Safety helmet shielding detection method, device, equipment, medium and product

Abstract

The invention discloses a safety helmet shielding detection method and device, equipment, a medium and a product. The invention relates to the technical field of target detection. The method comprises the following steps: acquiring a video frame of a construction area at the current moment; inputting the video frame of the construction area at the current moment into a pre-trained safety helmet detection model to obtain a safety helmet detection result at the current moment; under the condition that the safety helmet detection result at the current moment is that the target object does not wear the safety helmet, secondary verification is carried out on the safety helmet detection result at the current moment, and if the secondary verification result is that the target object wears the safety helmet, the safety helmet is detected. And if yes, updating the safety helmet detection result at the current moment as that the target object wears the safety helmet and the safety helmet is shielded. According to the technical scheme, through the secondary verification of the detection result that the target object does not wear the safety helmet, the shielding condition of the safety helmet is detected, the detection precision of whether the object wears the safety helmet is improved, and the condition of false alarm is avoided.

Description

Technical Field

[0001] The present disclosure relates to the technical field of object detection, and in particular, to a method, device, equipment, medium and product for detecting helmet occlusion. Background Art

[0002] With the rapid development of artificial intelligence technology, the application of artificial intelligence technology in object detection research has become increasingly widespread.

[0003] In the current helmet detection technology, the video stream of the construction area is often detected by a trained helmet detection model, resulting in inaccurate detection results and false alarms. Summary of the Invention

[0004] The present disclosure provides a method, device, equipment, medium and product for detecting helmet occlusion, which realizes the detection of the helmet occlusion situation, thereby improving the detection accuracy of whether an object wears a helmet.

[0005] According to one aspect of the present disclosure, a method for detecting helmet occlusion is provided, including:

[0006] Obtaining a video frame of the construction area at the current moment;

[0007] Inputting the video frame of the construction area at the current moment into a pre-trained helmet detection model to obtain a helmet detection result at the current moment;

[0008] In the case where the helmet detection result at the current moment is that the target object does not wear a helmet, performing secondary verification on the helmet detection result at the current moment. If the secondary verification result is that the target object wears a helmet, updating the helmet detection result at the current moment to that the target object wears a helmet and the helmet is occluded.

[0009] According to another aspect of the present disclosure, a device for detecting helmet occlusion is provided, including:

[0010] A video frame acquisition module at the current moment, configured to obtain a video frame of the construction area at the current moment;

[0011] A helmet detection result prediction module at the current moment, configured to input the video frame of the construction area at the current moment into a pre-trained helmet detection model to obtain a helmet detection result at the current moment;

[0012] The secondary verification module for the safety helmet detection result at the current moment is used to perform secondary verification on the safety helmet detection result at the current moment when the safety helmet detection result at the current moment is that the target object is not wearing a safety helmet. If the secondary verification result is that the target object is wearing a safety helmet, the safety helmet detection result at the current moment is updated to that the target object is wearing a safety helmet and the safety helmet is blocked.

[0013] According to another aspect of the present disclosure, there is provided an electronic device, which includes:

[0014] At least one processor;

[0015] And a memory communicatively connected to the at least one processor;

[0016] Wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the safety helmet occlusion detection method according to any embodiment of the present disclosure.

[0017] According to another aspect of the present disclosure, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the safety helmet occlusion detection method according to any embodiment of the present disclosure when executed.

[0018] According to another aspect of the present disclosure, there is provided a computer program product including a computer program, which implements the safety helmet occlusion detection method according to any one of the embodiments of the present disclosure when executed by a processor.

[0019] The technical solution of the embodiment of the present disclosure obtains the video frame of the construction area at the current moment, and then inputs the video frame of the construction area at the current moment into the pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment; when the safety helmet detection result at the current moment is that the target object is not wearing a safety helmet, perform secondary verification on the safety helmet detection result at the current moment. If the secondary verification result is that the target object is wearing a safety helmet, the safety helmet detection result at the current moment is updated to that the target object is wearing a safety helmet and the safety helmet is blocked. In the above technical solution, through the secondary verification of the detection result that the target object is not wearing a safety helmet, the detection of the situation where the safety helmet is blocked is realized, thereby improving the detection accuracy of whether the object wears a safety helmet and avoiding the occurrence of false alarms.

