CN114827466B - Human eye-like equipment image acquisition device and image acquisition method - Google Patents
Human eye-like equipment image acquisition device and image acquisition method Download PDFInfo
- Publication number
- CN114827466B CN114827466B CN202210413048.XA CN202210413048A CN114827466B CN 114827466 B CN114827466 B CN 114827466B CN 202210413048 A CN202210413048 A CN 202210413048A CN 114827466 B CN114827466 B CN 114827466B
- Authority
- CN
- China
- Prior art keywords
- image
- image acquisition
- equipment
- defect
- cradle head
- 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
- 238000000034 method Methods 0.000 title claims abstract description 56
- 230000007547 defect Effects 0.000 claims abstract description 63
- 230000008569 process Effects 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 19
- 230000002950 deficient Effects 0.000 claims description 17
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 5
- 230000001681 protective effect Effects 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 abstract description 19
- 238000001514 detection method Methods 0.000 abstract description 2
- 210000003128 head Anatomy 0.000 description 30
- 230000006872 improvement Effects 0.000 description 10
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/67—Focus control based on electronic image sensor signals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/695—Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/698—Control of cameras or camera modules for achieving an enlarged field of view, e.g. panoramic image capture
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/95—Computational photography systems, e.g. light-field imaging systems
- H04N23/951—Computational photography systems, e.g. light-field imaging systems by using two or more images to influence resolution, frame rate or aspect ratio
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a device image acquisition device and an image acquisition method for a human-eye-simulated device, which belong to the technical field of detection of the running state of the device. The device for acquiring the image of the equipment simulating the human eyes has a compact structure, and the method for acquiring the image is convenient and fast, so that the panoramic image acquisition and the local clear image acquisition of the equipment to be monitored can be realized, the image of the defect part on the equipment to be monitored can be accurately acquired, an accurate and enough judgment basis is provided for accurately judging the running state of the equipment to be monitored, the monitoring accuracy and reliability of the equipment are ensured, and sufficient guarantee is provided for the normal running of the corresponding equipment, so that the device has good practical value and application prospect.
Description
Technical Field
The invention belongs to the technical field of equipment operation state detection, and particularly relates to a device image acquisition device and an image acquisition method for a human-eye-simulated equipment.
Background
In the operation process of the high-voltage transformer substation, the operation states of various devices often directly influence the safety and reliability of the power grid. Therefore, in the actual operation of the substation, it is necessary to monitor the operation states of various kinds of devices, particularly, heavy-point components such as transformers, periodically or in real time.
At present, for monitoring operation equipment of a transformer substation, a method for setting a common camera device to perform overall shooting of the equipment is mostly adopted, so that state monitoring and judgment of related equipment are completed. Although the method can realize the monitoring of the substation equipment to a certain extent, the monitoring precision is poor, abnormal states of a local area and a core part of the equipment cannot be accurately observed, so that the interpretation error of the operation working conditions of the related equipment of the substation is large, and the monitoring of the operation state of the substation equipment cannot be rapidly and accurately realized.
Disclosure of Invention
Aiming at one or more of the defects or improvement demands of the prior art, the invention provides a device image acquisition device and an image acquisition method of a human-simulated eye, which can realize human-simulated eye monitoring of substation equipment, realize accurate acquisition of panoramic images and partial images and fully ensure the accuracy and reliability of equipment running state monitoring.
