Target recognition method, device, computer equipment and storage medium

Abstract

The present disclosure provides a target recognition method, apparatus, computer equipment and storage medium, wherein the method comprises: acquiring a first video stream captured by a target camera device, and determining a first to-be-processed video frame containing a first object in the first video stream; determining a target device type based on a capture area of ​​the first video stream; detecting a target device matching the target device type on the first object based on the first to-be-processed video frame to obtain a first device detection result, wherein the target device comprises a target operating device worn on the first object, and / or a target wearable device for wearing the target operating device on the first object; generating device warning information when it is determined based on the first device detection result that the target device is not detected on the first object.

Description

Technical Field

[0001] The present disclosure relates to the field of security technology, and in particular to a target recognition method, device, computer equipment and storage medium. Background Art

[0002] In daily life, many work scenarios require workers to wear corresponding work equipment to work to ensure the personal safety of workers. For example, in paint spraying operations, workers are required to wear protective masks; in pole operations, workers are required to wear safety belts; in underground operations, workers are required to wear oxygen cylinders, etc. At present, workers entering the work area are usually monitored by manual monitoring to ensure that each worker entering the work area wears the corresponding work equipment. However, this manual monitoring method not only wastes a lot of human resources, but is also prone to missed detection. Therefore, how to use computer technology to automatically identify whether workers are wearing work equipment has become an urgent problem to be solved. Summary of the invention

[0003] The embodiments of the present disclosure at least provide a target recognition method, apparatus, computer equipment, and storage medium.

[0004] An embodiment of the present disclosure provides a target recognition method, which includes: acquiring a first video stream captured by a target camera device, and determining a first to-be-processed video frame containing a first object in the first video stream; determining a target device type based on a capture area of ​​the first video stream; detecting a target device matching the target device type on the first object based on the first to-be-processed video frame to obtain a first device detection result, wherein the target device includes a target operating device worn on the first object, and / or a target wearable device for wearing the target operating device on the first object; generating device warning information when it is determined based on the first device detection result that the target device is not detected on the first object.

[0005] It can be known from the above description that, first, it is determined that the first video frame to be processed contains the first object in the first video stream collected by the target camera device, and the target device type is determined based on the collection area of ​​the first video stream. Afterwards, the target device matching the target device type is detected on the first object based on the first video frame to be processed to obtain a first device detection result. In the case where it is determined based on the obtained first device detection result that the target device is not detected on the first object, a device warning information can be generated. In the above implementation, the wearing condition of the target device on the first object can be automatically determined through the collected first video stream, so that automatic monitoring of whether the target object is wearing the target device can be achieved, thereby improving the detection efficiency of the target device on the first object. At the same time, in the case where the target device is not detected on the first object, a device warning information can be generated, so that the safety of the first object in performing operations can be further improved.

[0006] In an optional embodiment, determining the first video frame to be processed containing the first object in the first video stream includes: screening at least one initial video frame in the first video stream that contains the first object; determining a video frame in the at least one initial video frame that meets the detection requirements as the first video frame to be processed, wherein the detection requirements include at least one of the following: a size requirement for the first object, a clarity requirement for the initial video frame, a completeness requirement for the first object, and a position requirement for the first object in the initial video frame.

[0007] In the above implementation, after screening out at least one initial video frame containing the first object in the first video stream, the video frame that meets the detection requirements in the at least one initial video frame can be determined as the first video frame to be processed, thereby improving the image quality of the first video frame to be processed, and further improving the accuracy of target recognition in the first video frame to be processed.

[0008] In an optional embodiment, the method also includes: after generating the device warning information, obtaining a second video stream collected by the target camera device within a preset timing time; when it is determined based on the second video stream that the status information has changed, determining a second video frame to be processed in the second video stream; the status information is used to indicate the monitoring status of an object within the monitoring area of ​​the target camera device; based on the second video frame to be processed, detecting the target device on a second object to obtain a second device detection result, wherein the second object is an object whose monitoring status has changed in the monitoring area.

[0009] In the above implementation, after the device warning information is generated, the second video stream collected by the target camera device within the preset timing time can be obtained, so that the monitoring status of the object in the monitoring area of ​​the target camera device can be monitored in real time, and more accurate status information of the object in the monitoring area can be obtained. In the case where it is determined based on the second video stream that the status information has changed, the second video frame to be processed can be determined in the second video stream, and the target device can be detected on the second object based on the second video frame to be processed to obtain the second device detection result, so that the generated device warning information can be updated in time, thereby improving the accuracy and real-time performance of target recognition.

[0010] In an optional implementation, the method further includes: if no change in the status information is determined based on the second video stream, after the preset timing time expires, generating the device warning information again according to a preset generation mode.

[0011] In the above implementation, when no change in status information is determined based on the second video stream, the device warning information can be generated again according to the preset generation mode after the preset timing time expires, thereby making the display form of the device warning information richer and further improving the effectiveness of the device warning information.

[0012] In an optional implementation, the method further includes: after generating the device warning information, displaying target display information in a display interface of a terminal device, wherein the target display information includes the device warning information and / or the physical attribute information of the first object, and sending the target display information to the first object.

