CN114078234B - Detection method, system, storage medium and equipment for power supply area construction process - Google Patents
Detection method, system, storage medium and equipment for power supply area construction process Download PDFInfo
- Publication number
- CN114078234B CN114078234B CN202210012476.1A CN202210012476A CN114078234B CN 114078234 B CN114078234 B CN 114078234B CN 202210012476 A CN202210012476 A CN 202210012476A CN 114078234 B CN114078234 B CN 114078234B
- Authority
- CN
- China
- Prior art keywords
- identification object
- target
- type
- images
- objects
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a detection method, a detection system, a storage medium and a detection device for a power supply area construction process, wherein the method comprises the following steps: inputting each frame of image to be recognized into the trained semantic segmentation model, and extracting masks of target recognition objects of all the images to be recognized through the semantic segmentation model; extracting the outlines of the masks of the target identification objects of all the images to be identified to obtain the outlines of the target identification objects in all the images to be identified, and calculating the areas of the connected regions in the outlines of the target identification objects in all the images to be identified; screening out all effective regions of the target identification object according to the area of a connected region in the outline of the target identification object in all the images to be identified; and analyzing all effective areas of the target recognition object to judge whether the target recognition object is in a standard process state. The invention can realize the automatic detection of the construction process of the power supply area, improve the detection efficiency and obtain more accurate detection results.
Description
Technical Field
The invention relates to the field of construction process detection, in particular to a detection method, a detection system, a storage medium and detection equipment for a power supply platform area construction process.
Background
In an electric power system, a power supply transformer area is used as an important link of urban and rural power distribution and has an irreplaceable effect on promoting regional development. If the construction process of the power supply area does not reach the standard, frequent rush repair and transformation can be caused, the operation and maintenance cost in the later period is increased, and the normal production and life of residents are affected. Therefore, the power supply area needs to ensure that the orientation of the related equipment meets the process standard in construction and later maintenance. For example: the pole towers can not be inclined, the cross arms on the pole towers need to be horizontally placed, the transformer between the two pole towers and the ring main unit need to be placed in the middle, and the like.
At present, the detection of the power supply area construction process mainly depends on field checking by professionals and subjectively evaluates whether the process meets the standard or not according to the checking result, the traditional detection mode consumes manpower and has low efficiency, and the evaluation result is not accurate enough due to lack of an accurate analysis process of related processes. Therefore, a method for automatically detecting the power supply area construction process is needed to solve these problems.
Disclosure of Invention
The invention aims to provide a detection method, a detection system, a storage medium and a detection device for a power supply station area construction process, which are used for solving the problems that the existing detection method for the power supply station area construction process mainly depends on professional personnel to check on site and subjectively evaluates whether the process meets the standard according to the check result, the traditional detection mode consumes manpower and is low in efficiency, and the evaluation result is not accurate enough due to lack of an accurate analysis process for related processes.
The invention provides a detection method of a power supply area construction process, which comprises the following steps:
collecting a large number of monitoring images of a target platform area, and setting the monitoring images as images to be identified;
inputting each frame of image to be recognized into a trained semantic segmentation model, and extracting masks of target recognition objects of all the images to be recognized through the semantic segmentation model;
extracting the outlines of the masks of the target identification objects of all the images to be identified to obtain the outlines of the target identification objects in all the images to be identified, and calculating the areas of the connected regions in the outlines of the target identification objects in all the images to be identified;
screening out all effective regions of the target identification object according to the areas of the connected regions in the outline of the target identification object in all the images to be identified;
analyzing all effective areas of the target recognition object to judge whether the target recognition object is in a standard process state.
The detection method for the power supply area construction process provided by the invention has the following beneficial effects:
the method comprises the steps of collecting a large number of monitoring images of a target area, and setting the monitoring images as images to be identified; inputting each frame of image to be recognized into a trained semantic segmentation model, and extracting masks of target recognition objects of all the images to be recognized through the semantic segmentation model; extracting the outlines of the masks of the target identification objects of all the images to be identified to obtain the outlines of the target identification objects in all the images to be identified, and calculating the areas of the connected regions in the outlines of the target identification objects in all the images to be identified; screening out all effective regions of the target identification object according to the areas of the connected regions in the outline of the target identification object in all the images to be identified; analyzing all effective areas of the target identification object to judge whether the target identification object is in a standard process state or not, realizing automatic detection of a power supply platform area construction process, improving detection efficiency, and causing the outlines of some extracted masks of the target identification object to not truly reflect the outlines of the target identification object under the conditions that power equipment is influenced by illumination change, angle change, partial shielding, deformation, blurring, background interference and the like, and the invention acquires a large number of images to be identified of the target platform area, extracts the masks of the target identification objects in all the images to be identified, extracts the outlines of the target identification objects in all the images to be identified, screens out the effective areas according to the areas of communicated areas in the outlines to more closely screen out the real areas of the target identification object, and the contour comprising the effective area is closer to the real contour of the target identification object, all effective areas of the target identification object are analyzed to judge whether the target identification object is in a standard process state, and the process state of the target identification object can be more accurately judged by analyzing the effective area close to the real area of the target identification object so as to obtain a more accurate detection result.