In order to achieve the above object, according to one aspect of the present invention, there is provided a human eye-like apparatus image acquisition method, which is implemented by using an apparatus image acquisition device;
the equipment image acquisition device comprises a cradle head and an image acquisition main body which is rotatably connected to the cradle head through a bracket; the image acquisition main body comprises a variable-focus camera, an infrared camera and a binocular depth camera which are integrally arranged, and an image processing module is arranged in electrical connection with each camera, so that the image processing module can receive images acquired by the corresponding cameras and synthesize images for defect identification;
correspondingly, the method for acquiring the equipment image by using the equipment image acquisition device comprises the following steps:
s1, controlling the central axis of a shooting view field of an image acquisition main body to be consistent with the central axis of equipment to be monitored;
s2, initializing the position of the cradle head, and marking as P 0 The method comprises the steps of carrying out a first treatment on the surface of the Simultaneously, initializing the focal length of the variable-focus camera, so that a shooting picture is reduced to the minimum and the most objects can be shot in a camera view field;
s3, taking the region to be shot as P 0 Dividing the center into a nine-square grid, which is marked as P in turn 0 ～P 8 Starting the image acquisition main body to take a nine-grid photograph of the region to be photographed;
s4, transmitting the nine-grid pictures obtained in the step S3 to the image processing module, and carrying out picture fusion at the image processing module to obtain a panoramic large picture of the current shooting area of the equipment;
s5, controlling an image processing module to identify the panoramic large image by using a deep learning algorithm, finding out a defect part in the image, and marking the defect partDefect-recording rectangular frame r n N is an integer not less than 1;
s6, determining a defect rectangular frame r n Regional position P in plane Sudoku m The method comprises the steps of carrying out a first treatment on the surface of the m is an integer, and m is more than or equal to 0 and less than or equal to 8;
s7, controlling the cradle head to rotate to a nine-grid area corresponding to the defective rectangular frame, and calculating the physical distance between the current defective part and the device by a binocular depth camera according to the coordinates of the defective rectangular frame; controlling the variable-focus camera to adjust the focal length and amplifying the image of the defect part;
s8, controlling the cradle head to rotate to the region position P obtained in the step S6 respectively m And according to the defect rectangular frame r n The position of the cradle head is adjusted according to the starting point coordinates, the width and the height information, and the defect part is arranged at the center of a shooting picture of the image acquisition main body;
s9, shooting a defective rectangular frame r n Storing the image of the corresponding part to obtain a high-definition large image corresponding to the defect part; then, turning to the cradle head to reset to the initial position P 0 And (S8) repeating the step until the defect images corresponding to all the defect rectangular frames are acquired.
As a further improvement of the present invention, in step S3, the process of taking a picture of the nine squares includes:
s301: setting the rotation speed of the cradle head to be uniform, and rotating the cradle head to an initial position P 0 The method comprises the steps of carrying out a first treatment on the surface of the After that, shooting is started and stored to obtain P 0 A picture corresponding to the position;
s302: starting the cradle head to rotate, and starting the cradle head to rotate from P a Position starts and goes over a grid to P a+1 And (3) the positions a are integers, a is more than or equal to 0 and less than 8, and the pictures corresponding to the positions of all the areas in the nine-square lattice are sequentially obtained.
As a further improvement of the invention, the intervals between the adjacent two position areas of the nine Gong Gezhong are equal, and the rotation time of the cradle head between the adjacent two positions is equal.
As a further improvement of the present invention, in step S5, the process of obtaining the defective rectangular frame is as follows:
s501: collecting data pictures of various defects corresponding to equipment to be monitored;
s502: marking, cleaning and amplifying the data picture according to different defect types;
s503: training the processed sample by using a yolov5 training frame to obtain a high-precision equipment defect recognition model;
s504: the model is packaged and deployed on a computing platform with AI computing power in a micro-service mode, and defect identification service is provided for the outside;
s505: scaling, reducing the average value, normalizing and regularizing the synthesized panoramic big image;
s506: and uploading the preprocessed picture to the defect recognition service, and taking the rectangular frame exceeding a certain threshold value as an alarm rectangular frame to be output.
As a further improvement of the present invention, in step S6, the procedure of determining the position of the defective rectangular frame is as follows:
calculating the center coordinates (x ', y') of the defective rectangular frame according to the following formulas 1 and 2, and calculating the position of the region where the coordinates are located in the nine-square lattice according to the corresponding relation of the coordinates;
wherein x is imin The minimum x coordinate, x, of the square box i of the nine squares imax The maximum x coordinate of the square frame i; y is imin Is the minimum y coordinate, y of the square frame i of the nine squares imax The maximum y coordinate of the square box i of the nine squares.
As a further improvement of the present invention, in step S7, the variable-focus camera performs adjustment of the focal length by formula 3;
1/u+1/v=1/f (equation 3)
Where u is the object distance, v is the image distance, and f is the focal length.
As a further improvement of the invention, when the nine-grid shooting of the region to be shot is performed in step S3, the infrared camera is synchronously controlled to shoot the temperature state of each region in the nine-grid.
In another aspect of the present invention, a device image capturing apparatus for simulating a human eye is provided, which includes a pan-tilt and an image capturing body;
the image acquisition main body is rotatably connected to the cradle head through a bracket and can perform pitching adjustment relative to the cradle head;
the image acquisition main body comprises an integrally arranged variable-focus camera, an infrared camera and a binocular depth camera, which are respectively used for acquiring high-resolution images under the condition of zooming, acquiring the temperature state of a corresponding area and measuring the distance in the image acquisition process; meanwhile, an image processing module is arranged in the image acquisition main body and is electrically connected with each camera and used for receiving images acquired by the corresponding cameras and synthesizing images for defect identification.
As a further improvement of the invention, a moving mechanism is arranged corresponding to the cradle head; the cradle head is arranged on the moving mechanism in a carrying way, and can change shooting positions under the drive of the moving mechanism.
