CN115662083B - Unmanned inspection type alarm system and method based on acoustic imaging - Google Patents
Unmanned inspection type alarm system and method based on acoustic imaging Download PDFInfo
- Publication number
- CN115662083B CN115662083B CN202211587563.6A CN202211587563A CN115662083B CN 115662083 B CN115662083 B CN 115662083B CN 202211587563 A CN202211587563 A CN 202211587563A CN 115662083 B CN115662083 B CN 115662083B
- Authority
- CN
- China
- Prior art keywords
- abnormal sound
- detection point
- acoustic
- current detection
- imager
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 72
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000003384 imaging method Methods 0.000 title claims abstract description 20
- 230000002159 abnormal effect Effects 0.000 claims abstract description 221
- 238000001514 detection method Methods 0.000 claims abstract description 152
- 238000010191 image analysis Methods 0.000 claims description 13
- 230000000694 effects Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Landscapes
- Emergency Alarm Devices (AREA)
Abstract
The invention relates to the technical field of acoustic imaging, in particular to an unmanned inspection type alarm system and method based on acoustic imaging. An unmanned patrol type alarm system based on acoustic imaging comprises an acoustic imager; the bearing moving device is used for installing the acoustic imager and driving the acoustic imager to move along an inspection route in the area to be detected, and the inspection route is provided with a plurality of detection points; the acoustic imager comprises a data cache unit, and the data cache unit is used for storing M audio data collected by the acoustic imager at the current detection point position and deleting the M audio data collected by the acoustic imager last time. The unmanned patrol inspection type alarm system and the method can automatically detect and alarm abnormal sounds in a region to be detected through the acoustic imager and the bearing mobile device, save time and labor, have high safety, and can obtain accurate abnormal point detection results even if unmanned patrol inspection is carried out.
Description
Technical Field
The invention relates to the technical field of acoustic imaging, in particular to an unmanned inspection type alarm system and method based on acoustic imaging.
Background
The acoustic imager adopts a microphone array measuring technology, determines the distribution of a sound source by measuring the phase difference of signals of sound waves in a certain space reaching each microphone and applying a certain sound source position estimation method, and presents the distribution of the sound source by combining with an optical camera picture. The acoustic imager can be used for abnormal point detection in certain scenes, for example, for detecting a gas leak under pressure or for detecting partial discharge of electrical equipment.
In the prior art, there are two ways of detecting outliers in these scenes by using an acoustic imager: the first is that the manual hand-held acoustic imager detects the scene to be detected, and the second is that a plurality of acoustic imagers are fixedly arranged in the scene to be detected for detection. The drawbacks of the first approach are: each detection requires that a worker enters a scene to be detected and stays until the whole scene to be detected is completely detected, so that on one hand, a large amount of time is spent on the worker, on the other hand, the scene to be detected usually has certain danger, and the risk that the worker stays in the scene to be detected for a long time is high. The second mode has the defects that: the number of acoustic imagers required is large, the cost is high, and the monitoring range is limited, so that the detection result is not ideal.
Disclosure of Invention
The invention provides an unmanned inspection type alarm system and method based on acoustic imaging, aiming at the problems in the prior art, the system and method can automatically detect and alarm abnormal sounds in a detection area through an acoustic imager and a bearing mobile device, the use and operation are time-saving and labor-saving, the safety is high, and the accuracy of an abnormal point detection result is high.
The technical scheme adopted by the invention for solving the technical problems is as follows: unmanned patrol type alarm system based on acoustic imaging, comprising
An acoustic imager;
the bearing moving device is used for installing the acoustic imager and driving the acoustic imager to move along an inspection route in a region to be detected, and the inspection route is provided with a plurality of detection points;
wherein the acoustic imager comprises
The data cache unit is used for storing M audio data collected by the acoustic imager at the current detection point position and deleting the M audio data collected by the acoustic imager at the last time;
the abnormal sound judging unit is used for acquiring the audio data in the data caching unit, calculating to obtain a to-be-judged audio with the largest sound pressure value in the audio data, subtracting a background noise value at the current detection point position from the sound pressure value of the to-be-judged audio to obtain an abnormal sound judging value, comparing the abnormal sound judging value with a preset sound threshold value, judging that the current detection point position has abnormal sound when the abnormal sound judging value is larger than the preset sound threshold value, and otherwise judging that the current detection point position does not have the abnormal sound;
the abnormal sound validity determining unit is used for determining the validity of the abnormal sound according to the field range of the current detection point position when the current detection point position has the abnormal sound, judging that the abnormal sound is valid when the generation position of the abnormal sound is positioned in the field range of the current detection point position, and otherwise, judging that the abnormal sound is invalid;
and the abnormal sound alarm unit is used for recording the position of the current detection point, the current time and the actual position of the abnormal sound as alarm data when the abnormal sound is effective, wherein the actual position of the abnormal sound is a coordinate value of the abnormal sound in the three-dimensional scene model of the area to be detected.
Preferably, the acoustic imager further comprises
The image intercepting unit is used for intercepting a corresponding abnormal sound position point image on the three-dimensional scene model of the area to be detected through the actual position of the abnormal sound when the abnormal sound alarming unit records the actual position of the abnormal sound, and storing the abnormal sound position point image in a storage packet corresponding to the actual position of the abnormal sound;
and the image analysis unit is used for carrying out image analysis and identification on the abnormal sound position point pictures in the corresponding storage packets one by one according to the creation sequence of the storage packets so as to determine the types of the objects which emit abnormal sounds on the pictures in the corresponding storage packets.