[0013] In the above implementation, after the device warning information is generated, the target display information can be displayed in the display interface of the terminal device, and the target display information can be sent to the first object, so that the terminal device staff and the first object can obtain the target display information of the first object in a timely and clear manner, and then the first object can be reminded to perform safe operations in a timely manner. At the same time, the terminal device staff can also better perform data analysis on the physical attribute information of the first object to improve security performance.

[0014] In an optional embodiment, the method of detecting a target device matching the target device type on the first object based on the first video frame to be processed to obtain a first device detection result includes: determining device attributes of the device included in the first video frame to be processed, wherein the device attributes are used to indicate the device category and / or device location of the included device; obtaining the object location of the first object in the first video frame to be processed; and determining the first device detection result based on the object location and the device attributes.

[0015] In an optional embodiment, the determining that the target device is not detected on the first object based on the detection result of the first device includes: determining that the included device does not belong to the target device based on the device category, and / or determining that the object position and the device position do not meet preset position requirements, determining that the target device is not detected on the first object.

[0016] In the above implementation, the first device detection result may be determined based on the object position and the device attribute, so that the obtained first device detection result may be more accurate, thereby improving the accuracy of detecting the target device on the first object.

[0017] In an optional embodiment, when the first device detection result includes the confidence of the first object corresponding to each preset detection result, determining that the target device is not detected on the first object based on the first device detection result includes: when it is determined based on the first device detection result that the confidence of the first object corresponding to a first preset detection result is greater than a preset confidence threshold, determining that the target device is not detected on the first object, wherein the first preset detection result is used to indicate that the first object in the first video frame to be processed is wearing the target device.

[0018] In the above embodiment, when the detection result of the first device includes the confidence of each preset detection result corresponding to the first object, it can be determined that the target device is not detected on the first object when the confidence of the first preset detection result is greater than the preset confidence threshold based on the detection result of the first device. Thus, it can be determined more accurately that the target device is not detected on the first object, thereby reducing errors and improving the accuracy of detecting the target device on the first object.

[0019] In some embodiments, the embodiments of the present disclosure also provide a target recognition device, including: an acquisition unit, used to acquire a first video stream captured by a target camera device, and determine a first to-be-processed video frame containing a first object in the first video stream; a determination unit, used to determine a target device type based on a capture area of ​​the first video stream; a detection unit, used to detect a target device matching the target device type on the first object based on the first to-be-processed video frame, and obtain a first device detection result, wherein the target device includes a target operating device worn on the first object, and / or a target wearable device for wearing the target operating device on the first object; a generation unit, used to generate device warning information when it is determined based on the first device detection result that the target device is not detected on the first object.

[0020] In some embodiments, the present disclosure also provides a computer device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, when the computer device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the target recognition method of the above embodiment is performed, or the steps in any possible implementation of the target recognition method.

[0021] In some embodiments, the present disclosure also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the target recognition method of the above embodiment or the steps of any possible implementation manner is executed.

[0022] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.

[0024] Figure 1 A flow chart of a target recognition method provided by an embodiment of the present disclosure is shown;

[0025] Figure 2 A flowchart of a method for determining a first to-be-processed video frame containing a first object in a first video stream provided by an embodiment of the present disclosure is shown;

[0026] Figure 3 A flowchart of another target recognition method provided by an embodiment of the present disclosure is shown;

[0027] Figure 4 A schematic diagram showing a specific implementation process of an embodiment provided by the present disclosure;

[0028] Figure 5 A schematic diagram of a target recognition device provided by an embodiment of the present disclosure is shown;

[0029] Figure 6 A schematic diagram of a computer device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the present disclosure for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.

[0031] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0032] The term "and / or" herein only describes an association relationship, indicating that three relationships may exist. For example, A and / or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.

[0033] Research has found that in daily life, many work scenes require workers to wear corresponding work equipment to work to ensure the personal safety of workers. For example, in paint spraying operations, workers are required to wear protective masks; in pole operations, workers are required to wear safety belts; in underground operations, workers are required to wear oxygen cylinders, etc. At present, workers entering the work area are usually monitored by manual monitoring to ensure that each worker entering the work area wears the corresponding work equipment. However, this manual monitoring method not only wastes a lot of human resources, but is also prone to missed detection. Therefore, how to use computer technology to automatically identify whether workers are wearing work equipment has become an urgent problem to be solved.

[0034] Based on the above research, the present disclosure provides a target recognition method, device, computer device and storage medium. In an embodiment of the present disclosure, first determine the first video frame to be processed that contains the first object in the first video stream collected by the target camera device, and determine the target device type based on the collection area of ​​the first video stream. Afterwards, detect the target device that matches the target device type on the first object based on the first video frame to be processed to obtain a first device detection result. In the case where it is determined based on the obtained first device detection result that the target device is not detected on the first object, a device warning information can be generated. It can be seen from the above description that the wearing condition of the target device on the first object can be automatically determined through the collected first video stream, so that it is possible to automatically monitor whether the target object is wearing the target device, thereby improving the detection efficiency of the target device on the first object. At the same time, in the case where the target device is not detected on the first object, a device warning information can be generated, so that the safety of the first object in performing operations can be further improved.

[0035] To facilitate understanding of this embodiment, a target recognition method disclosed in the embodiment of the present disclosure is first introduced in detail. The executor of the target recognition method provided in the embodiment of the present disclosure is generally a computer device with certain computing capabilities.

