CN107610128A - The method for inspecting and device of a kind of oil level indicator - Google Patents
The method for inspecting and device of a kind of oil level indicator Download PDFInfo
- Publication number
- CN107610128A CN107610128A CN201710882264.8A CN201710882264A CN107610128A CN 107610128 A CN107610128 A CN 107610128A CN 201710882264 A CN201710882264 A CN 201710882264A CN 107610128 A CN107610128 A CN 107610128A
- Authority
- CN
- China
- Prior art keywords
- oil level
- image
- described image
- level indicator
- segmentation
- 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
Abstract
This application provides a kind of method for inspecting of oil level indicator and device, the image of oil level indicator to be inspected is specially obtained；Cluster segmentation is carried out to described image, obtains the label matrix of described image；Color space conversion processing is carried out to described image, obtains the eigenmatrix of described image；Locating segmentation is carried out to described image using the label matrix and the eigenmatrix, obtains the top edge line in region to be identified；When the top edge line exceedes preset oil level threshold value, actuation of an alarm is made.By the above-mentioned processing to image, crusing robot autonomous classification can be made to go out the oil level of oil level indicator, and can completed to alarm according to oil level threshold value, so that crusing robot can substitute artificial progress inspection.
Description
Technical field
The application is related to robotic technology field, more specifically to the method for inspecting and device of a kind of oil level indicator.
Background technology
The oil plant for playing radiating and insulation is filled with many equipment of transformer station, in these equipment or outside can install
There is the oil level indicator how many for showing these oil plants, can determine whether oil level is reasonable by the inspection to oil level indicator, if mistake
It is more or it is very few can have a negative impact to the equipment that it is protected, therefore to transformer station carry out inspection when, it is true by oil level indicator
The number of oil plant is important inspection content in locking equipment.
When using artificial progress inspection, observe the height of oil level in oil level indicator is not for the eyes of patrol officer
Problem, but in view of manual inspection is less efficient and dangerous, thus at present in many transformer stations applied robot to oil level
Meter carries out inspection, and when oil level is too high or too low and alarm, this just needs robot independently to determine oil level in oil level indicator
Height and complete to alarm.
The content of the invention
In view of this, the application provides a kind of method for inspecting and device of oil level indicator, and this method and device are applied to becoming
Power station carries out crusing robot, so that crusing robot can independently determine the height of oil level in oil level indicator and complete to alarm.
To achieve these goals, it is proposed that scheme it is as follows：
A kind of method for inspecting of oil level indicator, the oil level indicator method for inspecting include step：
Obtain the image of oil level indicator to be inspected；
Cluster segmentation is carried out to described image, obtains the label matrix of described image；
Color space conversion processing is carried out to described image, obtains the eigenmatrix of described image；
Locating segmentation is carried out to described image using the label matrix and the eigenmatrix, obtains battery limits in image
Top edge line in domain；
When the top edge line exceedes preset oil level threshold value, actuation of an alarm is made.
Optionally, it is described to described image progress cluster segmentation, including：
The equipment region of the oil level indicator is determined from described image using default image template；
Linear spectral cluster segmentation is carried out to the equipment region, obtains the label matrix of the equipment region；
Optionally, it is described to described image progress color space conversion processing, including：
The rgb space of the equipment region is converted into HSV space, and extracts H components, obtains H channel images；
By the contrast stretching of the H channel images；
The H channel images are split, and carry out gaussian filtering process；
The pixel extreme value of the image after segmentation is calculated, and coarse positioning segmentation is carried out using extreme value, obtains the equipment region
Eigenmatrix.
Optionally, the oil level threshold value is oil level higher limit or oil level lower limit, it is described when the top edge line exceed it is pre-
During the oil level threshold value put, actuation of an alarm is made, including：
When the top edge line is higher than the oil level higher limit, warning message is issued the user with；
Or when the top edge line is less than the oil level lower limit, issue the user with warning message.
Optionally, the determination step of the oil level threshold value includes：
The image template of oil level indicator is gathered, the bound coordinate of the oil level indicator is demarcated in described image template；
Characteristic matching calculating is carried out to described image and described image template, obtains longitudinal bias between the two；
The oil level threshold value is obtained according to the longitudinal bias and the bound coordinate.
