A method, device and equipment for detecting abnormal information in a mine

Abstract

The embodiment of the specification discloses a method, device and equipment for detecting abnormal information in a mine, which can comprise: acquiring a living body detection result in a preset area in the mine; if the living body detection result indicates that a target living body exists in the preset area, starting an image acquisition device in the preset area; acquiring a target image containing the target living body collected by the image acquisition device; inputting the target image into an abnormality detection model to obtain an abnormality detection result output by the abnormality detection model; the abnormality detection model is used for detecting whether abnormal information exists in the target image; the abnormal information includes that the target living body in the target image has a rule violation behavior. Based on the scheme of the application, the abnormal situation in the mine can be automatically and efficiently detected.

Description

Technical Field

[0001] This application relates to the field of information detection technology, and in particular to a method, apparatus and equipment for detecting abnormal information in mines. Background Technology

[0002] Underground mine construction is characterized by complex geological conditions, high construction difficulty, diverse types of disasters, wide distribution, and numerous hidden dangers such as unsafe acts by people and unsafe conditions of equipment. The complex underground environment makes it difficult for humans to maintain vigilance for extended periods due to visual fatigue. Safety supervisors or equipment operators cannot be present in every aspect and may be unable to promptly monitor violations. Therefore, intelligent monitoring equipment is needed to replace human eyes, automatically detect anomalies, generate alarms, and promptly alert on-site personnel and supervisors.

[0003] Therefore, there is a need to provide a solution that can automatically and efficiently detect abnormal information in mines. Summary of the Invention

[0004] In view of this, embodiments of this application provide a method, apparatus, and equipment for detecting abnormal information in mines, so as to provide a solution for automatically and efficiently detecting abnormal information in mines.

[0005] To address the aforementioned technical problems, embodiments of this specification provide a method for detecting abnormal information in mines, including:

[0006] Obtain the liveness detection results within a pre-defined area in the mine;

[0007] If the liveness detection result indicates that a target live body exists within the preset area, then the image acquisition device within the preset area is activated;

[0008] Acquire a target image containing the target living body, captured by the image acquisition device;

[0009] The target image is input into the anomaly detection model to obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image; the abnormal information includes the presence of illegal behavior by the target living being in the target image.

[0010] This specification also provides an embodiment of a device for detecting abnormal information in a mine, comprising:

[0011] The first acquisition module is used to acquire the liveness detection results within a preset area in the mine;

[0012] The device startup module is used to start the image acquisition device in the preset area if the liveness detection result indicates that there is a target live body in the preset area;

[0013] The second acquisition module is used to acquire a target image containing the target living body acquired by the image acquisition device;

[0014] An anomaly detection module is used to input the target image into an anomaly detection model and obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image; the abnormal information includes the presence of illegal behavior by the target living being in the target image.

[0015] This specification also provides an embodiment of a device for detecting abnormal information in a mine, including:

[0016] At least one processor; and,

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to:

[0019] Obtain the liveness detection results within a pre-defined area in the mine;

[0020] If the liveness detection result indicates that a target live body exists within the preset area, then the image acquisition device within the preset area is activated;

[0021] Acquire a target image containing the target living body, captured by the image acquisition device;

[0022] The target image is input into the anomaly detection model to obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image; the abnormal information includes the presence of illegal behavior by the target living being in the target image.

[0023] At least one embodiment provided in this specification can achieve the following beneficial effects:

[0024] In this embodiment, after obtaining the liveness detection result within a preset area in the mine, if the liveness detection result indicates the presence of a target live body within the preset area, an image acquisition device within the preset area can be activated; and a target image containing the target live body can be acquired by the image acquisition device; finally, the target image can be input into an anomaly detection model to obtain the anomaly detection result output by the anomaly detection model; wherein, the anomaly detection model is used to detect whether there is abnormal information in the target image; the abnormal information includes the target live body in the target image exhibiting illegal behavior. In this application, by utilizing an anomaly detection model to detect whether there is abnormal information in the target image of a preset area in the mine acquired by the image acquisition device, it is possible to determine whether there is an abnormal situation within the preset area in the mine based on image recognition technology, and to automatically detect illegal behavior by personnel in the mine, effectively avoiding safety hazards caused by negligence of safety supervisors or staff. Furthermore, by combining liveness detection technology, the image acquisition device in the preset area is only activated to acquire images when the liveness detection result indicates the presence of a target live body in the preset area of ​​the mine. This avoids continuously activating the image acquisition device to acquire images and using an anomaly detection model to analyze all acquired data from each time period. This effectively reduces the workload of the image acquisition device and the anomaly detection model, and helps to improve the detection efficiency for anomalies in the mine. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart illustrating a method for detecting abnormal information in a mine, provided as an embodiment of this specification.

[0027] Figure 2 This is a schematic diagram illustrating how a person obstructs the camera, as provided in the embodiments of this specification.

[0028] Figure 3 A schematic diagram of the camera arrangement in a mine tunnel provided for an embodiment of this specification;

[0029] Figure 4 The embodiments provided in this specification correspond to Figure 1 A schematic diagram of a device for detecting abnormal information in a mine;

[0030] Figure 5 The embodiments provided in this specification correspond to Figure 1 A schematic diagram of the structure of a device for detecting abnormal information in a mine. Detailed Implementation

[0031] Many specific details are set forth in the following description to provide a full understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this application; therefore, this application is not limited to the specific embodiments disclosed below.