Preferably, the data cache unit includes a first data cache region and a second data cache region, and when the first data cache region is in a first working mode, the second data cache region is in a second working mode; when the first data cache region is in a second working mode, the second data cache region is in a first working mode; the first working mode is to store M audio data collected by the acoustic imager at the current detection point position, and the second working mode is to delete M audio data collected by the acoustic imager last time.
Preferably, the acoustic imager further comprises
The background noise value determining unit is used for acquiring N pieces of background noise data of the current detection point position and calculating the root mean square value of the N pieces of background noise data to be used as the background noise value of the current detection point position; or, the method is used for acquiring N pieces of background noise data of the current detection point position, and selecting a median value of the N pieces of background noise data as a background noise value of the current detection point position;
the background noise value storage unit is used for storing the background noise value of the current detection point position;
the detection point view field range image acquisition and storage unit is used for acquiring and storing a view field range image of the current detection point position;
and the three-dimensional scene model determining unit is used for constructing and forming a three-dimensional scene model of the area to be detected through the view field range images of all detection point positions in the area to be detected.
Preferably, the carrying and moving device comprises
The inspection track is arranged in the area to be detected according to a required inspection route;
the movable trolley is connected with the inspection track and is used for moving on the inspection track;
a rotating member mounted on the mobile cart for mounting the acoustic imager;
and the controller is arranged on the moving trolley and is electrically connected with the moving trolley, the rotating part and the acoustic imager.
An unmanned inspection type alarm method based on acoustic imaging adopts the unmanned inspection type alarm system, and comprises the following steps:
l1, determining an inspection route in a region to be detected, and arranging a bearing mobile device and an acoustic imager in the region to be detected so that the bearing mobile device can drive the acoustic imager to move along the inspection route;
l2, the bearing moving device drives the acoustic imager to move along the routing inspection route, and when the acoustic imager reaches a new detection point, the acoustic imager enters L3; when the acoustic imager reaches the end point, ending;
l3, a data cache unit of the acoustic imager stores M audio data collected by the acoustic imager at the current detection point position and deletes M audio data collected by the acoustic imager last time;
l4, an abnormal sound judgment unit of the acoustic imager acquires the audio data in the data cache unit, calculates to obtain a to-be-judged audio with the maximum sound pressure value in the audio data, subtracts the background noise value of the current detection point position from the sound pressure value of the to-be-judged audio to obtain an abnormal sound judgment value, compares the abnormal sound judgment value with a preset sound threshold value, judges that the current detection point position has abnormal sound when the abnormal sound judgment value is larger than the preset sound threshold value, and enters L5; otherwise, judging that the position of the current detection point does not have abnormal sound, and returning to L2;
l5, an abnormal sound effectiveness determining unit of the acoustic imager determines the effectiveness of the abnormal sound according to the view field range of the current detection point position, and when the generation position of the abnormal sound is located in the view field range of the current detection point position, the abnormal sound is judged to be effective and enters L6; otherwise, judging that the abnormal sound is invalid, and returning to L2;
and L6, recording the position of the current detection point, the current time and the actual position of the abnormal sound as alarm data by an abnormal sound alarm unit of the acoustic imager, wherein the actual position of the abnormal sound is a coordinate value of the abnormal sound in the three-dimensional scene model of the area to be detected, and returning to L2.
Preferably, the L6 further includes the following steps before returning to L2:
l61, when the abnormal sound alarm unit records the actual position of the abnormal sound, an image intercepting unit of the acoustic imager intercepts a corresponding abnormal sound position point picture on a three-dimensional scene model of a region to be detected through the actual position of the abnormal sound, and stores the abnormal sound position point picture in a storage packet corresponding to the actual position of the abnormal sound;
after the L2 is finished, the method also comprises the following steps:
and L21, the image analysis unit of the acoustic imager is used for carrying out image analysis and identification on the abnormal sound position point pictures in the corresponding storage packets one by one according to the creation sequence of the storage packets so as to determine the types of the objects which emit abnormal sounds on the pictures in the corresponding storage packets.
Preferably, L3 specifically includes: when the first data cache region is in a first working mode and the second data cache region is in a second working mode, changing the first data cache region into the second working mode and changing the second data cache region into the first working mode; or when the first data cache region is in a second working mode and the second data cache region is in a first working mode, changing the first data cache region into the first working mode and changing the second data cache region into the second working mode; the first working mode is to store M audio data collected by the acoustic imager at the current detection point position, and the second working mode is to delete M audio data collected by the acoustic imager last time.
Preferably, L2 further comprises:
l01, the bearing moving device drives the acoustic imager to move along the routing inspection route, and when the acoustic imager reaches a new detection point, the acoustic imager enters L02; when the acoustic imager reaches the end point, entering L04;
l02, a background noise value determining unit of the acoustic imager acquires N pieces of background noise data of the current detection point position, and calculates root mean square values of the N pieces of background noise data to serve as background noise values of the current detection point position; or acquiring N pieces of background noise data of the current detection point position, and selecting a median value of the N pieces of background noise data as a background noise value of the current detection point position;
a background noise value storage unit of the acoustic imager stores a background noise value of the current detection point position;
l03, a detection point view field range image acquisition and storage unit of the acoustic imager acquires and stores a view field range image of the current detection point position and returns to L01;
and L04, constructing a three-dimensional scene model of the to-be-detected region by the three-dimensional scene model determining unit of the to-be-detected region through the view field range images of all detection point positions in the to-be-detected region.