A kind of inspection device of oil level indicator, the oil level indicator inspection device include：
Image collection module, for obtaining the image of oil level indicator to be inspected；
Cluster segmentation module, for carrying out cluster segmentation to described image, obtain the label matrix of described image；
Spatial transformation module, for carrying out color space conversion processing to described image, obtain the feature square of described image
Battle array；
Locating segmentation module, for carrying out positioning point to described image using the label matrix and the eigenmatrix
Cut, obtain the top edge line in region to be identified；
Alarm output module, for when the top edge line exceedes preset oil level threshold value, making actuation of an alarm.
Optionally, the cluster segmentation module includes：
Equipment region positioning unit, for determining setting for the oil level indicator from described image using default image template
Preparation area domain；
Linear spectral cluster segmentation unit, for carrying out the general cluster segmentation of linear spectral to the equipment region, obtain described set
The label matrix in preparation area domain；
Optionally, the spatial transformation module includes：
Spatial transformation unit, for the rgb space of the equipment region to be converted into HSV space, and H components are extracted, obtained
To H channel images；
Contrast stretching unit, for by the contrast stretching of the H components；
Channel image cutting unit, for splitting to the H channel images, and carry out gaussian filtering process；
Pixel extreme value computing unit, coarse positioning is carried out for calculating the pixel extreme value of the image after splitting, and using extreme value
Segmentation, obtains the eigenmatrix of the equipment region.
Optionally, oil level threshold value is oil level higher limit or oil level lower limit, and the alarm output module includes：
First alarm unit, for when the top edge line is higher than the oil level higher limit, issuing the user with alarm signal
Breath；
Second alarm unit, for when the top edge line is less than the oil level lower limit, issuing the user with alarm signal
Breath.
Optionally, in addition to threshold calculation module, the threshold calculation module include：
Image template collecting unit, for gathering the image template of oil level indicator, the oil is demarcated in described image template
The bound coordinate of position meter；
Longitudinal bias computing unit, for carrying out characteristic matching calculating to described image and described image template, obtain two
Longitudinal bias between person；
Oil level threshold computation unit, for obtaining the oil level threshold according to the longitudinal bias and the bound coordinate
Value.
It can be seen from the above technical scheme that this application discloses a kind of oil level indicator method for inspecting and device, this method
The image of oil level indicator to be inspected is specially obtained with device；Cluster segmentation is carried out to described image, obtains the mark of described image
Sign matrix；Color space conversion processing is carried out to described image, obtains the eigenmatrix of described image；Utilize the label matrix
Locating segmentation is carried out to described image with the eigenmatrix, obtains the top edge line in region to be identified；When the top edge line
During more than preset oil level threshold value, actuation of an alarm is made.By the processing of the above method and device to image, survey monitor can be made
Device people's autonomous classification goes out the oil level of oil level indicator, and can complete to alarm according to oil level threshold value, so that crusing robot can replace
Generation artificial progress inspection.
Brief description of the drawings
, below will be to embodiment or existing in order to illustrate more clearly of the embodiment of the present application or technical scheme of the prior art
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of application, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is a kind of step flow chart of the method for inspecting embodiment for oil level indicator that the application provides；
Fig. 1 a are the image of the equipment region for the oil level indicator that the application provides；
Fig. 1 b are the image after the processing of linear spectral cluster segmentation that the application provides；
Fig. 1 c are the H channel images that the application provides；
Fig. 1 d are the H channel images after contrast stretching is handled that the application provides；
Fig. 1 e are the image in the region to be identified that the application provides；
Fig. 1 f are the image for the eigenmatrix that the application provides；
Fig. 1 g are the recognition result image that the application provides；
Fig. 2 is a kind of inspection device example structure block diagram for oil level indicator that the application provides；
Fig. 2 a are the structured flowchart of the inspection device for another oil level indicator that the application provides.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present application, the technical scheme in the embodiment of the present application is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only some embodiments of the present application, rather than whole embodiments.It is based on
Embodiment in the application, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of the application protection.
Embodiment one
Fig. 1 is a kind of step flow chart of the method for inspecting embodiment of the special oil level indicator provided of the application.
As shown in figure 1, the oil level indicator method for inspecting that the present embodiment provides can apply to the crusing robot of transformer station, use
The oil level indicator of electrical equipment detects in crusing robot is enable to transformer station, determines that the oil level indicated by oil level indicator is
No normal, the oil level indicator method for inspecting specifically comprises the following steps：
S101：Obtain the image of oil level indicator to be inspected.