As a further improvement of the invention, a protective cover is provided on top of the image acquisition body.
The above-mentioned improved technical features can be combined with each other as long as they do not collide with each other.
In general, the above technical solutions conceived by the present invention have the beneficial effects compared with the prior art including:
(1) According to the human eye-simulated equipment image acquisition method, the image acquisition main body with the variable-focus camera, the binocular depth camera and the infrared camera is correspondingly arranged on the cradle head, the nine-grid division is carried out on the shooting area of equipment to be monitored, the image acquisition main body is controlled to carry out the nine-grid shooting, and then the panoramic image in the shooting area is accurately obtained; meanwhile, by arranging the image processing module, the defect positions in the panoramic image are determined in real time based on an AI technology, and the defect positions are marked and the amplified image is shot corresponding to the defect positions, so that the defect state of the equipment to be monitored is accurately acquired, the accurate monitoring of the corresponding equipment is realized, and the running reliability of the equipment is ensured.
(2) According to the human eye-like equipment image acquisition method, through the optimization design of the photographing process of the nine-square grid, the acquisition process of the defect rectangular frame, the position determination process of the defect rectangular frame and the like, the accurate completion of the equipment monitoring process can be realized, the determination and clear photographing of the defect part can be quickly realized, the accuracy and the efficiency of the state acquisition of the defect part are improved, the running state of related equipment is fed back in time, the normal operation of the related equipment is ensured, and sufficient basis is provided for the subsequent maintenance of the defect equipment.
(3) The device for acquiring the image of the equipment simulating the human eyes has a compact structure, is convenient to carry, can realize the shooting process of the human eyes through the integrated arrangement of the variable-focus camera, the infrared camera, the binocular depth camera and the image processing module, can realize the accurate acquisition of the panoramic large image and the regional defect image of the equipment to be monitored, can ensure the accuracy and the reliability of the monitoring of the running state of the equipment, can realize the accurate monitoring of the corresponding equipment, and can ensure the normal running of the corresponding system.
(4) The device for acquiring the device image imitating the human eyes has a compact structure, is convenient and fast in acquisition method, can acquire the panoramic image and the local clear image of the device to be monitored, can accurately acquire the image of the defect part on the device to be monitored, provides accurate and enough judgment basis for accurately judging the running state of the device to be monitored, ensures the accuracy and the reliability of the monitoring of the device, provides enough guarantee for the normal running of the corresponding device, and has good practical value and application prospect.
Drawings
FIG. 1 is a flow chart of an embodiment of the invention when an equipment image acquisition device is utilized to acquire equipment images;
fig. 2 is a schematic structural diagram of an image acquisition device of a human eye-simulated device in an embodiment of the present invention;
FIG. 3 is a trace diagram of a nine-grid image acquisition by an apparatus image acquisition device in an embodiment of the present invention;
like reference numerals denote like technical features throughout the drawings, in particular:
1. an image acquisition main body; 2. a cradle head; 3. a bracket; 4. a protective cover;
101. a variable focus camera; 102. an infrared camera; 103. binocular depth camera.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the device or element being referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.
In the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "secured," and the like are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; either directly or indirectly, through intermediaries, or both, may be in communication with each other or in interaction with each other, unless expressly defined otherwise. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
In the present invention, unless expressly stated or limited otherwise, a first feature "up" or "down" a second feature may be the first and second features in direct contact, or the first and second features in indirect contact via an intervening medium. Moreover, a first feature being "above," "over" and "on" a second feature may be a first feature being directly above or obliquely above the second feature, or simply indicating that the first feature is level higher than the second feature. The first feature being "under", "below" and "beneath" the second feature may be the first feature being directly under or obliquely below the second feature, or simply indicating that the first feature is less level than the second feature.
Examples:
referring to fig. 2, the device image capturing apparatus for a human eye-like device in a preferred embodiment of the present invention includes an image capturing main body 1 rotatably connected to a pan-tilt 2 through a bracket 3, so that the image capturing main body 1 can perform pitching rotation relative to the pan-tilt 2, thereby realizing accurate alignment of the image capturing main body 1 and an object to be photographed. Simultaneously, through the corresponding setting of cloud platform 2 for image acquisition main part 1 can carry out the rotation regulation in the horizontal plane along with cloud platform 2, realizes the adjustment of image acquisition main part 1 at horizontal annular upward shooting visual angle.
In addition, in the preferred embodiment, the pan-tilt 2 can perform 360 ° omni-directional rotation, and it is further preferably mounted on a mobile cart (not shown in the figure) capable of moving, and by using the mounting movement of the mobile cart, the conversion of the shooting position can be realized, so that the image acquisition main body 1 can monitor different positions of the same device or different devices, and correspondingly acquire images of different areas and different devices.