In addition, the detection method for the power supply area construction process provided by the invention can also have the following additional technical characteristics:
further, the target recognition object comprises a first class recognition object, and the step of analyzing all effective areas of the target recognition object to determine whether the target recognition object is in a standard process state comprises:
calculating first moments of all effective areas of the first type of recognition objects to obtain the gravity center positions of the first type of recognition objects;
calculating second moments for all effective areas of the first type of recognition objects to obtain the direction of the first type of recognition objects;
obtaining a straight line representing the direction and the center line of the first type identification object by using a point-oblique equation according to the gravity center position and the direction of the first type identification object, and setting the straight line as the straight line of the first type identification object;
fitting minimum circumscribed circles of the outlines of all effective areas of the first-class identification objects, and setting two intersection points of the minimum circumscribed circles and straight lines of the corresponding first-class identification objects as two end points of the first-class identification objects;
determining a line segment of the first-class identification object, which is obtained by intercepting two end points of the first-class identification object on a straight line of the first-class identification object;
merging the line segments of the first type identification objects between the different images to be identified according to the line segments of the first type identification objects and the positions of the two end points to obtain a target line segment of the first type identification object;
and judging whether the angle of the first type of identification object is in a standard process state or not according to the target line segment of the first type of identification object and the reference line.
Further, the step of judging whether the angle of the first type identification object is in the standard process state according to the target line segment of the first type identification object and the reference line further includes:
calculating first moments of all effective areas of the second type of recognition objects to obtain the gravity center positions of the second type of recognition objects;
and if the angle of the first class identification object is in a standard process state, judging whether the position of the second class identification object is in the standard process state or not according to the gravity center position of the second class identification object by taking the target line segment of the first class identification object as a reference.
Further, the first type of identification object includes a tower, the reference line is a horizontal line, and the step of screening out all effective regions of the target identification object according to the areas of connected regions in the contour of the target identification object in all the images to be identified includes:
if a single frame of image to be identified comprises more than two towers and the number of the connected regions of the contour of the tower, the area of which is larger than a first preset area threshold value, is more than two, determining the two connected regions with the largest area as effective regions of the towers;
if the number of the connected areas of the contour of the tower, with the area larger than a first preset area threshold value, in a single frame of the image to be identified is not more than two, determining the connected areas with the area larger than the first preset area threshold value as the effective areas of the tower;
the step of judging whether the angle of the first type identification object is in a standard process state according to the target line segment of the first type identification object and the reference line comprises the following steps:
and judging whether the tower is in a vertical state or not according to the included angle between the target line segment of the tower and the horizontal line.
Further, the first type of recognition object further includes a cross arm, and the step of screening out all effective regions of the target recognition object according to the areas of connected regions in the contour of the target recognition object in all the images to be recognized includes:
screening all connected regions with the area larger than a first preset area threshold value from the connected regions in the outlines of the cross arms of all the images to be recognized, and determining the connected regions as effective regions of the cross arms;
the step of judging whether the angle of the first type identification object is in a standard process state according to the target line segment of the first type identification object and the reference line comprises the following steps:
and judging whether the cross arm is in a horizontal state or not according to the included angle between the target line segment of the cross arm and the horizontal line.
Further, the second type recognition object includes a transformer, and the step of determining whether the position of the second type recognition object is in the standard process state or not based on the barycentric position of the second type recognition object with reference to the target line segment of the first type recognition object when the angle of the first type recognition object is in the standard process state includes:
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the transformer and the cross arms closest to the transformer according to the target line segments of the two towers closest to the transformer and the target line segments of the cross arms closest to the transformer, and setting the positions as the intersection points of the cross arms of the towers;
judging whether the transformer is centered or not according to the distance from the center of gravity of the transformer to the intersection point of the cross arms of the two towers;
and if the ratio of the gravity center of the transformer to the distance between the intersection points of the two tower cross arms is within a first preset range, determining that the transformer is centered.
Further, the second type of identification object further includes a ring main unit, and the step of judging whether the position of the second type of identification object is in the standard process state or not by taking the target line segment of the first type of identification object as a reference and according to the gravity center position of the second type of identification object if the angle of the first type of identification object is in the standard process state further includes:
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the ring main unit and the cross arms closest to the ring main unit according to the target line segments of the two towers closest to the ring main unit and the target line segments of the cross arms closest to the ring main unit, and setting the intersection points as tower cross arm intersection points;
judging whether the ring main unit is centered or not according to the distance between the gravity center of the ring main unit and the intersection point of the two pole tower cross arms;
and if the ratio of the gravity center of the ring main unit to the distance between the intersection points of the two tower cross arms is within a second preset range, the ring main unit is determined to be centered.
The invention also provides a detection system for the power supply area construction process, which comprises the following steps:
an acquisition module: the system comprises a monitoring system, a monitoring system and a recognition system, wherein the monitoring system is used for collecting a large number of monitoring images of a target platform area and setting the monitoring images as images to be recognized;
an input module: the system comprises a semantic segmentation model, a target recognition object extraction model and a target recognition object extraction model, wherein the semantic segmentation model is used for extracting masks of target recognition objects of images to be recognized;
a calculation module: the device comprises a mask, a detection unit, a processing unit and a display unit, wherein the mask is used for extracting the outline of a target identification object of all images to be identified so as to obtain the outline of the target identification object in all the images to be identified, and calculating the area of a communication region in the outline of the target identification object in all the images to be identified;
a screening module: screening out all effective regions of the target identification object according to the areas of the connected regions in the outline of the target identification object in all the images to be identified;
a judging module: the system is used for analyzing all effective areas of the target recognition object to judge whether the target recognition object is in a standard process state.
The invention also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor implements the above-mentioned method for detecting a power supply area construction process.
The invention also provides a detection device for the power supply station area construction process, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the detection method for the power supply station area construction process when executing the program.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of a detection method of a power supply area construction process according to a first embodiment of the present invention;
FIG. 2 is a system diagram of a detection system for a power supply area construction process according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a detection device of a power supply area construction process according to a third embodiment of the present invention.
Detailed Description
In order to make the objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. Several embodiments of the invention are presented in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Example 1
As shown in FIG. 1, an embodiment of the invention provides a method for detecting a power supply area construction process, which includes steps S101-S105.