[0032] The terminology used in one or more embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the scope of one or more embodiments of this application. The singular forms “a,” “the,” and “the” used in one or more embodiments of this application and in the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” used in one or more embodiments of this application refers to and includes any or all possible combinations of one or more associated listed items.

[0033] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this application, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this application, and similarly, second may also be referred to as first. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0034] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0035] In recent years, China's mining industry has achieved significant development in scale. Mine safety has also become particularly important, as mine accidents often occur deep underground, making rescue extremely difficult. Mine accidents are often caused by a combination of factors, including natural factors such as geological structures and gas outbursts, as well as human factors such as operational errors and mismanagement. Among these, violations by miners are the most preventable. Real-time monitoring and timely detection of miners' violations can directly or indirectly prevent major accidents.

[0036] In existing technologies, safety inspectors are typically appointed to supervise operations on-site, or they monitor operations using video footage collected by on-site cameras. However, the underground environment is complex, and due to visual fatigue, it is difficult for people to remain vigilant for extended periods. Consequently, safety inspectors may not be able to promptly monitor miners' violations.

[0037] To address the shortcomings in related technologies, this solution provides the following embodiments.

[0038] Figure 1 This is a flowchart illustrating a method for detecting abnormal information in a mine, provided as an embodiment of this specification. The entity executing this process can be an abnormal information detection platform. Figure 1 As shown, the process may include the following steps:

[0039] Step 102: Obtain the liveness detection results within the preset area in the mine.

[0040] In the embodiments of this specification, the preset area in the mine can refer to an area within a preset range in the mine. This area can be a construction area for employees, a rest area for employees to take temporary breaks, or a part of the walking passage. This application does not specifically limit the area function and area of ​​the preset area.

[0041] In practical applications, one or more liveness detection devices can be used to perform real-time detection on a predetermined area in the mine, obtaining liveness detection results for that area. Liveness detection devices refer to equipment capable of identifying the presence of living individuals. Specifically, liveness detection devices may include, but are not limited to: infrared thermal sensors, human activity monitors, and life detectors.

[0042] Infrared thermal sensors utilize infrared thermal imaging technology to create images by detecting the infrared radiation emitted by objects. Living and dead bodies exhibit differences in infrared radiation, which infrared thermal sensors can detect to determine the presence of a living body. Infrared thermal sensors can penetrate obstacles such as smoke and dust, making them suitable for various complex environments. Human activity monitors employ human motion monitoring algorithms, integrating advanced digital integrated circuit processing technology. They determine the presence of a living body by detecting its approach distance, area, and speed. Human activity monitors have adjustable detection distances, a large coverage area, and high environmental temperature adaptability, allowing them to operate in harsh environments. Life detectors include radar life detectors, infrared life detectors, and audio life detectors. They determine the presence of a living body by detecting subtle movements (such as breathing and heartbeat) or emitted signals such as heat and sound caused by human life activities.

[0043] In practical applications, if the liveness detection equipment detects a live body within a preset area in the mine, it can send the detection result to the anomaly detection platform. Data transmission between the liveness detection equipment and the anomaly detection platform can be achieved via a local area network (LAN), wide area network (WAN), internet connection, or other types of data network connections, or through other means; no specific limitations are imposed on this.

[0044] In practical applications, the liveness detection equipment can also periodically send liveness detection results for a preset area in the mine to the anomaly information detection platform at preset time intervals. If the liveness detection equipment does not detect a live body within the preset area, it can also send the liveness detection result indicating the absence of a live body within the preset area to the anomaly information detection platform. Thus, the anomaly information detection platform can monitor whether the liveness detection equipment is working properly.

[0045] Step 104: If the liveness detection result indicates that there is a target live body in the preset area, then start the image acquisition device in the preset area.

[0046] In the embodiments of this specification, the target living body can refer to a human body. Since there may also be live animals such as rats in the mine, even if the live body detection device detects a living body in the preset area, the detected living body may not necessarily be a human body, but may be other animals. Therefore, after the live body detection device detects a living body in the preset area, it can further determine whether the living body is the target living body (human body). If the determination result indicates that the living body detected in the preset area is the target living body, the live body detection result indicating the presence of the target living body in the preset area is then sent to the abnormal information detection platform.

[0047] Specifically, liveness detection equipment can determine whether a detected live body within a preset area is a human based on information such as its volume and the amount of heat it emits. For example, a liveness detection device can be an infrared thermal sensor, which can detect the difference between the infrared radiation emitted by a human body and that emitted by a small animal, thereby distinguishing whether the live body is human.

[0048] In the embodiments described in this specification, an image acquisition device is a device capable of converting external image information into digital or analog signals for storage, processing, analysis, and display by a computer system or other devices. Specifically, the image acquisition device may include, but is not limited to, cameras, video cameras, scanners, etc. After activating the image acquisition device within a preset area, the image acquisition device can acquire images of objects or living bodies within the preset area.

[0049] In practical applications, there may be multiple image acquisition devices within a preset area. The anomaly detection platform can activate one of these image acquisition devices, or it can activate all of the image acquisition devices within the preset area, or it can activate only some of the image acquisition devices. No specific limitation is made in this regard.

[0050] In practical applications, considering that image acquisition devices capable of acquiring images of a preset area may also be located outside the preset area, the abnormal information detection platform can also activate image acquisition devices outside the preset area capable of acquiring images of the preset area after detecting the presence of a live target within the preset area.