Specifically, the image capturing body 1 in the preferred embodiment has a box-like structure, in which a variable focus camera 101, an infrared camera 102, and a binocular depth camera 103 are integrally provided, and an image processing module is correspondingly provided in the image capturing body 1. Wherein the variable focus camera 101 is used to capture high resolution images with corresponding zooming; the infrared camera 102 is used for monitoring and shooting abnormal heating of the equipment; the binocular depth camera 103 is used for completing ranging in the image acquisition process, and the visual field centers of the variable-focus camera 101 and the binocular depth camera 103 are the same, and the visual fields of the variable-focus camera 101 and the binocular depth camera 103 are always consistent. Correspondingly, the image processing module is respectively and electrically connected with each camera and is used for receiving the images acquired by the corresponding camera and synthesizing the images for defect identification.
Preferably, for protecting the image capturing body 1, a protective cover 4 is preferably provided on top of it, which provides shielding and protection of the image capturing body 1 during use.
Further, in the preferred embodiment, the process of monitoring the running state of the device by using the device image acquisition device is as follows:
s1, the control equipment image acquisition device moves to one side of equipment to be monitored, and the central axis of the image acquisition main body 1 is ensured to be consistent with the central axis of the equipment to be monitored.
In actual operation, the method for keeping the central axes of the image acquisition main body 1 and the equipment to be monitored consistent mainly comprises the following steps: firstly, reducing the focal length of the variable-focus camera 101 to enable a shooting picture to contain the whole equipment to be monitored; after the image acquisition main body 1 recognizes the whole outline of the equipment, the central position of the whole outline of the equipment is taken, and the central position is coincident with the central position of the shot picture, namely the central axis is considered to be consistent. Thereafter, the focal length is enlarged and the pan/tilt head 2 is adjusted at a fixed point.
S2, initializing the position of the cradle head 2, namely P 0 The method comprises the steps of carrying out a first treatment on the surface of the At the same time, the focal length of the variable-focus camera 101 is initialized so that the frame is reduced to a minimum and the number of objects that can be captured within the camera field of view is maximized.
S3, dividing the region to be shot into nine-grid forms, and sequentially marking as P 0 ～P 8 Starting an image acquisition main body 1 to perform Sudoku photographing on equipment to be monitored, and acquiring an image of the equipment to be monitored;
during actual shooting, the holder 2 firstly generates a virtual nine-grid shooting area aiming at an object to be monitored; secondly, the holder 2 preferably starts with the center position of the nine squares, and the equipment is locally shot according to the nine directions of the center, the left upper part, the upper left part, the upper right part, the right lower part, the lower right part and the lower left part in sequence to obtain nine squares pictures; after shooting is completed, the cradle head 2 returns to the initial position P 0 。
It is apparent that the image capturing body 1 completes P during the shooting process 0 After the shooting of the point location, the subsequent shooting paths thereof are sequentially performed in the "clockwise" direction, sequentially completed (P 1 、P 2 、……、P 8 ) Image capturing of each point is shown in fig. 3. It will be appreciated that, when the movement path of the pan/tilt head 2 is actually set, the photographing path of the image capturing body 1 may be set to be counterclockwise, or any free scattered photographing path may be adopted, so long as the image capturing with the center P of the square grid can be obtained 0 Nine images in the center are needed.
Further specifically, the detailed implementation process of the nine-grid shooting method in step S3 is as follows:
s301: setting the rotation speed of the cradle head 2 to be uniform, and rotating the cradle head 2 to an initial position P 0 When shooting is started, the picture is saved and is P 0 And the picture corresponding to the position.
S302: starting the cradle head 2 to rotate from P 0 Starting from the position and starting timing, when the rotation time is t 0 When the representation passes through a lattice to P 1 After the position is reached, taking a photograph and storing to obtain P 1 Picture of spatial region.
S303: the rotation route of the control holder 2 is shown in fig. 3, and the like until all the areas corresponding to the nine squares are shot.
In the preferred embodiment, each spatial region in the nine Gong Geou domains is preferably square, i.e., adjacent latticesThe distances between the two are equal, so that the time length t of each rotation of the cradle head 2 can be ensured m Equal.
In addition, in actual operation, the size of one side of the device is often not very large, and in this case, the division of the nine boxes is performed so that the entire one-side area of the device can be included by the determination of the nine boxes. Of course, it will be appreciated that when the area of a single side of the device is large, a portion of the device may be selected and the selected area may be squared.