[0036] See also Figure 1 FIG. 1 is a flow chart of a target recognition method provided by an embodiment of the present disclosure, wherein the method includes steps S101 to S107, wherein:

[0037] S101: Acquire a first video stream captured by a target camera device, and determine a first to-be-processed video frame containing a first object in the first video stream.

[0038] Here, the target camera device may be a network camera device. At this time, the network camera may be installed at the entrance of the work area, or at the entrance of the park where the work area is located. The present disclosure does not specifically limit the installation position of the target camera device, and is subject to actual needs.

[0039] Here, the first video stream can be understood as a video stream captured by the target camera device after the first object enters the monitoring area corresponding to the target camera device.

[0040] In the embodiment of the present disclosure, the first object may correspond to one operator or multiple operators. The number of operators corresponding to the first object of the present disclosure is not specifically limited and is subject to achievable performance.

[0041] In the embodiment of the present disclosure, after obtaining the first video stream captured by the target camera device, the first video frame to be processed containing the first object can be determined in the first video stream. In specific implementation, multiple video frames contained in the first video stream captured by the target camera device can be first obtained, and at least one video frame containing the first object can be determined in the multiple video frames. Afterwards, the video frame that meets the detection requirements can be determined in the at least one video frame as the first video frame to be processed.

[0042] The number of video frames included in the first video stream may be 30 frames or 60 frames. The present disclosure does not limit the number of video frames included in the first video stream, and the acquisition time length and acquisition frequency of the target camera device are used as the basis. For example, when the acquisition time length of the target camera device is 1 second and the acquisition frequency of the target camera device is 30 frames / second, the number of video frames included in the first video stream is 30 frames.

[0043] S103: Determine the target device type based on the acquisition area of ​​the first video stream.

[0044] In an embodiment of the present disclosure, a target device type may be determined based on a capture area of ​​the first video stream, where the target device type is used to indicate the type of device that needs to be worn to perform work in a work scene corresponding to the capture area.

[0045] For example, when the operation scene corresponding to the collection area is paint spraying operation, the target device type can be protective equipment (for example, protective masks, protective clothing, etc.); when the operation scene corresponding to the collection area is electric pole operation, the target device type can be safety equipment (for example, safety belts); when the operation scene corresponding to the collection area is downhole operation, the target device type can be breathing equipment (for example, oxygen cylinders, carbon monoxide alarms, etc.).

[0046] The present disclosure does not specifically limit the above target device types, and is subject to meeting actual needs.

[0047] S105: Detect a target device matching the target device type on the first object based on the first video frame to be processed, and obtain a first device detection result, wherein the target device includes a target working device worn on the first object, and / or a target wearing device for wearing the target working device on the first object.

[0048] Here, the target device can be understood as the device that the first object (i.e., the operator) must wear when entering the working area to work (i.e., the above-mentioned target working device). For example, the target device can be an oxygen cylinder, or a strap for wearing the oxygen cylinder to the first object (i.e., the above-mentioned target wearing device). The present disclosure does not specifically limit the above-mentioned target device, which is subject to what can be realized.

[0049] In the embodiment of the present disclosure, after a first to-be-processed video frame including a first object is determined based on the first video stream, a target device may be detected on the first object based on the first to-be-processed video frame to obtain a first device detection result.

[0050] In the disclosed embodiment, the first device detection result may be: the first subject wears the target device (or, wears), or the first subject does not wear the target device (or, does not wear). The target device includes a target working device and / or a target wearing device.

[0051] S107: When it is determined based on the first device detection result that the target device is not detected on the first object, generate device warning information.

[0052] In the embodiment of the present disclosure, after obtaining the first device detection result, the wearing information of the target device on the first object may be determined based on the first device detection result.

[0053] For example, when it is determined based on the first device detection result that the target object is not wearing the target device, it can be determined that the target device is not detected on the first object; when it is determined based on the first device detection result that the target object is wearing the target device, it can be determined that the target device is detected on the first object.

[0054] In an embodiment of the present disclosure, when it is determined based on the first device detection result that the target device is not detected on the first object, device warning information may be generated. The device warning information may include at least the following information: object ID, information that the object is not wearing the target device, and a warning mark.

[0055] It can be known from the above description that, first, it is determined that the first video frame to be processed contains the first object in the first video stream collected by the target camera device, and the target device type is determined based on the collection area of ​​the first video stream. Afterwards, the target device matching the target device type is detected on the first object based on the first video frame to be processed to obtain a first device detection result. In the case where it is determined based on the obtained first device detection result that the target device is not detected on the first object, a device warning information can be generated. In the above implementation, the wearing condition of the target device on the first object can be automatically determined through the collected first video stream, so that automatic monitoring of whether the target object is wearing the target device can be achieved, thereby improving the detection efficiency of the target device on the first object. At the same time, in the case where the target device is not detected on the first object, a device warning information can be generated, so that the safety of the first object in performing operations can be further improved.

[0056] In an optional embodiment, if Figure 2 As shown, for S101, determining a first to-be-processed video frame containing a first object in the first video stream specifically includes the following steps:

[0057] Step S21: screening at least one initial video frame containing the first object in the first video stream;

[0058] Step S22: Determine a video frame in the at least one initial video frame that meets the detection requirements as the first video frame to be processed, wherein the detection requirements include at least one of the following: a size requirement for the first object, a clarity requirement for the initial video frame, a completeness requirement for the first object, and a position requirement for the first object in the initial video frame.