[0051] In the embodiments of this specification, the image acquisition device in the preset area is only activated to acquire images when the liveness detection result indicates that a target live body exists in the preset area of ​​the mine. This avoids continuously activating the image acquisition device to acquire images and using the anomaly detection model to analyze all the data acquired in each time period. This effectively reduces the workload of the image acquisition device and the anomaly detection model, and helps to improve the detection efficiency for anomaly information in the mine.

[0052] Step 106: Acquire a target image containing the target living body, captured by the image acquisition device.

[0053] In the embodiments described in this specification, after activating the image acquisition device within a preset area, the image acquisition device can acquire images of a target living body within the preset area to obtain a target image. For example, when employee A enters a preset area in a mine, the liveness detection device can report the liveness detection result indicating the presence of a target living body within the preset area to the anomaly information detection platform. After obtaining the liveness detection result, the anomaly information detection platform can activate the image acquisition device within the preset area, which can then acquire images of employee A within the preset area to obtain a target image containing employee A.

[0054] In practical applications, after acquiring a target image, the image acquisition device can send the target image to the anomaly detection platform. Data transmission between the image acquisition device and the anomaly detection platform can be achieved through a local area network (LAN), wide area network (WAN), internet connection, or other types of data network connections, or through other means; no specific limitations are imposed on this.

[0055] Step 108: Input the target image into the anomaly detection model to obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image; the abnormal information includes the presence of illegal behavior by the target living body in the target image.

[0056] In the embodiments of this specification, the anomaly detection model can be a model trained using sample data of images containing anomalous information and corresponding anomalous information labels. After training, the anomaly detection model can detect whether there is anomalous information in the image.

[0057] In practical applications, if a live target in a target image exhibits irregular behavior, the anomaly detection model can detect the presence of abnormal information in the target image. Furthermore, the anomaly detection model can also detect the specific nature of the irregular behavior. Irregular behaviors may include, but are not limited to: not wearing a safety helmet, not wearing work clothes, not carrying a miner's lamp, smoking in the mine, using heating equipment improperly, working under unsupported roofs, sitting on a conveyor belt improperly, and climbing over guardrails improperly.

[0058] In practical applications, anomaly information can also include detected equipment malfunctions in the mine within the target image, such as a broken conveyor belt. Anomaly information can also include the detection of smoke or other unusual phenomena in the target image.

[0059] In practical applications, abnormal information may also include images where the target image has an unrecognizable area exceeding a preset threshold. This could be due to malfunction of the image acquisition device or an employee obstructing the lens of the image acquisition device. Since subsequent embodiments in this specification will explain this situation in detail, it will not be repeated here.

[0060] Figure 1The method described above involves obtaining a liveness detection result within a preset area in the mine. If the liveness detection result indicates the presence of a target live body within the preset area, an image acquisition device within that area can be activated. The image acquisition device then acquires a target image containing the target live body. Finally, the target image is input into an anomaly detection model to obtain the anomaly detection result output by the model. The anomaly detection model is used to detect whether there is any abnormal information in the target image. This abnormal information includes instances where the target live body in the target image exhibits unauthorized behavior. By utilizing the anomaly detection model to detect whether there is any abnormal information in the target image within a preset area in the mine acquired by the image acquisition device, image recognition technology can be used to determine whether there are any abnormal situations within the preset area of ​​the mine. This can automatically detect unauthorized behavior by personnel in the mine, effectively preventing safety hazards caused by negligence of safety supervisors or staff. Furthermore, by combining liveness detection technology, the image acquisition device in the preset area is only activated to acquire images when the liveness detection result indicates the presence of a target live body in the preset area of ​​the mine. This avoids continuously activating the image acquisition device to acquire images and using an anomaly detection model to analyze all acquired data from each time period. This effectively reduces the workload of the image acquisition device and the anomaly detection model, and helps to improve the detection efficiency for anomalies in the mine.

[0061] based on Figure 1 In addition to the method described in the embodiments of this specification, some specific implementation schemes of the method are also provided, which will be described below.

[0062] Optional, Figure 1 The method described above, specifically acquiring a target image containing the target living body acquired by the image acquisition device, may include:

[0063] Acquire multiple video frame data containing the target living body captured by the camera.

[0064] In practical applications, the image acquisition device can be a camera. Once activated, the camera can capture video of a target living body within a preset area in the mine, obtaining multiple video frames containing the target living body. In practice, multiple target living bodies may exist within the preset area. In this case, the camera can capture multiple video frames for each target living body separately, or it can capture video frames for multiple living bodies simultaneously; that is, the captured video frames contain multiple target living bodies, without specific limitations. Furthermore, this application does not specifically limit the time interval or the number of video frames captured by the camera.

[0065] In practical applications, the multi-frame video data captured by the camera may contain some blank frames or blurry video frames. If these noisy data are input into the anomaly detection model, it may affect the accuracy of the anomaly detection model's output results and increase the burden on the anomaly detection model to process data. Therefore, it is necessary to filter out these noisy data.

[0066] Based on this, after acquiring the multi-frame video data containing the target living body captured by the camera, the process may further include:

[0067] The multi-frame video data is filtered to obtain target video frame data that meets preset quality requirements;

[0068] Correspondingly, Figure 1 The method described above, in which the target image is input into an anomaly detection model to obtain the anomaly detection result output by the anomaly detection model, may specifically include:

[0069] The target video frame data is input into the anomaly detection model to obtain the anomaly detection result output by the anomaly detection model.