[0020] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. Description of the Drawings

[0021] To more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0022] Figure 1 is a flowchart of a method for detecting helmet occlusion provided according to an embodiment of the present disclosure;

[0023] Figure 2 is a flowchart of another method for detecting helmet occlusion provided according to an embodiment of the present disclosure;

[0024] Figure 3 is a flowchart of another method for detecting helmet occlusion provided according to an embodiment of the present disclosure;

[0025] Figure 4 is a flowchart of another method for detecting helmet occlusion provided according to an embodiment of the present disclosure;

[0026] Figure 5 is a flowchart of another method for detecting helmet occlusion provided according to an embodiment of the present disclosure;

[0027] Figure 6 is a flowchart of another method for detecting helmet occlusion provided according to an embodiment of the present disclosure;

[0028] Figure 7 is a schematic structural diagram of a device for detecting helmet occlusion provided according to an embodiment of the present disclosure;

[0029] Figure 8 is a schematic structural diagram of an electronic device for implementing the method for detecting helmet occlusion in the embodiments of the present disclosure. Detailed implementation manners

[0030] To enable those skilled in the art to better understand the solutions of the present disclosure, the following will clearly and completely describe the technical solutions in the embodiments of the present disclosure in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only some, rather than all, of the embodiments of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present disclosure.

[0031] It should be noted that the terms "first", "second", etc. in the specification, claims and the above-mentioned drawings of the present disclosure are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. The acquisition, storage, use, processing, etc. of data in the technical solution of the present disclosure all comply with the relevant provisions of national laws and regulations.

[0032] Figure 1 FIG. is a flowchart of a method for detecting helmet occlusion provided by an embodiment of the present disclosure. This embodiment is applicable to the situation of detecting whether construction workers wear helmets. This method can be executed by a helmet occlusion detection device, which can be implemented in the form of hardware and / or software, and the helmet occlusion detection device can be configured in electronic devices such as terminals and servers. As Figure 1 shown, the method includes:

[0033] S110. Obtain a video frame of the construction area at the current moment.

[0034] Among them, the construction area refers to a specific range or site where building construction, maintenance, renovation or other engineering activities are being carried out. The video frame can be collected by a camera set in the construction area.

[0035] Exemplarily, through a camera set in the construction area, a video frame of the construction area at the current moment can be collected, or a video frame of the construction area at the current moment can be read from a preset storage path of the electronic device.

[0036] S120. Input the video frame of the construction area at the current moment into a pre-trained helmet detection model to obtain a helmet detection result at the current moment.

[0037] Among them, the helmet detection model can be a detection model trained based on YOLOV8 or an improved YOLOV8.

[0038] Specifically, the video frame of the construction area at the current moment is used as the input data of the model. Then, the video frame of the construction area at the current moment is input into the pre-trained safety helmet detection model. The safety helmet detection model outputs the safety helmet detection result at the current moment. The safety helmet detection result at the current moment can be that the target object is not wearing a safety helmet or the target object is wearing a safety helmet, etc.

[0039] S130. In the case where the safety helmet detection result at the current moment is that the target object is not wearing a safety helmet, perform a secondary verification on the safety helmet detection result at the current moment. If the secondary verification result is that the target object is wearing a safety helmet, update the safety helmet detection result at the current moment to that the target object is wearing a safety helmet and the safety helmet is blocked.

[0040] It should be noted that the first detection result is that the target object is not wearing a safety helmet, but the secondary verification result is that the target object is wearing a safety helmet, indicating that the safety helmet of the target object is blocked at the current moment. Therefore, the safety helmet detection result at the current moment is updated to that the target object is wearing a safety helmet and the safety helmet is blocked, so as to avoid false alarms and improve the detection accuracy of whether the object is wearing a safety helmet.

[0041] The technical solution of the embodiment of the present disclosure obtains the video frame of the construction area at the current moment, and then inputs the video frame of the construction area at the current moment into the pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment; in the case where the safety helmet detection result at the current moment is that the target object is not wearing a safety helmet, perform a secondary verification on the safety helmet detection result at the current moment. If the secondary verification result is that the target object is wearing a safety helmet, update the safety helmet detection result at the current moment to that the target object is wearing a safety helmet and the safety helmet is blocked. In the above technical solution, through the secondary verification of the detection result that the target object is not wearing a safety helmet, the detection of the situation where the safety helmet is blocked is realized, thereby improving the detection accuracy of whether the object is wearing a safety helmet and avoiding false alarms.

[0042] Figure 2 The figure is a flowchart of another safety helmet occlusion detection method provided by the embodiment of the present disclosure. The method of this embodiment can be combined with each optional solution in the safety helmet occlusion detection method provided in the above embodiment. On the basis of the above embodiments, this embodiment further refines the secondary verification process.