And S4, transmitting the nine-grid pictures obtained in the step S3 to an image processing module, and carrying out picture fusion at the image processing module to obtain a panoramic large picture of the current shooting area of the equipment.
In the actual implementation of the fusion of the panoramic large images, the fusion method preferably comprises the following steps: the single-layer image is fused firstly, and then the fusion between layers is carried out. Taking the form in FIG. 3 as an example, P is first set 2 、P 3 、P 4 Fused into a large graph P x The method comprises the steps of carrying out a first treatment on the surface of the And then P is added 1 、P 0 、P 5 Fused into a large graph P y The method comprises the steps of carrying out a first treatment on the surface of the And P is to 8 、P 7 、P 6 Fused into a large graph P z The method comprises the steps of carrying out a first treatment on the surface of the Thereafter, P is x 、P y 、P z Rotating anticlockwise by 90 degrees, and then sequentially fusing to obtain a complete large graph; and finally, rotating the obtained complete large image by 90 degrees clockwise to obtain the panoramic large image.
S5, controlling an image processing module to identify the panoramic large image by using a deep learning algorithm, finding out a defect part in the image, and marking a rectangular frame (r 1 、r 2 、……、r n The method comprises the steps of carrying out a first treatment on the surface of the Wherein n is an integer of not less than 1).
Specifically, the process of obtaining the rectangular frame of the defect part in the preferred embodiment includes the following steps:
s501: collecting data pictures of various defects corresponding to equipment to be monitored;
s502: marking, cleaning and amplifying the data picture according to different defect types;
s503: training the processed sample by using a yolov5 training frame to obtain a high-precision equipment defect recognition model;
s504: the model is packaged and deployed on a computing platform with AI computing power in a micro-service mode, and defect identification service is provided for the outside;
s505: and carrying out preprocessing such as scaling, average reduction, normalization, regularization and the like on the synthesized panoramic large image.
S506: and uploading the preprocessed picture to the defect recognition service, and taking the rectangular frame exceeding a certain threshold value as an alarm rectangular frame to be output.
S6, finding out the region position of the defect rectangular frame in the plane nine-square grid, and marking as (P 0 、P 1 、……、P m The method comprises the steps of carrying out a first treatment on the surface of the Wherein m is an integer, and m is more than or equal to 0 and less than or equal to 8).
Specifically, in the preferred embodiment, the process of determining the location of each defective rectangular box in the nine boxes is as follows:
according to the following formulas 1 and 2, the center coordinates (x ', y') of the defective rectangular frame are calculated, and the region where the coordinates are located in the nine-square lattice is calculated according to the correspondence of the coordinates.
Wherein x is imin The minimum x coordinate, x, of the square box i of the nine squares imax The maximum x coordinate of the square frame i; y is imin Is the minimum y coordinate, y of the square frame i of the nine squares imax The maximum y coordinate of the square box i of the nine squares.
S7, controlling the cradle head 2 to rotate to a nine-square area corresponding to the defect rectangular frame, calculating the physical distance dis between the current defect part and the device by the binocular depth camera 103 according to the coordinates of the defect rectangular frame, and adjusting the focal length and amplifying the image of the defect part by the variable-focus camera 101 according to a conversion formula (namely formula 3) of the distance and the focal length.
1/u+1/v=1/f (equation 3)
Where u is the object distance, v is the image distance, and f is the focal length.
S8, controlling the cradle head 2 to rotate to the position (P) obtained in the step S6 0 、P 1 、……、P n ) Corresponding region P in (1) n And according to a rectangular frame (r 1 、r 2 、……、r n ) The starting point coordinates, width and height information of the cradle head 2 are finely adjusted to the up-down and left-right amplitudes, and the defect part is arranged at the center of the shooting picture of the image acquisition main body 1.
S9, shooting and storing to obtain P m After the high-definition large map of the defect part of the position, the position is transferred to the cradle head 2 to an initial position P 0 . Thereafter, step S8 is repeated until a rectangular frame (r 1 、r 2 、……、r n ) And finishing the treatment.
After the steps are completed, the running state of the monitored object is judged according to the collected defect part image.
In more detail, in a specific preferred embodiment, the above-mentioned device image acquisition device is used for detecting the state of a transformer device in a transformer substation, for example, detecting the state of a transformer device in a high-voltage transformer substation, and in the detecting process, besides the above-mentioned monitoring content, the infrared camera 102 is used for shooting the temperature state in the region with the nine-grid, so that the image acquisition of the defect part in each region with the nine-grid can fully reflect the defect state of the corresponding part of the device, and the accuracy of the device monitoring is further improved. Through the monitoring of the equipment in each transformer substation, the abnormal state of the corresponding equipment can be found in real time, the normal operation of the transformer substation and the related power grid is ensured, and the reliability and stability of the power grid operation are improved.