I.e. the crusing robot inspection passes through video camera or photograph to being stopped when being provided with beside the electrical equipment of oil level indicator
Machine obtains the image of the oil level indicator of electrical equipment.When obtaining its image using video camera, it is necessary to the video obtained to video camera
Stream carries out real-time decoding, and video flowing is decomposed into the image sequence of independent frame, then carries out target location detection to image sequence,
Be disposed returning result after 3 seconds videos；Without being decoded to video flowing if camera is used, directly using camera
Obtained image.
S102：Cluster segmentation is carried out to image, obtains the label matrix of image.
After the image comprising corresponding oil level indicator is obtained, cluster segmentation processing is carried out to image, handled by cluster segmentation
Obtain the label matrix of the image.The matrix form that label matrix is made up of multiple labels, each label are used to express and above divided
Obtained each region is cut, multiple label composition label matrixs are consistent with artwork size；And it can be calculated according to label matrix
The area of each label area, i.e., in each label area pixel number.The detailed process of processing is as follows：
First, the equipment region of oil level indicator in image is determined using the image template of the oil level indicator gathered in advance, i.e., should
Oil level indicator position in the images, as shown in Figure 1a.
Then, linear spectral cluster segmentation is carried out to the equipment region, the label matrix of equipment region is obtained by segmentation, such as
Shown in Fig. 1 b.The label matrix is really to split what obtained multiple regions were formed by multiple, then after obtaining above-mentioned label matrix,
Calculate the area in the region representated by each label.
The step of linear spectral cluster segmentation, is as follows：
1st, for all picture element p in image be mapped to new feature space SP, SP (p)=(lp, ap, bp, xp, yp, 1,
1,1,1,1), lp, ap, bp, be that p is transformed into LAB spaces by rgb space, xp, yp are the coordinates of p spatially, behind all
Put 1, it is ensured that to higher dimensional space be linear separability after mapping；
2nd, sampled in x directions and y directions interval, k seed point is placed, according to the feature of oil level indicator, using x directions and y
Direction sampling interval is 1：2 ratio, k=18；
3rd, each seed is moved to the position of gradient minimum value in the range of 3*3；
4th, the mode of k mean clusters is copied, for carrying out initializing its weight center in the region of search of each seed point
Point and search center point；
5th, label L (p)=0, distance DIS (p)=infinity are made；
6th, it is every in zoning in its neighborhood in xp*yp (it is recommended that being 3*6) centered on p for each picture element p
The covariance of weight central point, seeks the minimum value of covariance in the region i.e. corresponding to individual picture element and each k cluster seeds point
It can determine that the point p affiliated k of label
All picture element p are carried out time completion cluster of traversal by the process for the 7th, repeating 3-6.
S103：Color space conversion processing is carried out to image, obtains the eigenmatrix of image.
After linear spectral cluster segmentation is carried out to equipment region or simultaneously, it is empty that color is carried out to the equipment region in image
Between conversion processing, obtain the eigenmatrix of equipment region.Specifically conversion process is：
First, the RGB color of equipment region is converted into HSV space, passes through the HSV space extraction equipment region
H components, so as to form H channel images, as illustrated in figure 1 c.
Then, the contrast of obtained H channel images is subjected to stretch processing.Histogram equalization is carried out to it, carried
Its high contrast and gray tone, H channel images are relatively sharp obtained from, as shown in Figure 1 d.
After again, coarse positioning is carried out to the H channel images of stretched processing, orients area-of-interest.I.e. by leading to H
Road image is split, and rejects fringe region therein, such as frame, to reduce interference；Then the image after segmentation is carried out
Gaussian filtering process, the mutation that gently pixel value occurs after equalization, obtained image is as shown in fig. le.
Finally, the pixel extreme value of the image after segmentation is calculated, coarse positioning segmentation is carried out using pixel extreme value；Carry out morphology
Operation, removes noise acnode therein, so as to obtain the mask MaskH, i.e. image of doubtful position eigenmatrix, such as Fig. 1 f institutes
Show.
S104：Locating segmentation is carried out to image using label matrix and eigenmatrix.
After above-mentioned label matrix and eigenmatrix is obtained, the equipment region using label matrix and eigenmatrix to image
Locating segmentation is carried out, so that it is determined that the top edge line in equipment region, as shown in Figure 1 g, the top edge line can be assumed that as oil level
Oil level in meter.