[0043] As Figure 2 shown, the method includes:

[0044] S210. Obtain the video frame of the construction area at the current moment.

[0045] S220. Input the video frame of the construction area at the current moment into the pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment.

[0046] S230. When the safety helmet detection result at the current moment indicates that the target object is not wearing a safety helmet, obtain the video frames of the construction area for a preset number before the current moment.

[0047] S240. Input the video frames of the construction area for a preset number before the current moment into the pre-trained safety helmet detection model respectively to obtain the safety helmet detection results at multiple historical moments.

[0048] S250. When the safety helmet detection results at the multiple historical moments include the detection result that the target object is wearing a safety helmet, determine that the secondary verification result is that the target object is wearing a safety helmet.

[0049] S260. If the secondary verification result is that the target object is wearing a safety helmet, update the safety helmet detection result at the current moment to that the target object is wearing a safety helmet and the safety helmet is blocked.

[0050] Wherein, the preset number can be one, two or more than two, and no specific limitation is made here.

[0051] Exemplarily, the video frame of the construction area at the current moment can be represented by P t ; when the preset number is 2, the video frames of the construction area before the current moment can be P t-1 and P t-2 ; input P t-1 and P t-2 into the pre-trained safety helmet detection model respectively to obtain the safety helmet detection result corresponding to P t-1 and the safety helmet detection result corresponding to P t-2 ; if there is a detection result that the target object is wearing a safety helmet in the safety helmet detection result corresponding to P t-1 and the safety helmet detection result corresponding to P t-2 , indicating that the safety helmet of the target object is blocked at the current moment, then update the safety helmet detection result corresponding to P t to that the target object is wearing a safety helmet and the safety helmet is blocked. If there is no detection result that the target object is wearing a safety helmet in the safety helmet detection result corresponding to P t-1 and the safety helmet detection result corresponding to P t-2 , then the safety helmet detection result corresponding to P t remains unchanged.

[0052] The technical solution of the embodiment of the present disclosure obtains video frames of a preset number of construction areas before the current moment, and then inputs the video frames of the preset number of construction areas before the current moment into a pre-trained safety helmet detection model respectively to obtain safety helmet detection results at multiple historical moments. When the safety helmet detection results at multiple historical moments include the detection result that the target object wears a safety helmet, it is determined that the secondary verification result is that the target object wears a safety helmet, realizing the secondary accurate verification of the safety helmet detection result.

[0053] Figure 3 FIG. 4 is a flowchart of another safety helmet occlusion detection method provided by an embodiment of the present disclosure. The method of this embodiment can be combined with each optional solution in the safety helmet occlusion detection method provided in the above embodiment. On the basis of the above embodiments, this embodiment further refines the secondary verification process.

[0054] As Figure 3 shown, the method includes:

[0055] S310. Obtain a video frame of the construction area at the current moment.

[0056] S320. Input the video frame of the construction area at the current moment into a pre-trained safety helmet detection model to obtain a safety helmet detection result at the current moment.

[0057] S330. When the safety helmet detection result at the current moment is that the target object does not wear a safety helmet, obtain video frames of a preset number of construction areas after the current moment.

[0058] S340. Input the video frames of the preset number of construction areas after the current moment into a pre-trained safety helmet detection model respectively to obtain safety helmet detection results at multiple future moments.

[0059] S350. When the safety helmet detection results at the multiple future moments include the detection result that the target object wears a safety helmet, determine that the secondary verification result is that the target object wears a safety helmet.

[0060] S360. If the secondary verification result is that the target object wears a safety helmet, update the safety helmet detection result at the current moment to that the target object wears a safety helmet and the safety helmet is occluded.

[0061] Wherein, the preset number can be one, two or more than two, and no specific limitation is made here.

[0062] Exemplarily, the video frame of the construction area at the current moment can be represented by P t When the preset number is 2, the video frames of the construction area after the current moment can be P t+1 and Pt+2 ; Input P t+1 and P t+2 into the pre-trained safety helmet detection model respectively, and obtain the safety helmet detection result corresponding to P t+1 and the safety helmet detection result corresponding to P t+2 . If there is a detection result that the target object wears a safety helmet in the safety helmet detection result corresponding to P t+1 and the safety helmet detection result corresponding to P t+2 , indicating that the safety helmet of the target object is blocked at the current moment, then update the safety helmet detection result corresponding to P t to the target object wears a safety helmet and the safety helmet is blocked. If there is no detection result that the target object wears a safety helmet in the safety helmet detection result corresponding to P t+1 and the safety helmet detection result corresponding to P t+2 , then the safety helmet detection result corresponding to P t remains unchanged.