S101, collecting a large number of monitoring images of a target platform area, and setting the monitoring images as images to be identified.
S102, inputting each frame of image to be recognized into the trained semantic segmentation model, and extracting the masks of the target recognition objects of all the images to be recognized through the semantic segmentation model.
S103, extracting the contours of the masks of the target identification objects of all the images to be identified to obtain the contours of the target identification objects in all the images to be identified, and calculating the areas of the connected regions in the contours of the target identification objects in all the images to be identified.
S104, screening out all effective regions of the target identification object according to the areas of the connected regions in the outline of the target identification object in the image to be identified.
S105, analyzing all effective areas of the target recognition object to judge whether the target recognition object is in a standard process state.
The step of analyzing all effective areas of the target identification object to judge whether the target identification object is in a standard process state comprises the following steps:
calculating first moments of all effective areas of the first type of recognition objects to obtain the gravity center positions of the first type of recognition objects;
calculating second moments for all effective areas of the first type of recognition objects to obtain the direction of the first type of recognition objects;
obtaining a straight line representing the direction and the center line of the first type identification object by using a point-oblique equation according to the gravity center position and the direction of the first type identification object, and setting the straight line as the straight line of the first type identification object;
fitting minimum circumscribed circles of the outlines of all effective areas of the first-class identification objects, and setting two intersection points of the minimum circumscribed circles and straight lines of the corresponding first-class identification objects as two end points of the first-class identification objects;
determining a line segment of the first-class identification object, which is obtained by intercepting two end points of the first-class identification object on a straight line of the first-class identification object;
merging the line segments of the first type identification objects between the different images to be identified according to the line segments of the first type identification objects and the positions of the two end points to obtain a target line segment of the first type identification object;
and judging whether the angle of the first type of identification object is in a standard process state according to the target line segment of the first type of identification object and the reference line, namely judging whether the first type of identification object is in a vertical or horizontal state.
The step of merging the line segments of the first-class identification objects between the different images to be identified according to the line segments of the first-class identification objects and the positions of the two end points to obtain the target line segment of the first-class identification object comprises the following steps:
extracting any two line segments of the first type identification object, and setting the line segments as an ab line segment and a cd line segment respectively;
respectively calculating the distances from the end point a and the end point b on the ab line segment to the cd line segment, and judging whether the distances from the end point a and the end point b to the cd line segment are both smaller than a preset pixel distance;
if so, merging the ab line segment and the cd line segment, taking two points with the farthest distances among four points of an end point a, an end point b, an end point c and an end point d as two end points of the merged line segment of the ab line segment and the cd line segment, and determining a connecting line of the two points with the farthest distances among four points of the end point a, the end point b, the end point c and the end point d as the merged line segment of the ab line segment and the cd line segment, wherein the preset pixel distance can be 20 pixel distances;
and sequentially merging all the line segments of the first type of identification object, merging all the merged line segments generated by merging to obtain a target line segment of the first type of identification object.
The first kind of recognition object in the embodiment of the present invention may be a rod-shaped device, and the embodiment of the present invention may calculate the first moment (i.e. obtaining the expected value) for all the effective areas of the target recognition object to obtain a more accurate barycentric position of the target recognition object, then calculate the second moment (i.e. obtaining the variance) for all the effective areas to determine the direction of the target recognition object, determine the straight line of the target recognition object by combining the barycentric position and the direction of the target recognition object and using a point-oblique equation, then take the minimum circumcircle of the contour, the intersection point of the minimum circumcircle and the straight line is the two end points of the target recognition object, the line segment between the two end points is the corresponding line segment of the first kind recognition object in each frame of image to be recognized, merge the line segments of the first kind recognition object in all the images to be recognized to obtain the target line segment of the first kind recognition object, the obtained target line segment of the first type of identification object can more accurately reflect the real line segment of the first type of identification object, and the method can well solve the problem that the identification object in the image is not clear caused by environment so as to ensure the reality and the accuracy of the extraction of the line segment of the identification object.
Further, the step of judging whether the angle of the first type identification object is in the standard process state according to the target line segment of the first type identification object and the reference line further includes:
calculating first moments of all effective areas of the second type of recognition objects to obtain the gravity center positions of the second type of recognition objects;
and if the angle of the first class of identification object is in a standard process state, judging whether the position of the second class of identification object is in the standard process state or not by taking the target line segment of the first class of identification object as a reference and according to the gravity center position of the second class of identification object.
In the above step, after the position of the center of gravity of the second type recognition object is obtained, it can be determined whether the position of the second type recognition object is in a standard process state, for example, centered, based on the position of the center of gravity of the second type recognition object with reference to the target line segment of the first type recognition object.
Further, the first type of identification object includes a tower, the reference line is a horizontal line, and the step of screening out all effective regions of the target identification object according to the areas of connected regions in the contour of the target identification object in all the images to be identified includes:
if a single frame of image to be identified comprises more than two towers and the number of the connected regions of the contour of the tower, the area of which is larger than a first preset area threshold value, is more than two, determining the two connected regions with the largest area as effective regions of the towers;
if the number of the connected areas of the contour of the tower, with the area larger than a first preset area threshold value, in a single frame of the image to be identified is not more than two, determining the connected areas with the area larger than the first preset area threshold value as the effective areas of the tower;
the step of judging whether the angle of the first type identification object is in a standard process state according to the target line segment of the first type identification object and the reference line comprises the following steps:
judging whether the tower is in a vertical state or not according to the included angle between the target line segment of the tower and the horizontal line;
and if the included angle between the target line segment of the tower and the horizontal line is 90 degrees, judging that the tower is in a vertical state.