[0059] In the embodiment of the present disclosure, the specific implementation process of determining the first to-be-processed video frame containing the first object in the first video stream can be described as the following process:

[0060] First, at least one initial video frame containing the first object is screened in the first video stream. The number of the at least one initial video frame may be one or more, and the present disclosure does not specifically limit the number of the at least one initial video frame, which is based on the number of video frames containing the first object in the first video stream.

[0061] During specific implementation, the first object detection can be performed on each video frame included in the first video stream through an object detection model to obtain a first object detection result, wherein the object detection model is a convolutional neural network model. The present disclosure does not specifically limit the model structure and type of the object detection model, which is subject to implementation. Afterwards, at least one video frame whose confidence of the first object meets the preset confidence requirement can be determined based on the first object detection result, and the at least one video frame can be determined as at least one initial video frame containing the first object in the first video stream. The preset confidence requirement can be: greater than a preset confidence threshold. For example, when the confidence of the first object is greater than 0.9, it can be determined that the confidence of the first object meets the preset confidence requirement.

[0062] Next, a video frame that meets the detection requirement may be determined in the at least one initial video frame as a first video frame to be processed. The detection requirement may include at least one of the following: a size requirement of the first object, a clarity requirement of the initial video frame, a completeness requirement of the first object, and a position requirement of the first object in the initial video frame.

[0063] In an embodiment of the present disclosure, before determining a video frame that meets the detection requirements in at least one initial video frame, the priority order of each detection requirement can be determined first, and then the video frame that meets the detection requirements can be determined in at least one initial video frame based on the priority order.

[0064] For example, the order of priority of the above-mentioned various detection requirements from high to low can be: size requirement of the first object, clarity requirement of the initial video frame, completeness requirement of the first object, and position requirement of the first object in the initial video frame. The present disclosure does not specifically limit the order of priority of various detection requirements, which is subject to what can be implemented.

[0065] At this time, when a video frame that meets the detection requirements is determined in at least one initial video frame, the initial video frames that meet the size requirements of the first object (for example, the size of the first object is greater than 60px*120px, where px can be understood as pixel) can be screened out in the at least one initial video frame, and the screened out initial video frames that meet the size requirements of the first object can be determined as the first group of initial video frames.

[0066] Afterwards, initial video frames that meet the definition requirement may be screened from the first group of initial video frames, and the initial video frames that meet the definition requirement may be determined as the second group of initial video frames.

[0067] Next, the initial video frames that meet the integrity requirement of the first object can be screened out from the second group of initial video frames, and the initial video frames that meet the integrity requirement of the first object can be determined as the third group of initial video frames, and the initial video frames that meet the integrity requirement of the first object can be screened out from the third group of initial video frames, and the initial video frames that meet the position requirement of the first object (for example, the position requirement of the first object can be that the first object is contained in the initial video frame, with the center point of the initial video frame as the center, and the initial video frame The area constitutes 2 / 3 of the area) is determined as the fourth group of initial video frames. After that, an initial video frame can be determined in the fourth group of initial video frames as the first video frame to be processed.

[0068] It should be noted that, when any one of the four groups of initial video frames contains only one initial video frame, the screening is stopped and the one initial video frame is determined as the first video frame to be processed.

[0069] In the above implementation, after screening out at least one initial video frame containing the first object in the first video stream, the video frame that meets the detection requirements in the at least one initial video frame can be determined as the first video frame to be processed, thereby improving the image quality of the first video frame to be processed, and further improving the accuracy of target recognition in the first video frame to be processed.

[0070] In an optional embodiment, if Figure 3 As shown, the embodiment of the present disclosure also includes the following steps:

[0071] Step S31: after generating the device warning information, obtaining a second video stream collected by the target camera device within a preset timing time;

[0072] Step S32: determining a second video frame to be processed in the second video stream when it is determined that the state information has changed based on the second video stream; the state information is used to indicate the monitoring state of the object in the monitoring area of ​​the target camera device;

[0073] Step S33: Detecting the target device on a second object based on the second video frame to be processed to obtain a second device detection result, wherein the second object is an object in which a monitoring status has changed in the monitoring area.

[0074] In an embodiment of the present disclosure, when it is determined based on the first device detection result that the target device is not detected on the first object, device warning information can be generated. At this time, in order to avoid repeatedly generating the device warning information in the same time, a preset timing time can be set, and the device warning information will not be regenerated during the preset timing time.

[0075] Among them, the preset timing time can be 10 seconds, that is, 20 seconds. The present disclosure does not make any specific limitation on the preset timing time, which is subject to meeting actual needs.

[0076] In the embodiment of the present disclosure, after the device warning information is generated, a second video stream collected by the target camera device within a preset timing time can be obtained. The second video stream can be a video stream collected by the target camera device within a preset timing time after the target camera device collects the first video stream.

[0077] After acquiring the second video stream captured by the target camera device within the preset timing time, if the state information can be determined based on the second video stream to change, the second video frame to be processed can be determined in the second video stream. The state information is used to indicate the monitoring state of the object in the monitoring area of ​​the target camera device.