Preferably, L1 specifically includes:
l11, arranging an inspection track in the area to be detected according to the required inspection route;
l12, connecting a movable trolley capable of moving on the inspection rail with the inspection rail;
l13, mounting a rotating part on the mobile trolley, and mounting the acoustic imager on the rotating part;
and L14, mounting a controller on the mobile trolley, and connecting the controller with the mobile trolley, the rotating part and the acoustic imager.
Advantageous effects
The unmanned patrol inspection type alarm system and the method can automatically detect and alarm abnormal sounds in a region to be detected through the acoustic imager and the bearing mobile device, are time-saving, labor-saving and high in safety, the acoustic imager judges whether abnormal sounds exist at the current detection point through the abnormal sound judgment unit, judges whether the abnormal sounds at the current detection point are effective through the abnormal sound effectiveness determination unit, and records alarm data through the abnormal sound alarm unit, so that accurate abnormal point detection results can be obtained even if unmanned patrol inspection is performed.
Drawings
FIG. 1 is a schematic diagram of an acoustic imager in an embodiment of the invention;
fig. 2 is a flowchart of an unmanned patrol type alarm method in the embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings.
Example 1: an unmanned inspection type alarm system based on acoustic imaging comprises an acoustic imager and a bearing mobile device.
The bearing moving device is used for installing the acoustic imager and driving the acoustic imager to move along an inspection route in a region to be detected, and a plurality of detection points are arranged on the inspection route. The acoustic imager can automatically detect the abnormity of a scene to be detected at a detection point, for example, the acoustic imager can detect the leakage of charged gas or detect the partial discharge of power equipment.
Specifically, the load-bearing moving device can comprise a patrol rail, a moving trolley, a rotating part and a controller. And the patrol track is arranged in the area to be detected according to a required patrol route. The movable trolley is connected with the inspection track and used for moving on the inspection track. The rotating part is installed on the mobile trolley and used for installing the acoustic imager, the rotating part can be a holder, and the holder can not only stably install the acoustic imager, but also drive the acoustic imager to rotate circumferentially. The controller is installed on the moving trolley and is electrically connected with the moving trolley, the rotating part and the acoustic imager.
The controller can control the moving trolley to move on the inspection track at a constant speed, and when the moving trolley reaches a detection point, the controller controls the moving trolley to pause. At this moment, if the acoustic imager needs to rotate in the circumferential direction, the controller controls the holder to rotate so as to drive the acoustic imager to rotate in the circumferential direction. For example, the controller may first control the pan/tilt head to rotate 60 degrees clockwise, then reset, then control the pan/tilt head to rotate 60 degrees counterclockwise, and finally reset.
In addition, each detection point is correspondingly provided with an infrared emitter, and the moving trolley is provided with an infrared receiver electrically connected with the controller. When the mobile trolley moves to the infrared receiver and receives the infrared rays emitted by the infrared emitter, the mobile trolley is indicated to reach a certain detection point, and at the moment, the controller controls the mobile trolley to pause. After the pause is finished, the controller controls the moving trolley to continue to advance until the infrared receiver on the moving trolley receives the infrared again, and the moving trolley reaches a new detection point.
Specifically, as shown in fig. 1, the acoustic imager includes a data caching unit, an abnormal sound determining unit, an abnormal sound validity determining unit, an abnormal sound alarm recording unit, an image capturing unit, an image analyzing unit, a background noise value determining unit, a background noise value storing unit, a detection point field range image obtaining unit, and a to-be-detected area three-dimensional scene model determining unit.
The data cache unit is configured to store M audio data (which may be 30) collected by the acoustic imager at the current detection point position, and delete the M audio data collected by the acoustic imager last time.
The data cache unit comprises a first data cache region and a second data cache region, and when the first data cache region is in a first working mode, the second data cache region is in a second working mode; when the first data cache region is in a second working mode, the second data cache region is in a first working mode; the first working mode is to store M audio data collected by the acoustic imager at the current detection point position, and the second working mode is to delete M audio data collected by the acoustic imager last time. In summary, when the first data buffer is in the first working mode, the second data buffer is necessarily in the second working mode; when the first data cache region is in the second working mode, the second data cache region is necessarily in the first working mode, and the working modes of the first data cache region and the second data cache region are necessarily opposite. For example, when the mobile cart reaches a new detection point, 30 audio data collected by the acoustic imager at the current detection point position can be directly stored in the empty first data buffer area, and the second data buffer area in which the 30 audio data collected at the previous detection point position are stored is subjected to data unloading and clearing, so that the mobile cart is in an empty standby state.