[0070] In practical applications, data filtering can be performed on multiple video frames captured by the camera based on preset quality requirements. Data that does not meet the preset quality requirements can be filtered out to obtain the target video frame data. The preset quality requirements can be set according to actual needs, such as requiring the image to be clear and complete.

[0071] In practical applications, after filtering multiple video frame data, the target video frame data that meets the preset quality requirements can be multiple frames or only one frame. After inputting one or more target video frame data into the anomaly detection model, the anomaly detection result output by the anomaly detection model can be obtained.

[0072] Optional, Figure 1 The method, wherein the method may further include:

[0073] Acquire training data; the training data includes images containing abnormal information and corresponding abnormal information labels; the abnormal information labels are used to indicate the presence of abnormal information in the image and the type of abnormal information; the type of abnormal information includes specific violations.

[0074] The neural network model is trained using the training data to obtain the anomaly detection model.

[0075] In the embodiments of this specification, the anomaly detection model can be a model obtained by training a neural network model. A neural network model is a computational model that mimics the structure and function of a biological neural network and is widely used in fields such as artificial intelligence, machine learning, and deep learning. A neural network model consists of a large number of artificial neurons (nodes), which are connected by weights and can learn and process complex data. The structure of a neural network model can generally be divided into an input layer, hidden layers, and an output layer. The input layer receives the raw input data, the hidden layers extract features from the data, and the output layer generates the final prediction result or classification label.

[0076] In the embodiments of this specification, the anomaly detection model may include, but is not limited to: YOLO (You Only LookOnce) model, Convolutional Neural Networks (CNN) model, Recurrent Neural Network (RNN) model, Bidirectional Recurrent Neural Network (Bi-RNN) model, Long Short-Term Memory (LSTM) model, Gated Recurrent Unit (GRU) model, etc.

[0077] In practical applications, anomaly detection models can be obtained by training a neural network model using training data. This training data can include images containing anomalies and corresponding anomaly labels. The anomaly labels indicate the presence and type of anomalies in the image, including specific violations. For example, a training dataset might include an image of an employee smoking and a label stating "Violation exists in the image; the violation is smoking." The neural network model can learn the features of the data during training. After training, the model can predict the presence and type of anomalies in an image. For instance, inputting an image of an employee not wearing a helmet into the anomaly detection model might result in an output indicating the presence of anomalies and specifically identifying the violation as "not wearing a helmet."

[0078] In practical applications, the process of training a neural network model using training data can terminate under conditions that include reaching a convergence condition or a preset threshold for the number of iterations. The convergence condition can include the model's loss function reaching the preset threshold. For example, if the preset threshold for the number of iterations is 500, then training the neural network model using training data will end after 500 iterations. As another example, if the convergence condition is set to the model's loss function being less than 0.01, then training the neural network model using training data will end when the model's loss function is less than 0.01.

[0079] In practical applications, training data samples can be divided into training and test sets. After training the model using samples from the training set, the performance of the trained model (accuracy, precision, recall, etc.) can be tested using samples from the test set. Training can end once the model's performance meets the target. If the model's performance is found to be unsatisfactory after testing, the training parameters can be adjusted and training can continue, or the model can be further trained using new samples.

[0080] Optional, Figure 1 The method, after inputting the target image into the anomaly detection model and obtaining the anomaly detection result output by the anomaly detection model, further includes:

[0081] If the anomaly detection result indicates that the target living person in the target image has engaged in illegal behavior, then the facial recognition model is used to determine the personnel identity information of the target living person based on the employee facial feature database; the employee facial feature database stores the facial feature data of employees who are permitted to work underground.

[0082] In practical applications, an employee facial feature database can be set up for employees who are allowed to work underground. Facial feature data of employees who are allowed to work underground can be collected in advance and added to the employee facial feature database.

[0083] In practical applications, identity verification devices can be installed at mine entrances. Before personnel enter the mine, their identities can be verified based on an employee facial feature database. Only those personnel authorized to work underground are allowed to pass through; otherwise, they are prohibited from entering. This can prevent unauthorized personnel from entering the mine and help ensure underground safety.

[0084] In the embodiments of this specification, the face recognition model can be a deep learning model with face recognition functionality. This application does not limit the specific category of the face recognition model. When the anomaly detection result indicates that a live target in the target image is engaging in illegal behavior, the anomaly detection platform can input the target image into the face recognition model. The face recognition model can extract the facial feature data of the live target in the target image and compare the extracted facial feature data with the facial feature data in the employee facial feature database. Finally, the employee identity information corresponding to the facial feature data with the highest similarity to the extracted facial feature data in the employee facial feature database is determined as the identity information of the live target. Therefore, once an employee engaging in illegal behavior while working underground, the identity of the violator can be determined through the face recognition model.

[0085] Optional, Figure 1 The method described above includes an anomaly detection result that may include an anomaly information type; the anomaly information type may include specific violation information; correspondingly, after inputting the target image into the anomaly detection model and obtaining the anomaly detection result output by the anomaly detection model, the method may further include:

[0086] If the anomaly detection result indicates that there is abnormal information in the target image, an alarm message is generated; the alarm message includes at least one of the following: the location information of the area where the abnormality occurred, the alarm time point information, the type information of the abnormality, the identity information of the person in violation, and the handling suggestion information;

[0087] The alarm information is sent to the terminal of the mine management personnel so that the mine management personnel can handle the abnormal situation in the mine based on the alarm information.