Further, the first type of recognition object further includes a cross arm, and the step of screening out all effective regions of the target recognition object according to the areas of connected regions in the contour of the target recognition object in all the images to be recognized includes:
screening all connected regions with the area larger than a first preset area threshold value from the connected regions in the outlines of the cross arms of all the images to be recognized, and determining the connected regions as effective regions of the cross arms;
the step of judging whether the angle of the first type identification object is in a standard process state according to the target line segment of the first type identification object and the reference line comprises the following steps:
judging whether the cross arm is in a horizontal state or not according to the included angle between the target line segment of the cross arm and the horizontal line;
and if the included angle between the target line segment of the cross arm and the horizontal line is 0 degree, judging that the cross arm is in a horizontal state.
Further, the step of determining whether the position of the second type of recognition object is in the standard process state according to the barycentric position of the second type of recognition object with the target line segment of the first type of recognition object as a reference if the angle of the first type of recognition object is in the standard process state includes:
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the transformer and the cross arms closest to the transformer according to the target line segments of the two towers closest to the transformer and the target line segments of the cross arms closest to the transformer, and setting the positions as the intersection points of the cross arms of the towers;
judging whether the transformer is centered or not according to the distance from the center of gravity of the transformer to the intersection point of the cross arms of the two towers;
and if the ratio of the gravity center of the transformer to the distance between the intersection points of the two tower cross arms is within a first preset range, determining that the transformer is centered.
The first preset range is used for limiting the distance between the center of gravity of the transformer and the intersection point of the two tower cross arms to be approximately equal, for example, between 0.95 and 1.05, so as to ensure that the transformer is in a centered position and allow a certain construction allowance.
Further, the second type of identification object further includes a ring main unit, and the step of judging whether the position of the second type of identification object is in the standard process state or not by taking the target line segment of the first type of identification object as a reference and according to the gravity center position of the second type of identification object if the angle of the first type of identification object is in the standard process state further includes:
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the ring main unit and the cross arms closest to the ring main unit according to the target line segments of the two towers closest to the ring main unit and the target line segments of the cross arms closest to the ring main unit, and setting the intersection points as tower cross arm intersection points;
judging whether the ring main unit is centered or not according to the distance between the gravity center of the ring main unit and the intersection point of the two pole tower cross arms;
and if the ratio of the gravity center of the ring main unit to the distance between the intersection points of the two tower cross arms is within a second preset range, the ring main unit is determined to be centered.
The second preset range is used for limiting the distance between the center of gravity of the ring main unit and the intersection point of the two tower cross arms to be approximately equal, for example, 0.95-1.05, so that the ring main unit is ensured to be in a centered position, and a certain construction allowance is allowed.
And inputting each frame of image to be recognized into the trained semantic segmentation model, and extracting masks of all towers, cross arms, transformers and ring main units of the image to be recognized through the semantic segmentation model.
Taking a tower as an example, the specific implementation mode is as follows:
and extracting the outlines of the masks of the towers of all the images to be identified to obtain the outlines of the towers in all the images to be identified, and calculating the areas of the communication areas in the outlines of the towers in all the images to be identified.
And screening out all effective regions of the tower according to the areas of the connected regions in the profile of the tower in all the images to be identified.
Calculating a first moment of all effective areas of the tower to obtain the gravity center position of the tower;
calculating a second moment for all effective areas of the tower to obtain the direction of the tower;
obtaining a straight line representing the direction and the center line of the tower by using a point-oblique equation according to the position and the direction of the center of gravity of the tower, and setting the straight line as the straight line of the tower;
fitting minimum circumscribed circles of the outlines of all effective areas of the tower, and determining two intersection points of the minimum circumscribed circles and the corresponding straight lines of the tower as upper and lower end points of the tower;
determining line segments of two end points of the tower, which are intercepted on a straight line of the tower, as line segments of the tower;
merging the line segments of the first type identification objects between the different images to be identified according to the line segments of the first type identification objects and the positions of the two end points to obtain a target line segment of the tower;
judging whether the angle of the tower is in a standard process state according to the target line segment of the tower and the reference line, namely judging whether the tower is in a vertical state;
and if the included angle between the target line segment of the tower and the horizontal line is 90 degrees, judging that the tower is in a vertical state.
Taking a transformer as an example, the specific implementation mode is as follows:
and extracting the contours of the masks of the transformers of all the images to be identified to obtain the contours of the transformers in all the images to be identified, and calculating the areas of the communication regions in the contours of the transformers in all the images to be identified.
And screening out all effective regions of the transformer according to the areas of the communication regions in the outline of the transformer in all the images to be identified.
Calculating a first moment for all effective areas of the transformer to obtain the gravity center position of the transformer;
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the transformer and the cross arms closest to the transformer according to the target line segments of the two towers closest to the transformer and the target line segments of the cross arms closest to the transformer, and setting the positions as the intersection points of the cross arms of the towers;
judging whether the transformer is centered or not according to the distance from the center of gravity of the transformer to the intersection point of the cross arms of the two towers;
and if the ratio of the gravity center of the transformer to the distance between the intersection points of the two tower cross arms is within a first preset range, determining that the transformer is centered.
In summary, the detection method for the power supply area construction process provided by the invention has the beneficial effects that: the method comprises the steps of collecting a large number of monitoring images of a target area, and setting the monitoring images as images to be identified; inputting each frame of image to be recognized into a trained semantic segmentation model, and extracting masks of target recognition objects of all the images to be recognized through the semantic segmentation model; extracting the outlines of the masks of the target identification objects of all the images to be identified to obtain the outlines of the target identification objects in all the images to be identified, and calculating the areas of the connected regions in the outlines of the target identification objects in all the images to be identified; screening out all effective regions of the target identification object according to the areas of the connected regions in the contour of the target identification object in all the images to be identified; analyzing all effective areas of the target identification object to judge whether the target identification object is in a standard process state or not, realizing automatic detection of a power supply platform area construction process, improving detection efficiency, and causing the outlines of some extracted masks of the target identification object to not truly reflect the outlines of the target identification object under the conditions that power equipment is influenced by illumination change, angle change, partial shielding, deformation, blurring, background interference and the like, and the invention acquires a large number of images to be identified of the target platform area, extracts the masks of the target identification objects in all the images to be identified, extracts the outlines of the target identification objects in all the images to be identified, screens out the effective areas according to the areas of communicated areas in the outlines to more closely screen out the real areas of the target identification object, and the contour containing the effective area is closer to the real contour of the target identification object, all effective areas of the target identification object are analyzed to judge whether the target identification object is in a standard process state, and the process state of the target identification object can be more accurately judged by analyzing the effective area close to the real area of the target identification object so as to obtain a more accurate detection result.
Example 2
Referring to fig. 2, the present embodiment provides a system for detecting a power supply area construction process, including:
an acquisition module: the system comprises a monitoring system, a monitoring system and a recognition system, wherein the monitoring system is used for collecting a large number of monitoring images of a target platform area and setting the monitoring images as images to be recognized;
an input module: the system comprises a semantic segmentation model, a target recognition object extraction model and a target recognition object extraction model, wherein the semantic segmentation model is used for extracting masks of target recognition objects of images to be recognized;
a calculation module: the device comprises a mask, a detection unit, a processing unit and a display unit, wherein the mask is used for extracting the outline of a target identification object of all images to be identified so as to obtain the outline of the target identification object in all the images to be identified, and calculating the area of a communication region in the outline of the target identification object in all the images to be identified;
a screening module: screening out all effective regions of the target identification object according to the areas of the connected regions in the outline of the target identification object in all the images to be identified;
a judging module: the system is used for analyzing all effective areas of the target recognition object to judge whether the target recognition object is in a standard process state.
Wherein the target identification object comprises a first type identification object, and the determining module is further configured to:
calculating first moments of all effective areas of the first type of recognition objects to obtain the gravity center positions of the first type of recognition objects;
calculating second moments for all effective areas of the first type of recognition objects to obtain the direction of the first type of recognition objects;
obtaining a straight line representing the direction and the center line of the first type identification object by using a point-oblique equation according to the gravity center position and the direction of the first type identification object, and setting the straight line as the straight line of the first type identification object;
fitting minimum circumscribed circles of the outlines of all effective areas of the first-class identification objects, and setting two intersection points of the minimum circumscribed circles and straight lines of the corresponding first-class identification objects as two end points of the first-class identification objects;
determining a line segment of the first-class identification object, which is obtained by intercepting two end points of the first-class identification object on a straight line of the first-class identification object;
merging the line segments of the first type identification objects between the different images to be identified according to the line segments of the first type identification objects and the positions of the two end points to obtain a target line segment of the first type identification object;
and judging whether the angle of the first type of identification object is in a standard process state according to the target line segment of the first type of identification object and the reference line, namely judging whether the first type of identification object is in a vertical or horizontal state.
The judging module is further configured to:
extracting any two line segments of the first type identification object, and setting the line segments as an ab line segment and a cd line segment respectively;
respectively calculating the distances from the end point a and the end point b on the ab line segment to the cd line segment, and judging whether the distances from the end point a and the end point b to the cd line segment are both smaller than a preset pixel distance;
if so, merging the ab line segment and the cd line segment, taking two points with the farthest distances from four points of an endpoint a, an endpoint b, an endpoint c and an endpoint d as two endpoints of the merged line segment of the ab line segment and the cd line segment, and determining a connecting line of the two points with the farthest distances from four points of the endpoint a, the endpoint b, the endpoint c and the endpoint d as the merged line segment of the ab line segment and the cd line segment, wherein the preset pixel distance can be 20 pixels;
and sequentially merging all the line segments of the first type of identification object, merging all the merged line segments generated by merging to obtain a target line segment of the first type of identification object.
Further, the determining module is further configured to:
calculating first moments of all effective areas of the second type of recognition objects to obtain the gravity center positions of the second type of recognition objects;
and if the angle of the first class of identification object is in a standard process state, judging whether the position of the second class of identification object is in the standard process state or not by taking the target line segment of the first class of identification object as a reference and according to the gravity center position of the second class of identification object.
In the above step, after obtaining the center of gravity position of the second type recognition object, it can be determined whether the position of the second type recognition object is in a standard process state, for example, centered, by using the target line segment of the first type recognition object as a reference and according to the center of gravity position of the second type recognition object.
Further, the first type of identification object includes a tower, and the determining module is further configured to:
if a single frame of image to be identified comprises more than two towers and the number of the connected regions of the contour of the tower, the area of which is larger than a first preset area threshold value, is more than two, determining the two connected regions with the largest area as effective regions of the towers;
if the number of the connected areas of the contour of the tower, with the area larger than a first preset area threshold value, in a single frame of the image to be identified is not more than two, determining the connected areas with the area larger than the first preset area threshold value as the effective areas of the tower;
the step of judging whether the angle of the first type identification object is in a standard process state according to the target line segment of the first type identification object and the reference line comprises the following steps:
judging whether the tower is in a vertical state or not according to the included angle between the target line segment of the tower and the horizontal line;
and if the included angle between the target line segment of the tower and the horizontal line is 90 degrees, judging that the tower is in a vertical state.
Further, the first type identification object further includes a cross arm, and the determining module is further configured to:
screening all connected regions with the area larger than a first preset area threshold value from the connected regions in the outlines of the cross arms of all the images to be recognized, and determining the connected regions as effective regions of the cross arms;
the step of judging whether the angle of the first type identification object is in a standard process state according to the target line segment of the first type identification object and the reference line comprises the following steps:
judging whether the cross arm is in a horizontal state or not according to the included angle between the target line segment of the cross arm and the horizontal line;
and if the included angle between the target line segment of the cross arm and the horizontal line is 0 degree, judging that the cross arm is in a horizontal state.
Further, the second class identification object includes a transformer, and the determining module is further configured to:
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the transformer and the cross arms closest to the transformer according to the target line segments of the two towers closest to the transformer and the target line segments of the cross arms closest to the transformer, and setting the positions as the intersection points of the cross arms of the towers;
judging whether the transformer is centered or not according to the distance from the center of gravity of the transformer to the intersection point of the cross arms of the two towers;
and if the ratio of the gravity center of the transformer to the distance between the intersection points of the two tower cross arms is within a first preset range, determining that the transformer is centered.
Further, the second type identification object further includes a ring main unit, and the determining module is further configured to:
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the ring main unit and the cross arms closest to the ring main unit according to the target line segments of the two towers closest to the ring main unit and the target line segments of the cross arms closest to the ring main unit, and setting the intersection points as tower cross arm intersection points;
judging whether the ring main unit is centered or not according to the distance between the center of gravity of the ring main unit and the intersection point of the cross arms of the two towers;
and if the ratio of the gravity center of the ring main unit to the distance between the intersection points of the two tower cross arms is within a second preset range, the ring main unit is determined to be centered.
Example 3
Referring to fig. 3, the present invention further provides a detection apparatus for a power supply area construction process, which is shown as a detection apparatus for a power supply area construction process in a third embodiment of the present invention, and includes a memory 20, a processor 10, and a computer program 30 stored in the memory and running on the processor, where when the processor 10 executes the computer program 30, the detection method for the power supply area construction process is implemented.
The detection device of the power supply station area construction process may specifically be a computer, a server, an upper computer, and the like, and the processor 10 may be a Central Processing Unit (CPU), a controller, a microcontroller, a microprocessor, or other data Processing chips in some embodiments, and is configured to run program codes or Processing data stored in the memory 20, for example, execute an access restriction program and the like.
The memory 20 includes at least one type of readable storage medium including flash memory, hard disks, multimedia cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, and the like. The memory 20 may be, in some embodiments, an internal storage unit of the inspection apparatus for the power supply bay construction process, such as a hard disk of the inspection apparatus for the power supply bay construction process. The memory 20 may also be an external storage device of the power supply station area construction process detection device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the power supply station area construction process detection device. Further, the memory 20 may also include both an internal storage unit and an external storage device of the inspection equipment for the power bay construction process. The memory 20 may be used not only to store application software installed in the inspection equipment for the power supply station construction process and various kinds of data, but also to temporarily store data that has been output or will be output.
It should be noted that the configuration shown in fig. 3 does not constitute a limitation of the detection device of the power supply bay construction process, and in other embodiments, the detection device of the power supply bay construction process may include fewer or more components than those shown, or some components may be combined, or a different arrangement of components.
The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the above detection method for a power supply platform area construction process.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (9)
1. A detection method for a power supply area construction process is characterized by comprising the following steps:
collecting a large number of monitoring images of a target platform area, and setting the monitoring images as images to be identified;
inputting each frame of image to be recognized into a trained semantic segmentation model, and extracting masks of target recognition objects of all the images to be recognized through the semantic segmentation model;
extracting the outlines of the masks of the target identification objects of all the images to be identified to obtain the outlines of the target identification objects in all the images to be identified, and calculating the areas of the connected regions in the outlines of the target identification objects in all the images to be identified;
screening out all effective regions of the target identification object according to the areas of the connected regions in the outline of the target identification object in all the images to be identified;
analyzing all effective areas of the target identification object to judge whether the target identification object is in a standard process state;
the step of analyzing all effective areas of the target identification object to judge whether the target identification object is in a standard process state comprises the following steps:
calculating first moments of all effective areas of the first class of recognition objects to obtain the barycentric position of the first class of recognition objects;
calculating second moments for all effective areas of the first type of recognition objects to obtain the direction of the first type of recognition objects;
obtaining a straight line representing the direction and the center line of the first type identification object by using a point-oblique equation according to the gravity center position and the direction of the first type identification object, and setting the straight line as the straight line of the first type identification object;
fitting minimum circumscribed circles of the outlines of all effective areas of the first-class identification objects, and setting two intersection points of the minimum circumscribed circles and straight lines of the corresponding first-class identification objects as two end points of the first-class identification objects;
determining line segments of the first type of identification objects, which are obtained by intercepting two end points of the first type of identification objects on a straight line of the first type of identification objects, as the line segments of the first type of identification objects;
merging the line segments of the first type identification objects between the different images to be identified according to the line segments of the first type identification objects and the positions of the two end points to obtain a target line segment of the first type identification object;
and judging whether the angle of the first type of identification object is in a standard process state or not according to the target line segment of the first type of identification object and the reference line.
2. The method for detecting the power supply area construction process according to claim 1, wherein the target identification object further comprises a second type identification object, and the step of judging whether the angle of the first type identification object is in the standard process state according to the target line segment of the first type identification object and the reference line further comprises:
calculating first moments of all effective areas of the second type of recognition objects to obtain the gravity center positions of the second type of recognition objects;
and if the angle of the first class identification object is in a standard process state, judging whether the position of the second class identification object is in the standard process state or not according to the gravity center position of the second class identification object by taking the target line segment of the first class identification object as a reference.
3. The method for detecting the power supply area construction process according to claim 2, wherein the first type of identification object includes a tower, the reference line is a horizontal line, and the step of screening out all effective regions of the target identification object according to the areas of connected regions in the outline of the target identification object in all the images to be identified includes:
if a single frame of image to be identified comprises more than two towers and the number of the connected regions of the contour of the tower, the area of which is larger than a first preset area threshold value, is more than two, determining the two connected regions with the largest area as effective regions of the towers;
if the number of the connected areas of the contour of the tower, with the area larger than a first preset area threshold value, in a single frame of the image to be identified is not more than two, determining the connected areas with the area larger than the first preset area threshold value as the effective areas of the tower;
the step of judging whether the angle of the first type identification object is in a standard process state according to the target line segment of the first type identification object and the reference line comprises the following steps:
and judging whether the tower is in a vertical state or not according to the included angle between the target line segment of the tower and the horizontal line.
4. The method for detecting the power supply area construction process according to claim 3, wherein the first type of identification object further comprises a cross arm, and the step of screening out all effective regions of the target identification object according to the areas of connected regions in the outline of the target identification object in all the images to be identified comprises:
screening all connected regions with the area larger than a first preset area threshold value from the connected regions in the outlines of the cross arms of all the images to be recognized, and determining the connected regions as effective regions of the cross arms;
the step of judging whether the angle of the first type identification object is in a standard process state according to the target line segment of the first type identification object and the reference line comprises the following steps:
and judging whether the cross arm is in a horizontal state or not according to the included angle between the target line segment of the cross arm and the horizontal line.
5. The method for detecting the power supply area construction process according to claim 4, wherein the second type identification object comprises a transformer, and the step of determining whether the position of the second type identification object is in the standard process state or not based on the gravity center position of the second type identification object with reference to the target line segment of the first type identification object if the angle of the first type identification object is in the standard process state comprises:
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the transformer and the cross arms closest to the transformer according to the target line segments of the two towers closest to the transformer and the target line segments of the cross arms closest to the transformer, and setting the positions as the intersection points of the cross arms of the towers;
judging whether the transformer is centered or not according to the distance from the center of gravity of the transformer to the intersection point of the cross arms of the two towers;
and if the ratio of the gravity center of the transformer to the distance between the intersection points of the two tower cross arms is within a first preset range, determining that the transformer is centered.
6. The method for detecting the power supply area construction process according to claim 4, wherein the second type identification object further includes a ring main unit, and the step of determining whether the position of the second type identification object is in the standard process state or not according to the gravity center position of the second type identification object with reference to the target line segment of the first type identification object if the angle of the first type identification object is in the standard process state further includes:
if the towers are in a vertical state and the cross arms are in a horizontal state, determining the positions of the intersection points of the target line segments of the two towers closest to the ring main unit and the cross arms closest to the ring main unit according to the target line segments of the two towers closest to the ring main unit and the target line segments of the cross arms closest to the ring main unit, and setting the intersection points as tower cross arm intersection points;
judging whether the ring main unit is centered or not according to the distance between the center of gravity of the ring main unit and the intersection point of the cross arms of the two towers;
and if the ratio of the gravity center of the ring main unit to the distance between the intersection points of the two tower cross arms is within a second preset range, the ring main unit is determined to be centered.
7. A detection system for a power supply area construction process is characterized by comprising:
an acquisition module: the system comprises a monitoring system, a monitoring system and a recognition system, wherein the monitoring system is used for collecting a large number of monitoring images of a target platform area and setting the monitoring images as images to be recognized;
an input module: the system comprises a semantic segmentation model, a target recognition object extraction model and a target recognition object extraction model, wherein the semantic segmentation model is used for extracting masks of target recognition objects of images to be recognized;
a calculation module: the device comprises a mask, a detection unit, a processing unit and a display unit, wherein the mask is used for extracting the outline of a target identification object of all images to be identified so as to obtain the outline of the target identification object in all the images to be identified, and calculating the area of a communication region in the outline of the target identification object in all the images to be identified;
a screening module: screening out all effective regions of the target identification object according to the areas of the connected regions in the outline of the target identification object in all the images to be identified;
a judging module: the system comprises a target identification object, a control unit and a display unit, wherein the target identification object is used for identifying the target identification object in a standard process state;
wherein the target identification object comprises a first type of identification object, and the determining module is further configured to:
calculating first moments of all effective areas of the first type of recognition objects to obtain the gravity center positions of the first type of recognition objects;
calculating second moments for all effective areas of the first type of recognition objects to obtain the direction of the first type of recognition objects;
obtaining a straight line representing the direction and the center line of the first type identification object by using a point-oblique equation according to the gravity center position and the direction of the first type identification object, and setting the straight line as the straight line of the first type identification object;
fitting minimum circumscribed circles of the outlines of all effective areas of the first-class identification objects, and setting two intersection points of the minimum circumscribed circles and straight lines of the corresponding first-class identification objects as two end points of the first-class identification objects;
determining a line segment of the first-class identification object, which is obtained by intercepting two end points of the first-class identification object on a straight line of the first-class identification object;
merging the line segments of the first type identification objects between the different images to be identified according to the line segments of the first type identification objects and the positions of the two end points to obtain a target line segment of the first type identification object;
and judging whether the angle of the first type of identification object is in a standard process state or not according to the target line segment of the first type of identification object and the reference line.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the method for detecting a power supply bay construction process according to any one of claims 1 to 6.
9. An apparatus for detecting a power supply bay construction process, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for detecting a power supply bay construction process as claimed in any one of claims 1 to 6 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210012476.1A CN114078234B (en) | 2022-01-07 | 2022-01-07 | Detection method, system, storage medium and equipment for power supply area construction process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210012476.1A CN114078234B (en) | 2022-01-07 | 2022-01-07 | Detection method, system, storage medium and equipment for power supply area construction process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114078234A CN114078234A (en) | 2022-02-22 |
| CN114078234B true CN114078234B (en) | 2022-05-31 |
Family
ID=80284416
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210012476.1A Active CN114078234B (en) | 2022-01-07 | 2022-01-07 | Detection method, system, storage medium and equipment for power supply area construction process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114078234B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116485759A (en) * | 2023-04-25 | 2023-07-25 | 什维新智医疗科技(上海)有限公司 | Method, system and electronic equipment for identifying target boundary in ultrasonic image |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101082592A (en) * | 2006-06-01 | 2007-12-05 | 大日本网目版制造株式会社 | Uneven checking method and device and recording medium |
| CN111091527A (en) * | 2018-10-24 | 2020-05-01 | 华中科技大学 | Method and system for automatically detecting pathological change area in pathological tissue section image |
| CN112967287A (en) * | 2021-01-29 | 2021-06-15 | 平安科技（深圳）有限公司 | Gastric cancer focus identification method, device, equipment and storage medium based on image processing |
| CN113313013A (en) * | 2021-05-26 | 2021-08-27 | 国网电力科学研究院武汉南瑞有限责任公司 | Transformer bushing target detection method based on infrared image processing technology |
| CN113390880A (en) * | 2021-06-04 | 2021-09-14 | 国网江苏省电力有限公司常州供电分公司 | Transmission conductor defect detection device and method |
| CN113569819A (en) * | 2021-09-23 | 2021-10-29 | 泰豪软件股份有限公司 | Method and system for identifying opening and closing states of disconnecting link, storage medium and computer equipment |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4490468B2 (en) * | 2007-10-10 | 2010-06-23 | シーケーディ株式会社 | Solder printing inspection device |
| DE102008062916A1 (en) * | 2008-12-23 | 2010-06-24 | Continental Safety Engineering International Gmbh | Method for determining a collision probability of a vehicle with a living being |
-
2022
- 2022-01-07 CN CN202210012476.1A patent/CN114078234B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101082592A (en) * | 2006-06-01 | 2007-12-05 | 大日本网目版制造株式会社 | Uneven checking method and device and recording medium |
| CN111091527A (en) * | 2018-10-24 | 2020-05-01 | 华中科技大学 | Method and system for automatically detecting pathological change area in pathological tissue section image |
| CN112967287A (en) * | 2021-01-29 | 2021-06-15 | 平安科技（深圳）有限公司 | Gastric cancer focus identification method, device, equipment and storage medium based on image processing |
| CN113313013A (en) * | 2021-05-26 | 2021-08-27 | 国网电力科学研究院武汉南瑞有限责任公司 | Transformer bushing target detection method based on infrared image processing technology |
| CN113390880A (en) * | 2021-06-04 | 2021-09-14 | 国网江苏省电力有限公司常州供电分公司 | Transmission conductor defect detection device and method |
| CN113569819A (en) * | 2021-09-23 | 2021-10-29 | 泰豪软件股份有限公司 | Method and system for identifying opening and closing states of disconnecting link, storage medium and computer equipment |
Non-Patent Citations (2)
| Title |
|---|
| 《Automatic Detection of Power Transmission Lines and Risky Object Location Using UAV LiDAR Data》;Mustafa D et al;《Springer Link》;20210321;全文 * |
| 《基于计算机视觉变电站绝缘子闪络监测系统的设计与实现》;周伟浩;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》;20210715(第2021年第07期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114078234A (en) | 2022-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112837290B (en) | Crack image automatic identification method based on seed filling algorithm | |
| CN111091544B (en) | Method for detecting breakage fault of side integrated framework of railway wagon bogie | |
| CN108537154A (en) | Transmission line of electricity Bird's Nest recognition methods based on HOG features and machine learning | |
| CN111782583B (en) | Intelligent drawing examining method and related device | |
| CN114998852A (en) | Intelligent detection method for road pavement diseases based on deep learning | |
| CN108711148A (en) | A kind of wheel tyre defect intelligent detecting method based on deep learning | |
| CN113221889A (en) | Anti-interference recognition method and device for chip characters | |
| CN114078234B (en) | Detection method, system, storage medium and equipment for power supply area construction process | |
| CN115995056A (en) | Automatic bridge disease identification method based on deep learning | |
| CN112102296A (en) | Power equipment target identification method based on human concept | |
| CN111900694A (en) | Relay protection equipment information acquisition method and system based on automatic identification | |
| CN116168351A (en) | Inspection method and device for power equipment | |
| CN110287967B (en) | Image-based digit and mechanical watch digit identification method | |
| CN112967224A (en) | Electronic circuit board detection system, method and medium based on artificial intelligence | |
| CN115760854A (en) | Deep learning-based power equipment defect detection method and device and electronic equipment | |
| CN115811655A (en) | Inspection method, device and equipment for transformer substation and readable storage medium | |
| CN115761613A (en) | Automatic tunnel crack detection method based on convolutional network | |
| CN115205747A (en) | Automatic road detection system based on image recognition | |
| CN114240924A (en) | Power grid equipment quality evaluation method based on digitization technology | |
| CN116152674A (en) | Dam unmanned aerial vehicle image crack intelligent recognition method based on improved U-Net model | |
| CN112115876A (en) | Water-soluble method experimental process identification method based on 3D convolutional neural network | |
| CN116682188B (en) | Method and device for attendance-free perception-free card punching of constructors on borderless working surface of hydropower station | |
| CN117726239B (en) | Engineering quality acceptance actual measurement method and system | |
| CN117496443B (en) | Highway power line fault detection method, device, electronic equipment and medium | |
| CN114581890B (en) | Method and device for determining lane line, electronic equipment and storage medium |
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 |