[0063] The technical solution of the embodiment of the present disclosure obtains video frames of a preset number of construction areas after the current moment, and then inputs the video frames of the preset number of construction areas after the current moment into the pre-trained safety helmet detection model respectively to obtain safety helmet detection results at multiple future moments. Furthermore, in the case where the safety helmet detection results at multiple future moments include the detection result that the target object wears a safety helmet, it is determined that the secondary verification result is that the target object wears a safety helmet, realizing secondary precise verification of the safety helmet detection result.

[0064] Figure 4 FIG. is a flowchart of another safety helmet occlusion detection method provided by the embodiment of the present disclosure. The method of this embodiment can be combined with each optional solution in the safety helmet occlusion detection method provided in the above embodiment. On the basis of the above embodiments, this embodiment further refines the secondary verification process.

[0065] As Figure 4 shown, the method includes:

[0066] S410. Obtain a video frame of the construction area at the current moment.

[0067] S420. Input the video frame of the construction area at the current moment into the pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment.

[0068] S430. When the safety helmet detection result at the current moment is that the target object does not wear a safety helmet, obtain the camera identifier corresponding to the video frame of the construction area.

[0069] Among them, the camera identifier refers to the identifier of the camera used to capture the video frame, which is unique. For example, the camera identifier can be 0012 or 0015, etc.

[0070] Exemplarily, the camera identifier can be extracted from the name of the video frame file or the name of the folder storing the video frame.

[0071] S440. Increment the camera identifier by 1 to obtain the first associated perspective camera identifier, and obtain the video frame corresponding to the first associated perspective camera identifier; decrement the camera identifier by 1 to obtain the second associated perspective camera identifier, and obtain the video frame corresponding to the second associated perspective camera identifier.

[0072] In the embodiments of the present disclosure, there may be cameras with multiple shooting perspectives in the construction area. There is an overlapping field of view area for adjacent cameras, and the camera identifiers of adjacent cameras are also adjacent. In other words, the video frames of the associated perspectives can be obtained according to the increased or decreased camera identifier.

[0073] Specifically, incrementing the camera identifier by 1 can obtain the identifier of the next adjacent camera, that is, the first associated perspective camera identifier, and then the video frame corresponding to the first associated perspective camera identifier can be obtained; decrementing the camera identifier by 1 can obtain the identifier of the previous adjacent camera, that is, the second associated perspective camera identifier, and then the video frame corresponding to the second associated perspective camera identifier can be obtained.

[0074] S450. Input the video frames corresponding to the first associated perspective camera identifier and the video frames corresponding to the second associated perspective camera identifier into the pre-trained safety helmet detection model respectively to obtain the safety helmet detection results of multiple associated perspectives.

[0075] S460. In the case that the detection result of the target object wearing a safety helmet is included in the safety helmet detection results of the multiple associated perspectives, determine that the secondary verification result is that the target object wears a safety helmet.

[0076] S470. If the secondary verification result is that the target object wears a safety helmet, update the safety helmet detection result at the current moment to that the target object wears a safety helmet and the safety helmet is blocked.

[0077] Exemplarily, the video frame corresponding to the first associated perspective camera identifier can be represented by P1, and the video frame corresponding to the second associated perspective camera identifier can be represented by P2. P1 and P2 are respectively input into a pre-trained safety helmet detection model to obtain the safety helmet detection result corresponding to P1 and the safety helmet detection result corresponding to P2. If there is a detection result that the target object wears a safety helmet in the safety helmet detection result corresponding to P1 and the safety helmet detection result corresponding to P2, it indicates that the safety helmet of the target object is blocked at the current moment, then the safety helmet detection result at the current moment is updated to that the target object wears a safety helmet and the safety helmet is blocked. If there is no detection result that the target object wears a safety helmet in the safety helmet detection result corresponding to P1 and the safety helmet detection result corresponding to P2, then the safety helmet detection result at the current moment remains unchanged.