[0088] In the embodiments of this specification, if the anomaly detection model outputs an anomaly detection result for a target image indicating the presence of abnormal information, the anomaly detection platform needs to generate an alarm message and send it to the mine management personnel's terminal to prompt them to handle the anomaly within a preset area of ​​the mine. The alarm message may include at least one of the following: the location of the anomaly, the alarm time, the type of anomaly, the identity of the violating personnel, and handling suggestions. The location and alarm time information help mine management personnel quickly understand the time and location of the anomaly; the type and identity information help them quickly identify which employee committed the violation and the specific violation; and the handling suggestions assist in decision-making. For example, if employee A is detected not wearing a safety helmet in the target image, an alarm message containing the location of the anomaly, the alarm time, the violation being "not wearing a safety helmet," employee A's identity, and handling suggestions can be sent to the mine management personnel's terminal so they can promptly handle the situation on-site.

[0089] In practical applications, after generating an alarm, the anomaly detection platform can also send the alarm to the terminal of the violator to remind them to stop the violation as soon as possible. If the violator did not actually commit the violation, they can appeal, submitting their appeal to the anomaly detection platform. The appeal can include an explanation and evidence. The explanation can be a written or audio statement from the person clarifying that they did not commit the violation and that the detection was erroneous; the evidence can be proof submitted by the person to demonstrate their innocence. After receiving the appeal submitted by the person using their terminal, the anomaly detection platform can re-capture the person's image using image acquisition equipment for re-detection; or, it can send an instruction to the terminal of the suspected violator, directing them to a designated camera for re-examination.

[0090] In practical applications, if the liveness detection equipment used to detect a preset area malfunctions, the image acquisition equipment within that area may be unable to start for an extended period, preventing timely detection of anomalies. To address this issue, in practice, in addition to activating the image acquisition equipment within the preset area to acquire images when the liveness detection result indicates the presence of a target live body in the mine, the equipment can also be activated periodically. If an image acquisition equipment within the preset area is not activated by the liveness detection equipment but is periodically activated and the acquired images show a target live body, it indicates an malfunction in the liveness detection equipment used to detect that preset area, triggering an alarm so that mine management personnel can address the issue promptly. The activation cycle of the image acquisition equipment can be set and adjusted according to actual needs; specifically, it can be activated every hour or every three hours, without a specific limitation.

[0091] In practical applications, periodically activating the image acquisition equipment allows for periodic checks on its operational status. If any abnormalities are detected, an alarm can be triggered promptly, enabling mine management personnel to address the issue by replacing or repairing the malfunctioning equipment. Specifically, abnormalities may include, but are not limited to: the equipment failing to start, acquiring a completely black image, or acquiring an image where most of the data is obscured and unrecognizable. In practice, these abnormalities may be caused by human interference or obstruction, or by a malfunction in the equipment itself. In such cases, an alarm can be sent to the mine management personnel's terminal, allowing them to promptly investigate the cause of the abnormality and take appropriate action, thereby enhancing safety in the mine.

[0092] In practical applications, mine workers may deliberately obstruct the camera and then perform illegal operations. In this case, although the camera is activated, it cannot capture the illegal operation after being obstructed. The captured image may be a completely black image or an image that is unrecognizable due to most of the area being obstructed. Figure 2 This is a schematic diagram illustrating how a person obstructs the camera, as provided in the embodiments of this specification; for example... Figure 2 As shown, employees can use objects to block or cover the camera, making it unable to function properly.

[0093] In practical applications, image acquisition devices such as cameras may malfunction. In such cases, even if the camera is not obstructed, some areas of the captured image may be unrecognizable. Therefore, for situations such as obstructed image acquisition devices or malfunctions of the image acquisition devices themselves, an alarm can be triggered to notify mine management personnel to investigate the specific anomaly.

[0094] Based on this Figure 1 The method in the text, wherein the anomaly information may include the target image being an image where the unrecognizable region exceeds a preset threshold; correspondingly, after inputting the target image into the anomaly detection model and obtaining the anomaly detection result output by the anomaly detection model, it may further include:

[0095] If the anomaly detection result indicates that the target image is an image with an unrecognizable area exceeding a preset threshold, a first alarm message is sent to the terminal of the mine management personnel; the first alarm message includes at least one of the following: device number information of the image acquisition device that acquired the target image, and location information of the area where the image acquisition device that acquired the target image is located.

[0096] In the embodiments of this specification, the unrecognizable area in the target image can refer to the area where specific information cannot be identified. For example, if half of the lens of camera A is covered by an opaque black cloth by an employee, then half of the target image captured by camera A is pure black. This part of the area cannot be identified and is therefore the unrecognizable area in the target image.

[0097] In practical applications, if the anomaly detection model outputs an anomaly detection result indicating that the target image contains unidentifiable areas exceeding a preset threshold, the anomaly detection platform can send a first alarm message to the mine management personnel's terminal. The preset threshold can be a pre-defined percentage value, set and adjusted according to actual needs, and its specific value is not limited. For example, if the preset threshold is 30%, and the unidentifiable areas in the target image account for 45% of the entire image, then the target image is considered to have unidentifiable areas exceeding the preset threshold.

[0098] In practical applications, the first alarm message sent by the anomaly detection platform to the mine management personnel's terminal may include information such as the device number of the image acquisition equipment collecting the target image and the location information of the area where the image acquisition equipment is located. Upon receiving the first alarm message, the mine management personnel's terminal can issue a voice alarm or use other prompts to alert them to the anomaly in the mine. The mine management personnel can then determine which image acquisition equipment is malfunctioning based on the device number and location information included in the first alarm message, and proceed to the site to investigate the specific anomaly.