[0078] In the disclosed embodiment, the monitoring state of the object in the monitoring area may be the number of objects in the monitoring area and / or the posture of the first object in the monitoring area. At this time, when the number of objects in the monitoring area changes (for example, the number increases) and / or the posture of the first object in the monitoring area changes, it can be determined that the state information has changed, and at this time, the second video frame to be processed can be determined in the second video stream.

[0079] Here, the increase in the number of objects in the monitoring area can be understood as new objects entering the monitoring area corresponding to the target camera device; the change in the posture of the first object in the monitoring area can be understood as the posture of the first object in the monitoring area changing from a posture that cannot be recognized by the target device to a posture that can be recognized by the target device.

[0080] In the embodiment of the present disclosure, the second video frame to be processed may be determined in the second video stream in the manner described below, specifically including the following process:

[0081] Filter at least one initial video frame containing the second object in the second video stream; here, the first object can be the same as the second object, or can be different from the second object; then determine that a video frame in the at least one initial video frame that meets the detection requirements is a second video frame to be processed, wherein the detection requirements include at least one of the following: a size requirement for the second object, a clarity requirement for the initial video frame, a completeness requirement for the second object, and a position requirement for the second object in the initial video frame.

[0082] Here, the principle of screening at least one initial video frame containing the second object in the second video stream is the same as the screening principle of the above step S21. Here, the principle of determining the video frame that meets the detection requirement in at least one initial video frame as the second video frame to be processed is the same as the principle of determining the first video frame to be processed in the above step S22, and will not be described in detail here.

[0083] After determining the second video frame to be processed in the second video stream, the target device can be detected on the second object based on the second video frame to be processed to obtain a second device detection result. The second object can be an object whose monitoring status has changed in the monitoring area.

[0084] In the above implementation, after the device warning information is generated, the second video stream collected by the target camera device within the preset timing time can be obtained, so that the monitoring status of the object in the monitoring area of ​​the target camera device can be monitored in real time, and more accurate status information of the object in the monitoring area can be obtained. In the case where it is determined based on the second video stream that the status information has changed, the second video frame to be processed can be determined in the second video stream, and the target device can be detected on the second object based on the second video frame to be processed to obtain the second device detection result, so that the generated device warning information can be updated in time, thereby improving the accuracy and real-time performance of target recognition.

[0085] In an optional implementation, with respect to the above steps S31 to S33, the embodiment of the present disclosure specifically includes the following steps:

[0086] When no change in the status information is determined based on the second video stream, the device warning information is generated again according to a preset generation mode after the preset timing time expires.

[0087] In the embodiment of the present disclosure, when no change in the status information is determined based on the second video stream, the device warning information can be generated again according to the preset generation mode after the preset timing time expires.

[0088] The preset generation mode can be used to indicate the display form of the regenerated device warning information. For example, the preset generation mode can include at least the following information: the display color of the device warning information, the font style of the device warning information, the flashing frequency of the device warning information, the display sound effect of the device warning information, etc. The present disclosure does not limit the specific content of the above preset generation mode, which is subject to implementation.

[0089] For example, when the equipment warning information is generated again according to the preset generation mode, the equipment warning information can be displayed as: the display color of the equipment warning information is red, the font style of the equipment warning information is Songti, the flashing frequency of the equipment warning information is 3 times per second, and the display sound effect of the equipment warning information is an alarm sound effect.

[0090] In addition, in the disclosed embodiment, the preset generation mode can be determined based on the number of times the device warning information is generated again. For example, when the number of times the device warning information is generated again is the second time, the flashing frequency of the device warning information in the device preset generation mode can be set to flash 3 times per second, and the display sound effect of the device warning information can be set to 30 decibels; when the number of times the device warning information is generated again is the third time, the flashing frequency of the device warning information in the device preset generation mode can be set to flash 6 times per second, and the display sound effect of the device warning information can be set to 40 decibels, etc., which are not listed here one by one.

[0091] In the above implementation, when no change in status information is determined based on the second video stream, the device warning information can be generated again according to the preset generation mode after the preset timing time expires, thereby making the display form of the device warning information richer and further improving the effectiveness of the device warning information.

[0092] In an optional embodiment, the embodiment of the present disclosure also includes: after generating the device warning information, displaying target display information in a display interface of a terminal device, wherein the target display information includes the device warning information and / or the physical attribute information of the first object, and sending the target display information to the first object.

[0093] In the disclosed embodiment, the target display information includes device warning information and / or physical attribute information of the first object.

[0094] As can be seen from the above description, the device warning information may include at least the following information: object ID, information that the object is not wearing the target device, and warning mark; the physical attribute information of the first object may include at least the following information: work clothes information of the first object, head wearing information of the first object, target device wearing information of the first object, reflective clothing attribute information of the first object, gender information of the first object, age information of the first object, height information of the first object, etc. The present disclosure does not limit the specific content of the physical attribute information of the first object, which is subject to what can be realized.

[0095] In an embodiment of the present disclosure, after the above-mentioned device warning information is generated, the above-mentioned target display information can be displayed in the display interface of the terminal device, and the target display information can be sent to the first object.

[0096] In the embodiment of the present disclosure, when the target display information includes device warning information, the device warning information can be sent to the first object to remind the first object to wear the target device. The device warning information can be sent to the first object in the following manner.

[0097] Method 1: A text message containing the target display information may be sent to the first object based on the contact information (eg, mobile phone number) reserved by the first object on the terminal device.

