Inspection robot detection and identification method, device and equipment and storage medium

Abstract

The invention discloses an inspection robot detection and identification method, device and equipment and a storage medium, and relates to the technical field of park inspection. The method comprises the following steps: starting a video stream decoding process when an initialization starting signal is detected, and writing a to-be-detected video frame collected by a camera of an inspection robot into a cache queue; after the to-be-detected video frame is taken out and subjected to format conversion, whether human shape data information exists in the to-be-detected video frame or not is detected; if yes, calculating a first azimuth angle of the human-shaped target in the human-shaped data information, and locking the human-shaped target according to the first azimuth angle; and performing face recognition detection on the human-shaped target, and pushing alarm data information to a monitoring background when judging that the human-shaped target is a strange person according to a detection result. Whether the video data collected by the park inspection robot during the inspection task contains the human-shaped data is judged, the camera is aligned with the human-shaped target by calculating the azimuth angle, and then the human face recognition judgment is performed again, so that the accuracy of personnel recognition judgment is improved.

Description

Technical Field

[0001] The present application relates to the field of park inspection technology, and in particular to a method, device, equipment and storage medium for inspection robot detection and identification. Background Art

[0002] The park inspection robot is an electronic device used for security, environmental monitoring and facility maintenance in the park. It is usually equipped with a variety of sensors, such as visible light cameras and infrared thermal imagers, which can sense the environmental information in the park in real time and transmit the data to the control center for analysis. The robot can move in the park inspection area and monitor various safety hazards in the park in real time.

[0003] In the existing technology, the inspection robot has weak detection and identification capabilities for safety hazards in complex environments. It is unable to accurately determine personnel information when performing inspection tasks, resulting in misjudgment or missed judgment, leaving safety hazards for park inspections.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present application is to provide a patrol robot detection and identification method, device, equipment and storage medium, aiming to solve the technical problem of misjudgment and missed judgment caused by the weak detection and identification ability of the patrol robot in the prior art.

[0006] To achieve the above objectives, the present application proposes a patrol robot detection and identification method, the patrol robot detection and identification method comprising: When the initialization start signal is detected, the video stream decoding process is started, and the video frames to be detected collected by the inspection robot camera are written into the cache queue; After taking out the video frame to be detected and converting the format, detecting whether there is human figure data information in the video frame to be detected; If yes, calculating a first azimuth angle of the humanoid target in the humanoid data information, so as to lock the humanoid target according to the first azimuth angle; Perform face recognition detection on the humanoid target, and push alarm data information to the monitoring background when it is determined that the humanoid target is a stranger based on the detection result.

[0007] In one embodiment, after the step of taking out the video frame to be detected and converting the format, detecting whether there is human figure data information in the video frame includes: Determining whether the number of human-shaped targets in the to-be-detected video frame exceeds one according to the human-shaped data information; If yes, the step of calculating the first azimuth of the humanoid target in the humanoid data information to lock the humanoid target according to the first azimuth includes: Calculating the first azimuth angle of the humanoid target with the largest area in the humanoid data information; A pan / tilt rotation instruction is set according to the first azimuth angle to drive the pan / tilt to lock onto the humanoid target.

[0008] In one embodiment, the step of performing face recognition detection on the humanoid target and pushing alarm data information to the monitoring background when judging that the humanoid target is a stranger according to the detection result includes: Collecting facial data information after the gimbal locks onto the human-shaped target; Analyze the facial feature points of the facial data information, and calculate the feature similarity between the facial feature points and the preset personnel data; When the feature similarity is lower than a preset judgment threshold, the human-shaped target is judged to be a stranger; The inspection task of the inspection robot is suspended, the pan / tilt head is kept locked on the humanoid target, and the alarm data information is pushed to the monitoring background.

[0009] In one embodiment, after the steps of pausing the inspection task of the inspection robot, keeping the pan / tilt head locked on the humanoid target, and pushing the alarm data information to the monitoring background, the following steps are further included: After the inspection robot receives the processing instruction from the background, it performs corresponding actions according to the processing instruction.

[0010] In one embodiment, after the steps of analyzing the facial feature points of the facial data information and calculating the feature similarity between the facial feature points and the preset personnel data, the method further includes: When the feature similarity is higher than a preset judgment threshold, the humanoid target is judged to be a compliant person; Unlock the pan / tilt platform and drive the inspection robot to continue to perform the inspection task.