[0099] In practical applications, employees may only briefly obstruct the image acquisition equipment, and when mine management personnel arrive, they may not be able to determine which employee had obstructed the image acquisition equipment.

[0100] To address the aforementioned problems, this application also proposes an implementation method in which, after inputting the target image into the anomaly detection model and obtaining the anomaly detection result output by the anomaly detection model, the method may further include:

[0101] If the anomaly detection result indicates that the target image is an image whose unrecognizable area exceeds a preset threshold, then target video data collected by other image acquisition devices within a preset distance range from the image acquisition device that acquired the target image within the preset area during a target time period is obtained; the end time of the target time period is the current time.

[0102] Based on the target video data, a second alarm message is generated; the second alarm message includes at least one of the following: device number information of the image acquisition device that acquired the target image, location information of the area where the image acquisition device that acquired the target image is located, and the target video data;

[0103] The second alarm information is sent to the terminal of the mine management personnel.

[0104] In the embodiments of this specification, when the anomaly detection result indicates that the target image is an image with an unrecognizable area exceeding a preset threshold, it may be that the image acquisition device that acquired the target image is obstructed, or that the image acquisition device itself is malfunctioning. Therefore, the anomaly information detection platform can obtain target video data collected by other image acquisition devices near the image acquisition device that acquired the target image within a preset area where the image acquisition device is located in the recent period, and add this part of the target video data to the alarm information and send it to the mine management personnel, so that the mine management personnel can determine which employee obstructed the image acquisition device or determine the specific abnormal situation of the image acquisition device based on the target video data.

[0105] For example, Figure 3 This is a schematic diagram illustrating the arrangement of cameras in a mine tunnel, as provided in an embodiment of this specification. Figure 3 As shown, four cameras are installed in this section of the mine tunnel: camera 1, camera 2, camera 3, and camera 4. If the unidentifiable area in the image captured by camera 2 exceeds a preset threshold, and cameras 1 and 3 are within a preset range of camera 2, and the distance between camera 1 and camera 2 is less than the distance between camera 3 and camera 2, then video data captured by camera 1 in the area where camera 2 is located in the most recent period can be obtained first. If camera 1 malfunctions, video data captured by camera 3 in the area where camera 2 is located in the most recent period can also be obtained.

[0106] In practical applications, when the mine management personnel's terminal receives the second alarm information, it can issue a voice alarm prompt or use other prompting methods to notify the mine management personnel of an abnormal situation in the mine. The mine management personnel can determine which image acquisition device is malfunctioning based on information included in the second alarm information, such as the device number of the image acquisition device capturing the target image and the location information of the image acquisition device. The mine management personnel can also determine the specific abnormal situation of the image acquisition device based on the target video data included in the second alarm information. For example, they can determine whether someone is obstructing the image acquisition device based on the target video data, and if so, which employee is obstructing the image acquisition device.

[0107] In practical applications, employees typically need to approach a camera before they can block it. During this approach, the camera can capture video of the employee, meaning that even if the camera is blocked, the most recently captured video before the blocking is still highly valuable. Therefore, to address the problem of mine management personnel being unable to determine which employee blocked the image acquisition equipment upon arrival, this application also proposes another implementation method. Correspondingly, after inputting the target image into the anomaly detection model and obtaining the anomaly detection result output by the model, the method may further include:

[0108] If the anomaly detection result indicates that the target image is an image whose unrecognizable area exceeds a preset threshold, then the first video data collected by the image acquisition device that acquired the target image within the preset area during the target time period is obtained; the end time of the target time period is the current time.

[0109] Based on the first video data, a third alarm message is generated; the third alarm message includes at least one of the following: device number information of the image acquisition device that acquired the target image, location information of the area where the image acquisition device that acquired the target image is located, and the first video data;

[0110] The third alarm information is sent to the terminal of the mine management personnel.

[0111] In the embodiments of this specification, the end time of the target time period is the current time, and the target time period is the most recent period of time. The duration of the target time period can be set and adjusted according to actual needs, and there is no specific limitation on it. For example, the current time is 10:00:00, and the target time period is 10 minutes from 9:50:00 to 10:00:00.

[0112] In practical applications, the anomaly detection platform can obtain the first video data collected by the image acquisition device within a preset area during the target time period from the data storage unit of the image acquisition device, or it can obtain it from other databases that store video data collected by the image acquisition device, without any specific limitation.

[0113] In practical applications, if the malfunction is caused by camera obstruction, then the first video data is the most recently captured video data before the camera was obstructed. This first video data is very likely to include data captured by the camera showing an employee approaching the camera. For example, if the first video data includes data captured by the camera showing employee A approaching the camera, mine managers can determine based on this first video data that employee A is very likely the person obstructing the camera.

[0114] In the embodiments of this specification, by sending a third alarm message containing video data collected by the image acquisition device within a preset area during a target time period to the terminal of the mine management personnel, it is helpful to assist the mine management personnel in quickly and accurately identifying the suspects who are obstructing the image acquisition device.

[0115] Figure 4 The embodiments provided in this specification correspond to Figure 1 A schematic diagram of a device for detecting abnormal information in a mine, as shown below. Figure 4 As shown, the device may include:

[0116] The first acquisition module 402 is used to acquire the liveness detection results within a preset area in the mine;

[0117] The device startup module 404 is used to start the image acquisition device in the preset area if the liveness detection result indicates that there is a target live body in the preset area;

[0118] The second acquisition module 406 is used to acquire a target image containing the target living body acquired by the image acquisition device;

[0119] An anomaly detection module 408 is used to input the target image into an anomaly detection model and obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image; the abnormal information includes the presence of illegal behavior by the target living body in the target image.

[0120] based on Figure 4 The embodiments of this specification also provide some specific implementations of the device, which will be described below.

[0121] Optionally, the second acquisition module 406 may specifically include:

[0122] The data acquisition unit is used to acquire multiple video frame data containing the target living body captured by the camera.

[0123] Optionally, the device may further include:

[0124] The data filtering module is used to filter the multi-frame video data to obtain target video frame data that meets preset quality requirements.

[0125] Correspondingly, the anomaly detection module 408 may specifically include:

[0126] An anomaly detection unit is used to input the target video frame data into an anomaly detection model and obtain the anomaly detection result output by the anomaly detection model.

[0127] Optionally, the violation may include at least one of the following: not wearing a safety helmet, not wearing work clothes, not carrying a miner's lamp, smoking in the mine, using heating equipment in violation of regulations, working under an open ceiling, sitting on a conveyor belt in violation of regulations, and climbing over a guardrail in violation of regulations.

[0128] Optionally, the device may further include:

[0129] The training data acquisition module is used to acquire training data; the training data includes images containing abnormal information and corresponding abnormal information labels; the abnormal information labels are used to indicate the presence of abnormal information in the image and the type of abnormal information; the type of abnormal information includes specific violations.

[0130] The model training module is used to train a neural network model using the training data to obtain the anomaly detection model.

[0131] Optionally, the anomaly detection result may include anomaly information type information; the anomaly information type information may include specific violation information; correspondingly, the device may further include:

[0132] An alarm information generation module is used to generate alarm information if the anomaly detection result indicates that there is abnormal information in the target image; the alarm information includes at least one of the following: the location information of the area where the abnormal situation occurred, the alarm time point information, the type information of the abnormal information, the identity information of the violator, and the handling suggestion information;

[0133] An alarm information sending module is used to send the alarm information to the terminal of the mine management personnel, so that the mine management personnel can handle the abnormal situation in the mine based on the alarm information.

[0134] Optionally, the device may further include:

[0135] The personnel identity information determination module is used to determine the personnel identity information of the target living person in the target image by using a face recognition model based on an employee face feature database if the anomaly detection result indicates that the target living person has violated regulations. The employee face feature database stores the face feature data of employees who are allowed to work underground.

[0136] Optionally, the abnormal information may include the target image being an image where the unrecognizable area exceeds a preset threshold; correspondingly, the device may further include:

[0137] The first alarm information sending module is used to send a first alarm information to the terminal of the mine management personnel if the abnormal detection result indicates that the target image is an image with an unrecognizable area exceeding a preset threshold; the first alarm information includes at least one of the following: device number information of the image acquisition device that acquires the target image and location information of the area where the image acquisition device that acquires the target image is located;

[0138] Optionally, the abnormal information may include the target image being an image where the unrecognizable area exceeds a preset threshold; correspondingly, the device may further include:

[0139] The target video data acquisition module is used to acquire target video data collected by other image acquisition devices within a preset distance range from the image acquisition device that acquired the target image within a preset time period if the anomaly detection result indicates that the target image is an image with an unrecognizable area exceeding a preset threshold; the end time of the target time period is the current time.

[0140] The second alarm information generation module is used to generate a second alarm information based on the target video data; the second alarm information includes at least one of the following: device number information of the image acquisition device that acquired the target image, location information of the area where the image acquisition device that acquired the target image is located, and the target video data;

[0141] The second alarm information sending module is used to send the second alarm information to the terminal of the mine management personnel.

[0142] It is understood that the modules mentioned above refer to computer programs or program segments used to perform one or more specific functions. Furthermore, the distinction between these modules does not imply that the actual program code must also be separate.

[0143] Based on the same idea, this specification also provides devices corresponding to the above methods in its embodiments.

[0144] Figure 5 The embodiments provided in this specification correspond to Figure 1 A schematic diagram of a device for detecting abnormal information in a mine. Figure 5 As shown, device 500 may include:

[0145] At least one processor 510; and,

[0146] Memory 530 communicatively connected to the at least one processor;

[0147] The memory 530 stores instructions 520 that can be executed by the at least one processor 510, and the instructions, when executed by the at least one processor 510, enable the at least one processor 510 to:

[0148] Obtain the liveness detection results within a pre-defined area in the mine;

[0149] If the liveness detection result indicates that a target live body exists within the preset area, then the image acquisition device within the preset area is activated;

[0150] Acquire a target image containing the target living body, captured by the image acquisition device;

[0151] The target image is input into the anomaly detection model to obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image; the abnormal information includes the presence of illegal behavior by the target living being in the target image.

[0152] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, for... Figure 5 As the device shown is basically similar to the corresponding method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0153] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0154] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0155] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0156] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0157] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0158] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0159] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0160] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0161] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0162] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0163] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0164] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0165] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0166] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0167] The above description is merely an embodiment of this specification and is not intended to limit this specification. Various modifications and variations can be made to this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for detecting abnormal information in a mine, comprising: Obtain the liveness detection results within a pre-defined area in the mine; If the liveness detection result indicates that a target live body exists within the preset area, then the image acquisition device within the preset area is activated; Acquire a target image containing the target living body, captured by the image acquisition device; The target image is input into the anomaly detection model to obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image. The abnormal information includes violations by the target living being in the target image; the abnormality detection model is obtained by training a neural network model using training data; the training data includes images containing abnormal information and corresponding abnormal information labels; the abnormal information labels are used to indicate the presence of abnormal information in the image and the type of abnormal information; the types of abnormal information include not wearing a safety helmet, not wearing work clothes, not carrying a miner's lamp, smoking in the mine, using heating equipment in violation of regulations, working under an open ceiling, sitting on a conveyor belt in violation of regulations, or climbing over a guardrail in violation of regulations; If the anomaly detection result indicates that the target image is an image where the area of ​​the occluded region exceeds a preset threshold, then the first video data collected by the image acquisition device within the preset region during the target time period is obtained; the end time of the target time period is the current time. Based on the first video data, a third alarm message is generated; the third alarm message includes at least one of the following: the device number information of the image acquisition device, the location information of the area where the image acquisition device is located, and the first video data. The third alarm message is sent to the terminal of the mine management personnel.

2. The method as described in claim 1, wherein acquiring the target image containing the target living body acquired by the image acquisition device specifically includes: Acquire multiple video frame data containing the target living body captured by the camera.

3. The method as described in claim 2, further comprising, after acquiring the multi-frame video data containing the target living body captured by the camera, the method further includes: The multi-frame video data is filtered to obtain target video frame data that meets preset quality requirements; The step of inputting the target image into the anomaly detection model and obtaining the anomaly detection result output by the anomaly detection model specifically includes: The target video frame data is input into the anomaly detection model to obtain the anomaly detection result output by the anomaly detection model.

4. The method as described in claim 1, wherein the anomaly detection result includes anomaly information type information; the anomaly information type information includes specific violation behavior information; after inputting the target image into the anomaly detection model and obtaining the anomaly detection result output by the anomaly detection model, the method further includes: If the anomaly detection result indicates that there is abnormal information in the target image, an alarm message is generated; The alarm information includes at least one of the following: the location of the abnormal situation, the time of the alarm, the type of abnormal information, the identity information of the person in violation, and the handling suggestion information; The alarm information is sent to the terminal of the mine management personnel so that the mine management personnel can handle the abnormal situation in the mine based on the alarm information.

5. The method as described in claim 1, further comprising, after inputting the target image into the anomaly detection model and obtaining the anomaly detection result output by the anomaly detection model: If the anomaly detection result indicates that the target living person in the target image has engaged in illegal behavior, then the facial recognition model is used to determine the personnel identity information of the target living person based on the employee facial feature database; the employee facial feature database stores the facial feature data of employees who are permitted to work underground.

6. A device for detecting abnormal information in a mine, comprising: The first acquisition module is used to acquire the liveness detection results within a preset area in the mine; The device startup module is used to start the image acquisition device in the preset area if the liveness detection result indicates that there is a target live body in the preset area; The second acquisition module is used to acquire a target image containing the target living body acquired by the image acquisition device; An anomaly detection module is used to input the target image into an anomaly detection model and obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image. The abnormal information includes violations by the target living being in the target image; the abnormality detection model is obtained by training a neural network model using training data; the training data includes images containing abnormal information and corresponding abnormal information labels; the abnormal information labels are used to indicate the presence of abnormal information in the image and the type of abnormal information; the types of abnormal information include not wearing a safety helmet, not wearing work clothes, not carrying a miner's lamp, smoking in the mine, using heating equipment in violation of regulations, working under an open ceiling, sitting on a conveyor belt in violation of regulations, or climbing over a guardrail in violation of regulations; The first video data acquisition module is used to acquire first video data collected by the image acquisition device within the preset area during the target time period if the anomaly detection result indicates that the target image is an image in which the area of ​​the occluded region exceeds a preset threshold. The end time of the target time period is the current time; The third alarm information generation module is used to generate third alarm information based on the first video data; the third alarm information includes at least one of the following: the device number information of the image acquisition device, the location information of the area where the image acquisition device is located, and the first video data. The third alarm information sending module is used to send the third alarm information to the terminal of the mine management personnel.

7. A device for detecting abnormal information in a mine, comprising: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to: Obtain the liveness detection results within a pre-defined area in the mine; If the liveness detection result indicates that a target live body exists within the preset area, then the image acquisition device within the preset area is activated; Acquire a target image containing the target living body, captured by the image acquisition device; The target image is input into an anomaly detection model to obtain the anomaly detection result output by the anomaly detection model; the anomaly detection model is used to detect whether there is abnormal information in the target image; the abnormal information includes the target living being in the target image exhibiting illegal behavior; the anomaly detection model is obtained by training a neural network model using training data; the training data includes images with abnormal information and corresponding abnormal information labels; the abnormal information labels are used to indicate the presence of abnormal information in the image and the type of abnormal information; the types of abnormal information include not wearing a safety helmet, not wearing work clothes, not carrying a miner's lamp, smoking in the mine, illegally using heating equipment, working under an open ceiling, illegally sitting on a conveyor belt, or illegally climbing over a guardrail; If the anomaly detection result indicates that the target image is an image where the area of ​​the occluded region exceeds a preset threshold, then the first video data collected by the image acquisition device within the preset region during the target time period is obtained; the end time of the target time period is the current time. Based on the first video data, a third alarm message is generated; the third alarm message includes at least one of the following: the device number information of the image acquisition device, the location information of the area where the image acquisition device is located, and the first video data. The third alarm message is sent to the terminal of the mine management personnel.

Images