Specifically in locating segmentation, the region of the label matrix by eigenmatrix is searched, calculates what each label passed through
The area in region, and the area of this area and label area above is contrasted, then regard as it more than certain proportion
For region to be identified, so-called region to be identified refers to oil column region in oil level indicator, it is can determine that after the region is obtained
The oil level of top edge line, i.e. oil level indicator.
S105：When top edge line exceeds oil level threshold value, actuation of an alarm is made.
Because the standard of some equipment is can not to exceed oil level higher limit, some is then that cannot be below oil level lower limit, because
Here oil level threshold value includes oil level higher limit and oil level lower limit.Top edge line then refers to the top edge beyond oil level threshold value
Line is higher than oil level higher limit or less than oil level lower limit.
Specifically, if the standard of certain equipment is can not to exceed oil level higher limit, edge line is higher than on the oil level on top
Warning message is issued the user with during limit value；On the contrary, if the standard of certain equipment is to cannot be below oil level lower limit, work as top edge
Line issues the user with warning message when being less than the oil level lower limit.
It can be seen from the above technical proposal that present embodiments providing a kind of oil level indicator method for inspecting, specially obtain and treat
The image of the oil level indicator of inspection；Cluster segmentation is carried out to described image, obtains the label matrix of described image；Described image is entered
The processing of row color space conversion, obtains the eigenmatrix of described image；Using the label matrix and the eigenmatrix to institute
State image and carry out locating segmentation, obtain the top edge line in region to be identified；When the top edge line exceedes preset oil level threshold value
When, make actuation of an alarm.By the above-mentioned processing to image, crusing robot autonomous classification can be made to go out the oil level of oil level indicator,
And can complete to alarm according to oil level threshold value, so that crusing robot can substitute artificial progress inspection.
In addition, the oil level threshold value being previously mentioned in the present embodiment can be obtained by following method：
First, the image template of collection oil level indicator, and determine the bound coordinate of oil level indicator in the image template subscript in advance.
Then, the image to the oil level indicator of actual acquisition and image template carry out characteristic matching calculating, obtain between the two
Longitudinal bias；
Finally, bound coordinate is adjusted according to longitudinal bias, so as to obtain oil level threshold value, the oil level threshold value includes
Oil level higher limit and oil level lower limit.
Embodiment two
Fig. 2 is a kind of structured flowchart of the inspection device embodiment of the special oil level indicator provided of the application.
As shown in Fig. 2 the oil level indicator inspection device that the present embodiment provides can apply to the crusing robot of transformer station, use
The oil level indicator of electrical equipment detects in crusing robot is enable to transformer station, determines that the oil level indicated by oil level indicator is
It is no normal, the oil level indicator inspection device specifically include image collection module 10, cluster segmentation module 20, spatial transformation module 30,
Locating segmentation module 40 and alarm output module 50.
Image collection module is used for the image for obtaining oil level indicator to be inspected.
I.e. the crusing robot inspection passes through video camera or photograph to being stopped when being provided with beside the electrical equipment of oil level indicator
Machine obtains the image of the oil level indicator of electrical equipment.When obtaining its image using video camera, it is necessary to the video obtained to video camera
Stream carries out real-time decoding, and video flowing is decomposed into the image sequence of independent frame, then carries out target location detection to image sequence,
Be disposed returning result after 3 seconds videos；Without being decoded to video flowing if camera is used, directly using camera
Obtained image.
Cluster segmentation module is used to carry out cluster segmentation to image, obtains the label matrix of image.
After the image comprising corresponding oil level indicator is obtained, cluster segmentation processing is carried out to image, handled by cluster segmentation
Obtain the label matrix of the image.The module specifically includes equipment region positioning unit and linear spectral cluster segmentation unit.
Equipment region positioning unit is used to determine oil level indicator in image using the image template of the oil level indicator gathered in advance
Equipment region, i.e., the oil level indicator position in the images, as shown in Figure 1a.
Linear spectral cluster segmentation unit is used to carry out linear spectral cluster segmentation to the equipment region, and equipment is obtained by segmentation
The label matrix in region, as shown in Figure 1 b.The label matrix is really to split what obtained multiple regions were formed by multiple, then
To after above-mentioned label matrix, the area in the region representated by each label is calculated.
Spatial transformation module is used to carry out color space conversion processing to image, obtains the eigenmatrix of image.
After linear spectral cluster segmentation is carried out to equipment region or simultaneously, it is empty that color is carried out to the equipment region in image
Between conversion processing, obtain the eigenmatrix of equipment region.The module specifically include spatial transformation unit, contrast stretching unit,
Channel image cutting unit and pixel extreme value computing unit.