[0011] In one embodiment, after the step of starting the video stream decoding process when the initialization start signal is detected and writing the video frame to be detected collected by the inspection robot camera into the cache queue, it also includes: After taking out the video frame to be detected and converting the format, detecting whether there is vehicle data information in the video frame to be detected; If yes, the alarm data information is pushed to the monitoring background, driving the inspection robot to continue to perform the inspection task, and starting the timer at the same time; After the timer reaches a preset time, the to-be-detected video frame is redetected until the to-be-detected video frame does not contain the vehicle data information.

[0012] In one embodiment, after the step of starting the video stream decoding process when the initialization start signal is detected and writing the video frame to be detected collected by the inspection robot camera into the cache queue, it also includes: After taking out the video frame to be detected and converting the format, detecting whether there is abnormal heat source data information in the video frame to be detected; If yes, then calculating the second azimuth of the abnormal heat source in the abnormal heat source data information, so as to set a pan / tilt rotation instruction according to the second azimuth, so as to drive the pan / tilt to lock the abnormal heat source; The inspection task of the inspection robot is suspended, the pan / tilt head is kept locked on the abnormal heat source, and the alarm data information is pushed to the monitoring background.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a patrol robot detection and identification device, the patrol robot detection and identification device comprising: The data acquisition module is used to start the video stream decoding process when an initialization start signal is detected, and write the video frames to be detected collected by the inspection robot camera into the cache queue; An information judgment module, after taking out the video frame to be detected and converting the format, detects whether there is human figure data information in the video frame to be detected; The information judgment module is further used to calculate a first azimuth angle of the humanoid target in the humanoid data information, so as to lock the humanoid target according to the first azimuth angle; The alarm processing module is used to perform face recognition detection on the human-shaped target, and push alarm data information to the monitoring background when it is determined that the human-shaped target is a stranger based on the detection result.

[0014] In addition, to achieve the above-mentioned objectives, the present application also proposes a patrol robot detection and identification device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the patrol robot detection and identification method described above.

[0015] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the inspection robot detection and identification method described above are implemented.

[0016] The present application provides a detection and identification method for a patrol robot. The present application first starts the video stream decoding process when an initialization start signal is detected, and writes the video frames to be detected collected by the patrol robot camera into a cache queue; after taking out the video frames to be detected and converting the format, detect whether there is humanoid data information in the video frames to be detected; if so, calculate the first azimuth angle of the humanoid target in the humanoid data information to lock the humanoid target according to the first azimuth angle; perform face recognition detection on the humanoid target, and push alarm data information to the monitoring background when judging that the humanoid target is a stranger based on the detection result. By judging whether the video data collected by the park patrol robot during the patrol task contains humanoid data, and re-performing face recognition judgment after aiming the camera at the humanoid target by calculating the azimuth angle, the accuracy of personnel recognition judgment is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 This is a flow chart of the first embodiment of the inspection robot detection and identification method of the present application; Figure 2 A flow chart of the second embodiment of the inspection robot detection and identification method of the present application; Figure 3 A flowchart diagram of the third embodiment of the inspection robot detection and identification method of the present application; Figure 4 A schematic diagram of the abnormal heat source detection process provided in Example 3 of the inspection robot detection and identification method of the present application; Figure 5 This is a schematic diagram of the module structure of the inspection robot detection and identification device according to an embodiment of the present application; Figure 6 This is a schematic diagram of the structure of the inspection robot detection and identification equipment of the hardware operating environment involved in the inspection robot detection and identification method in the embodiment of the present application.

[0020] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0022] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0023] The main solution of the embodiment of the present application is: starting the video stream decoding process when an initialization start signal is detected, and writing the video frames to be detected collected by the inspection robot camera into the cache queue; taking out the video frames to be detected and converting the format, detecting whether there is human data information in the video frames to be detected; if so, calculating the first azimuth angle of the human target in the human data information to lock the human target according to the first azimuth angle; performing face recognition detection on the human target, and pushing the alarm data information to the monitoring background when judging that the human target is a stranger according to the detection result.

[0024] Inspection robots have weak capabilities for detecting and identifying safety hazards in complex environments and are unable to accurately determine personnel information when performing inspection tasks, resulting in misjudgments or missed judgments, leaving safety hazards for park inspections.

[0025] The present application provides a solution, which improves the accuracy of personnel recognition by determining whether the video data collected by the park inspection robot during the inspection mission contains human data, and by calculating the azimuth angle to aim the camera at the human target and then re-perform face recognition judgment.