The abnormal sound judging unit can acquire 30 pieces of audio data in a first data buffer area under a first working mode, can calculate to obtain a to-be-judged audio with the largest sound pressure value in the 30 pieces of audio data, subtracts a background noise value at the current detection point position from the sound pressure value of the to-be-judged audio to obtain an abnormal sound judging value, compares the abnormal sound judging value with a preset sound threshold value, judges that the current detection point position has abnormal sound when the abnormal sound judging value is larger than the preset sound threshold value, and judges that the current detection point position does not have the abnormal sound otherwise. If the abnormal sound judging unit judges that the position of the current detection point does not have the abnormal sound, the controller controls the mobile trolley to continue to move to the next detection point.
If the abnormal sound judging unit finds that the 30 audio data are abnormal when calculating the 30 audio data, and the corresponding audio data cannot be calculated to obtain the corresponding sound pressure value, the abnormal sound judging unit sends a data acquisition abnormal signal to the data caching unit. At the moment, the data buffer unit exchanges the working modes of the first data buffer area and the second data buffer area, 30 pieces of audio data are collected again through the acoustic imager and stored in the empty second data buffer area, and meanwhile, the first data buffer area is subjected to data unloading and clearing so as to be in an empty standby state. Then, the abnormal sound determination unit obtains 30 audio data in the second data buffer under the first working mode, and calculates to obtain a to-be-determined audio with the largest sound pressure value in the 30 audio data. If the abnormal sound determination unit can successfully calculate the sound pressure values of the 30 audio data, the abnormal sound determination unit continues to process the data downward, otherwise, the step is repeated until the abnormal sound determination unit can successfully calculate the sound pressure values of the 30 audio data.
Before formal inspection, the mobile trolley carries an acoustic imager to walk along an inspection route for one time, and when the mobile trolley moves to a detection point, a background noise value determining unit acquires N background noise data (which can be 100) of the current detection point position, and calculates the root mean square value of the 100 background noise data to serve as the background noise value of the current detection point position; or, the method is used for acquiring 100 pieces of background noise data of the current detection point position, and selecting a median value of the 100 pieces of background noise data to serve as a background noise value of the current detection point position. The background noise value storage unit is used for storing the background noise value of the current detection point position, so that the abnormal sound judgment unit can directly call the background noise value of the corresponding detection point position at a certain detection point position in formal inspection.
In addition, when the mobile trolley moves to a detection point, the detection point view field range image acquisition and storage unit acquires a view field range image of the current detection point position and stores the view field range image of the current detection point position. When the view field range image of the current detection point position is obtained, the controller is only required to control the holder to drive the acoustic imager to rotate circumferentially within a certain angle range, and the corresponding view field range image is shot through the camera of the acoustic imager. After the travelling trolley finishes walking the routing inspection route, the three-dimensional scene model determining unit of the area to be detected can construct and form a three-dimensional scene model of the area to be detected through the view field range images of all detection point positions in the area to be detected (specifically, how to construct the three-dimensional scene model is the prior art).
When formal routing inspection is carried out, when the current detection point position has abnormal sound, the abnormal sound effectiveness determining unit determines the effectiveness of the abnormal sound according to the view field range of the current detection point position, when the generation position of the abnormal sound is located in the view field range of the current detection point position, the abnormal sound is judged to be effective, and otherwise, the abnormal sound is judged to be ineffective. Specifically, when the abnormal sound determination unit determines that the current detection point position has abnormal sound, the controller controls the cradle head to drive the acoustic imager to circumferentially rotate according to a preset angle range to determine the field range of the current detection point position, if the generation position of the abnormal sound is not within the field range, the abnormal sound is invalid, and the controller controls the mobile trolley to continuously advance to the next detection point (specifically, it is the prior art how to determine whether the generation position of the abnormal sound is within the field range of the current detection point position).
When the abnormal sound is valid, the abnormal sound alarm unit records the position of the current detection point (such as several detection points), the current time and the actual position of the abnormal sound as an alarm datum, wherein the actual position of the abnormal sound is the coordinate value of the abnormal sound in the three-dimensional scene model of the area to be detected (the coordinate value of the abnormal sound in the three-dimensional scene model of the area to be detected is calculated in the prior art). After patrolling and examining, the staff can directly look over alarm data on the display screen of acoustic imager, and then knows and treat the abnormal conditions in the detection area.
When the abnormal sound alarm unit records the actual position of the abnormal sound, the image intercepting unit intercepts a corresponding abnormal sound position point picture on the three-dimensional scene model of the area to be detected according to the actual position of the abnormal sound, and stores the abnormal sound position point picture in a storage packet corresponding to the actual position of the abnormal sound.
And (3) placing pictures with similar actual positions of the abnormal sounds (namely the difference value x, the difference value y and the difference value z of the coordinate values of the two positions are smaller than the preset difference value) into the same storage packet. For example, when the acoustic imager moves to the detection point No. 10, the abnormal sound determination unit determines that there is an abnormal sound, and the abnormal sound validity determination unit determines that the abnormal sound is valid, the abnormal sound alarm unit records the position of the current detection point (detection point No. 10), the current time (8 points 43 minutes 20 seconds), and the actual position of the abnormal sound (coordinate axes (x 1, y1, z 1)), and the image capture unit captures an abnormal sound position dot image of the coordinate axes (x 1, y1, z 1) on the three-dimensional scene model of the area to be detected, and stores the abnormal sound position dot image in a storage packet corresponding to the coordinate axes (x 1, y1, z 1).