[0078] The technical solution of the embodiments of the present disclosure obtains the camera identifier corresponding to the video frame of the construction area, adds 1 to the camera identifier to obtain the first associated perspective camera identifier, and obtains the video frame corresponding to the first associated perspective camera identifier; subtracts 1 from the camera identifier to obtain the second associated perspective camera identifier, and obtains the video frame corresponding to the second associated perspective camera identifier. Then, the video frame corresponding to the first associated perspective camera identifier and the video frame corresponding to the second associated perspective camera identifier are respectively input into a pre-trained safety helmet detection model to obtain the safety helmet detection results of multiple associated perspectives. In the case where the safety helmet detection results of multiple associated perspectives include the detection result that the target object wears a safety helmet, it is determined that the secondary verification result is that the target object wears a safety helmet, realizing the secondary accurate verification of the safety helmet detection result.

[0079] Figure 5 It is a flowchart of another safety helmet occlusion detection method provided by the embodiments of the present disclosure. The method of this embodiment can be combined with each optional solution in the safety helmet occlusion detection method provided in the above embodiments. On the basis of the above embodiments, this embodiment further refines the secondary verification process.

[0080] As Figure 5 shown, the method includes:

[0081] S510. Obtain the video frame of the construction area at the current moment.

[0082] S520. Input the video frame of the construction area at the current moment into a pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment.

[0083] S530. When the safety helmet detection result at the current moment is that the target object does not wear a safety helmet, obtain the camera identifier corresponding to the video frame of the construction area.

[0084] S540. Determine the location information of the construction area based on the camera identifier; send the location information of the construction area to the UAV control terminal to control the UAV to fly to the construction area.

[0085] Among them, the location information of the construction area refers to the map coordinates of the construction area where the camera is located.

[0086] Specifically, a mapping relationship table between the camera identifier and the map coordinates is preset. The mapping relationship table contains multiple camera identifiers and the coordinates of the installation positions of the cameras corresponding to the camera identifiers. Further, the camera identifier corresponding to the video frame of the construction area can be matched in the mapping relationship table to obtain the location information of the construction area.

[0087] In the embodiments of the present disclosure, the electronic device is communicatively connected to the UAV control terminal. The electronic device can send the location information of the construction area to the UAV control terminal. Then, the UAV control terminal controls the UAV to fly near the construction area according to the location information of the construction area to perform tracking shooting on the target object and obtain the video frames of the construction area collected by the UAV. The UAV control terminal can be a personal computer or a mobile terminal, etc., which is not specifically limited herein.

[0088] S550. Receive the video frames of the construction area collected by the UAV; input the video frames of the construction area collected by the UAV into a pre-trained safety helmet detection model to obtain the safety helmet detection result from the perspective of the UAV.

[0089] Among them, the number of video frames of the construction area collected by the UAV can be one or more. When the number of video frames of the construction area collected by the UAV is more than one, multiple safety helmet detection results from the perspective of the UAV can be predicted through the safety helmet detection model. Then, the secondary verification result is determined according to the multiple safety helmet detection results from the perspective of the UAV.

[0090] S560. When the safety helmet detection result from the perspective of the UAV includes the detection result that the target object wears a safety helmet, determine that the secondary verification result is that the target object wears a safety helmet.

[0091] S570. If the secondary verification result is that the target object wears a safety helmet, update the safety helmet detection result at the current moment to that the target object wears a safety helmet and the safety helmet is blocked.

[0092] Exemplarily, the video frame of the construction area collected by the drone can be represented by Pf. The Pf is input into a pre-trained safety helmet detection model to obtain the safety helmet detection result corresponding to Pf. If there is a detection result that the target object wears a safety helmet in the safety helmet detection result corresponding to Pf, it indicates that the safety helmet of the target object is blocked at the current moment. Then, the safety helmet detection result at the current moment is updated to that the target object wears a safety helmet and the safety helmet is blocked. If there is no detection result that the target object wears a safety helmet in the safety helmet detection result corresponding to Pf, the safety helmet detection result at the current moment remains unchanged.

[0093] The technical solution of the embodiment of the present disclosure obtains the camera identifier corresponding to the video frame of the construction area, and then determines the location information of the construction area based on the camera identifier, and sends the location information of the construction area to the drone control terminal to control the drone to fly to the construction area. Furthermore, it receives the video frame of the construction area collected by the drone, inputs the video frame of the construction area collected by the drone into a pre-trained safety helmet detection model to obtain the safety helmet detection result from the perspective of the drone. When the safety helmet detection result from the perspective of the drone includes the detection result that the target object wears a safety helmet, it is determined that the secondary verification result is that the target object wears a safety helmet, realizing the secondary accurate verification of the safety helmet detection result.