[0026] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or a patrol robot detection and identification device that can realize the above functions, etc. The following takes the patrol robot detection and identification device as an example to illustrate this embodiment and the following embodiments.

[0027] Based on this, the present application embodiment provides a patrol robot detection and identification method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the inspection robot detection and identification method of the present application.

[0028] In this embodiment, the inspection robot detection and identification method includes steps S10 to S40: Step S10: When the initialization start signal is detected, the video stream decoding process is started, and the video frames to be detected collected by the inspection robot camera are written into the cache queue.

[0029] It should be noted that the initialization start signal can be a start signal that drives the inspection robot to perform the inspection task, it can be a hardware signal controlled by a local electronic switch for power supply, or it can be a communication signal transmitted by the monitoring background when the inspection robot communicates with the remote monitoring background through a wireless transmission device (such as Bluetooth, WIFI, etc.).

[0030] It should be understood that starting the video stream decoding process can first decode the compressed video frame data collected by the camera into a process that can be used for frame processing data. At this time, the panoramic camera thread is controlled to be locked to prevent the panoramic camera from writing new video data to the cache queue at this time. After the video frame to be detected is written to the cache queue, the panoramic camera is unlocked to write real-time video data. The video frame to be detected can be a video frame segment that needs to be detected and checked for safety hazards in the video data continuously collected by the panoramic camera. Among them, the cameras included in the inspection robot may include a panoramic camera and a high-definition camera. The panoramic camera can be used to collect frame data within a 360-degree range, identify and detect human targets and vehicle targets, and calculate azimuth angles. The high-definition camera can collect high-resolution frame data for face recognition detection.

[0031] Step S20: after taking out the video frame to be detected and converting its format, detecting whether there is human figure data information in the video frame to be detected.

[0032] It should be noted that the panoramic camera thread needs to be locked before taking out the video frame to be detected from the cache queue, and unlocked after the video frame to be detected is taken out to avoid the interference of the continuously collected video data on the video frame to be detected. The decoded video frame data is format converted to ensure that the quality of the converted data meets the requirements of subsequent judgment using the detection and recognition algorithm.

[0033] It should be understood that the detection of human figures can be based on a deep learning target detection algorithm (YOLO, etc.), pre-training or loading a pre-trained model to perform human figure detection, and then determine whether there is human figure data information in the video frame to be detected.

[0034] Step S30: If yes, then calculate the first azimuth angle of the human-shaped target in the human-shaped data information to lock the human-shaped target according to the first azimuth angle.

[0035] It should be noted that when it is detected that the video frame to be detected contains human data information, it can be set to write the human alarm information into the cache queue, and send the human alarm information to the monitoring background, and display the video frame and the human target's annotation box in real time through the window. When no human data information is detected in the video frame to be detected, the inspection robot continues to patrol the park.

[0036] It should be understood that after the inspection robot detects that the video frame to be detected contains human data information, it can also verify the identity of the human target corresponding to the human data information. The first azimuth angle can be the azimuth angle of the human target in the video frame to be detected relative to the inspection robot. The camera set on the inspection robot can be brought as close to the face recognition distance as possible by driving the pan-tilt head to rotate or driving the robot to move, and aim and lock the human target. When performing face recognition, the video frame data containing face information can be sampled by a high-definition camera.

[0037] Step S40: Performing face recognition detection on the humanoid target, and pushing alarm data information to the monitoring background when judging that the humanoid target is a stranger according to the detection result.

[0038] It should be noted that a face detection algorithm (such as SCRFD, etc.) can be used to analyze the face image reacquired after the camera locks the human-shaped target, and then judge whether the human-shaped target is a pre-set and saved compliant person or a stranger who has not been stored in the database according to the algorithm detection result. Specifically, the face feature points of the face data information can be analyzed, and the feature similarity between the face feature points and the preset personnel data can be calculated; when the feature similarity is lower than the preset judgment threshold, the human-shaped target is judged to be a stranger, and the alarm data information is pushed to the monitoring background. And when the feature similarity is higher than the preset judgment threshold, the human-shaped target is judged to be a compliant person; at the same time, the pan-tilt is unlocked, and the inspection robot is driven to continue to perform the inspection task. Among them, the preset personnel data can be the registered face image data pre-set in the memory of the inspection robot. The preset judgment threshold can be the similarity value between the face image reacquired after the camera locks the human-shaped target and the preset personnel data.