When the acoustic imager moves to the detection point No. 11, the abnormal sound judging unit judges that abnormal sound exists at the position, and the abnormal sound effectiveness determining unit judges that the abnormal sound is effective, the abnormal sound alarming unit records the position (the detection point No. 11) of the current detection point, the current time (8 points 44 minutes 20 seconds) and the actual position (coordinate axes (x 2, y2, z 2)) of the abnormal sound, the image intercepting unit intercepts the abnormal sound position point picture of the coordinate axes (x 2, y2, z 2) on the three-dimensional scene model of the area to be detected, when the x difference value, the y difference value and the z difference value of the coordinate axes (x 2, y2, z 2) and the coordinate axes (x 1, y1, z 1) are smaller than the preset difference value, the two abnormal sound actual positions are close to each other, and the abnormal sound position point picture of the coordinate axes (x 2, y2, z 2) is directly stored in the storage packet No. 1 corresponding to the coordinate axes (x 1, y1, z 1).
When the acoustic imager moves to the No. 12 detection point and the abnormal sound judgment unit judges that no abnormal sound exists, the acoustic imager continues to move forward along with the moving trolley.
When the acoustic imager moves to the No. 13 detection point, the abnormal sound judging unit judges that the abnormal sound exists, but the abnormal sound validity determining unit judges that the abnormal sound is invalid, the acoustic imager continues to move forward along with the moving trolley.
When the acoustic imager moves to the detection point number 14, the abnormal sound judging unit judges that abnormal sound exists, and the abnormal sound effectiveness determining unit judges that the abnormal sound is effective, the abnormal sound alarming unit records the position (the detection point number 14) of the current detection point, the current time (8 points 46 minutes 40 seconds) and the actual position (coordinate axes (x 3, y3, z 3)) of the abnormal sound, the image intercepting unit intercepts and stores the abnormal sound position point picture of the coordinate axes (x 3, y3, z 3) on the three-dimensional scene model of the area to be detected, and when the x difference value, the y difference value and the z difference value of the coordinate axes (x 3, y3, z 3) and the coordinate axes (x 1, y1, z 1) are larger than preset difference values, the two abnormal sound actual positions are different, and the image intercepting unit stores the abnormal sound position point picture of the coordinate axes (x 3, y3, z 3) in the storage packet number 2 corresponding to the coordinate axes (x 3, y3, z 3).
When the acoustic imager displays the alarm data, the alarm data with the same and similar actual positions of the abnormal sounds can be displayed by using the same color, and the alarm data with different actual positions of the abnormal sounds can be displayed by using different colors (because the objects with the same and similar actual positions of the abnormal sounds indicate that the abnormal sounds are generated by the same object, namely, the abnormal sounds are the same abnormal sound source), so that a worker can more conveniently know the abnormal condition in the area to be detected.
After the inspection is finished, the image analysis unit performs image analysis and identification on the abnormal sound position point pictures in the corresponding storage packets one by one according to the creation sequence of the storage packets so as to determine the types of the objects which emit abnormal sounds on the pictures in the corresponding storage packets. The objects which give off abnormal sounds on the pictures in the same storage packet are the same by default. Compared with single picture identification, the number of pictures in the storage bag is large, the shooting angle is large and comprehensive, and therefore the identification effect of the object which gives off abnormal sound is good, and the identification accuracy is high. In addition, when the identified object is not the abnormal generating object in the scene to be detected, for example, the scene to be detected is detected with pressurized gas leakage, and the identified object is not an object such as a pipeline or a valve, the alarm can be determined to be invalid and corresponding alarm data can be deleted, because the sound is possibly a normal sound emitted by a testing device placed in the field. The image analysis unit can remove non-abnormal sound sources, and the alarm accuracy of the unmanned patrol inspection type alarm system is further improved.
Example 2: as shown in fig. 2, an unmanned inspection type alarm method based on acoustic imaging, which adopts the unmanned inspection type alarm system of embodiment 1, includes the following steps:
and L1, determining an inspection route in a region to be detected, and arranging a bearing mobile device and an acoustic imager in the region to be detected so that the bearing mobile device can drive the acoustic imager to move along the inspection route. The L1 specifically includes: and L11, arranging the inspection track in the area to be detected according to the required inspection route. And L12, connecting a movable trolley capable of moving on the inspection track with the inspection track. L13. Mounting the rotating part on the mobile cart, and mounting the acoustic imager on the rotating part. And L14, mounting a controller on the mobile trolley, and connecting the controller with the mobile trolley, the rotating part and the acoustic imager.
L2, the bearing moving device drives the acoustic imager to move along the routing inspection route, and when the acoustic imager reaches a new detection point, the acoustic imager enters L3; and ending when the acoustic imager reaches the end point.
And L3, a data cache unit of the acoustic imager stores the M audio data collected by the acoustic imager at the current detection point position, and simultaneously deletes the M audio data collected by the acoustic imager last time. The L3 specifically includes: when the first data cache region is in a first working mode and the second data cache region is in a second working mode, changing the first data cache region into the second working mode and changing the second data cache region into the first working mode; or when the first data cache region is in a second working mode and the second data cache region is in a first working mode, changing the first data cache region into the first working mode and changing the second data cache region into the second working mode; the first working mode is to store M audio data collected by the acoustic imager at the current detection point position, and the second working mode is to delete M audio data collected by the acoustic imager last time.