The device for acquiring the device image imitating the human eyes has a compact structure, is convenient and fast in acquisition method, can acquire the panoramic image and the local clear image of the device to be monitored, can accurately acquire the image of the defect part on the device to be monitored, provides accurate and enough judgment basis for accurately judging the running state of the device to be monitored, ensures the accuracy and the reliability of the monitoring of the device, provides enough guarantee for the normal running of the corresponding device, and has good practical value and application prospect.
It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (10)
1. The equipment image acquisition method for the human eye imitation is characterized by being realized by an equipment image acquisition device;
the equipment image acquisition device comprises a cradle head and an image acquisition main body which is rotatably connected to the cradle head through a bracket; the image acquisition main body comprises a variable-focus camera, an infrared camera and a binocular depth camera which are integrally arranged, and an image processing module is arranged in electrical connection with each camera, so that the image processing module can receive images acquired by the corresponding cameras and synthesize images for defect identification;
correspondingly, the method for acquiring the equipment image by using the equipment image acquisition device comprises the following steps:
s1, controlling the central axis of a shooting view field of an image acquisition main body to be consistent with the central axis of equipment to be monitored;
s2, initializing the position of the cradle head, and marking asP 0 The method comprises the steps of carrying out a first treatment on the surface of the Simultaneously, initializing the focal length of the variable-focus camera, so that a shooting picture is reduced to the minimum and the most objects can be shot in a camera view field;
s3, the region to be shotP 0 Dividing the center into a nine-square grid, which is marked asP 0 ~P 8 Starting the image acquisition main body to take a nine-grid photograph of the region to be photographed;
s4, transmitting the nine-grid pictures obtained in the step S3 to the image processing module, and carrying out picture fusion at the image processing module to obtain a panoramic large picture of the current shooting area of the equipment;
s5, controlling an image processing module to identify a panoramic large image by using a deep learning algorithm, and finding out the panoramic large imageAnd marking the defective rectangular frame with the defective portionr n N is an integer not less than 1;
s6, determining a defect rectangular framer n Location of regions in a planar SudokuP m ；mIs an integer of 0 to less than or equal tom≤8；
S7, controlling the cradle head to rotate to a nine-grid area corresponding to the defective rectangular frame, and calculating the physical distance between the current defective part and the device by a binocular depth camera according to the coordinates of the defective rectangular frame; controlling the variable-focus camera to adjust the focal length and amplifying the image of the defect part;
s8, controlling the cradle head to rotate to the region positions obtained in the step S6 respectivelyP m And according to the defect rectangular framer n The position of the cradle head is adjusted according to the starting point coordinates, the width and the height information, and the defect part is arranged at the center of a shooting picture of the image acquisition main body;
s9, shooting a defect rectangular framer n Storing the image of the corresponding part to obtain a high-definition large image corresponding to the defect part; then, turning to the cradle head to reset to the initial positionP 0 And (S8) repeating the step until the defect images corresponding to all the defect rectangular frames are acquired.
2. The method for capturing an image of a human eye-like apparatus according to claim 1, wherein in step S3, the process of taking a picture of a nine-grid comprises:
s301: setting the rotation speed of the cradle head to be uniform, and rotating the cradle head to an initial positionP 0 The method comprises the steps of carrying out a first treatment on the surface of the After that, shooting is started and stored to obtainP 0 A picture corresponding to the position;
s302: starting the cradle head to rotate, therebyP a The position starts and goes beyond a latticeP a+1 The position of the device is determined by the position,ais an integer of 0 to less than or equal toaAnd (3) obtaining pictures corresponding to the positions of all the areas in the nine-square grid in sequence.
3. The method for collecting an image of a device according to claim 2, wherein the intervals between the two adjacent position areas of the nine Gong Gezhong are equal, and the rotation time of the holder between the two adjacent positions is equal.
4. The method for collecting an image of a human-eye-like device according to any one of claims 1 to 3, wherein in step S5, a process of obtaining a defective rectangular frame is as follows:
s501: collecting data pictures of various defects corresponding to equipment to be monitored;
s502: marking, cleaning and amplifying the data picture according to different defect types;
s503: training the processed sample by using a yolov5 training frame to obtain a high-precision equipment defect recognition model;
s504: the model is packaged and deployed on a computing platform with AI computing power in a micro-service mode, and defect identification service is provided for the outside;
s505: scaling, reducing the average value, normalizing and regularizing the synthesized panoramic big image;
s506: and uploading the preprocessed picture to the defect recognition service, and taking the rectangular frame exceeding a certain threshold value as an alarm rectangular frame to be output.