Spatial transformation unit is used to the RGB color of equipment region being converted to HSV space, is carried by the HSV space
The H components in taking equipment region, so as to form H channel images, as illustrated in figure 1 c.
The contrast that contrast stretching unit is used for the H channel images that will be obtained carries out stretch processing.It is carried out directly
Side's figure equalization, improves its contrast and gray tone, H channel images are relatively sharp obtained from, as shown in Figure 1 d.
Channel image cutting unit is used to carry out coarse positioning to the H channel images of stretched processing, orients interested
Region.I.e. by splitting to H channel images, fringe region therein, such as frame are rejected, to reduce interference；Then it is right
Image after segmentation carries out gaussian filtering process, the mutation that gently pixel value occurs after equalization, obtained image such as Fig. 1 e institutes
Show.
Pixel extreme value computing unit is used for the pixel extreme value for calculating the image after splitting, and utilizes pixel extreme value to carry out coarse positioning
Segmentation；Morphological operation is carried out, noise acnode therein is removed, so as to obtain the mask MaskH, i.e. image of doubtful position spy
Matrix is levied, as shown in Figure 1 f.
Locating segmentation module is used to carry out locating segmentation to image using label matrix and eigenmatrix.
After above-mentioned label matrix and eigenmatrix is obtained, the equipment region using label matrix and eigenmatrix to image
Locating segmentation is carried out, so that it is determined that the top edge line in equipment region, as shown in Figure 1 g, the top edge line can be assumed that as oil level
Oil level in meter.
Specifically in locating segmentation, the region of the label matrix by eigenmatrix is searched, calculates what each label passed through
The area in region, and the area of this area and label area above is contrasted, then regard as it more than certain proportion
For region to be identified, so-called region to be identified refers to oil column region in oil level indicator, it is can determine that after the region is obtained
The oil level of top edge line, i.e. oil level indicator.
Alarm output module is used for when top edge line exceeds oil level threshold value, makes actuation of an alarm.
Because the standard of some equipment is can not to exceed oil level higher limit, some is then that cannot be below oil level lower limit, because
Here oil level threshold value includes oil level higher limit and oil level lower limit.Top edge line then refers to the top edge beyond oil level threshold value
Line is higher than oil level higher limit or less than oil level lower limit.
The module includes the first alarm unit and the second alarm unit, specifically, the first alarm unit is used in certain equipment
Standard be that when can not exceed oil level higher limit, warning message is issued the user with when edge line is higher than the oil level higher limit on top；
If the second alarm unit is then that cannot be below oil level lower limit for the standard of certain equipment, when top edge line is less than the oil level
Warning message is issued the user with during lower limit.
It can be seen from the above technical proposal that present embodiments providing a kind of oil level indicator inspection device, specially obtain and treat
The image of the oil level indicator of inspection；Cluster segmentation is carried out to described image, obtains the label matrix of described image；Described image is entered
The processing of row color space conversion, obtains the eigenmatrix of described image；Using the label matrix and the eigenmatrix to institute
State image and carry out locating segmentation, obtain the top edge line in region to be identified；When the top edge line exceedes preset oil level threshold value
When, make actuation of an alarm.By the above-mentioned processing to image, crusing robot autonomous classification can be made to go out the oil level of oil level indicator,
And can complete to alarm according to oil level threshold value, so that crusing robot can substitute artificial progress inspection.
In addition, the oil level indicator polling module provided in the present embodiment also includes threshold calculation module 60, such as Fig. 2 a institutes
Show.The threshold calculation module is used to calculate oil level threshold value above-mentioned, specifically includes image template collecting unit, indulges
To deviation computing unit and oil level threshold computation unit.
Image template collecting unit is used for the image template for gathering oil level indicator in advance, and in the image template subscript stand oil position
The bound coordinate of meter.
Longitudinal bias computing unit is based on image and image template the progress characteristic matching of the oil level indicator to actual acquisition
Calculate, obtain longitudinal bias between the two；
Oil level threshold computation unit is used to be adjusted bound coordinate according to longitudinal bias, so as to obtain oil level threshold
Value, the oil level threshold value include oil level higher limit and oil level lower limit.