[0039] It should be understood that two different sets of face detection algorithms can be used for initial recognition and strong recognition to improve recognition accuracy and reduce false alarms. At the same time, after determining that the humanoid target is a compliant person, the face detection function is suspended for a period of time (for example, 30 seconds) and then restarted to avoid repeated reporting caused by the target still being within the visual range of the inspection robot.

[0040] In this embodiment, by starting the video stream decoding process when the initialization start signal is detected, the video frames to be detected collected by the inspection robot camera are written into the cache queue; after taking out the video frames to be detected and converting the format, it is detected whether there is human data information in the video frames to be detected; if so, the first azimuth angle of the human target in the human data information is calculated to lock the human target according to the first azimuth angle; face recognition detection is performed on the human target, and when the human target is judged to be a stranger according to the detection result, the alarm data information is pushed to the monitoring background. By judging whether the video data collected by the park inspection robot during the inspection task contains human data, and by calculating the azimuth angle, the camera is aimed at the human target and the face recognition judgment is re-performed, the accuracy of personnel recognition judgment is improved.

[0041] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 2 , Figure 2 A flowchart of the second embodiment of the inspection robot detection and identification method of the present application is provided. After step S20, the inspection robot detection and identification method includes: Step S201: determining whether the number of human-shaped targets in the to-be-detected video frame exceeds one according to the human-shaped data information.

[0042] It should be noted that when the inspection robot is collecting video data with its camera, there may be a situation where a video to be detected contains more than one humanoid target. It can be determined whether the number of humanoid targets in the video frame to be detected exceeds one based on the humanoid data information.

[0043] At this time, step S30 includes: Step S202: Calculate the first azimuth angle of the humanoid target with the largest area in the humanoid data information.

[0044] Step S203: setting a pan / tilt rotation instruction according to the first azimuth angle to drive the pan / tilt to lock onto the humanoid target.

[0045] It should be noted that the humanoid target with the largest area in the video frame to be detected is the closest to the inspection robot, and the azimuths of the humanoid targets are obtained in order from near to far. The first azimuth of the humanoid target with the largest area is determined first, and the pan-tilt rotation command is set according to the first azimuth to drive the pan-tilt to lock the humanoid target.

[0046] It should be understood that the distance conditions of different human-shaped targets can also be judged to determine whether they are fellow travelers. When the compliant person exists among the fellow travelers, it is determined that all fellow travelers are compliant persons, the gimbal is unlocked, the face detection function is suspended, and the inspection robot is driven to continue to perform the inspection task.

[0047] Furthermore, after the alarm data information is pushed to the monitoring background when it is determined that the humanoid target is a stranger, the monitoring background can obtain the video images collected by the inspection robot in real time when receiving the alarm data information. At the same time, the inspection robot adjusts the pan-tilt to keep the camera locked on the stranger and waits for the monitoring background to return a disposal instruction (such as a release or expulsion instruction). Expulsion can be achieved by the electronic control equipment set on the inspection robot emitting light, sound and other signals. Among them, if the inspection robot does not receive the disposal instruction returned by the monitoring background within a period of time (for example, 3 minutes), the pan-tilt is unlocked, the face detection function is suspended, and the inspection robot is driven to continue to perform the inspection task. To avoid delays in normal inspection tasks, the monitoring background retains the received alarm data information.

[0048] In this embodiment, it is determined whether the number of human-shaped targets in the video frame to be detected is more than one according to the human-shaped data information. When there are more than one human-shaped targets, the first azimuth angle of the human-shaped target with the largest area in the human-shaped data information is calculated; and a pan-tilt rotation instruction is set according to the first azimuth angle to drive the pan-tilt to lock the human-shaped target. When there are multiple targets to be detected, they are detected in sequence from near to far, so as to further improve the efficiency and accuracy of personnel recognition and judgment.

[0049] Based on the first and second embodiments of this application, please refer to Figure 3 , Figure 3 A flowchart of the third embodiment of the inspection robot detection and identification method of the present application is provided. After step S10, the inspection robot detection and identification method includes: Step S301: after taking out the video frame to be detected and converting its format, detecting whether there is vehicle data information in the video frame to be detected.

[0050] It should be noted that when the inspection robot collects video data with its camera, in addition to collecting and identifying human targets, it can also detect other safety risk conditions, such as vehicle information or abnormal heat source information.

[0051] Step S302: If yes, push the alarm data information to the monitoring background, drive the inspection robot to continue to perform the inspection task, and start the timer at the same time.

[0052] Step S303: after the timer reaches a preset time, re-detect the video frame to be detected until the video frame to be detected does not contain the vehicle data information.

[0053] It should be noted that when a vehicle target is detected in the video frame to be detected, the alarm data information is pushed to the monitoring background. At this time, there is no need to suspend the current inspection task of the inspection robot to prevent the robot from entering a waiting loop state due to the presence of parked vehicles in the field of view. The timer can be a time detection software stored in a robot-readable memory and edited by a program, or a timing circuit set on the robot, which is not specifically limited in this embodiment.

[0054] It should be understood that the preset time may be the interval time (eg, may be set to 10 seconds) for reacquiring and determining whether the video frame to be detected contains vehicle data information, so as to prevent repeated alarms for the same target vehicle from interfering with risk judgment.

[0055] Reference Figure 4 FIG. 4 is a schematic diagram of a flow chart of abnormal heat source detection provided by the third embodiment of the inspection robot detection and identification method of the present application. After step S10, the inspection robot detection and identification method further includes: Step S401: after taking out the video frame to be detected and converting its format, detecting whether there is abnormal heat source data information in the video frame to be detected.

[0056] It should be noted that, when the inspection robot collects video data with a camera, an infrared thermal imager can be used in conjunction with a camera to achieve the fusion of video frame data and temperature data. The abnormal heat source data information can be information about an area whose temperature exceeds a preset detection temperature. When the temperature of a target area in the collected video frame data exceeds the preset detection temperature, it is determined that abnormal heat source data information exists. When the temperature of a target area in the collected video frame data does not exceed the preset detection temperature, it is determined that abnormal heat source data information does not exist.

[0057] Step S402: If yes, then the second azimuth angle of the abnormal heat source in the abnormal heat source data information is calculated to set a gimbal rotation instruction according to the second azimuth angle to drive the gimbal to lock the abnormal heat source.

[0058] It should be noted that the second azimuth angle can be the angle information of the abnormal heat source from the inspection robot, and the second azimuth angle is obtained according to the position of the target area in the video frame to be detected. The infrared thermal imager is aligned with the center of the target area where the abnormal heat source is located by adjusting the pan / tilt.

[0059] Step S403: suspend the inspection task of the inspection robot, keep the pan / tilt head locked on the abnormal heat source, and push the alarm data information to the monitoring background.

[0060] It should be noted that after detecting the presence of an abnormal heat source, the inspection robot's other inspection tasks are suspended, the pan-tilt keeps locked on the abnormal heat source, and the alarm data information is pushed to the monitoring background. After the alarm data information is pushed to the monitoring background when it is determined that there is an abnormal heat source, the monitoring background can obtain the video images collected by the inspection robot in real time when receiving the alarm data information. At the same time, the inspection robot adjusts the pan-tilt to keep the camera locked on the center area of ​​the abnormal heat source, waiting for the monitoring background to return a disposal instruction (such as an inspection instruction). After receiving the disposal instruction returned by the monitoring background to keep the inspection, the inspection robot unlocks the pan-tilt and continues to perform the inspection task. In addition, if the inspection robot does not receive the disposal instruction returned by the monitoring background within a period of time (for example, 3 minutes), the pan-tilt is unlocked, the detection function of the abnormal heat source is suspended, and the inspection robot is driven to continue to perform the inspection task. To avoid delays in normal inspection tasks, the monitoring background retains the received alarm data information.

[0061] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the inspection robot detection and identification method of the present application. More simple transformations based on this technical concept are all within the protection scope of the present application.

[0062] This application also provides a patrol robot detection and identification device, please refer to Figure 5 , the inspection robot detection and identification device comprises: The data acquisition module 10 is used to start the video stream decoding process when an initialization start signal is detected, and write the video frames to be detected collected by the inspection robot camera into the cache queue; An information determination module 20 is used to take out the video frame to be detected, perform format conversion, and then detect whether there is human figure data information in the video frame to be detected; The information judgment module is further used to calculate a first azimuth angle of the humanoid target in the humanoid data information, so as to lock the humanoid target according to the first azimuth angle; The alarm processing module 30 is used to perform face recognition detection on the human-shaped target, and push alarm data information to the monitoring background when it is determined that the human-shaped target is a stranger based on the detection result.

[0063] Optionally, the information judgment module 20 is also used to judge whether the number of human-shaped targets in the video frame to be detected exceeds one based on the human-shaped data information, and calculate the first azimuth angle of the human-shaped target with the largest area in the human-shaped data information when the number of human-shaped targets exceeds one; and set the gimbal rotation instruction based on the first azimuth angle to drive the gimbal to lock the human-shaped target.

[0064] Optionally, the alarm processing module 30 is also used to collect facial data information after the pan-tilt head locks the humanoid target; parse the facial feature points of the facial data information, and calculate the feature similarity between the facial feature points and the preset personnel data; when the feature similarity is lower than a preset judgment threshold, judge that the humanoid target is a stranger; suspend the inspection task of the inspection robot, keep the pan-tilt head locked on the humanoid target, and push the alarm data information to the monitoring background.

[0065] Optionally, the alarm processing module 30 is further configured to execute corresponding actions according to the processing instructions after the inspection robot receives the processing instructions from the background.

[0066] Optionally, the alarm processing module 30 is further configured to, when the feature similarity is higher than a preset judgment threshold, determine that the humanoid target is a compliant person; unlock the gimbal, and drive the inspection robot to continue to perform the inspection task.

[0067] Optionally, the information judgment module 20 is also used to take out the video frame to be detected and convert its format, and then detect whether there is vehicle data information in the video frame to be detected; if so, push the alarm data information to the monitoring background, drive the inspection robot to continue to perform the inspection task, and start the timer at the same time; after the timer reaches a preset time, re-detect the video frame to be detected until the video frame to be detected does not contain the vehicle data information.

[0068] Optionally, the information judgment module 20 is also used to take out the video frame to be detected and convert its format, and then detect whether there is abnormal heat source data information in the video frame to be detected; if so, calculate the second azimuth angle of the abnormal heat source in the abnormal heat source data information, and set the pan-tilt rotation instruction according to the second azimuth angle to drive the pan-tilt to lock the abnormal heat source; suspend the inspection task of the inspection robot, keep the pan-tilt to lock the abnormal heat source, and push the alarm data information to the monitoring background.

[0069] The inspection robot detection and identification device provided by the present application adopts the inspection robot detection and identification method in the above embodiment, which can solve the technical problem of misjudgment and missed judgment caused by the inspection robot's weak detection and identification ability. Compared with the prior art, the inspection robot detection and identification device provided by the present application has the same beneficial effects as the inspection robot detection and identification method provided by the above embodiment, and the other technical features in the inspection robot detection and identification device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0070] The present application provides a patrol robot detection and identification device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the patrol robot detection and identification method in the above-mentioned embodiment one.

[0071] Reference below Figure 6 , which shows a schematic diagram of the structure of the inspection robot detection and identification device suitable for implementing the embodiment of the present application. The inspection robot detection and identification device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The inspection robot detection and identification device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0072] like Figure 6As shown, the inspection robot detection and identification device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the inspection robot detection and identification device are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the inspection robot detection and identification device to communicate with other devices wirelessly or by wire to exchange data. Although the inspection robot detection and identification device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0073] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0074] The inspection robot detection and identification device provided by the present application adopts the inspection robot detection and identification method in the above embodiment, which can solve the technical problem of misjudgment and missed judgment caused by the inspection robot's weak detection and identification ability. Compared with the prior art, the inspection robot detection and identification device provided by the present application has the same beneficial effects as the inspection robot detection and identification method provided by the above embodiment, and the other technical features in the inspection robot detection and identification device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0075] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0076] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0077] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the inspection robot detection and identification method in the above-mentioned embodiment.

[0078] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0079] The computer-readable storage medium may be included in the inspection robot detection and identification device; or may exist independently without being assembled into the inspection robot detection and identification device.

[0080] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate 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 may 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 may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0081] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0082] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0083] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned inspection robot detection and identification method, and can solve the technical problem of misjudgment and missed judgment caused by the inspection robot's weak detection and identification ability. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the inspection robot detection and identification method provided by the above-mentioned embodiment, and will not be repeated here.