[0098] Method 2: The device warning information can be sent to the first object through a sound playback device installed in the monitoring area of ​​the target camera device. The sound playback device can be any of the following: an intelligent robot, a loudspeaker, a speaker, etc. The present disclosure does not specifically limit the sound playback device, which is subject to implementation.

[0099] Method three: The first party may be contacted by the staff of the terminal device, and the device warning information may be sent to the first party.

[0100] The present disclosure does not specifically limit the above-mentioned method of sending device warning information to the first object, which is subject to what can be implemented.

[0101] In the above implementation, after the device warning information is generated, the target display information can be displayed in the display interface of the terminal device, and the target display information can be sent to the first object, so that the terminal device staff and the first object can obtain the target display information of the first object in a timely and clear manner, and then the first object can be reminded to perform safe operations in a timely manner. At the same time, the terminal device staff can also better perform data analysis on the physical attribute information of the first object to improve security performance.

[0102] In an optional implementation, with respect to S103, detecting a target device matching the target device type on the first object based on the first video frame to be processed to obtain a first device detection result specifically includes the following steps:

[0103] Step S41: determining device attributes of devices included in the first to-be-processed video frame, wherein the device attributes are used to indicate device categories and / or device locations of the included devices;

[0104] Step S42: obtaining the object position of the first object in the first to-be-processed video frame;

[0105] Step S43: Determine the first device detection result based on the object position and the device attributes.

[0106] In an embodiment of the present disclosure, when detecting a target device on a first object based on a first video frame to be processed, device properties of all devices contained in the first video frame to be processed may be obtained first, and the object position of the first object in the first video frame to be processed may be obtained.

[0107] Among them, the device attribute is used to indicate the device category of the device included in the first video frame to be processed, and / or the device position of the included device in the first video frame to be processed, and the object position of the first object is used to indicate the position of the first object in the first video frame to be processed.

[0108] Among them, the device position can be the position of the center point of the area where the device is located, or the position of the vertex of the area where the device is located. The object position can be the position of the center point of the area where the first object is located, or the position of the vertex of the area where the first object is located. The present disclosure does not make specific limitations on the above-mentioned device positions and object positions, which are subject to what can be implemented.

[0109] Afterwards, a first device detection result may be determined based on the object position and the device attributes.

[0110] In an embodiment of the present disclosure, when it is determined that the device included in the first video frame to be processed belongs to the target device based on the device category of the device included in the first video frame to be processed, and / or when it is determined that the object position of the first object in the first video frame to be processed and the device position of the device included in the first video frame to be processed meet preset position requirements, it is determined that the target device is detected on the first object.

[0111] The preset position requirement can be understood as the distance between the object position and the device position being within a preset range. For example, when the distance between the object position and the preset position is within the range of [20, 40], it can be determined that the object position and the device position meet the preset position requirement. The present disclosure does not specifically limit the preset position requirement, which is subject to achievable implementation.

[0112] At this time, the determined first device detection result may be: the first subject wears the target device, or wears.

[0113] In an optional embodiment, when it is determined based on the device category that the included device does not belong to the target device, and / or when it is determined that the object position and the device position do not meet preset position requirements, it is determined that the target device is not detected on the first object.

[0114] At this time, the determined first device detection result may be: the first subject does not wear the target device, or does not wear it.

[0115] In the above implementation, the first device detection result may be determined based on the object position and the device attribute, so that the obtained first device detection result may be more accurate, thereby improving the accuracy of detecting the target device on the first object.

[0116] In an optional implementation, when the first device detection result includes the confidence level of each preset detection result corresponding to the first object, for S105, determining that the target device is not detected on the first object based on the first device detection result specifically includes:

[0117] When it is determined based on the first device detection result that the confidence level of the first object corresponding to the first preset detection result is greater than a preset confidence threshold, it is determined that the target device is not detected on the first object, wherein the first preset detection result is used to indicate that the first object in the first video frame to be processed is wearing the target device.

[0118] In the embodiment of the present disclosure, the preset detection result corresponding to the first object may include at least the following results: the first object wears the target device (or, wears), the first object does not wear the target device (or, does not wear). The present disclosure does not limit the specific content of the preset detection result corresponding to the first object, which is subject to what can be implemented.

[0119] In the embodiment of the present disclosure, the first preset detection result may be: the first object wears the target device (or, wears).

[0120] In the embodiment of the present disclosure, when the first device detection result includes the confidence of each preset detection result corresponding to the first object, the confidence of the first preset detection result in the first device detection result can be determined. When the confidence of the first preset detection result is greater than the preset confidence threshold, it can be determined that the target device is not detected on the first object.

[0121] The preset confidence threshold may be set to 0.9 or 0.8. The present disclosure does not impose any specific limitation on the preset confidence threshold, which is subject to implementation.

[0122] In the above embodiment, when the detection result of the first device includes the confidence of each preset detection result corresponding to the first object, it can be determined that the target device is not detected on the first object when the confidence of the first preset detection result is greater than the preset confidence threshold based on the detection result of the first device. Thus, it can be determined more accurately that the target device is not detected on the first object, thereby reducing errors and improving the accuracy of detecting the target device on the first object.

[0123] The following will combine Figure 4 The embodiment shown introduces the implementation process of the above target identification in detail.