Each embodiment is described by the way of progressive in this specification, what each embodiment stressed be and other
The difference of embodiment, between each embodiment identical similar portion mutually referring to.To the upper of the disclosed embodiments
State bright, professional and technical personnel in the field is realized or using the application.A variety of modifications to these embodiments are to ability
It will be apparent for the professional and technical personnel in domain, generic principles defined herein can not depart from the application's
In the case of spirit or scope, realize in other embodiments.Therefore, the application be not intended to be limited to it is shown in this article these
Embodiment, and it is to fit to the most wide scope consistent with principles disclosed herein and features of novelty.
Claims (10)
1. a kind of method for inspecting of oil level indicator, it is characterised in that the oil level indicator method for inspecting includes step：
Obtain the image of oil level indicator to be inspected；
Cluster segmentation is carried out to described image, obtains the label matrix of described image；
Color space conversion processing is carried out to described image, obtains the eigenmatrix of described image；
Locating segmentation is carried out to described image using the label matrix and the eigenmatrix, obtained in image in equipment region
Top edge line；
When the top edge line exceedes preset oil level threshold value, actuation of an alarm is made.
2. method for inspecting as claimed in claim 1, it is characterised in that it is described to described image progress cluster segmentation, including：
The equipment region of the oil level indicator is determined from described image using default image template；
Linear spectral cluster segmentation is carried out to the equipment region, obtains the label matrix of the equipment region.
3. method for inspecting as claimed in claim 2, it is characterised in that described to be carried out to described image at color space conversion
Reason, including：
The rgb space of the equipment region is converted into HSV space, and extracts H components, obtains H channel images；
By the contrast stretching of the H channel images；
The H channel images are split, and carry out gaussian filtering process；
The pixel extreme value of the image after segmentation is calculated, and coarse positioning segmentation is carried out using extreme value, obtains the spy of the equipment region
Levy matrix.
4. method for inspecting as claimed in claim 1, it is characterised in that the oil level threshold value is oil level higher limit or oil level lower limit
Value, it is described to make actuation of an alarm when the top edge line exceedes preset oil level threshold value, including：
When the top edge line is higher than the oil level higher limit, warning message is issued the user with；
Or when the top edge line is less than the oil level lower limit, issue the user with warning message.
5. method for inspecting as claimed in claim 1, it is characterised in that the determination step of the oil level threshold value includes：
The image template of oil level indicator is gathered, the bound coordinate of the oil level indicator is demarcated in described image template；
Characteristic matching calculating is carried out to described image and described image template, obtains longitudinal bias between the two；
The oil level threshold value is obtained according to the longitudinal bias and the bound coordinate.
6. a kind of inspection device of oil level indicator, it is characterised in that the oil level indicator inspection device includes：
Image collection module, for obtaining the image of oil level indicator to be inspected；
Cluster segmentation module, for carrying out cluster segmentation to described image, obtain the label matrix of described image；
Spatial transformation module, for carrying out color space conversion processing to described image, obtain the eigenmatrix of described image；
Locating segmentation module, for carrying out locating segmentation to described image using the label matrix and the eigenmatrix, obtain
To the top edge line in region to be identified；
Alarm output module, for when the top edge line exceedes preset oil level threshold value, making actuation of an alarm.
7. inspection device as claimed in claim 6, it is characterised in that the cluster segmentation module includes：
Equipment region positioning unit, for determining the battery limits of the oil level indicator from described image using default image template
Domain；
Linear spectral cluster segmentation unit, for carrying out the general cluster segmentation of linear spectral to the equipment region, obtain the battery limits
The label matrix in domain.
8. inspection device as claimed in claim 7, it is characterised in that the spatial transformation module includes：
Spatial transformation unit, for the rgb space of the equipment region to be converted into HSV space, and H components are extracted, obtain H and lead to
Road image；
Contrast stretching unit, for by the contrast stretching of the H components；
Channel image cutting unit, for splitting to the H channel images, and carry out gaussian filtering process；
Pixel extreme value computing unit, coarse positioning segmentation is carried out for calculating the pixel extreme value of the image after splitting, and using extreme value,
Obtain the eigenmatrix of the equipment region.
9. inspection device as claimed in claim 6, it is characterised in that oil level threshold value is oil level higher limit or oil level lower limit,
The alarm output module includes：
First alarm unit, for when the top edge line is higher than the oil level higher limit, issuing the user with warning message；
Second alarm unit, for when the top edge line is less than the oil level lower limit, issuing the user with warning message.