5. A method for acquiring an image of a human eye-like device according to any one of claims 1 to 3, wherein in step S6, the position of the defective rectangular frame is determinedPmThe process of (2) is as follows:
calculating the center coordinates of the defective rectangular frame according to the following formulas 1 and 2
Calculating the position of the region with the coordinate in the nine-square lattice according to the corresponding relation of the coordinate;
Wherein,,x m min is a square frame of a nine-square latticemIs the minimum of (2)xThe coordinates of the two points of the coordinate system,x m max is a square frame of a nine-square latticemMaximum of (2)xCoordinates;y m min is a square frame of a nine-square latticemIs the minimum of (2)yThe coordinates of the two points of the coordinate system,y m max is a square frame of a nine-square latticemMaximum of (2)yCoordinates.
6. A method for acquiring an image of a device simulating a human eye according to any one of claims 1 to 3, wherein in step S7, the variable-focus camera adjusts a focal length according to formula 3;
Where u is the object distance, v is the image distance, and f is the focal length.
7. The method for collecting an image of a device simulating a human eye according to any one of claims 1 to 3, wherein when the nine-grid shooting of the region to be shot is performed in step S3, the infrared camera is synchronously controlled to shoot the temperature state of each region in the nine-grid.
8. The device image acquisition device for the human-eye-simulated device is characterized by comprising a holder and an image acquisition main body, wherein the device image acquisition is carried out by utilizing the device image acquisition method for the human-eye-simulated device according to any one of claims 1-7;
the image acquisition main body is rotatably connected to the cradle head through a bracket and can perform pitching adjustment relative to the cradle head;
the image acquisition main body comprises an integrally arranged variable-focus camera, an infrared camera and a binocular depth camera, which are respectively used for acquiring high-resolution images under the condition of zooming, acquiring the temperature state of a corresponding area and measuring the distance in the image acquisition process; meanwhile, an image processing module is arranged in the image acquisition main body and is electrically connected with each camera and used for receiving images acquired by the corresponding cameras and synthesizing images for defect identification.
9. The human eye-like device image acquisition apparatus according to claim 8, wherein a moving mechanism is provided corresponding to the pan-tilt; the cradle head is arranged on the moving mechanism in a carrying way, and can change shooting positions under the drive of the moving mechanism.
10. The human eye-like apparatus image capturing device according to claim 8 or 9, wherein a protective cover is provided on top of the image capturing body.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210413048.XA CN114827466B (en) | 2022-04-20 | 2022-04-20 | Human eye-like equipment image acquisition device and image acquisition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210413048.XA CN114827466B (en) | 2022-04-20 | 2022-04-20 | Human eye-like equipment image acquisition device and image acquisition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114827466A CN114827466A (en) | 2022-07-29 |
| CN114827466B true CN114827466B (en) | 2023-07-04 |
Family
ID=82505922
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210413048.XA Active CN114827466B (en) | 2022-04-20 | 2022-04-20 | Human eye-like equipment image acquisition device and image acquisition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114827466B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106809710A (en) * | 2015-11-30 | 2017-06-09 | 福建省特种设备检验研究院泉州分院 | Elevator magnetic buffering protection device |
| CN109979468A (en) * | 2019-03-05 | 2019-07-05 | 武汉三江中电科技有限责任公司 | A kind of lightning stroke optical path monitoring system and method |
| CN111698413A (en) * | 2019-03-13 | 2020-09-22 | 杭州海康威视数字技术股份有限公司 | Object image acquisition method and device and electronic equipment |
| CN111951237A (en) * | 2020-08-04 | 2020-11-17 | 上海微亿智造科技有限公司 | Visual appearance detection method |
| CN112396116A (en) * | 2020-11-24 | 2021-02-23 | 武汉三江中电科技有限责任公司 | Thunder and lightning detection method and device, computer equipment and readable medium |
| CN113702777A (en) * | 2021-07-27 | 2021-11-26 | 广西电网有限责任公司电力科学研究院 | High-voltage equipment on-line inspection monitoring system |
| WO2022000300A1 (en) * | 2020-06-30 | 2022-01-06 | 深圳市大疆创新科技有限公司 | Image processing method, image acquisition apparatus, unmanned aerial vehicle, unmanned aerial vehicle system, and storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011114788A (en) * | 2009-11-30 | 2011-06-09 | Sanyo Electric Co Ltd | Imaging apparatus and imaging method |