L4, an abnormal sound judgment unit of the acoustic imager acquires the audio data in the data cache unit, calculates to obtain a to-be-judged audio with the largest sound pressure value in the audio data, subtracts the background noise value of the current detection point position from the sound pressure value of the to-be-judged audio to obtain an abnormal sound judgment value, compares the abnormal sound judgment value with a preset sound threshold value, judges that the current detection point position has abnormal sound when the abnormal sound judgment value is larger than the preset sound threshold value, and enters L5; otherwise, judging that the current detection point position has no abnormal sound, and returning to L2.
L5, an abnormal sound effectiveness determining unit of the acoustic imager determines the effectiveness of the abnormal sound according to the view field range of the current detection point position, and when the generation position of the abnormal sound is located in the view field range of the current detection point position, the abnormal sound is judged to be effective and enters L6; otherwise, judging that the abnormal sound is invalid, and returning to L2.
And L6, an abnormal sound alarm unit of the acoustic imager records the position of the current detection point, the current time and the actual position of the abnormal sound as alarm data, wherein the actual position of the abnormal sound is a coordinate value of the abnormal sound in the three-dimensional scene model of the area to be detected, and the L2 is returned.
Before returning to L2, the L6 further comprises the following steps: and L61, when the abnormal sound alarm unit records the actual position of the abnormal sound, an image intercepting unit of the acoustic imager intercepts a corresponding abnormal sound position point picture on the three-dimensional scene model of the area to be detected according to the actual position of the abnormal sound, and stores the abnormal sound position point picture in a storage packet corresponding to the actual position of the abnormal sound.
After the L2 is finished, the method also comprises the following steps: and L21, the image analysis unit of the acoustic imager is used for carrying out image analysis and identification on the abnormal sound position point pictures in the corresponding storage packets one by one according to the creation sequence of the storage packets so as to determine the types of the objects which emit abnormal sounds on the pictures in the corresponding storage packets.
The L2 further comprises the following steps: the L01 bearing mobile device drives the acoustic imager to move along the routing inspection route, and when the acoustic imager reaches a new detection point, the L02 is started; when the acoustic imager reaches the end point, L04 is entered. A background noise value determining unit of the L02 acoustic imager acquires N pieces of background noise data of the current detection point position, and calculates the root mean square value of the N pieces of background noise data to be used as the background noise value of the current detection point position; or acquiring N pieces of background noise data of the current detection point position, and selecting a median value of the N pieces of background noise data to serve as a background noise value of the current detection point position. A background noise value storage unit of the acoustic imager stores a background noise value at a current detection point position. And a detection point view field range image acquisition and storage unit of the L03 acoustic imager acquires and stores a view field range image of the current detection point position and returns to L01. The L04 to-be-detected area three-dimensional scene model determining unit constructs a three-dimensional scene model of the to-be-detected area through the view field range images of all detection point positions in the to-be-detected area.
The above-described embodiments are merely illustrative of the preferred embodiments of the present invention and do not limit the spirit and scope of the present invention. Various modifications and improvements of the technical solutions of the present invention that are made by those skilled in the art without departing from the design concept of the present invention shall fall within the protection scope of the present invention, and the technical contents of the present invention that are claimed shall be fully described in the claims.
Claims (8)
1. The utility model provides an unmanned formula alarm system that patrols and examines based on acoustic imaging which characterized in that: the method comprises the following steps:
an acoustic imager;
the bearing moving device is used for installing the acoustic imager and driving the acoustic imager to move along an inspection route in the area to be detected, and the inspection route is provided with a plurality of detection points;
wherein the acoustic imager comprises:
the data cache unit is used for storing M pieces of audio data acquired by the acoustic imager at the current detection point position and deleting M pieces of audio data acquired by the acoustic imager last time;
the abnormal sound judging unit is used for acquiring the audio data in the data cache unit, calculating to obtain a to-be-judged audio with the largest sound pressure value in the audio data, subtracting the background noise value at the current detection point position from the sound pressure value of the to-be-judged audio to obtain an abnormal sound judging value, comparing the abnormal sound judging value with a preset sound threshold value, judging that the current detection point position has abnormal sound when the abnormal sound judging value is larger than the preset sound threshold value, and otherwise judging that the current detection point position does not have the abnormal sound;
the abnormal sound validity determining unit is used for determining the validity of the abnormal sound according to the field range of the current detection point position when the current detection point position has the abnormal sound, judging that the abnormal sound is valid when the generation position of the abnormal sound is positioned in the field range of the current detection point position, and otherwise, judging that the abnormal sound is invalid;
the abnormal sound alarm unit is used for recording the position of a current detection point, the current time and the actual position of the abnormal sound as alarm data when the abnormal sound is effective, wherein the actual position of the abnormal sound is a coordinate value of the abnormal sound in the three-dimensional scene model of the area to be detected;
the data cache unit comprises a first data cache region and a second data cache region, and when the first data cache region is in a first working mode, the second data cache region is in a second working mode; when the first data cache region is in a second working mode, the second data cache region is in a first working mode; the first working mode is to store M audio data collected by the acoustic imager at the current detection point position, and the second working mode is to delete M audio data collected by the acoustic imager last time.