[0094] Figure 6 The flowchart of another safety helmet occlusion detection method provided by the embodiment of the present disclosure. The method of this embodiment can be combined with each optional solution in the safety helmet occlusion detection method provided in the above embodiment. On the basis of the above embodiments, this embodiment adds a step of warning for not wearing a safety helmet.

[0095] As Figure 6 shown, the method includes:

[0096] S610. Obtain the video frame of the construction area at the current moment.

[0097] S620. Input the video frame of the construction area at the current moment into a pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment.

[0098] S630. When the safety helmet detection result at the current moment is that the target object does not wear a safety helmet, perform secondary verification on the safety helmet detection result at the current moment. If the secondary verification result is that the target object wears a safety helmet, update the safety helmet detection result at the current moment to that the target object wears a safety helmet and the safety helmet is blocked.

[0099] S640. If the secondary verification result indicates that the target object is not wearing a safety helmet, an alarm message for not wearing a safety helmet is generated; the camera identifier corresponding to the video frame of the construction area is obtained, and the alarm message for not wearing a safety helmet is sent to the player associated with the camera identifier to play the alarm message for not wearing a safety helmet.

[0100] Among them, the alarm message for not wearing a safety helmet is used to alarm the person not wearing a safety helmet, which can be a voice broadcast "Please wear a safety helmet for operation" or a warning sound "Didi Di", etc. The player can be a speaker or other voice player.

[0101] In the embodiments of the present disclosure, a player is integrated in each camera of the collected video frames, or a player is set near the camera to achieve targeted reminder or alarm for the person not wearing a safety helmet under the camera.

[0102] The technical solution of the embodiments of the present disclosure realizes targeted alarm for the object not wearing a safety helmet and improves the accuracy of the alarm for not wearing a safety helmet by obtaining the camera identifier corresponding to the video frame of the construction area and sending the alarm message for not wearing a safety helmet to the player associated with the camera identifier to play the alarm message for not wearing a safety helmet.

[0103] Figure 7 This is a schematic structural diagram of a safety helmet occlusion detection device provided by the embodiments of the present disclosure. As Figure 7 shown, the device includes:

[0104] A video frame acquisition module 710 at the current moment, configured to acquire a video frame of the construction area at the current moment;

[0105] A safety helmet detection result prediction module 720 at the current moment, configured to input the video frame of the construction area at the current moment into a pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment;

[0106] A secondary verification module 730 for the safety helmet detection result at the current moment, configured to perform secondary verification on the safety helmet detection result at the current moment when the safety helmet detection result at the current moment indicates that the target object is not wearing a safety helmet. If the secondary verification result indicates that the target object is wearing a safety helmet, the safety helmet detection result at the current moment is updated to that the target object is wearing a safety helmet and the safety helmet is occluded.

[0107] The technical solution of the embodiment of the present disclosure obtains the video frame of the construction area at the current moment, and then inputs the video frame of the construction area at the current moment into the pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment; in the case that the safety helmet detection result at the current moment is that the target object is not wearing a safety helmet, the safety helmet detection result at the current moment is verified twice. If the double-verification result is that the target object is wearing a safety helmet, the safety helmet detection result at the current moment is updated to that the target object is wearing a safety helmet and the safety helmet is blocked. In the above technical solution, by double-verifying the detection result that the target object is not wearing a safety helmet, the detection of the situation where the safety helmet is blocked is realized, thereby improving the detection accuracy of whether the object is wearing a safety helmet and avoiding false alarms.

[0108] Based on any optional technical solution in the embodiment of the present disclosure, optionally, the current moment safety helmet detection result double-verification module 730 includes:

[0109] The first double-verification unit is used to obtain a preset number of video frames of the construction area before the current moment; input the preset number of video frames of the construction area before the current moment into the pre-trained safety helmet detection model respectively to obtain the safety helmet detection results at multiple historical moments; in the case that the safety helmet detection results at the multiple historical moments include the detection result that the target object is wearing a safety helmet, determine that the double-verification result is that the target object is wearing a safety helmet.

[0110] Based on any optional technical solution in the embodiment of the present disclosure, optionally, the current moment safety helmet detection result double-verification module 730 includes:

[0111] The second double-verification unit is used to obtain a preset number of video frames of the construction area after the current moment; input the preset number of video frames of the construction area after the current moment into the pre-trained safety helmet detection model respectively to obtain the safety helmet detection results at multiple future moments; in the case that the safety helmet detection results at the multiple future moments include the detection result that the target object is wearing a safety helmet, determine that the double-verification result is that the target object is wearing a safety helmet.