[0084] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A patrol robot detection and identification method, characterized in that: The inspection robot detection and identification method comprises: When the initialization start signal is detected, the video stream decoding process is started, and the video frames to be detected collected by the inspection robot camera are written into the cache queue; After taking out the video frame to be detected and converting the format, detecting whether there is human figure data information in the video frame to be detected; If yes, calculating a first azimuth angle of the humanoid target in the humanoid data information, so as to lock the humanoid target according to the first azimuth angle; Perform face recognition detection on the humanoid target, and push alarm data information to the monitoring background when it is determined that the humanoid target is a stranger based on the detection result.

2. The inspection robot detection and identification method according to claim 1, characterized in that: After the step of taking out the video frame to be detected and converting the format, and then detecting whether there is human figure data information in the video frame, the method includes: Determining whether the number of human-shaped targets in the to-be-detected video frame exceeds one according to the human-shaped data information; If yes, the step of calculating the first azimuth of the humanoid target in the humanoid data information to lock the humanoid target according to the first azimuth includes: Calculating the first azimuth angle of the humanoid target with the largest area in the humanoid data information; A pan / tilt rotation instruction is set according to the first azimuth angle to drive the pan / tilt to lock onto the humanoid target.

3. The inspection robot detection and identification method according to claim 2, characterized in that: The step of performing face recognition detection on the humanoid target and pushing alarm data information to the monitoring background when judging that the humanoid target is a stranger according to the detection result includes: Collecting facial data information after the gimbal locks onto the human-shaped target; Analyze the facial feature points of the facial data information, and calculate the feature similarity between the facial feature points and the preset personnel data; When the feature similarity is lower than a preset judgment threshold, the human-shaped target is judged to be a stranger; The inspection task of the inspection robot is suspended, the pan / tilt head is kept locked on the humanoid target, and the alarm data information is pushed to the monitoring background.

4. The inspection robot detection and identification method according to claim 3, characterized in that: After the steps of pausing the inspection task of the inspection robot, keeping the pan / tilt head locked on the humanoid target, and pushing the alarm data information to the monitoring background, the method further includes: After the inspection robot receives the processing instruction from the background, it performs corresponding actions according to the processing instruction.

5. The inspection robot detection and identification method according to claim 3, characterized in that: After the steps of analyzing the facial feature points of the facial data information and calculating the feature similarity between the facial feature points and the preset personnel data, the method further includes: When the feature similarity is higher than a preset judgment threshold, the humanoid target is judged to be a compliant person; Unlock the pan / tilt platform and drive the inspection robot to continue to perform the inspection task.

6. The inspection robot detection and identification method according to claim 1, characterized in that: After the step of starting the video stream decoding process when the initialization start signal is detected and writing the video frame to be detected collected by the inspection robot camera into the cache queue, the method further includes: After taking out the video frame to be detected and converting the format, detecting whether there is vehicle data information in the video frame to be detected; If yes, the alarm data information is pushed to the monitoring background, driving the inspection robot to continue to perform the inspection task, and starting the timer at the same time; After the timer reaches a preset time, the to-be-detected video frame is redetected until the to-be-detected video frame does not contain the vehicle data information.

7. The inspection robot detection and identification method according to claim 1, characterized in that: After the step of starting the video stream decoding process when the initialization start signal is detected and writing the video frame to be detected collected by the inspection robot camera into the cache queue, the method further includes: After taking out the video frame to be detected and converting the format, detecting whether there is abnormal heat source data information in the video frame to be detected; If yes, then calculating the second azimuth of the abnormal heat source in the abnormal heat source data information, so as to set a pan / tilt rotation instruction according to the second azimuth, so as to drive the pan / tilt to lock the abnormal heat source; The inspection task of the inspection robot is suspended, the pan / tilt head is kept locked on the abnormal heat source, and the alarm data information is pushed to the monitoring background.

8. A patrol robot detection and identification device, characterized in that: The device comprises: The data acquisition module is used to start the video stream decoding process when an initialization start signal is detected, and write the video frames to be detected collected by the inspection robot camera into the cache queue; An information judgment module, after taking out the video frame to be detected and converting the format, detects whether there is human figure data information in the video frame to be detected; The information judgment module is further used to calculate a first azimuth angle of the humanoid target in the humanoid data information, so as to lock the humanoid target according to the first azimuth angle; The alarm processing module is used to perform face recognition detection on the human-shaped target, and push alarm data information to the monitoring background when it is determined that the human-shaped target is a stranger based on the detection result.

9. A patrol robot detection and identification device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the inspection robot detection and identification method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the inspection robot detection and identification method according to any one of claims 1 to 7 are implemented.

Images