2. The unmanned inspection type alarm system based on acoustic imaging according to claim 1, characterized in that: the acoustic imager further comprises:
the image intercepting unit is used for intercepting a corresponding abnormal sound position point image on the three-dimensional scene model of the area to be detected through the actual position of the abnormal sound when the abnormal sound alarming unit records the actual position of the abnormal sound, and storing the abnormal sound position point image in a storage packet corresponding to the actual position of the abnormal sound;
and the image analysis unit is used for carrying out image analysis and identification on the abnormal sound position point pictures in the corresponding storage packets one by one according to the creation sequence of the storage packets so as to determine the types of the objects which emit abnormal sounds on the pictures in the corresponding storage packets.
3. The unmanned inspection type alarm system based on acoustic imaging according to claim 1, characterized in that: the acoustic imager further comprises:
the background noise value determining unit is used for acquiring N pieces of background noise data of the current detection point position and calculating the root mean square value of the N pieces of background noise data to be used as the background noise value of the current detection point position; or, the detecting device is used for acquiring N pieces of background noise data of the current detecting point position, and selecting a median value of the N pieces of background noise data as a background noise value of the current detecting point position;
the background noise value storage unit is used for storing the background noise value of the current detection point position;
the detection point view field range image acquisition and storage unit is used for acquiring and storing a view field range image of the current detection point position;
and the three-dimensional scene model determining unit is used for constructing and forming a three-dimensional scene model of the area to be detected through the view field range images of all detection point positions in the area to be detected.
4. The unmanned inspection type alarm system based on acoustic imaging according to claim 1, characterized in that: the bearing mobile device comprises:
the inspection track is arranged in the area to be detected according to a required inspection route;
the movable trolley is connected with the inspection track and is used for moving on the inspection track;
a rotating member mounted on the mobile cart for mounting the acoustic imager;
and the controller is arranged on the moving trolley and is electrically connected with the moving trolley, the rotating part and the acoustic imager.
5. An unmanned inspection type alarm method based on acoustic imaging adopts the unmanned inspection type alarm system of any one of claims 1 to 4, and is characterized in that: the method comprises the following steps:
l1, determining an inspection route in a region to be detected, and arranging a bearing mobile device and an acoustic imager in the region to be detected so that the bearing mobile device can drive the acoustic imager to move along the inspection route;
l2, the bearing moving device drives the acoustic imager to move along the routing inspection route, and when the acoustic imager reaches a new detection point, the acoustic imager enters L3; when the acoustic imager reaches the end point, ending;
l3, a data cache unit of the acoustic imager stores M audio data collected by the acoustic imager at the current detection point position and deletes M audio data collected by the acoustic imager last time;
l4, an abnormal sound judgment unit of the acoustic imager acquires the audio data in the data cache unit, calculates to obtain a to-be-judged audio with the maximum sound pressure value in the audio data, subtracts the background noise value of the current detection point position from the sound pressure value of the to-be-judged audio to obtain an abnormal sound judgment value, compares the abnormal sound judgment value with a preset sound threshold value, judges that the current detection point position has abnormal sound when the abnormal sound judgment value is larger than the preset sound threshold value, and enters L5; otherwise, judging that the position of the current detection point does not have abnormal sound, and returning to L2;
l5, an abnormal sound effectiveness determining unit of the acoustic imager determines the effectiveness of the abnormal sound according to the view field range of the current detection point position, and when the generation position of the abnormal sound is located in the view field range of the current detection point position, the abnormal sound is judged to be effective and enters L6; otherwise, judging that the abnormal sound is invalid, and returning to L2;
l6, an abnormal sound alarm unit of the acoustic imager records the position of a current detection point, the current time and the actual position of the abnormal sound as alarm data, wherein the actual position of the abnormal sound is a coordinate value of the abnormal sound in the three-dimensional scene model of the area to be detected, and the L2 is returned;
the L3 specifically includes: when the first data cache region is in a first working mode and the second data cache region is in a second working mode, changing the first data cache region into the second working mode and changing the second data cache region into the first working mode; or when the first data cache region is in a second working mode and the second data cache region is in a first working mode, changing the first data cache region into the first working mode and changing the second data cache region into the second working mode; the first working mode is to store M audio data collected by the acoustic imager at the current detection point position, and the second working mode is to delete M audio data collected by the acoustic imager last time.
6. The unmanned inspection type alarm method based on acoustic imaging according to claim 5, characterized in that: before returning to L2, the L6 also comprises the following steps:
l61, when the abnormal sound alarm unit records the actual position of the abnormal sound, an image intercepting unit of the acoustic imager intercepts a corresponding abnormal sound position point picture on a three-dimensional scene model of the area to be detected through the actual position of the abnormal sound, and stores the abnormal sound position point picture in a storage packet corresponding to the actual position of the abnormal sound;
after the L2 is finished, the method further comprises the following steps:
and L21, the image analysis unit of the acoustic imager is used for carrying out image analysis and identification on the abnormal sound position point pictures in the corresponding storage packets one by one according to the creation sequence of the storage packets so as to determine the types of the objects which emit abnormal sounds on the pictures in the corresponding storage packets.