[0112] Based on any optional technical solution in the embodiment of the present disclosure, optionally, the current moment safety helmet detection result double-verification module 730 includes:

[0113] A third secondary verification unit, configured to, when the hard hat detection result at the current moment indicates that the target object is not wearing a hard hat, obtain the camera identifier corresponding to the video frame of the construction area; increment the camera identifier by 1 to obtain a first associated perspective camera identifier, and obtain the video frame corresponding to the first associated perspective camera identifier; decrement the camera identifier by 1 to obtain a second associated perspective camera identifier, and obtain the video frame corresponding to the second associated perspective camera identifier; input the video frame corresponding to the first associated perspective camera identifier and the video frame corresponding to the second associated perspective camera identifier into a pre-trained hard hat detection model respectively to obtain hard hat detection results of multiple associated perspectives; and when the hard hat detection results of the multiple associated perspectives include a detection result that the target object is wearing a hard hat, determine that the secondary verification result is that the target object is wearing a hard hat.

[0114] Based on any optional technical solution in the embodiments of the present disclosure, optionally, the hard hat detection result secondary verification module 730 at the current moment includes:

[0115] A fourth secondary verification unit, configured to, when the hard hat detection result at the current moment indicates that the target object is not wearing a hard hat, obtain the camera identifier corresponding to the video frame of the construction area; determine the construction area location information based on the camera identifier; send the construction area location information to the unmanned aerial vehicle control terminal to control the unmanned aerial vehicle to fly to the construction area; receive the video frame of the construction area collected by the unmanned aerial vehicle; input the video frame of the construction area collected by the unmanned aerial vehicle into a pre-trained hard hat detection model to obtain the hard hat detection result from the perspective of the unmanned aerial vehicle; and when the hard hat detection result from the perspective of the unmanned aerial vehicle includes a detection result that the target object is wearing a hard hat, determine that the secondary verification result is that the target object is wearing a hard hat.

[0116] Based on any optional technical solution in the embodiments of the present disclosure, optionally, the hard hat occlusion detection device includes:

[0117] A non-wearing hard hat warning module, configured to generate a non-wearing hard hat warning message if the secondary verification result is that the target object is not wearing a hard hat; obtain the camera identifier corresponding to the video frame of the construction area, and send the non-wearing hard hat warning message to the player associated with the camera identifier to play the non-wearing hard hat warning message.

[0118] The hard hat occlusion detection device provided in the embodiments of the present disclosure can execute the hard hat occlusion detection method provided in any embodiment of the present disclosure, and has corresponding functional modules and beneficial effects for executing the method.

[0119] Figure 8The structural schematic diagram of the electronic device 10 that can be used to implement the embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0120] As Figure 8 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. The memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other via a bus 14. The I / O interface 15 is also connected to the bus 14.

[0121] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0122] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the hard hat occlusion detection method, which includes:

[0123] Obtaining a video frame of the construction area at the current moment;

[0124] Input the video frame of the construction area at the current moment into the pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment;

[0125] In the case that the safety helmet detection result at the current moment is that the target object is not wearing a safety helmet, perform a secondary verification on the safety helmet detection result at the current moment. If the secondary verification result is that the target object is wearing a safety helmet, update the safety helmet detection result at the current moment to that the target object is wearing a safety helmet and the safety helmet is blocked.

[0126] In some embodiments, the safety helmet occlusion detection method can be implemented as a computer program, which is tangibly included in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the safety helmet occlusion detection method described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the safety helmet occlusion detection method by any other suitable means (e.g., by means of firmware).

[0127] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs, which can be executed and / or interpreted on a programmable system including at least one programmable processor. The programmable processor can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0128] The computer program for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a dedicated computer, or other programmable data processing devices, such that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0129] In the context of this disclosure, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0130] To provide for interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0131] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0132] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0133] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added, or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions of this disclosure can be achieved, and no limitation is made herein.

[0134] The embodiments of this disclosure also provide a computer program product, including a computer program which, when executed by a processor, implements the hard hat occlusion detection method provided in any embodiment of this disclosure.

[0135] In the process of implementing the computer program product, the computer program code for performing the operations of this disclosure can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0136] In particular, according to the embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments of the present invention include a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through the communication unit 19, or installed from the storage unit 18, or installed from the ROM 12. When the computer program is executed by the processor 11, the above-mentioned functions defined in the methods of the embodiments of the present invention are performed.