10. inspection device as claimed in claim 6, it is characterised in that also including threshold calculation module, the threshold calculations mould
Block includes：
Image template collecting unit, for gathering the image template of oil level indicator, the oil level indicator is demarcated in described image template
Bound coordinate；
Longitudinal bias computing unit, for carrying out characteristic matching calculating to described image and described image template, obtain both it
Between longitudinal bias；
Oil level threshold computation unit, for obtaining the oil level threshold value according to the longitudinal bias and the bound coordinate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710882264.8A CN107610128A (en) | 2017-09-26 | 2017-09-26 | The method for inspecting and device of a kind of oil level indicator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710882264.8A CN107610128A (en) | 2017-09-26 | 2017-09-26 | The method for inspecting and device of a kind of oil level indicator |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107610128A true CN107610128A (en) | 2018-01-19 |
Family
ID=61058654
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710882264.8A Pending CN107610128A (en) | 2017-09-26 | 2017-09-26 | The method for inspecting and device of a kind of oil level indicator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107610128A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108844594A (en) * | 2018-04-17 | 2018-11-20 | 广州博通信息技术有限公司 | The alarm method and system of compressor of air conditioner liquid level identification |
| CN109344768A (en) * | 2018-09-29 | 2019-02-15 | 南京理工大学 | Pointer breaker recognition methods based on crusing robot |
| CN109344766A (en) * | 2018-09-29 | 2019-02-15 | 南京理工大学 | Slide block type breaker recognition methods based on crusing robot |
| CN109359646A (en) * | 2018-09-29 | 2019-02-19 | 南京理工大学 | Liquid level type Meter recognition method based on crusing robot |
| CN109389165A (en) * | 2018-09-29 | 2019-02-26 | 南京理工大学 | Oil level gauge for transformer recognition methods based on crusing robot |
| CN109447062A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Pointer-type gauges recognition methods based on crusing robot |
| CN109447949A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Insulated terminal defect identification method based on crusing robot |
| CN109447061A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Reactor oil level indicator recognition methods based on crusing robot |
| CN109446916A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Discharge counter recognition methods based on crusing robot |
| CN109934542A (en) * | 2019-04-19 | 2019-06-25 | 湖州三滴油科技有限公司 | A kind of lubricating oil product warehouse intelligent management system |
| CN112395972A (en) * | 2020-11-16 | 2021-02-23 | 中国科学院沈阳自动化研究所 | Electric power system insulator string identification method based on unmanned aerial vehicle image processing |
| CN113052823A (en) * | 2021-03-26 | 2021-06-29 | 东莞市科研世智能科技有限公司 | Oil level and oil color detection method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104408459A (en) * | 2014-10-27 | 2015-03-11 | 国家电网公司 | Image identification method applied to power equipment monitoring |
| CN105608703A (en) * | 2015-12-28 | 2016-05-25 | 浙江国自机器人技术有限公司 | Current transformer oil level detection method of intelligent substation inspection robot |
| CN106485728A (en) * | 2016-09-09 | 2017-03-08 | 电子科技大学 | A kind of automatic measure of bar shaped main transformer oil level indicator |
-
2017
- 2017-09-26 CN CN201710882264.8A patent/CN107610128A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104408459A (en) * | 2014-10-27 | 2015-03-11 | 国家电网公司 | Image identification method applied to power equipment monitoring |
| CN105608703A (en) * | 2015-12-28 | 2016-05-25 | 浙江国自机器人技术有限公司 | Current transformer oil level detection method of intelligent substation inspection robot |
| CN106485728A (en) * | 2016-09-09 | 2017-03-08 | 电子科技大学 | A kind of automatic measure of bar shaped main transformer oil level indicator |
Non-Patent Citations (4)
| Title |
|---|
| J. CHEN: ""Linear Spectral Clustering Superpixel"", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 * |
| REDDY NS: ""Image Segmentation by Using Linear Spectral Clustering"", 《JOURNAL OF TELECOMMUNICATIONS SYSTEM &》 * |
| ZHENGQIN LI: ""Superpixel segmentation using Linear Spectral Clustering"", 《2015 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)》 * |
| 许晓丽: ""基于聚类分析的图像分割算法研究"", 《万方》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108844594A (en) * | 2018-04-17 | 2018-11-20 | 广州博通信息技术有限公司 | The alarm method and system of compressor of air conditioner liquid level identification |
| CN108844594B (en) * | 2018-04-17 | 2021-03-12 | 广州博通信息技术有限公司 | Alarm method and system for liquid level identification of air conditioner compressor |
