CN111571598A - Intelligent inspection method of explosion-proof inspection robot - Google Patents
Intelligent inspection method of explosion-proof inspection robot Download PDFInfo
- Publication number
- CN111571598A CN111571598A CN202010474081.4A CN202010474081A CN111571598A CN 111571598 A CN111571598 A CN 111571598A CN 202010474081 A CN202010474081 A CN 202010474081A CN 111571598 A CN111571598 A CN 111571598A
- Authority
- CN
- China
- Prior art keywords
- image
- inspection
- holder
- target object
- target
- 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.)
- Pending
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 187
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000033001 locomotion Effects 0.000 claims abstract description 49
- 230000008569 process Effects 0.000 claims abstract description 13
- 238000001514 detection method Methods 0.000 claims abstract description 11
- 239000013598 vector Substances 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 7
- 230000000007 visual effect Effects 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 2
- 238000011161 development Methods 0.000 description 4
- 230000004888 barrier function Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000009432 framing Methods 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 238000011897 real-time detection Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000009347 mechanical transmission Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
- B25J9/1697—Vision controlled systems
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- Manipulator (AREA)
Abstract
The invention discloses an intelligent inspection method of an explosion-proof inspection robot, which comprises a detection machine body, wherein a holder body control module, an image acquisition module and an industrial personal computer are arranged on the detection machine body, the holder body control module comprises a holder actuator, a holder sensor, a holder driver and a holder controller, and the holder controller comprises: the control system is used for realizing the function of the control holder, communicating with the industrial personal computer, analyzing the control instruction of the industrial personal computer and issuing control instructions such as motion and the like; the image acquisition module acquires and transmits inspection information through the pan-tilt camera; the industrial personal computer processes the acquired information, generates a control command to communicate with the holder controller, and controls the mobile holder to perform cooperative motion so as to complete high-quality image acquisition; the invention ensures that the inspection robot runs more flexibly and smoothly, improves the inspection speed, enhances the image acquisition quality and ensures the safety and efficiency of the inspection work.
Description
Technical Field
The invention relates to the field of inspection, in particular to an intelligent inspection method of an explosion-proof inspection robot.
Background
The development of the power grid in China provides powerful power guarantee for the economic development of China, the operation level of a transformer substation is higher and higher, the requirement on a safe operation environment is stricter and stricter, and the transformer substation serving as a core hub in the power grid can cause irreparable loss to the economic development even if a small safety accident happens once. To ensure the normal operation of the transformer substation, the transformer substation equipment needs to be regularly inspected, and how to improve the efficiency of the inspection work is a topic with great research significance.
The robot that patrols and examines that present development gets up still samples the most basic preset position mode in the aspect of patrolling and examining equipment registration collection, and the robot that patrols and examines moves the fixed point and stops, calls cloud platform preset position, makes the camera of cloud platform load change to fixed position, later carries out the camera and shoots the collection. There are several major problems:
the labor cost of routing inspection is high. The number of stations of the power distribution station is large, and the distribution range is wide, so that inspection personnel need to shuttle in a large range for a long time; and part of the stations are complicated in-out procedures and unsmooth in traffic in part of areas, so that most of the time of inspection personnel is lost in the process of going to the distribution station, and the inspection efficiency is low.
The requirement of professional skills is high. The power distribution station belongs to a high-voltage place, part of inspection projects have the specialty, the requirements for skills and experience of personnel are high, meanwhile, in a traditional inspection mode, the technical personnel are caused to work with low technical skills for a long time, and the waste of human resources is serious.
Disclosure of Invention
The invention aims to provide an intelligent inspection method of an explosion-proof inspection robot aiming at the defects of the prior art.