[0124] After acquiring the first video stream captured by the target camera device, a first to-be-processed video frame containing the first object may be determined in the first video stream, such as Figure 4 As shown. Afterwards, the target device type can be determined based on the acquisition area of ​​the first video stream, and a target device matching the target device type can be detected on the first object based on the first video frame to be processed to obtain a first device detection result. In the case where it is determined based on the first device detection result that the target device is not detected on the first object, device warning information is generated. In the case where it is determined based on the first device detection result that the target device is detected on the first object, the physical attribute information of the first object can be displayed on the display interface of the terminal device.

[0125] Next, the second video stream captured by the target camera device within the preset timing time can be obtained, and when it is determined based on the second video stream that the state information has changed, the second video frame to be processed is determined in the second video stream, and the target device is detected on the second object based on the second video frame to be processed to obtain a second device detection result. When it is determined based on the second device detection result that the target device is not detected on the second object, device warning information is generated. When it is determined based on the second device detection result that the target device is detected on the second object, the physical attribute information of the second object can be pushed to the monitoring party.

[0126] When it is determined based on the second video stream that the status information has not changed, the target device can be detected on the first object after the preset timing time expires, and when the target device is not detected on the first object, the device warning information can be generated again according to the preset generation mode.

[0127] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0128] Based on the same inventive concept, a target identification device corresponding to the target identification method is also provided in the embodiment of the present disclosure. Since the principle of solving the problem by the device in the embodiment of the present disclosure is similar to the above-mentioned target identification method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0129] Reference Figure 5 FIG. 1 is a schematic diagram of a target recognition device provided by an embodiment of the present disclosure, wherein the device comprises: an acquisition unit 51, a determination unit 52, a detection unit 53, and a generation unit 54; wherein:

[0130] An acquisition unit 51 is used to acquire a first video stream captured by a target camera device, and determine a first to-be-processed video frame containing a first object in the first video stream;

[0131] A determination unit 52, configured to determine a target device type based on a capture area of ​​the first video stream;

[0132] A detection unit 53 is used to detect a target device matching the target device type on the first object based on the first video frame to be processed, and obtain a first device detection result, wherein the target device includes a target operating device worn on the first object, and / or a target wearing device for wearing the target operating device on the first object;

[0133] The generating unit 54 is configured to generate device warning information when it is determined based on the first device detection result that the target device is not detected on the first object.

[0134] It can be known from the above description that, first, it is determined that the first video frame to be processed contains the first object in the first video stream collected by the target camera device, and the target device type is determined based on the collection area of ​​the first video stream. Afterwards, the target device matching the target device type is detected on the first object based on the first video frame to be processed to obtain a first device detection result. In the case where it is determined based on the obtained first device detection result that the target device is not detected on the first object, a device warning information can be generated. In the above implementation, the wearing condition of the target device on the first object can be automatically determined through the collected first video stream, so that automatic monitoring of whether the target object is wearing the target device can be achieved, thereby improving the detection efficiency of the target device on the first object. At the same time, in the case where the target device is not detected on the first object, a device warning information can be generated, so that the safety of the first object in performing operations can be further improved.

[0135] In one possible implementation, the determination unit is further used to: screen at least one initial video frame containing the first object in the first video stream; determine a video frame in the at least one initial video frame that meets the detection requirements as the first video frame to be processed, wherein the detection requirements include at least one of the following: a size requirement for the first object, a clarity requirement for the initial video frame, a completeness requirement for the first object, and a position requirement for the first object in the initial video frame.

[0136] In one possible implementation, the determination unit is further used to: after generating the device warning information, obtain a second video stream collected by the target camera device within a preset timing time; when it is determined based on the second video stream that the status information has changed, determine a second video frame to be processed in the second video stream; the status information is used to indicate the monitoring status of an object within the monitoring area of ​​the target camera device; based on the second video frame to be processed, detect the target device on a second object to obtain a second device detection result, wherein the second object is an object whose monitoring status has changed in the monitoring area.

[0137] In a possible implementation, the generation unit is further configured to: if no change in the status information is determined based on the second video stream, generate the device warning information again according to a preset generation mode after the preset timing time expires.

[0138] In a possible implementation, the generation unit is further used to: after generating the device warning information, display target display information in a display interface of a terminal device, wherein the target display information includes the device warning information and / or the physical attribute information of the first object, and send the target display information to the first object.

[0139] In a possible implementation, the detection unit is further used to: determine device attributes of a device included in the first video frame to be processed, wherein the device attributes are used to indicate a device category and / or a device location of the included device; obtain the object location of the first object in the first video frame to be processed; and determine the first device detection result based on the object location and the device attributes.

[0140] In a possible implementation, the detection unit is further used to: determine that the target device is not detected on the first object when it is determined based on the device category that the included device does not belong to the target device, and / or when it is determined that the object position and the device position do not meet preset position requirements.

[0141] In one possible implementation, the generation unit is further used to: when the first device detection result includes the confidence of the first object corresponding to each preset detection result, and when it is determined based on the first device detection result that the confidence of the first object corresponding to the first preset detection result is greater than a preset confidence threshold, determine that the target device is not detected on the first object, wherein the first preset detection result is used to indicate that the first object in the first video frame to be processed is wearing the target device.

[0142] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference may be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.

[0143] Corresponds to Figure 1 The target recognition method in the present disclosure also provides a computer device 600, such as Figure 6 FIG. 6 is a schematic diagram of the structure of a computer device 600 provided in an embodiment of the present disclosure, including:

[0144] Processor 61, memory 62, and bus 63; memory 62 is used to store execution instructions, including internal memory 621 and external memory 622; the internal memory 621 is also called internal memory, which is used to temporarily store the operation data in the processor 61 and the data exchanged with the external memory 622 such as a hard disk. The processor 61 exchanges data with the external memory 622 through the internal memory 621. When the computer device 600 is running, the processor 61 communicates with the memory 62 through the bus 63, so that the processor 61 executes the following instructions:

[0145] Acquire a first video stream captured by a target camera device, and determine a first to-be-processed video frame containing a first object in the first video stream;

[0146] Determining a target device type based on a capture area of ​​the first video stream;

[0147] Detecting a target device matching the target device type on the first object based on the first video frame to be processed, and obtaining a first device detection result, wherein the target device includes a target operating device worn on the first object, and / or a target wearing device for wearing the target operating device on the first object;

[0148] When it is determined based on the first device detection result that the target device is not detected on the first object, device warning information is generated.

[0149] The present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the target recognition method described in the above method embodiment are executed. The storage medium can be a volatile or non-volatile computer-readable storage medium.

[0150] The embodiments of the present disclosure also provide a computer program product, which carries a program code. The instructions included in the program code can be used to execute the steps of the target identification method described in the above method embodiment. For details, please refer to the above method embodiment, which will not be repeated here.

[0151] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is implemented as a computer storage medium. In another optional embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0152] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0153] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0154] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0155] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0156] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A target recognition method, characterized in that: include: Acquire a first video stream captured by a target camera device, and determine a first to-be-processed video frame containing a first object in the first video stream; Determining a target device type based on a capture area of ​​the first video stream; Detecting a target device matching the target device type on the first object based on the first video frame to be processed, and obtaining a first device detection result, wherein the target device includes a target operating device worn on the first object, and / or a target wearing device for wearing the target operating device on the first object; When it is determined based on the first device detection result that the target device is not detected on the first object, device warning information is generated.

2. The method according to claim 1, characterized in that The step of determining a first to-be-processed video frame containing a first object in the first video stream includes: Filtering the first video stream to include at least one initial video frame of the first object; Determine a video frame among the at least one initial video frame that meets the detection requirements as the first video frame to be processed, wherein the detection requirements include at least one of the following: a size requirement for the first object, a clarity requirement for the initial video frame, a completeness requirement for the first object, and a position requirement for the first object in the initial video frame.

3. The method according to claim 1 or 2, characterized in that: The method further comprises: After generating the device warning information, obtaining a second video stream captured by the target camera device within a preset timing time; In the case where it is determined based on the second video stream that the state information has changed, determining a second video frame to be processed in the second video stream; the state information is used to indicate a monitoring state of an object in a monitoring area of ​​the target camera device; The target device is detected on a second object based on the second video frame to be processed, and a second device detection result is obtained, wherein the second object is an object in which a monitoring status of the monitoring area changes.

4. The method according to claim 3, characterized in that The method further comprises: When no change in the status information is determined based on the second video stream, the device warning information is generated again according to a preset generation mode after the preset timing time expires.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: After the device warning information is generated, target display information is displayed in a display interface of a terminal device, wherein the target display information includes the device warning information and / or the physical attribute information of the first object, and the target display information is sent to the first object.

6. The method according to any one of claims 1 to 5, characterized in that The detecting, based on the first to-be-processed video frame, a target device matching the target device type on the first object to obtain a first device detection result includes: Determining device attributes of devices included in the first to-be-processed video frame, wherein the device attributes are used to indicate a device category and / or a device location of the included devices; Acquire an object position of the first object in the first to-be-processed video frame; The first device detection result is determined based on the object position and the device attributes.

7. The method according to claim 6, characterized in that The determining, based on the first device detection result, that the target device is not detected on the first object includes: When it is determined based on the device category that the included device does not belong to the target device and / or when it is determined that the object position and the device position do not meet a preset position requirement, it is determined that the target device is not detected on the first object.

8. The method according to any one of claims 1 to 7, characterized in that The first device detection result includes the confidence level of each preset detection result corresponding to the first object; The determining, based on the first device detection result, that the target device is not detected on the first object includes: When it is determined based on the first device detection result that the confidence level of the first object corresponding to the first preset detection result is greater than a preset confidence threshold, it is determined that the target device is not detected on the first object, wherein the first preset detection result is used to indicate that the first object in the first video frame to be processed is wearing the target device.

9. A target recognition device, characterized in that: include: An acquisition unit, configured to acquire a first video stream captured by a target camera device, and determine a first to-be-processed video frame containing a first object in the first video stream; A determination unit, configured to determine a target device type based on a capture area of ​​the first video stream; a detection unit, configured to detect a target device matching the target device type on the first object based on the first video frame to be processed, and obtain a first device detection result, wherein the target device includes a target operating device worn on the first object, and / or a target wearing device for wearing the target operating device on the first object; A generating unit is used to generate device warning information when it is determined based on the first device detection result that the target device is not detected on the first object.

10. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the target recognition method as described in any one of claims 1 to 8 are performed.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the target recognition method according to any one of claims 1 to 8 are executed.

Images