| CN109447949A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Insulated terminal defect identification method based on crusing robot |
| CN109359646A (en) * | 2018-09-29 | 2019-02-19 | 南京理工大学 | Liquid level type Meter recognition method based on crusing robot |
| CN109389165A (en) * | 2018-09-29 | 2019-02-26 | 南京理工大学 | Oil level gauge for transformer recognition methods based on crusing robot |
| CN109447062A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Pointer-type gauges recognition methods based on crusing robot |
| CN109344766A (en) * | 2018-09-29 | 2019-02-15 | 南京理工大学 | Slide block type breaker recognition methods based on crusing robot |
| CN109447061A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Reactor oil level indicator recognition methods based on crusing robot |
| CN109446916A (en) * | 2018-09-29 | 2019-03-08 | 南京理工大学 | Discharge counter recognition methods based on crusing robot |
| CN109344768A (en) * | 2018-09-29 | 2019-02-15 | 南京理工大学 | Pointer breaker recognition methods based on crusing robot |
| CN109934542A (en) * | 2019-04-19 | 2019-06-25 | 湖州三滴油科技有限公司 | A kind of lubricating oil product warehouse intelligent management system |
| CN112395972A (en) * | 2020-11-16 | 2021-02-23 | 中国科学院沈阳自动化研究所 | Electric power system insulator string identification method based on unmanned aerial vehicle image processing |
| CN113052823A (en) * | 2021-03-26 | 2021-06-29 | 东莞市科研世智能科技有限公司 | Oil level and oil color detection method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107610128A (en) | The method for inspecting and device of a kind of oil level indicator | |
| CN105069746B (en) | Video real-time face replacement method and its system based on local affine invariant and color transfer technology | |
| CN103925927B (en) | A kind of traffic mark localization method based on Vehicular video | |
| CN103186904B (en) | Picture contour extraction method and device | |
| CN108334847A (en) | A kind of face identification method based on deep learning under real scene | |
| CN109635875A (en) | A kind of end-to-end network interface detection method based on deep learning | |
| CN106023185A (en) | Power transmission equipment fault diagnosis method | |
| CN108648169A (en) | The method and device of high voltage power transmission tower defects of insulator automatic identification | |
| CN101620676A (en) | Fast image recognition method of insulator contour | |
| CN106709950A (en) | Binocular-vision-based cross-obstacle lead positioning method of line patrol robot | |
| CN105718912B (en) | A kind of vehicle characteristics object detecting method based on deep learning | |
| CN105574527A (en) | Quick object detection method based on local feature learning | |
| CN103927758B (en) | Saliency detection method based on contrast ratio and minimum convex hull of angular point | |
| CN102662949A (en) | Method and system for retrieving specified object based on multi-feature fusion | |
| CN109118548A (en) | A kind of comprehensive intelligent water quality recognition methods | |
| CN114387528A (en) | Pine nematode disease monitoring space-air-ground integrated monitoring method | |
| CN106127205A (en) | A kind of recognition methods of the digital instrument image being applicable to indoor track machine people | |
| CN107818303A (en) | Unmanned plane oil-gas pipeline image automatic comparative analysis method, system and software memory | |
| CN106548131A (en) | A kind of workmen's safety helmet real-time detection method based on pedestrian detection | |
| CN107264570A (en) | steel rail light band distribution detecting device and method | |
| CN112102296A (en) | Power equipment target identification method based on human concept | |
| CN107292328A (en) | The remote sensing image shadow Detection extracting method and system of multiple dimensioned multiple features fusion | |
| Klein et al. | Quantification of carbon sequestration in urban forests | |
| CN106709432A (en) | Binocular stereoscopic vision based head detecting and counting method | |
| CN108921844A (en) | A kind of defects of insulator detection method and device |
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 | ||
| CB02 | Change of applicant information |
Address after: 250000 Electric Power Intelligent Robot Production Project 101 in Jinan City, Shandong Province, South of Feiyue Avenue and East of No. 26 Road (ICT Industrial Park) Applicant after: National Network Intelligent Technology Co., Ltd. Address before: Xinluo Avenue high tech Zone of Ji'nan City, Shandong province 250101 silver bearing No. 2008 building Applicant before: Shandong Luneng Intelligent Technology Co., Ltd. |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180119 |
|
| RJ01 | Rejection of invention patent application after publication |