[0137] The above specific embodiments do not constitute a limitation to the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. A method for detecting helmet occlusion, characterized in that, including: Obtain a video frame of the construction area at the current moment; Input the video frame of the construction area at the current moment into a pre-trained safety helmet detection model to obtain the safety helmet detection result at the current moment; In the case that the safety helmet detection result at the current moment is that the target object is not wearing a safety helmet, perform secondary verification on the safety helmet detection result at the current moment. If the secondary verification result is that the target object is wearing a safety helmet, update the safety helmet detection result at the current moment to that the target object is wearing a safety helmet and the safety helmet is blocked.

2. The method according to claim 1, characterized in that, The secondary verification of the safety helmet detection result at the current moment includes: Obtain video frames of the construction area in a preset number before the current moment; Input the video frames of the construction area in a preset number before the current moment into a pre-trained safety helmet detection model respectively to obtain safety helmet detection results at multiple historical moments; In the case that the detection results of the target object wearing a safety helmet are included in the safety helmet detection results at the multiple historical moments, determine that the secondary verification result is that the target object is wearing a safety helmet.

3. The method according to claim 1, wherein The secondary verification of the safety helmet detection result at the current moment includes: Obtain video frames of the construction area in a preset number after the current moment; Input the video frames of the construction area in a preset number after the current moment into a pre-trained safety helmet detection model respectively to obtain safety helmet detection results at multiple future moments; In the case that the detection results of the target object wearing a safety helmet are included in the safety helmet detection results at the multiple future moments, determine that the secondary verification result is that the target object is wearing a safety helmet.

4. The method according to claim 1, wherein The secondary verification of the safety helmet detection result at the current moment includes: Obtain the camera identifier corresponding to the video frame of the construction area; Increment the camera identifier by 1 to obtain the first associated perspective camera identifier, and obtain the video frame corresponding to the first associated perspective camera identifier; Decrement the camera identifier by 1 to obtain the second associated perspective camera identifier, and obtain the video frame corresponding to the second associated perspective camera identifier; Input the video frame corresponding to the first associated perspective camera identifier and the video frame corresponding to the second associated perspective camera identifier into a pre-trained safety helmet detection model respectively to obtain safety helmet detection results at multiple associated perspectives; In the case that the detection results of the target object wearing a safety helmet are included in the safety helmet detection results at the multiple associated perspectives, determine that the secondary verification result is that the target object is wearing a safety helmet.

5. The method according to claim 1, wherein The secondary verification of the safety helmet detection result at the current moment includes: Obtain the camera identifier corresponding to the video frame of the construction area; Determine the construction area location information based on the camera identifier; Send the construction area location information to the UAV control terminal to control the UAV to fly to the construction area; Receive the video frame of the construction area collected by the UAV; Input the video frame of the construction area collected by the UAV into a pre-trained safety helmet detection model to obtain the safety helmet detection result from the UAV perspective; When the helmet detection result from the perspective of the UAV includes the detection result that the target object is wearing a helmet, determine that the secondary verification result is that the target object is wearing a helmet.

6. According to the method described in any one of claims 1-5, characterized in that, After the secondary verification of the helmet detection result at the current moment, it further includes: If the secondary verification result is that the target object is not wearing a helmet, generate a warning message for not wearing a helmet. Obtain the camera identifier corresponding to the video frame of the construction area, and send the warning message for not wearing a helmet to the player associated with the camera identifier to play the warning message for not wearing a helmet.

7. A safety helmet occlusion detection device, characterized in that, It includes: A video frame acquisition module at the current moment, which is used to acquire the video frame of the construction area at the current moment. A helmet detection result prediction module at the current moment, which is used to input the video frame of the construction area at the current moment into a pre-trained helmet detection model to obtain the helmet detection result at the current moment. A secondary verification module for the helmet detection result at the current moment, which is used to perform secondary verification on the helmet detection result at the current moment when the helmet detection result at the current moment is that the target object is not wearing a helmet. If the secondary verification result is that the target object is wearing a helmet, update the helmet detection result at the current moment to that the target object is wearing a helmet and the helmet is blocked.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; And a memory communicatively connected to the at least one processor; Wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the helmet occlusion detection method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the helmet occlusion detection method according to any one of claims 1-6 when executed by a processor.

10. A computer program product, characterized in that, The computer program product includes a computer program, and the computer program implements the helmet occlusion detection method according to any one of claims 1-6 when executed by a processor.

Images