| CN101883261B (en) * | 2010-05-26 | 2012-12-12 | 中国科学院自动化研究所 | Method and system for abnormal target detection and relay tracking under large-range monitoring scene |
| CN107040725B (en) * | 2017-05-15 | 2021-04-30 | 惠科股份有限公司 | Coordinate correction method of image acquisition device and image acquisition device |
| CN111614875A (en) * | 2020-05-13 | 2020-09-01 | 深圳供电局有限公司 | Remote photography holder control system for transformer substation inspection robot |
| CN113796064A (en) * | 2020-06-28 | 2021-12-14 | 深圳市大疆创新科技有限公司 | Shooting control method and device, holder and shooting system |
| CN111935412B (en) * | 2020-10-19 | 2021-04-13 | 广东科凯达智能机器人有限公司 | Method, system and robot for automatically identifying and tracking inspection target |
| CN112947567A (en) * | 2021-03-02 | 2021-06-11 | 广州科腾信息技术有限公司 | Method for inspecting power transmission line by using multi-rotor unmanned aerial vehicle |
-
2022
- 2022-04-20 CN CN202210413048.XA patent/CN114827466B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106809710A (en) * | 2015-11-30 | 2017-06-09 | 福建省特种设备检验研究院泉州分院 | Elevator magnetic buffering protection device |
| CN109979468A (en) * | 2019-03-05 | 2019-07-05 | 武汉三江中电科技有限责任公司 | A kind of lightning stroke optical path monitoring system and method |
| CN111698413A (en) * | 2019-03-13 | 2020-09-22 | 杭州海康威视数字技术股份有限公司 | Object image acquisition method and device and electronic equipment |
| WO2022000300A1 (en) * | 2020-06-30 | 2022-01-06 | 深圳市大疆创新科技有限公司 | Image processing method, image acquisition apparatus, unmanned aerial vehicle, unmanned aerial vehicle system, and storage medium |
| CN111951237A (en) * | 2020-08-04 | 2020-11-17 | 上海微亿智造科技有限公司 | Visual appearance detection method |
| CN112396116A (en) * | 2020-11-24 | 2021-02-23 | 武汉三江中电科技有限责任公司 | Thunder and lightning detection method and device, computer equipment and readable medium |
| CN113702777A (en) * | 2021-07-27 | 2021-11-26 | 广西电网有限责任公司电力科学研究院 | High-voltage equipment on-line inspection monitoring system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114827466A (en) | 2022-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112164015B (en) | Monocular vision autonomous inspection image acquisition method and device and power inspection unmanned aerial vehicle | |
| CN105118055B (en) | Camera position amendment scaling method and system | |
| CN110142785A (en) | A kind of crusing robot visual servo method based on target detection | |
| EP3089449B1 (en) | Method for obtaining light-field data using a non-light-field imaging device, corresponding device, computer program product and non-transitory computer-readable carrier medium | |
| CN112949478A (en) | Target detection method based on holder camera | |
| CN110246175A (en) | Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera | |
| CN106643699A (en) | Space positioning device and positioning method in VR (virtual reality) system | |
| CN109739239B (en) | Planning method for uninterrupted instrument recognition of inspection robot | |
| CN111242988B (en) | Method for tracking target by linkage of wide-angle camera and long-focus camera and double cradle head | |
| CN106625673A (en) | Narrow space assembly system and assembly method | |
| CN106403900B (en) | Flying object tracking location system and method | |
| CN106871787A (en) | Large space line scanning imagery method for three-dimensional measurement | |
| CN110146030A (en) | Side slope surface DEFORMATION MONITORING SYSTEM and method based on gridiron pattern notation | |
| CN113850137A (en) | Power transmission line image online monitoring method, system and equipment | |
| CN206460516U (en) | A kind of multichannel fisheye camera binocular calibration device | |
| CN109584309A (en) | A kind of twin-lens emergency cloth ball-handling of rifle ball linkage | |
| CN110022444A (en) | The panorama photographic method of unmanned flight's machine and the unmanned flight's machine for using it | |
| CN206460515U (en) | A kind of multichannel fisheye camera caliberating device based on stereo calibration target | |
| CN111862169B (en) | Target follow-up method and device, cradle head camera and storage medium | |
| CN110059511A (en) | A kind of scan code system and barcode scanning method | |
| CN208653473U (en) | Image capture device, 3D information comparison device, mating object generating means | |
| CN110533728A (en) | Scaling method, device and the medium of binocular stereo vision camera based on Zhang Zhengyou calibration | |
| CN114827466B (en) | Human eye-like equipment image acquisition device and image acquisition method | |
| CN112907647A (en) | Three-dimensional space size measurement method based on fixed monocular camera | |
| CN112764052A (en) | Air defense missile flight monitoring system |
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 |