is the target object, the distance between the image acquisition device and the pedestrian walkway space is determined based on the video surveillance image, and the distance is used as the feature data of the pedestrian walkway space; The state information of the target object is updated based on the feature data; The generation module is used to generate a display image based on the updated status information, the number of the target object, the video surveillance image, and the physical number of the hydraulic support.

Images