In order to solve the technical problems, the following technical scheme is adopted:
an intelligent inspection method of an explosion-proof inspection robot comprises a detection machine body, wherein a holder body control module, an image acquisition module and an industrial personal computer are arranged on the detection machine body, the holder body control module comprises a holder actuator, a holder sensor, a holder driver and a holder controller,
the holder actuator: the device is used for executing the pose adjustment of the holder and ensuring the motion precision of the holder;
cloud platform sensor: the device is used for determining the limit angle of the actuator and ensuring the motion range of the holder;
the holder driver: the device is used for driving a tripod head motor to carry out accurate tripod head position positioning;
the cloud platform controller: the control system is used for realizing the function of the control holder, communicating with the industrial personal computer, analyzing the control instruction of the industrial personal computer and issuing control instructions such as motion and the like;
the image acquisition module acquires and transmits inspection information through the pan-tilt camera;
the industrial personal computer processes the acquired information, generates a control command to communicate with the holder controller, and controls the mobile holder to perform cooperative motion so as to complete high-quality image acquisition;
the industrial personal computer controls the cradle head body control module and the image acquisition module to acquire images as follows:
step 1: the explosion-proof inspection robot performs inspection operation, controls the image acquisition module to start scanning, acquires the current frame image and judges whether target information exists or not; if the collected image contains a target image, extracting the position information of the target in the image;
step 2: acquiring the motion information of the mobile platform of the explosion-proof inspection robot, and determining the relative motion of a target image in the visual field of the holder; sending a control instruction to a pan-tilt motor according to the difference and the relative motion information of the target image so that the pan-tilt motor can adjust the angle of a pan-tilt camera until the tracking target is located at the central position of the current frame image;
and step 3: adjusting the magnification and focusing of a lens of the pan-tilt camera to enable the size of a target in an image to meet the requirement of an acquisition standard, acquiring the target image, and receiving a patrol inspection image p (x, y) acquired by the pan-tilt camera; x is more than or equal to 1 and less than or equal to X, Y is more than or equal to 1 and less than or equal to Y, and the pixel quantity of the inspection image is X multiplied by Y; extracting characteristic points representing the edges of the inspection target object, and acquiring coordinates T (x, y) of the characteristic points of the edges of the inspection target object; calculating the position of an image area surrounded by the characteristic points T (x, y) of the edge of the inspection target object in the inspection image collected by the pan-tilt camera to form a new inspection target object image f (x, y);
and 4, step 4: mapping the characteristic points of the image f (x, y) of the inspection target object to a comparison database of the industrial personal computer to obtain
In the formula XA、YAIs mapped to a contrast databaseCoordinate of (2), Xa、YaIs the coordinate m in the image of the inspection target object11...m22To rotate the transformation matrix, h1And h2Converting the matrix for translation so as to obtain a mapped inspection target object image A (x, y);
and 5: matching the mapped inspection target object image A (x, y) with an inspection target object image B (x, y) stored in a comparison database, and firstly determining the shape similarity E of the regional characteristics of the inspection target object imageABThen by inspecting the point similarity R of the regional characteristics of the target imageABObtaining a matching similarity g value, and calculating a threshold value
And G is a matching similarity comparison coefficient of the comparison database, if T is less than or equal to a value T, the inspection target meets the standard, and if not, the inspection target does not meet the standard, wherein the range of T is 0 to 0.15.
Further, the judging step of the shape similarity of the regional characteristics of the image of the routing inspection target object comprises the following steps: assuming that the shape frame of the area feature of the i-th area of the mapped patrol inspection target object image a (x, y) is EA 
EAThe shape frame of the area characteristics of the j area of the inspection target object image B (x, y) stored in the comparison database is EB 
Wherein
And
the coordinates of the lower left corner and the upper right corner of the shape box that is the region feature of the ith region,
and
the coordinates of the lower left corner and the upper right corner of the shape frame of the area characteristic of the jth area, wherein the similarity of the shapes of the area characteristics of the inspection target object image is determined by EABTo express that the expression (A) is,
wherein
Further, the judging step of the point similarity of the regional characteristics of the patrol inspection target object image is as follows: suppose that the region feature vector of the i-th region of the mapped image A (x, y) of the inspection target object is
Comparing the i area characteristic vector of the inspection target object image B (x, y) stored in the database into
And
by R, the cosine similarity of the feature vectors ofABTo express that the expression (A) is,
wherein i is 1,2,3.
Further, according to the shape similarity E of the area characteristics of the image of the inspection target objectABPoint similarity R with regional features of inspection target imageABThe calculation formula for obtaining the matching similarity g, g is as follows:
wherein is nAAnd nBIs the number of feature vectors of the selected region.
Further, in the step (1), the position information of the target in the image includes coordinates of four vertices of a rectangular frame for framing the detection target and a target center coordinate, and a difference between the target center coordinate and the image center coordinate is calculated.
Further, in step (2), determining the relative motion of the target image includes acquiring motion information of the mobile platform, including a motion speed in a horizontal X, Y direction and a rotation speed around a Z axis, and determining the relative motion direction of the target image with respect to the center of the camera view.
Further, in the step (2), the angular direction and the angular velocity of the pan-tilt camera which need to be rotated are determined according to the difference and the relative motion information of the target image.
Further, the two sides of the holder camera are provided with an LED lamp supplementary light source and a demister fog removing lamp.
Due to the adoption of the technical scheme, the method has the following beneficial effects:
the invention relates to an intelligent inspection method of an explosion-proof inspection robot, which is characterized in that when the explosion-proof inspection robot of a transformer substation is used for inspection, an image acquisition module is controlled by a holder body control module to carry out intelligent scanning, the pose state of target identification is automatically adjusted, the interference of external factors such as human and environment can be reduced, and better image acquisition quality can be obtained; the inspection robot does not need to frequently stop and go to perform information acquisition and detection in the image information acquisition process, and the inspection efficiency is improved by performing real-time detection in the operation process. The invention ensures that the inspection robot runs more flexibly and smoothly, improves the inspection speed, enhances the image acquisition quality and ensures the safety and efficiency of the inspection work.
The method breaks through the image information acquisition mode of the preset position of the holder of the original inspection robot, realizes real-time target identification and tracking of the inspection robot in the motion process, acquires high-quality images at better positions, reduces the influence of external factors such as holder self positioning, light intensity and barrier shielding, and ensures the image quality. The method provided by the embodiment of the invention does not need to make the inspection robot shoot stop-and-go, ensures the quality of the acquired image by utilizing the image acquisition standard and quality evaluation, has higher robustness and accuracy, and makes the image acquisition more convenient and flexible.
Drawings
The invention will be further described with reference to the accompanying drawings in which:
FIG. 1 is a schematic structural diagram of an intelligent inspection system of an explosion-proof inspection robot according to the present invention;
fig. 2 is a schematic structural diagram of an explosion-proof inspection robot in the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood, however, that the description herein of specific embodiments is only intended to illustrate the invention and not to limit the scope of the invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.
As shown in fig. 1-2, an explosion-proof inspection robot, including detecting the organism, is equipped with cloud platform body control module, image acquisition module and industrial computer on detecting the organism, and cloud platform body control module includes cloud platform executor, cloud platform sensor, cloud platform driver and cloud platform controller.
The holder actuator: the method is used for executing the pose adjustment of the holder and ensuring the motion precision of the holder.
Specifically, the holder actuator comprises a holder motor, a mechanical transmission mechanism, a limit switch and the like, and is mainly responsible for realizing the command issuing of the holder controller, realizing accurate movement and feeding back movement information; and outputting feedback information such as position information.
Cloud platform sensor: the limiting angle is used for determining the limiting angle of the actuator, and the motion range of the holder is ensured.
The holder driver: the device is used for driving the motor of the holder to carry out accurate position positioning of the holder.
The cloud platform controller: the control device is used for realizing the function of the control holder, communicating with the industrial personal computer, analyzing the control instruction of the industrial personal computer and issuing control instructions such as movement and the like.
Specifically, the pan-tilt controller is responsible for analyzing the control instruction of the industrial personal computer, controlling the movement of a pan-tilt motor and realizing the position and orientation adjustment and positioning of the pan-tilt camera; and reading data backtransmission of the pan-tilt motor, the pan-tilt camera and the like, and performing motion limitation and position determination of the pan-tilt camera.
The image acquisition module acquires and transmits inspection information through the pan-tilt camera;
specifically, the two sides of the holder camera are provided with an LED lamp supplementary light source and a demister defogging lamp.
The industrial personal computer processes the acquired information, generates a control command to communicate with the holder controller, and controls the mobile holder to perform cooperative motion so as to complete high-quality image acquisition;
the industrial personal computer is also used for positioning and navigating the explosion-proof inspection robot; processing the collected image information and sending a control instruction to carry out image registration; the system is responsible for evaluating the quality of the collected image; and storing the image information and transmitting the image information back to the background system.
Specifically, the industrial personal computer controls the cradle head body control module and the image acquisition module to acquire images as follows:
step 1: the explosion-proof inspection robot performs inspection operation, controls the image acquisition module to start scanning, acquires the current frame image and judges whether target information exists or not; if the collected image contains a target image, extracting the position information of the target in the image; the position information of the target in the image comprises coordinates of four vertexes of a rectangular frame for framing the detection target and a target center coordinate, and a difference value between the target center coordinate and the image center coordinate is calculated.
Step 2: acquiring the motion information of the mobile platform of the explosion-proof inspection robot, and determining the relative motion of a target image in the visual field of the holder; determining the relative motion of the target image based on the difference and the target image relative motion information includes acquiring motion information of the mobile platform including a speed of motion in a horizontal X, Y direction and a speed of rotation about the Z-axis, and determining a direction of relative motion of the target image with respect to a center of the field of view of the camera. Finally, sending a control command to a pan-tilt motor so that the pan-tilt motor can adjust the angle of a pan-tilt camera until the tracking target is located at the central position of the current frame image;
specifically, the angular direction and the angular velocity of the pan/tilt camera to be rotated are determined according to the difference and the relative motion information of the target image.
In the step 2, the explosion-proof inspection robot can complete the acquisition of images in the normal movement inspection process without the shooting of stop and go, break through the original acquisition mode of the image information of the preset position of the holder of the inspection robot, realize the real-time target identification and tracking of the inspection robot in the movement process, acquire high-quality images at better positions, reduce the influence of external factors such as holder self positioning, light intensity, barrier shielding and the like, and ensure the image quality.
And step 3: adjusting the magnification and focusing of a lens of the pan-tilt camera to enable the size of a target in an image to meet the requirement of an acquisition standard, acquiring the target image, and receiving a patrol inspection image p (x, y) acquired by the pan-tilt camera; x is more than or equal to 1 and less than or equal to X, Y is more than or equal to 1 and less than or equal to Y, and the pixel quantity of the inspection image is X multiplied by Y; extracting characteristic points representing the edges of the inspection target object, and acquiring coordinates T (x, y) of the characteristic points of the edges of the inspection target object; calculating the position of an image area surrounded by the characteristic points T (x, y) of the edge of the inspection target object in the inspection image collected by the pan-tilt camera to form a new inspection target object image f (x, y);
and 4, step 4: mapping the characteristic points of the image f (x, y) of the inspection target object to a comparison database of the industrial personal computer to obtain
In the formula XA、YAIs a coordinate, X, mapped to a comparison databasea、YaIs the coordinate m in the image of the inspection target object11...m22To rotate the transformation matrix, h1And h2The matrix is translated to obtain a mapped inspection target image a (x, y).
In the step 4, the characteristic points of the inspection target object image f (x, y) are mapped to a comparison database of the industrial personal computer, and a part of background or obstacle images which do not belong to the inspection target object are removed, so that the subsequent image processing difficulty is reduced, and the image acquisition and the robot inspection can be completed better and faster.
And 5: matching the mapped inspection target object image A (x, y) with an inspection target object image B (x, y) stored in a comparison database, and firstly determining the shape similarity E of the regional characteristics of the inspection target object imageABThen by inspecting the point similarity R of the regional characteristics of the target imageABObtaining a matching similarity g value, and calculating a threshold value
And G is a matching similarity comparison coefficient of the comparison database, if T is less than or equal to a value T, the inspection target meets the standard, and if not, the inspection target does not meet the standard, wherein the range of T is 0 to 0.15.
In the step 5, the comparison database stores the inspection target image and the panoramic image of the inspection target, the industrial personal computer calls an inspection target image B (x, y) of a corresponding angle and position through the comparison database according to the mapped inspection target image A (x, y), and the shape similarity E of the area characteristics of the inspection target image is firstly comparedABAnd obtaining the approximate shape and the characteristics of the routing inspection target object. Then comparing the point similarity R of the regional characteristics of the inspection target object imageABObtaining more fine local point characteristics of the inspection target object, and then passing through the threshold value
If the t value is smaller, the acquired image is more in accordance with the standard, and generally, if the t value is less than 0.15, the acquired image is in accordance with the standard.
Specifically, the step of judging the shape similarity of the regional features of the inspection target image is as follows: assuming that the shape frame of the area feature of the i-th area of the mapped patrol inspection target object image a (x, y) is EA 
EAComparing j areas of the inspection target object image B (x, y) stored in the databaseThe shape of the region feature of (1) is framed byB 
Wherein
And
the coordinates of the lower left corner and the upper right corner of the shape box that is the region feature of the ith region,
and
the coordinates of the lower left corner and the upper right corner of the shape frame of the area characteristic of the jth area, wherein the similarity of the shapes of the area characteristics of the inspection target object image is determined by EABTo express that the expression (A) is,
wherein,
i and j are 0,1, 2,3, 4 … … n.
Wherein, in the judgment of the shape similarity of the regional characteristics of the inspection target object image, the calculated EABThe smaller the value is, the higher the similarity degree of the image acquisition module and the inspection target object is, and the more the inspection target object acquired by the image acquisition module meets the standard.
Specifically, the step of judging the point similarity of the regional features of the inspection target image is as follows: suppose that the region feature vector of the i-th region of the mapped image A (x, y) of the inspection target object is
Comparing the inspection target images stored in the databaseThe region feature vector of the i region of B (x, y) is
And
by R, the cosine similarity of the feature vectors ofABTo express that the expression (A) is,
wherein i is 1,2,3.
Wherein, in the judgment of the point similarity of the regional characteristics of the image of the inspection target object, the calculated RABThe closer the value is to 1, the higher the similarity degree between the value and the value is, and further, the more standard the inspection target object acquired by the image acquisition module meets.
Specifically, the shape similarity E according to the region feature of the patrol target imageABPoint similarity R with regional features of inspection target imageABThe calculation formula for obtaining the matching similarity g, g is as follows:
wherein is nAAnd nBFor the number of the selected regional characteristic vectors, the larger the g is, the higher the total similarity degree is, and the more the inspection target object acquired by the image acquisition module meets the standard.
In particular, n is generallyAAnd nBAnd taking a basic integer more than or equal to 25. G is calculated to obtain a similarity matrix of the whole data set, and a formula is calculated
If the t value is smaller, the acquired image is more in accordance with the standard, and generally, if the t value is less than 0.15, the acquired image is in accordance with the standard.
When the explosion-proof inspection robot of the transformer substation is used for inspection, the cloud platform body control module controls the image acquisition module to carry out intelligent scanning, the pose state of target identification is automatically adjusted, and people can be reducedAnd interference of external factors such as environment and the like can be realized, and better image acquisition quality can be obtained. Specifically, motion information of a mobile platform of the explosion-proof inspection robot is obtained firstly, relative motion of a target image in a visual field of a holder is determined, and a holder motor is enabled to adjust the angle of a holder camera until a tracking target is located at the central position of the current frame image; then adjusting the magnification and focusing of a lens of the pan-tilt camera to enable the size of a target in the image to meet the requirement of an acquisition standard, acquiring the target image, and receiving a patrol inspection image p (x, y) acquired by the pan-tilt camera; and then, the characteristic points of the inspection target object image f (x, y) are mapped into a comparison database of the industrial personal computer, and a part of background or obstacle images which do not belong to the inspection target object are removed, so that the subsequent image processing difficulty is reduced, and the image acquisition and the robot inspection can be completed better and faster. The comparison database stores the inspection target object image and is a panoramic image of the inspection target object, and according to the mapped inspection target object image A (x, y), the industrial personal computer calls an inspection target object image B (x, y) of a corresponding angle and position through the comparison database, and compares the shape similarity E of the area characteristics of the inspection target object imageABAnd obtaining the approximate shape and the characteristics of the routing inspection target object. Then comparing the point similarity R of the regional characteristics of the inspection target object imageABObtaining more fine local point characteristics of the inspection target object, and then passing through the threshold value
If the t value is smaller, the acquired image is more in accordance with the standard, and generally, if the t value is less than 0.15, the acquired image is in accordance with the standard. In the inspection mode, generally, an inspection target object can finish the acquisition work only by shooting one image, more importantly, the inspection robot does not need to frequently stop and go to perform information acquisition and detection in the image information acquisition process, the inspection robot performs real-time detection in the operation process, and the inspection efficiency is improved. The invention ensures that the inspection robot runs more flexibly and smoothly, improves the inspection speed, enhances the image acquisition quality and ensures the safety and efficiency of the inspection work.
The inspection method of the explosion-proof inspection robot breaks through the image information acquisition mode of the preset position of the holder of the original inspection robot, realizes real-time target identification and tracking of the inspection robot in the motion process, acquires high-quality images at better positions, reduces the influence of external factors such as holder self positioning, light intensity and barrier shielding, and ensures the image quality. The method provided by the embodiment of the invention does not need to make the inspection robot shoot stop-and-go, ensures the quality of the acquired image by utilizing the image acquisition standard and quality evaluation, has higher robustness and accuracy, and makes the image acquisition more convenient and flexible.
The above is only a specific embodiment of the present invention, but the technical features of the present invention are not limited thereto. Any simple changes, equivalent substitutions or modifications made on the basis of the present invention to solve the same technical problems and achieve the same technical effects are all covered in the protection scope of the present invention.
Claims (8)
1. An intelligent inspection method of an explosion-proof inspection robot comprises a detection machine body, wherein a holder body control module, an image acquisition module and an industrial personal computer are arranged on the detection machine body, the holder body control module comprises a holder actuator, a holder sensor, a holder driver and a holder controller,
the holder actuator: the device is used for executing the pose adjustment of the holder and ensuring the motion precision of the holder;
the holder sensor: the device is used for determining the limit angle of the actuator and ensuring the motion range of the holder;
the holder driver: the device is used for driving a tripod head motor to carry out accurate tripod head position positioning;
the holder controller: the control system is used for realizing the function of the control holder, communicating with the industrial personal computer, analyzing the control instruction of the industrial personal computer and issuing control instructions such as motion and the like;
the image acquisition module acquires and transmits inspection information through a pan-tilt camera;
the industrial personal computer processes the acquired information, generates a control command to communicate with the holder controller, and controls the mobile holder to perform cooperative motion so as to complete high-quality image acquisition;
the industrial personal computer controls the cradle head body control module and the image acquisition module to acquire images as follows:
step 1: the explosion-proof inspection robot performs inspection operation, controls the image acquisition module to start scanning, acquires the current frame image and judges whether target information exists or not; if the collected image contains a target image, extracting the position information of the target in the image;
step 2: acquiring the motion information of the mobile platform of the explosion-proof inspection robot, and determining the relative motion of a target image in the visual field of the holder; sending a control instruction to a pan-tilt motor according to the difference and the relative motion information of the target image so that the pan-tilt motor can adjust the angle of a pan-tilt camera until the tracking target is located at the central position of the current frame image;
and step 3: adjusting the magnification and focusing of a lens of the pan-tilt camera to enable the size of a target in an image to meet the requirement of an acquisition standard, acquiring the target image, and receiving a patrol inspection image p (x, y) acquired by the pan-tilt camera; x is more than or equal to 1 and less than or equal to X, Y is more than or equal to 1 and less than or equal to Y, and the pixel quantity of the inspection image is X multiplied by Y; extracting characteristic points representing the edges of the inspection target object, and acquiring the coordinates of the characteristic points of the edges of the inspection target object as T (x, y); calculating the position of an image area surrounded by the characteristic points T (x, y) of the edge of the inspection target object in the inspection image collected by the pan-tilt camera to form a new inspection target object image f (x, y);
and 4, step 4: mapping the characteristic points of the inspection target object image f (x, y) to a comparison database of an industrial personal computer to obtain
In the formula XA、YAIs a coordinate, X, mapped to a comparison databasea、YaIs the coordinate m in the image of the inspection target object11...m22To rotate the transformation matrix, h1And h2Converting the matrix for translation so as to obtain a mapped inspection target object image A (x, y);
and 5: matching the mapped inspection target object image A (x, y) with an inspection target object image B (x, y) stored in a comparison database, and firstly determining the shape similarity E of the regional characteristics of the inspection target object imageABThen by inspecting the point similarity R of the regional characteristics of the target imageABObtaining a matching similarity g value, and calculating a threshold value
And G is a matching similarity comparison coefficient of the comparison database, if T is less than or equal to a value T, the inspection target meets the standard, and if not, the inspection target does not meet the standard, wherein the range of T is 0 to 0.15.
2. The intelligent inspection method of the explosion-proof inspection robot according to claim 1, wherein: the judging step of the shape similarity of the regional characteristics of the patrol target object image comprises the following steps: assuming that the shape frame of the area feature of the i-th area of the mapped patrol inspection target object image a (x, y) is EA 
EAThe shape frame of the area characteristics of the j area of the inspection target object image B (x, y) stored in the comparison database is
Wherein
And
the coordinates of the lower left corner and the upper right corner of the shape box that is the region feature of the ith region,
and
the coordinates of the lower left corner and the upper right corner of the shape frame of the area characteristic of the jth area are obtained, wherein the similarity of the shapes of the area characteristics of the inspection target object image is determined by EABTo express that the expression (A) is,
wherein
3. The intelligent inspection method of the explosion-proof inspection robot according to claim 2, wherein: the judging step of the point similarity of the regional characteristics of the patrol target object image comprises the following steps: suppose that the region feature vector of the i-th region of the mapped image A (x, y) of the inspection target object is
The i area characteristic vector of the inspection target object image B (x, y) stored in the comparison database is
And
by R, the cosine similarity of the feature vectors ofABTo express that the expression (A) is,
wherein i is 1,2,3.
4. The intelligent inspection method of the explosion-proof inspection robot according to claim 3, wherein: according to the similarity of the shapes of the regional characteristics of the inspection target object imagesProperty EABPoint similarity R with regional features of the inspection target imageABThe calculation formula for obtaining the matching similarity g, g is as follows:
wherein is nAAnd nBIs the number of feature vectors of the selected region.
5. The intelligent inspection method of the explosion-proof inspection robot according to claim 1, wherein: in the step (1), the position information of the target in the image includes coordinates of four vertexes of a rectangular frame of the frame-selected detection target and a target center coordinate, and a difference value between the target center coordinate and the image center coordinate is calculated.
6. The intelligent inspection method of the explosion-proof inspection robot according to claim 1, wherein: in the step (2), the determining of the relative motion of the target image comprises acquiring motion information of the mobile platform, wherein the motion information comprises a motion speed in a horizontal X, Y direction and a rotation speed around a Z axis, and the relative motion direction of the target image relative to the center of the camera visual field is determined.
7. The intelligent inspection method of the explosion-proof inspection robot according to claim 1, wherein: in the step (2), an angular direction and an angular velocity of rotation that the pan/tilt head camera needs to rotate are determined according to the difference and the relative motion information of the target image.
8. The intelligent inspection method of the explosion-proof inspection robot according to claim 1, wherein: and LED lamp supplementary light sources and demister defogging lamps are arranged on two sides of the holder camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010474081.4A CN111571598A (en) | 2020-05-29 | 2020-05-29 | Intelligent inspection method of explosion-proof inspection robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010474081.4A CN111571598A (en) | 2020-05-29 | 2020-05-29 | Intelligent inspection method of explosion-proof inspection robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111571598A true CN111571598A (en) | 2020-08-25 |
Family
ID=72109720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010474081.4A Pending CN111571598A (en) | 2020-05-29 | 2020-05-29 | Intelligent inspection method of explosion-proof inspection robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111571598A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112543272A (en) * | 2020-12-07 | 2021-03-23 | 杭州申昊科技股份有限公司 | Transformer substation inspection camera device with light regulation function and method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1926575A (en) * | 2004-03-03 | 2007-03-07 | 日本电气株式会社 | Image similarity calculation system, image search system, image similarity calculation method, and image similarity calculation program |
| CN109664301A (en) * | 2019-01-17 | 2019-04-23 | 中国石油大学(北京) | Method for inspecting, device, equipment and computer readable storage medium |
| CN109773808A (en) * | 2019-03-20 | 2019-05-21 | 杭州申昊科技股份有限公司 | A kind of crusing robot |
| KR20190126607A (en) * | 2018-05-02 | 2019-11-12 | 주식회사 마로로봇 테크 | Autonomous driving logistics robot equipped with multiple cameras for QR code recognition |
| CN209699082U (en) * | 2018-10-12 | 2019-11-29 | 国机智能(苏州)有限公司 | A kind of intelligent inspection robot |
| CN110695958A (en) * | 2019-09-09 | 2020-01-17 | 上海朗驰佰特智能技术有限公司 | Be applicable to explosion-proof robot of patrolling and examining of chemical industry |
| CN111161446A (en) * | 2020-01-10 | 2020-05-15 | 浙江大学 | Image acquisition method of inspection robot |
-
2020
- 2020-05-29 CN CN202010474081.4A patent/CN111571598A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1926575A (en) * | 2004-03-03 | 2007-03-07 | 日本电气株式会社 | Image similarity calculation system, image search system, image similarity calculation method, and image similarity calculation program |
| KR20190126607A (en) * | 2018-05-02 | 2019-11-12 | 주식회사 마로로봇 테크 | Autonomous driving logistics robot equipped with multiple cameras for QR code recognition |
| CN209699082U (en) * | 2018-10-12 | 2019-11-29 | 国机智能(苏州)有限公司 | A kind of intelligent inspection robot |
| CN109664301A (en) * | 2019-01-17 | 2019-04-23 | 中国石油大学(北京) | Method for inspecting, device, equipment and computer readable storage medium |
| CN109773808A (en) * | 2019-03-20 | 2019-05-21 | 杭州申昊科技股份有限公司 | A kind of crusing robot |
| CN110695958A (en) * | 2019-09-09 | 2020-01-17 | 上海朗驰佰特智能技术有限公司 | Be applicable to explosion-proof robot of patrolling and examining of chemical industry |
| CN111161446A (en) * | 2020-01-10 | 2020-05-15 | 浙江大学 | Image acquisition method of inspection robot |
Non-Patent Citations (4)
| Title |
|---|
| 周咏梅等: "一种基于区域的图像相似性计算方法", 《一种基于区域的图像相似性计算方法》 * |
| 栾悉道等著: "《多媒体情报处理技术》", 30 May 2016, 国防工业出版社 * |
| 潘锡英等: "基于图像感兴趣区域的机器人闭环检测算法", 《基于图像感兴趣区域的机器人闭环检测算法》 * |
| 熊有伦等编著: "《机器人学：建模、控制与视觉》", 31 March 2018, 华中科技大学出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112543272A (en) * | 2020-12-07 | 2021-03-23 | 杭州申昊科技股份有限公司 | Transformer substation inspection camera device with light regulation function and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111161446B (en) | Image acquisition method of inspection robot | |
| CN110614638B (en) | Transformer substation inspection robot autonomous acquisition method and system | |
| CN111958592B (en) | Image semantic analysis system and method for transformer substation inspection robot | |
| CA2950791C (en) | Binocular visual navigation system and method based on power robot | |
| CN106680290B (en) | Multifunctional detection vehicle in narrow space | |
| CN109270534A (en) | A kind of intelligent vehicle laser sensor and camera online calibration method | |
| CN110246175A (en) | Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera | |
| CN109532522A (en) | A kind of unmanned charging system of automobile based on 3D vision technique and its application method | |
| CN102608998A (en) | Vision guiding AGV (Automatic Guided Vehicle) system and method of embedded system | |
| CN114905512B (en) | Panoramic tracking and obstacle avoidance method and system for intelligent inspection robot | |
| CN113177918B (en) | Intelligent and accurate inspection method and system for electric power tower by unmanned aerial vehicle | |
| CN113900436B (en) | Inspection control method, inspection control device, inspection control equipment and storage medium | |
| CN112819943A (en) | Active vision SLAM system based on panoramic camera | |
| CN113031462A (en) | Port machine inspection route planning system and method for unmanned aerial vehicle | |
| CN111571598A (en) | Intelligent inspection method of explosion-proof inspection robot | |
| CN106444774B (en) | Vision navigation method of mobile robot based on indoor illumination | |
| CN113743286A (en) | Target monitoring system and method for multi-source signal fusion | |
| CN115454138B (en) | Construction violation determination method and system based on unmanned aerial vehicle image recognition technology | |
| CN112504263A (en) | Indoor navigation positioning device based on multi-view vision and positioning method thereof | |
| CN117110214A (en) | Water quality analysis system and method based on hyperspectral imaging of unmanned aerial vehicle | |
| CN112797893A (en) | Method for measuring position parameters of long-distance cable | |
| WO2024035918A1 (en) | Autonomous solar installation using artificial intelligence | |
| CN114998444B (en) | Robot high-precision pose measurement system based on two-channel network | |
| CN108227689A (en) | A kind of design method of Agriculture Mobile Robot independent navigation | |
| CN116242319A (en) | High-precision binocular vision measurement method and device for large-range moving object |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200825 |
|
| RJ01 | Rejection of invention patent application after publication |