7. The unmanned patrol inspection type alarm method based on the acoustic imaging is characterized by comprising the following steps of: the L2 further comprises the following steps:
l01, the bearing moving device drives the acoustic imager to move along the routing inspection route, and when the acoustic imager reaches a new detection point, the acoustic imager enters L02; when the acoustic imager reaches the end point, entering L04;
l02, a background noise value determining unit of the acoustic imager acquires N pieces of background noise data of the current detection point position, and calculates root mean square values of the N pieces of background noise data to serve as background noise values of the current detection point position; or acquiring N pieces of background noise data of the current detection point position, and selecting a median value of the N pieces of background noise data as a background noise value of the current detection point position;
a background noise value storage unit of the acoustic imager stores a background noise value of the current detection point position;
l03, a detection point view field range image acquisition and storage unit of the acoustic imager acquires and stores a view field range image of the current detection point position and returns to L01;
and L04, constructing a three-dimensional scene model of the to-be-detected region by the three-dimensional scene model determining unit of the to-be-detected region through the view field range images of all detection point positions in the to-be-detected region.
8. The unmanned patrol inspection type alarm method based on the acoustic imaging is characterized by comprising the following steps of: the L1 specifically includes:
l11, arranging an inspection track in the area to be detected according to the required inspection route;
l12, connecting a movable trolley capable of moving on the inspection rail with the inspection rail;
l13, mounting a rotating part on the mobile trolley, and mounting the acoustic imager on the rotating part;
and L14, mounting a controller on the mobile trolley, and connecting the controller with the mobile trolley, the rotating part and the acoustic imager.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211587563.6A CN115662083B (en) | 2022-12-12 | 2022-12-12 | Unmanned inspection type alarm system and method based on acoustic imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211587563.6A CN115662083B (en) | 2022-12-12 | 2022-12-12 | Unmanned inspection type alarm system and method based on acoustic imaging |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115662083A CN115662083A (en) | 2023-01-31 |
| CN115662083B true CN115662083B (en) | 2023-04-07 |
Family
ID=85019772
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211587563.6A Active CN115662083B (en) | 2022-12-12 | 2022-12-12 | Unmanned inspection type alarm system and method based on acoustic imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115662083B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115327480A (en) * | 2022-08-09 | 2022-11-11 | 合肥智能语音创新发展有限公司 | Sound source positioning method and system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109854965A (en) * | 2017-11-30 | 2019-06-07 | 中国科学院声学研究所 | One kind being used for natural gas station pipeline leakage checking system and its method |
| CN108597164B (en) * | 2018-04-28 | 2021-11-12 | 侯苏华 | Anti-theft method, anti-theft device, anti-theft terminal and computer readable medium |
| KR20190125232A (en) * | 2019-07-18 | 2019-11-06 | 주식회사 경림이앤지 | Peculiar sound detection system and method in a peculiar sound detection terminal including camera, array microphone and A/V controller, and CCTV system |
| CN114414963A (en) * | 2022-01-20 | 2022-04-29 | 西安交通大学 | Acoustic imaging positioning system and method for intelligent monitoring of substation domain faults |
| CN114964650B (en) * | 2022-08-01 | 2022-11-18 | 杭州兆华电子股份有限公司 | Gas leakage alarm method and device based on acoustic imaging |
-
2022
- 2022-12-12 CN CN202211587563.6A patent/CN115662083B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115327480A (en) * | 2022-08-09 | 2022-11-11 | 合肥智能语音创新发展有限公司 | Sound source positioning method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115662083A (en) | 2023-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101028060B1 (en) | Apparatus and method for measuring degradation of power equipment using thermal image recognition | |
| KR100898061B1 (en) | Hybrid tunnel scanning device | |
| CN104079881A (en) | Monitoring device and monitoring method related to monitoring device | |
| KR101871937B1 (en) | Device and Method for measuring flight data of flying objects using high speed video camera and computer readable recording medium having program the same | |
| US20010012016A1 (en) | Three-dimensional shape measuring system | |
| JP4056777B2 (en) | Autonomous mobile object traveling system and autonomous mobile object position correction method | |
| CN102445194B (en) | Measuring method and surveying instrument | |
| KR101063565B1 (en) | Vibration Measurement System and Method Using Real-time Image Processing | |
| GB2202104A (en) | Ranging by correlation | |
| CN115662083B (en) | Unmanned inspection type alarm system and method based on acoustic imaging | |
| KR101923314B1 (en) | Unmanned air wehicle for tunnel structure investigation, and system using the same | |
| CN212635748U (en) | Power distribution network inspection robot | |
| CN112384751B (en) | Optical measuring method and optical measuring device | |
| CN109559333B (en) | Tracking recording device | |
| CN110702016A (en) | Power transmission line icing measurement system and method | |
| CN118224050B (en) | Unmanned on duty formula wind turbine external monitoring robot system | |
| KR20110121426A (en) | Day and night combined moving object observation system | |
| CN109842796B (en) | Fault detection system and method of video shooting equipment and video shooting equipment | |
| KR101497396B1 (en) | A system for measuring target location and method for measuring target location using the same | |
| CN111046765B (en) | Dangerous early warning method and system for high-speed rail | |
| CN214308618U (en) | Device for acquiring vibration data of flexible bridge by remote sensing and machine vision | |
| KR20090035763A (en) | Ultrasonic Inspection Device and Measuring Jig Using 3D Camera Device | |
| KR101128967B1 (en) | Apparatus for detecting flight of object | |
| KR100627483B1 (en) | Emission Recognition and Alarm System and Method through Image Analysis | |
| KR101471187B1 (en) | System and method for controlling movement of camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |