CN115830106B - Auxiliary positioning method for electrified cleaning of equipment in machine room - Google Patents
Auxiliary positioning method for electrified cleaning of equipment in machine room Download PDFInfo
- Publication number
- CN115830106B CN115830106B CN202310120232.XA CN202310120232A CN115830106B CN 115830106 B CN115830106 B CN 115830106B CN 202310120232 A CN202310120232 A CN 202310120232A CN 115830106 B CN115830106 B CN 115830106B
- Authority
- CN
- China
- Prior art keywords
- heat radiation
- area
- target
- heat source
- cleaned
- 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
Landscapes
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image data processing, in particular to an auxiliary positioning method for electrified cleaning of equipment in a machine room, which comprises the following steps: acquiring a target thermal infrared image of equipment in a machine room to be cleaned, and determining a gray level histogram of the target thermal infrared image; image data processing is carried out on the gray level histogram, and an adaptive cutting threshold value is determined; screening a highlight pixel point set from the target thermal infrared image; screening and growing treatment is carried out on the highlight pixel points; screening edge pixel points from a heat source area and a radiation range area, and determining a radiation length set; screening out a heat radiation area set to be cleaned from the heat radiation area set; and determining target cleaning position information corresponding to the equipment in the machine room to be cleaned, and cleaning the position corresponding to the target cleaning position information. According to the invention, through data processing of the target thermal infrared image, the accuracy of cleaning the equipment in the machine room is improved, and the method is mainly applied to cleaning the equipment in the machine room.
Description
Technical Field
The invention relates to the technical field of image data processing, in particular to an auxiliary positioning method for electrified cleaning of equipment in a machine room.
Background
In recent years, abnormal operation of equipment in a machine room is continuously caused by surface pollution, and the surface pollution is one of main hazards affecting the working state of the equipment in the machine room, so that the equipment becomes an unavoidable potential safety hazard of the machine room. Among others, surface contamination may include, but is not limited to: dust, soot, moisture, accumulated static electricity, and various charged particles. When the surface pollution attached to the equipment in the machine room is more, the temperature of the equipment in the machine room is always hotter, so that the equipment in the machine room is easy to fail, and potential safety hazard is caused. Therefore, the machine room equipment is often required to be cleaned, and the electrified cleaning is a relatively common method for cleaning the machine room equipment. At present, when cleaning machine room equipment, the mode generally adopted is: and shooting images of the machine room equipment through an infrared thermal image sensor, identifying the temperature of the machine room equipment, and cleaning the machine room equipment according to the temperature of the machine room equipment.
However, when the above manner is adopted, there are often the following technical problems:
factors causing heating of the machine room equipment often not only increase the upper surface pollution of the machine room equipment, but also heat caused by long-time use of the machine room equipment, therefore, the surface pollution area to be cleaned on the machine room equipment is often difficult to accurately position only according to the temperature of the machine room equipment, thereby often causing low accuracy of cleaning the machine room equipment and further causing waste of clean resources.
Disclosure of Invention
The summary of the invention is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. The summary of the invention is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
In order to solve the technical problem of low accuracy in cleaning machine room equipment, the invention provides an auxiliary positioning method for electrified cleaning of the machine room equipment.
The invention provides an auxiliary positioning method for electrified cleaning of equipment in a machine room, which comprises the following steps:
acquiring a target thermal infrared image of equipment in a machine room to be cleaned, and determining a gray level histogram of the target thermal infrared image;
performing image data processing on the gray level histogram, and determining an adaptive cutting threshold;
screening a highlight pixel point set from the target thermal infrared image according to the self-adaptive cutting threshold;
screening and growing the highlight pixel points in the highlight pixel point set to obtain a heat radiation area set, wherein the heat radiation area in the heat radiation area set comprises: a heat source region and a radiation range region;
For each heat radiation area in the heat radiation area set, selecting edge pixel points from a heat source area and a radiation range area included in the heat radiation area, and determining a radiation length set corresponding to the heat radiation area according to the selected edge pixel points;
according to the heat source areas included in each heat radiation area in the heat radiation area set and the corresponding radiation length sets of the heat radiation areas, selecting a heat radiation area set to be cleaned from the heat radiation area set;
and determining target cleaning position information corresponding to the equipment in the machine room to be cleaned according to the heat radiation area set to be cleaned, and cleaning the position corresponding to the target cleaning position information.
Further, the image data processing is performed on the gray histogram, and determining an adaptive cutting threshold includes:
determining an average brightness index and a deviation brightness index according to the gray level histogram;
determining the ratio of the average brightness index to the deflection brightness index as an overall deflection brightness index;
when the overall deflection brightness index is larger than or equal to a preset brightness uniformity value, determining the deflection brightness index as a self-adaptive cutting threshold;
And when the overall deviation brightness index is smaller than the brightness uniformity value, determining the average brightness index as an adaptive cutting threshold.
Further, the determining an average brightness index and a biased brightness index according to the gray level histogram includes:
determining a gray value at a target position in the gray histogram as an average brightness index, wherein the absolute value of the difference between the number of pixel points before the target position and the number of pixel points after the target position is smaller than a preset number threshold;
and determining half of the maximum gray value in the gray histogram as a biased brightness index.
Further, the screening the highlight pixel point set from the target thermal infrared image according to the adaptive cutting threshold includes:
and determining the pixel point with the gray value larger than or equal to the self-adaptive cutting threshold value in the target thermal infrared image as a highlight pixel point.
Further, the screening and growing treatment is performed on the highlight pixel points in the highlight pixel point set to obtain a heat radiation area set, including:
median filtering is carried out on the highlight pixel points in the highlight pixel point set, and a filtering value corresponding to the highlight pixel points is obtained;
When the filtering value corresponding to the highlight pixel is equal to the gray value corresponding to the highlight pixel, determining the highlight pixel as a seed pixel;
determining a corresponding extremely poor value of the seed pixel point according to a preset target window corresponding to the seed pixel point;
performing region growth according to the corresponding extreme difference value of each seed pixel point to obtain a target region set;
determining the average value of gray values corresponding to pixel points in each target area in the target area set as the gray average value corresponding to the target area;
screening a target area group set of a preset target arrangement mode from the target area set, wherein each target area group in the target area group set comprises two target areas;
for each target region group in the target region group set, determining a target region with a larger gray average value in the target region group as a heat source region, determining a target region with a smaller gray average value in the target region group as a radiation range region, and determining regions where two target regions in the target region group are located as heat radiation regions.
Further, the screening edge pixel points from the heat source area and the radiation range area included in the heat radiation area, and determining the radiation length set corresponding to the heat radiation area according to the screened edge pixel points, includes:
Determining the edge pixel points screened from the heat source region included in the heat radiation region as heat source edge pixel points, and obtaining a heat source edge pixel point set corresponding to the heat radiation region;
the edge pixel points screened from the radiation range area included in the heat radiation area are determined to be radiation edge pixel points, and a radiation edge pixel point set corresponding to the heat radiation area is obtained;
determining the mass center of a heat source area included in the heat radiation area as the mass center corresponding to the heat radiation area;
connecting the mass centers of each heat source edge pixel point in the heat source edge pixel point set corresponding to the heat radiation area and the heat radiation area to obtain a heat source line segment set corresponding to the heat radiation area;
for each heat source line segment in the heat source line segment set corresponding to the heat radiation area, extending one end of the heat source line segment where a heat source edge pixel point is located until the radiation edge pixel points in the radiation edge pixel point set corresponding to the heat radiation area intersect to obtain an intersection point, and taking a line segment between the centroid in the heat source line segment and the intersection point as the heat radiation line segment corresponding to the heat source line segment;
And for each heat source line segment in the heat source line segment set corresponding to the heat radiation area, determining the difference value between the length of the heat radiation line segment corresponding to the heat source line segment and the length of the heat source line segment as the radiation length.
Further, the selecting, according to the heat source region and the radiation length set corresponding to the heat radiation region included in each heat radiation region in the heat radiation region set, the heat radiation region set to be cleaned from the heat radiation region set includes:
determining an identification index to be cleaned corresponding to the heat radiation area according to a heat source area included in each heat radiation area in the heat radiation area set and a radiation length set corresponding to the heat radiation area;
and screening the heat radiation area set to be cleaned from the heat radiation area set according to a preset threshold to be cleaned and the recognition indexes to be cleaned corresponding to each heat radiation area in the heat radiation area set.
Further, the formula corresponding to the identification index to be cleaned corresponding to the heat radiation area is determined as follows:
wherein F is an identification index to be cleaned corresponding to the heat radiation area,th() Is a function of the hyperbolic tangent, NIs the number of gray values in the heat source region included in the heat radiation region,vis the number of gray values in the heat source region included in the heat radiation region,
the corresponding gray value in the heat source region included in the heat radiation region is equal to the firstvThe number of pixels of the individual gray values,His the number of pixel points in the heat source region included in the heat radiation region,ln() Is a logarithmic function based on natural constant, < ->
Is the standard deviation of the radiation length in the radiation length set corresponding to the heat radiation area, < + >>
To take the European norm function, ++>
In order to take the function of the absolute value,AandBis a preset value,/->
Is the average value of gray values corresponding to pixel points in a heat source region included in the heat radiation region,/or>
Is the maximum value of the gray values corresponding to the pixel points in the heat source region included in the heat radiation region.
Further, the selecting the heat radiation area set to be cleaned from the heat radiation area set according to a preset threshold to be cleaned and the recognition indexes to be cleaned corresponding to each heat radiation area in the heat radiation area set, includes:
and when the identification index to be cleaned corresponding to the heat radiation area in the heat radiation area set is smaller than or equal to the threshold to be cleaned, determining the heat source area included in the heat radiation area as the heat radiation area to be cleaned.
Further, determining, according to the set of heat radiation areas to be cleaned, target cleaning position information corresponding to the equipment in the machine room to be cleaned, includes:
and combining the positions of the heat radiation areas to be cleaned in the heat radiation area set to be cleaned into target cleaning position information corresponding to the equipment in the machine room to be cleaned.
The invention has the following beneficial effects:
according to the auxiliary positioning method for the electrified cleaning of the machine room equipment, the technical problem of low accuracy of cleaning the machine room equipment is solved by performing data processing on the target thermal infrared image, and the accuracy of cleaning the machine room equipment is improved. First, surface contamination attached to the machine room equipment, which may be contamination to be cleaned, tends to cause the machine room equipment to heat up. Therefore, the target thermal infrared image of the equipment to be cleaned, which is generated based on thermal radiation, is acquired, and the surface pollution attached to the equipment to be cleaned can be conveniently and subsequently positioned. Then, the machine room equipment tends to be heated due to surface contamination attached to the machine room equipment, and when a heated area and an unheated area of the machine room equipment are presented on the target thermal infrared image, the brightness and the gradation value corresponding in the gradation histogram of the two tend to be different. Further, when the heat of the heat generating region is different, the luminance and the gradation value corresponding to the gradation histogram are often also different. Therefore, the self-adaptive cutting threshold value is determined through the gray level histogram, so that whether the areas with different heating degrees are to be cleaned or not can be conveniently distinguished later. Then, surface contamination attached to the machine room equipment tends to cause the machine room equipment to heat, and tends to correspond to a pixel point with a higher gray value in the target thermal infrared image. Therefore, the highlight pixel points are collected and screened out, and the surface pollution area on the machine room equipment can be conveniently positioned later. And only the highlight pixel point set can be processed later, so that the redundant calculation amount can be reduced, and the occupation of calculation resources can be reduced. Continuing, the heat source tends to have a range of radiation to the surroundings. The heat source area may often be a surface contamination area. The radiation field area is often the area where the heat source radiates around. Therefore, distinguishing the heat source region and the radiation range region can prevent the heat source region and the radiation range region from being entirely used as the surface contamination region, and the surface contamination region can be further precisely located. Then, when the equipment in the machine room operates normally, the range of heat radiation from the edge of the heat source area to the periphery is unified, that is, the difference of the radiation lengths in the radiation length set corresponding to the heat radiation area is small. But surface contamination (e.g., dust) tends to accumulate randomly on the machine room equipment and the bulk density tends to be different. The stacking densities are different, and the interference degrees on the internal circuits of the machine room equipment are different, so that the machine room equipment generated by dust stacking is overheated, and the heat radiation ranges from the edges of the heat source areas to the periphery are often different, namely the radiation lengths in the radiation length sets corresponding to the heat radiation areas are often different greatly. Therefore, the radiation length set corresponding to the heat radiation area is determined, so that the follow-up judgment of whether the heat source area included in the heat radiation area is the surface pollution area to be cleaned can be facilitated. And finally, determining target cleaning position information corresponding to the equipment in the machine room to be cleaned according to the set of the heat radiation areas to be cleaned, and cleaning the position corresponding to the target cleaning position information. The heat source areas included in the heat radiation areas to be cleaned in the heat radiation area set to be cleaned are often surface pollution areas to be cleaned, so that the positions of the heat source areas included in the heat radiation areas to be cleaned are often positions to be cleaned. Therefore, the invention solves the technical problem of low accuracy of cleaning the equipment in the machine room by processing the data of the target thermal infrared image, improves the accuracy of cleaning the equipment in the machine room, and can avoid the waste of cleaning resources.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an auxiliary positioning method for electrified cleaning of equipment in a machine room according to the present invention;
FIG. 2 is a schematic diagram of a target arrangement according to the present invention;
fig. 3 is a schematic diagram of a heat source line segment and a radiation line segment according to the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description is given below of the specific implementation, structure, features and effects of the technical solution according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides an auxiliary positioning method for electrified cleaning of equipment in a machine room, which comprises the following steps:
acquiring a target thermal infrared image of equipment in a machine room to be cleaned, and determining a gray level histogram of the target thermal infrared image;
image data processing is carried out on the gray level histogram, and an adaptive cutting threshold value is determined;
screening a highlight pixel point set from the target thermal infrared image according to the self-adaptive cutting threshold value;
screening and growing treatment is carried out on the highlight pixel points in the highlight pixel point set to obtain a heat radiation area set;
for each heat radiation area in the heat radiation area set, selecting edge pixel points from a heat source area and a radiation range area included in the heat radiation area, and determining a radiation length set corresponding to the heat radiation area according to the selected edge pixel points;
according to the heat source areas included in each heat radiation area in the heat radiation area set and the corresponding radiation length sets of the heat radiation areas, selecting the heat radiation area set to be cleaned from the heat radiation area set;
And determining target cleaning position information corresponding to the equipment in the machine room to be cleaned according to the set of the heat radiation areas to be cleaned, and cleaning the position corresponding to the target cleaning position information.
The following detailed development of each step is performed:
referring to fig. 1, a flow of some embodiments of an assisted positioning method for live cleaning of equipment in a machine room according to the present invention is shown. The auxiliary positioning method for the electrified cleaning of the equipment in the machine room comprises the following steps:
step S1, acquiring a target thermal infrared image of equipment in a machine room to be cleaned, and determining a gray level histogram of the target thermal infrared image.
In some embodiments, a target thermal infrared image of the equipment in the machine room to be cleaned may be acquired, and a gray level histogram of the target thermal infrared image may be determined.
The equipment in the machine room to be cleaned can be equipment in the machine room to be cleaned. For example, the machine room equipment to be cleaned may be a cabinet. The target thermal infrared image may be an image after image preprocessing. Image preprocessing may include, but is not limited to: graying, denoising, and image enhancement.
As an example, this step may include the steps of:
firstly, acquiring a thermal infrared image of equipment in a machine room to be cleaned through a thermal infrared scanner (thermal infrared imager).
And secondly, filtering and denoising the thermal infrared image to obtain a denoised image.
For example, the filtered denoising may be a 5×5 mean filtered denoising.
In practical cases, complicated electromagnetic interference is often received under a machine room environment, electromagnetic signals often interfere equipment in a conduction, induction, radiation and other modes, so that a thermal infrared scanner often causes a great deal of Gaussian noise to be generated in a thermal infrared image when the thermal infrared scanner collects images, most of the thermal infrared image is low-frequency information, and therefore, fine filtering parameter adjustment is often not needed when filtering and denoising are carried out, and 5×5 mean filtering can be adopted for denoising.
And thirdly, graying the denoising image to obtain the target thermal infrared image.
Because the thermal infrared image is often a monochromatic color image, the thermal infrared image after noise reduction can be subjected to gray processing in order to facilitate calculation and reduce other redundant information such as image saturation.
In actual conditions, at present, a server and network equipment all often generate a lot of heat in the operation process, in order to radiate the heat, the heat is usually discharged in an active radiating mode, and because the space of a machine room is narrow, the equipment usually radiates heat in an air cooling mode, and radiating holes are matched with convection air, so that dust is often brought into the machine room equipment. Dust often entrains moisture and corrosive substances to enter the machine room equipment, and covers the electronic components, so that the heat dissipation capacity of the electronic components is reduced, and the equipment is often unstable in operation due to long-term accumulation of a large amount of heat. When maintenance personnel use special cleaning agents to clean in an electrified manner, dust at dead angles inside equipment (such as a cabinet) in a machine room is difficult to observe, so that the cleaning difficulty is often high. The infrared thermal imager mounted by the temperature anomaly alarming system mounted in the machine room can detect the specific heating position of the machine room equipment, the more dust is attached to the position, the more easily heating position is, and further the dust attaching dead angle can be accurately positioned by using the thermal infrared image, but the thermal infrared image is based on thermal radiation imaging, so that the heat source has a radiation range on the periphery, the boundary of the heat source tends to be fuzzy, the position of the heat source tends to be unable to be accurately positioned, and the heat accumulation of the heat source is difficult to judge due to dust accumulation or other reasons, wherein the temperature anomaly alarming system is generally mounted at the top of the machine room. Therefore, the invention can reduce redundant calculation by carrying out gray level cutting on the target thermal infrared image, set seed pixel points and growth rules on the cut image by utilizing median filtering to realize the segmentation of the fuzzy region, obtain a heat source region, construct a formula for determining the identification index to be cleaned, distinguish the heat source forming reason, accurately position the dust accumulation region, enable the electrified cleaning of the machine room equipment to be more efficient and save clean resources.
And S2, performing image data processing on the gray level histogram, and determining an adaptive cutting threshold.
In some embodiments, the gray level histogram may be subjected to image data processing to determine an adaptive cut threshold.
Wherein the adaptive cut threshold may be an adaptive luminance cut threshold.
As an example, this step may include the steps of:
first, determining an average brightness index and a biased brightness index according to the gray level histogram.
For example, this step may include the sub-steps of:
and a first sub-step of determining a gray value at a target position in the gray histogram as an average brightness index.
Wherein the absolute value of the difference between the number of pixels before the target position and the number of pixels after the target position may be smaller than a preset number threshold. The target position may be a center position of a gray histogram set in advance. For example, the last pixel point of half the pixels in the gray level histogram is located. The number threshold may be a preset number. For example, the number threshold may be 2.
For example, the number of pixels in the target thermal infrared image may be 50. The gray histogram may include: 10 pixels with a gray level of 100, 20 pixels with a gray level of 180, and 20 pixels with a gray level of 200. The target position may be a position where the 25 th pixel point is counted from left to right in the gray level histogram. The position of the 25 th pixel from left to right in the gray level histogram may be the position of the pixel having a gray level value of 180. The average brightness index may be 180.
As another example, the average luminance index may be from left to right to the gray level histogram
At the gray level at which the gray level, wherein,Gis the number of pixels in the target thermal infrared image.
And a second sub-step of determining half of the maximum gray value in the gray histogram as a biased brightness index.
For example, the formula for determining the bias brightness index may be:
wherein, the liquid crystal display device comprises a liquid crystal display device,
is a biased brightness index. />
Is the largest gray value in the gray histogram.
In practice, the maximum gray value in the gray histogram
The larger the bias brightness index +.>
The larger.
And a second step of determining the ratio of the average brightness index to the biased brightness index as an overall biased brightness index.
For example, the formula for determining the overall biased brightness index may be:
wherein, the liquid crystal display device comprises a liquid crystal display device,Kis an overall biased brightness index.
Is an average brightness index. />
Is a biased brightness index.
In practice, the number of the sensors, due to,
can be characterized from left to right to +.>
At the gray level at which the gray level, wherein,Gis the number of pixels in the target thermal infrared image. />
Can be used forIs half the maximum gray value in the gray histogram. Therefore, when->
In this case, the center of gravity of the target thermal infrared image may be biased toward the highlight gray level, or the highlight region on the target thermal infrared image may be more. When- >
When the center of gravity of the target thermal infrared image is biased towards a gray level, the center of gravity of the target thermal infrared image can be represented, and the highlight area on the target thermal infrared image is less.
And thirdly, determining the deviation brightness index as an adaptive cutting threshold when the whole deviation brightness index is larger than or equal to a preset brightness uniformity value.
The luminance uniformity value may be a preset value. For example, the luminance uniformity value may be 1.
And fourth, when the overall deviation brightness index is smaller than the brightness uniformity value, determining the average brightness index as an adaptive cutting threshold.
And S3, screening out a highlight pixel point set from the target thermal infrared image according to the self-adaptive cutting threshold.
In some embodiments, the set of highlight pixels may be screened from the target thermal infrared image based on an adaptive cut threshold.
As an example, a pixel point in the target thermal infrared image having a gray value greater than or equal to the adaptive cutting threshold may be determined as a highlight pixel point. The gray value corresponding to the pixel point in the target thermal infrared image with the gray value smaller than the adaptive cutting threshold value can be set to 0.
And S4, screening and growing the highlight pixel points in the highlight pixel point set to obtain a heat radiation area set.
In some embodiments, the highlight pixel points in the highlight pixel point set may be subjected to screening growth processing, so as to obtain a heat radiation area set.
Wherein, the heat radiation area in the heat radiation area set may include: a heat source region and a radiation range region. The heat source region may be a region where a heat source is located. The radiation-range region may be a region where the heat source region radiates outward.
As an example, this step may include the steps of:
the first step, median filtering is carried out on the highlight pixel points in the highlight pixel point set, and a filtering value corresponding to the highlight pixel points is obtained.
For example, a median filter of 5×5 may be performed on the highlight pixel, to obtain a filtered value corresponding to the highlight pixel.
In practice, the reason for adopting the median filtering is that when an isolated highlight pixel is at the center of a 5×5 window, since there are no or fewer other highlight pixels around the highlight pixel, the filter value corresponding to the highlight pixel is often 0. The isolated highlight pixels are often not pixels in the heat source region or the radiation range region, and therefore, the isolated highlight pixels can be avoided by adopting median filtering. When the highlight pixel is at the center of the 5×5 window, if there are more other highlight pixels in the 5×5 window, the filter value corresponding to the highlight pixel is often not 0.
And secondly, determining the highlight pixel point as a seed pixel point when the filtering value corresponding to the highlight pixel point is equal to the gray value corresponding to the highlight pixel point.
In practical situations, the gray values corresponding to the pixels in the heat source region or the radiation range region tend to be relatively stable, that is, the gray values corresponding to the pixels in the heat source region or the radiation range region tend to not change greatly, so when the filter value corresponding to the highlighted pixel is equal to the gray value corresponding to the highlighted pixel, the highlighted pixel is relatively stable, and may be the pixel in the heat source region or the radiation range region, and therefore, the seed pixel may be the pixel in the heat source region or the radiation range region.
And thirdly, determining the corresponding extremely poor value of the seed pixel point according to a preset target window corresponding to the seed pixel point.
The target window may be a preset window. For example, the target window may be a 5×5 window.
For example, the maximum gray value and the minimum gray value can be screened out from the target window corresponding to the seed pixel point, and the difference value between the maximum gray value and the minimum gray value is determined as the corresponding polar difference value of the seed pixel point.
And step four, carrying out region growth according to the corresponding extreme values of the seed pixel points to obtain a target region set.
For example, when the absolute value of the difference between the gray value corresponding to the seed pixel and the gray value corresponding to the adjacent pixel is less than or equal to the polar difference corresponding to the seed pixel, the two pixels may be grown to the same region. That is, the extremely poor value corresponding to the seed pixel may be used as a growth rule for the seed pixel. When multiple seed pixels in the same area are grown, the respective growth rules are often compatible with each other. If the growth rules differ greatly, the growth is stopped and edges are formed.
And fifthly, determining the average value of the gray values corresponding to the pixel points in each target area in the target area set as the gray average value corresponding to the target area.
And sixthly, screening a target area group set of a preset target arrangement mode from the target area set.
Wherein each target region group in the set of target region groups comprises two target regions. The target arrangement may be such that there is an edge overlap of the two target areas. The target arrangement may be such that the relationship between the connected domains of the two target areas is an inclusion relationship.
As shown in fig. 2, 3 target area groups can be obtained. The region where two concentric circles are located may represent a target region group. The region in which the two concentric rectangles reside may represent a target region group. The region in which the two concentric ellipses are located may represent a target region group. The smaller of the two concentric circles may characterize the target area and the area between the two concentric circles may characterize the target area. The smaller of the two concentric rectangles may characterize the target area and the area between the two concentric rectangles may characterize the target area. The smaller of the two concentric ellipses may characterize the target area and the area between the two concentric ellipses may characterize the target area. Wherein fig. 2 may be a schematic diagram of a target arrangement.
Seventh, for each target region group in the target region group set, determining a target region with a larger gray average value in the target region group as a heat source region, determining a target region with a smaller gray average value in the target region group as a radiation range region, and determining regions where two target regions in the target region group are located as heat radiation regions.
And S5, screening edge pixel points from a heat source region and a radiation range region included in the heat radiation region for each heat radiation region in the heat radiation region set, and determining a radiation length set corresponding to the heat radiation region according to the screened edge pixel points.
In some embodiments, for each heat radiation area in the heat radiation area set, edge pixel points may be screened from a heat source area and a radiation area included in the heat radiation area, and a radiation length set corresponding to the heat radiation area may be determined according to the screened edge pixel points.
The edge pixel points may be pixel points on edges in a heat source region or a radiation range region.
As an example, this step may include the steps of:
and determining the edge pixel points screened from the heat source region included in the heat radiation region as heat source edge pixel points to obtain a heat source edge pixel point set corresponding to the heat radiation region.
And secondly, determining the edge pixel points screened from the radiation range area included in the heat radiation area as radiation edge pixel points, and obtaining a radiation edge pixel point set corresponding to the heat radiation area.
And thirdly, determining the mass center of the heat source area included in the heat radiation area as the mass center corresponding to the heat radiation area.
And fourthly, connecting the mass centers of each heat source edge pixel point in the heat source edge pixel point set corresponding to the heat radiation area and the heat radiation area to obtain a heat source line segment set corresponding to the heat radiation area.
And fifthly, for each heat source line segment in the heat source line segment set corresponding to the heat radiation area, extending one end of the heat source line segment where the heat source edge pixel point is located until the radiation edge pixel point in the radiation edge pixel point set corresponding to the heat radiation area intersects, so as to obtain an intersection point, and taking the line segment between the centroid in the heat source line segment and the intersection point as the heat radiation line segment corresponding to the heat source line segment.
And sixthly, determining the difference between the length of the heat radiation line segment corresponding to the heat source line segment and the length of the heat source line segment as the radiation length for each heat source line segment in the heat source line segment set corresponding to the heat radiation area.
Wherein the radiating length may be the length of a radiating line segment. The radiation line segment may be a line segment other than the heat source line segment in the heat radiation line segment. As shown in fig. 3, the smaller of the two concentric circles may characterize the heat source region and the region between the two concentric circles may characterize the radiation range region. Solid points may characterize the centroid of the heat source region. The dashed line segments may characterize the heat source line segments. The real line segment may characterize the radiation line segment. The segment obtained by connecting the virtual segment with the solid line segment may be a heat radiation segment.
Optionally, the centroid point of the heat source area may be marked, the line direction from the centroid point to each edge pixel point of the heat source area is taken as the outward radiation direction of the edge pixel point, each edge pixel point is searched outwards according to the radiation direction until the searching of 0 pixel point is stopped, and the searching distance of each edge pixel point is recorded as the radiation length.
And S6, screening the heat radiation area set to be cleaned from the heat radiation area set according to the heat source areas and the corresponding radiation length sets of the heat radiation areas in each heat radiation area in the heat radiation area set.
In some embodiments, the set of heat radiation areas to be cleaned may be selected from the set of heat radiation areas according to the heat source area and the set of radiation lengths corresponding to the heat radiation areas included in each of the set of heat radiation areas.
The heat radiation area to be cleaned in the heat radiation area set to be cleaned may be an area to be cleaned.
As an example, this step may include the steps of:
the first step, determining an identification index to be cleaned corresponding to the heat radiation area according to a heat source area included in each heat radiation area in the heat radiation area set and a radiation length set corresponding to the heat radiation area.
For example, the formula corresponding to the identification index to be cleaned corresponding to the heat radiation area may be determined as follows:
f is the identification index to be cleaned corresponding to the heat radiation area.th() Is a hyperbolic tangent function.NIs the number of gray values in the heat source region included in the heat radiation region. The number of gray values in the heat source region may be the number of different gray values in the heat source region. For example, the different gray values in the heat source regions may be 180, 200, 210, and 240. At this time, the number of gray values in the heat source region may be 4.vIs the number of the gradation value in the heat source region included in the heat radiation region.
The corresponding gray value in the heat source region included in the heat radiation region is equal to the firstvThe number of pixels of the individual gray values.HIs the number of pixel points in the heat source region included in the heat radiation region.ln() Is a logarithmic function based on natural constants. />
Is the standard deviation of the radiation length in the radiation length set corresponding to the heat radiation area. />
To take the euclidean norm function. />
To take an absolute function.AAndBis a preset value. For example, a->
。/>
。/>
Is the average value of the gray values corresponding to the pixel points in the heat source region included in the heat radiation region. / >
Is the maximum value of the gray values corresponding to the pixel points in the heat source region included in the heat radiation region.
In actual situations, as the corresponding condition that the thermal infrared image is brighter is often existed when the components in the machine room equipment are in normal operation, the condition of overheat generated by dust accumulation can be confused. When the components are in normal operation, the gray value of the heat source region of the thermal infrared image tends to be uniform, and the heat radiation range from the edge of the heat source region to the periphery tends to be uniform, but when dust is randomly accumulated on the components, the different accumulation densities tend to have different interference degrees on the internal circuits of the components, so that the components generated by dust accumulation are overheated, the inside of the heat source region tends to be uneven, and the heat radiation range from the edge of the heat source region to the periphery tends to be more different. The smaller the identification index F to be cleaned corresponding to the heat radiation area, the more likely the heat source area included in the heat radiation area is a dust accumulation area.
May be entropy valueThe square of the gray level in the heat source region may be represented as the expansion entropy of the gray level in the heat source region, since the gray level corresponding to the pixel point having the gray level smaller than the adaptive cut threshold in the target thermal infrared image is set to 0 in the foregoing, the entropy may be squared in order to emphasize the uniformity of the gray level in the heat source region, when the interior of the heat source region is almost completely uniform, and the entropy is smaller than 1, the expansion entropy after the square is smaller, but when the heat source region has relatively large unevenness, the expansion entropy after the square tends to be larger as the entropy is larger. The greater the expansion entropy, the more it tends to overheat the components caused by dust accumulation. / >
For normalization with the hyperbolic tangent function. Standard deviation of radiation length in radiation length set corresponding to heat radiation area +.>
The larger the heat radiation range from the edge of the heat source region to the periphery, the larger the difference, and the overheat of the component caused by dust accumulation tends to be more prone. />
For normalization with the hyperbolic tangent function.
Can represent the Euclidean norm calculated by the expansion entropy and the standard deviation of the radiation lengths in the set of radiation lengths, with a maximum of +.>
。/>
The larger the gray value in the heat source region, the more the gray value tends to be biased, which means that the higher the gray value in the heat source region, the more the abnormal heat dissipation due to dust accumulation tends to be likely. />
The more toward 1, the more likely the heat source region is the component region of dust accumulation. Therefore (S)>
The smaller the heat source area is, the more likely it is that dust accumulation area, i.e., the more likely it is that the heat source area is a heat radiation area to be cleaned.
And a second step of screening a heat radiation area set to be cleaned from the heat radiation area set according to a preset threshold value to be cleaned and a recognition index to be cleaned corresponding to each heat radiation area in the heat radiation area set.
The threshold to be cleaned may be a preset threshold. For example, the threshold to be cleaned may be 0.2.
For example, when the identification index to be cleaned corresponding to the heat radiation area in the heat radiation area set is less than or equal to the threshold to be cleaned, determining the heat source area included in the heat radiation area as the heat radiation area to be cleaned.
And S7, determining target cleaning position information corresponding to the equipment in the machine room to be cleaned according to the heat radiation area set to be cleaned, and cleaning the position corresponding to the target cleaning position information.
In some embodiments, the target cleaning position information corresponding to the equipment in the machine room to be cleaned may be determined according to the set of heat radiation areas to be cleaned, and the position corresponding to the target cleaning position information may be cleaned.
As an example, the positions of the heat radiation areas to be cleaned in the heat radiation area set to be cleaned are combined to form target cleaning position information corresponding to the equipment in the machine room to be cleaned, and the positions corresponding to the target cleaning position information are cleaned. For example, the heat radiation area to be cleaned may be cleaned directly.
In summary, the invention proposes that the boundary of the heat source is fuzzy due to the radiation range of the heat source on the target thermal infrared image of the equipment to be cleaned, the position of the heat source cannot be accurately positioned, and the heat accumulation of the heat source is difficult to judge due to dust accumulation or other reasons. According to the invention, through cutting the image gray scale, redundant calculation can be reduced, seed pixel points and growth rules are set on the cut image by utilizing median filtering to realize the segmentation of the fuzzy region, the heat source region is obtained, a formula for determining the identification index to be cleaned is constructed, the heat source forming reason can be distinguished, the dust accumulation region can be accurately positioned, the electrified cleaning of the machine room equipment is more efficient, and clean resources are saved.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting thereof; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the scope of the embodiments of the present application, and are intended to be included within the scope of the present application.
Claims (9)
1. The auxiliary positioning method for the electrified cleaning of the equipment in the machine room is characterized by comprising the following steps of:
acquiring a target thermal infrared image of equipment in a machine room to be cleaned, and determining a gray level histogram of the target thermal infrared image;
performing image data processing on the gray level histogram, and determining an adaptive cutting threshold;
screening a highlight pixel point set from the target thermal infrared image according to the self-adaptive cutting threshold;
screening and growing the highlight pixel points in the highlight pixel point set to obtain a heat radiation area set, wherein the heat radiation area in the heat radiation area set comprises: a heat source region and a radiation range region;
For each heat radiation area in the heat radiation area set, selecting edge pixel points from a heat source area and a radiation range area included in the heat radiation area, and determining a radiation length set corresponding to the heat radiation area according to the selected edge pixel points;
according to the heat source areas included in each heat radiation area in the heat radiation area set and the corresponding radiation length sets of the heat radiation areas, selecting a heat radiation area set to be cleaned from the heat radiation area set;
determining target cleaning position information corresponding to the equipment in the machine room to be cleaned according to the heat radiation area set to be cleaned, and cleaning a position corresponding to the target cleaning position information;
the step of screening and growing the highlight pixel points in the highlight pixel point set to obtain a heat radiation area set comprises the following steps:
median filtering is carried out on the highlight pixel points in the highlight pixel point set, and a filtering value corresponding to the highlight pixel points is obtained;
when the filtering value corresponding to the highlight pixel is equal to the gray value corresponding to the highlight pixel, determining the highlight pixel as a seed pixel;
determining a corresponding extremely poor value of the seed pixel point according to a preset target window corresponding to the seed pixel point;
Performing region growth according to the corresponding extreme difference value of each seed pixel point to obtain a target region set;
determining the average value of gray values corresponding to pixel points in each target area in the target area set as the gray average value corresponding to the target area;
screening a target area group set of a preset target arrangement mode from the target area set, wherein each target area group in the target area group set comprises two target areas;
for each target region group in the target region group set, determining a target region with a larger gray average value in the target region group as a heat source region, determining a target region with a smaller gray average value in the target region group as a radiation range region, and determining regions where two target regions in the target region group are located as heat radiation regions.
2. The auxiliary positioning method for live cleaning of equipment in a machine room according to claim 1, wherein the performing image data processing on the gray level histogram to determine an adaptive cutting threshold includes:
determining an average brightness index and a deviation brightness index according to the gray level histogram;
Determining the ratio of the average brightness index to the deflection brightness index as an overall deflection brightness index;
when the overall deflection brightness index is larger than or equal to a preset brightness uniformity value, determining the deflection brightness index as a self-adaptive cutting threshold;
and when the overall deviation brightness index is smaller than the brightness uniformity value, determining the average brightness index as an adaptive cutting threshold.
3. The auxiliary positioning method for live cleaning of equipment in a machine room according to claim 2, wherein determining an average brightness index and a biased brightness index according to the gray level histogram comprises:
determining a gray value at a target position in the gray histogram as an average brightness index, wherein the absolute value of the difference between the number of pixel points before the target position and the number of pixel points after the target position is smaller than a preset number threshold;
and determining half of the maximum gray value in the gray histogram as a biased brightness index.
4. The auxiliary positioning method for live cleaning of equipment in a machine room according to claim 1, wherein the screening the set of highlight pixels from the target thermal infrared image according to the adaptive cutting threshold includes:
And determining the pixel point with the gray value larger than or equal to the self-adaptive cutting threshold value in the target thermal infrared image as a highlight pixel point.
5. The auxiliary positioning method for live cleaning of equipment in a machine room according to claim 1, wherein the step of screening edge pixel points from a heat source area and a radiation range area included in the heat radiation area, and determining a radiation length set corresponding to the heat radiation area according to the screened edge pixel points includes:
determining the edge pixel points screened from the heat source region included in the heat radiation region as heat source edge pixel points, and obtaining a heat source edge pixel point set corresponding to the heat radiation region;
the edge pixel points screened from the radiation range area included in the heat radiation area are determined to be radiation edge pixel points, and a radiation edge pixel point set corresponding to the heat radiation area is obtained;
determining the mass center of a heat source area included in the heat radiation area as the mass center corresponding to the heat radiation area;
connecting the mass centers of each heat source edge pixel point in the heat source edge pixel point set corresponding to the heat radiation area and the heat radiation area to obtain a heat source line segment set corresponding to the heat radiation area;
For each heat source line segment in the heat source line segment set corresponding to the heat radiation area, extending one end of the heat source line segment where a heat source edge pixel point is located until the radiation edge pixel points in the radiation edge pixel point set corresponding to the heat radiation area intersect to obtain an intersection point, and taking a line segment between the centroid in the heat source line segment and the intersection point as the heat radiation line segment corresponding to the heat source line segment;
and for each heat source line segment in the heat source line segment set corresponding to the heat radiation area, determining the difference value between the length of the heat radiation line segment corresponding to the heat source line segment and the length of the heat source line segment as the radiation length.
6. The auxiliary positioning method for live cleaning of equipment in a machine room according to claim 1, wherein the step of screening the set of heat radiation areas to be cleaned from the set of heat radiation areas according to the heat source areas included in each heat radiation area in the set of heat radiation areas and the set of radiation lengths corresponding to the heat radiation areas includes:
determining an identification index to be cleaned corresponding to the heat radiation area according to a heat source area included in each heat radiation area in the heat radiation area set and a radiation length set corresponding to the heat radiation area;
And screening the heat radiation area set to be cleaned from the heat radiation area set according to a preset threshold to be cleaned and the recognition indexes to be cleaned corresponding to each heat radiation area in the heat radiation area set.
7. The auxiliary positioning method for live cleaning of equipment in a machine room according to claim 6, wherein the formula corresponding to the identification index to be cleaned corresponding to the heat radiation area is determined as follows:
wherein F is an identification index to be cleaned corresponding to the heat radiation area, th () is a hyperbolic tangent function, N is the number of gray values in the heat source area included in the heat radiation area, v is the number of gray values in the heat source area included in the heat radiation area,
is the number of pixels in the heat source region included in the heat radiation region, the corresponding gray value of which is equal to the v-th gray value, H is the number of pixels in the heat source region included in the heat radiation region, ln () is a logarithmic function based on a natural constant, and>
is the standard deviation of the radiation length in the radiation length set corresponding to the heat radiation area, < + >>
To take the European norm function, ++>
To take the absolute function, A and B are preset values, < > >
Is the average value of gray values corresponding to pixel points in a heat source region included in the heat radiation region,/or>
Is the maximum value of the gray values corresponding to the pixel points in the heat source region included in the heat radiation region.
8. The auxiliary positioning method for live cleaning of equipment in a machine room according to claim 6, wherein the selecting the set of heat radiation areas from the set of heat radiation areas according to the preset threshold to be cleaned and the identification indicators to be cleaned corresponding to each heat radiation area in the set of heat radiation areas includes:
and when the identification index to be cleaned corresponding to the heat radiation area in the heat radiation area set is smaller than or equal to the threshold to be cleaned, determining the heat source area included in the heat radiation area as the heat radiation area to be cleaned.
9. The auxiliary positioning method for live cleaning of equipment in a machine room according to claim 8, wherein the determining, according to the set of heat radiation areas to be cleaned, target cleaning position information corresponding to the equipment in the machine room to be cleaned includes:
and combining the positions of the heat radiation areas to be cleaned in the heat radiation area set to be cleaned into target cleaning position information corresponding to the equipment in the machine room to be cleaned.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310120232.XA CN115830106B (en) | 2023-02-16 | 2023-02-16 | Auxiliary positioning method for electrified cleaning of equipment in machine room |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310120232.XA CN115830106B (en) | 2023-02-16 | 2023-02-16 | Auxiliary positioning method for electrified cleaning of equipment in machine room |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115830106A CN115830106A (en) | 2023-03-21 |
| CN115830106B true CN115830106B (en) | 2023-04-25 |
Family
ID=85521535
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310120232.XA Active CN115830106B (en) | 2023-02-16 | 2023-02-16 | Auxiliary positioning method for electrified cleaning of equipment in machine room |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115830106B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116862918B (en) * | 2023-09-05 | 2023-11-24 | 浙江奔月电气科技有限公司 | Real-time detection method for condensation inside ring main unit based on artificial intelligence |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1489557A2 (en) * | 2003-06-10 | 2004-12-22 | hema electronic GmbH | Method for adaptive edge detection |
| CN115633259A (en) * | 2022-11-15 | 2023-01-20 | 深圳市泰迅数码有限公司 | Automatic regulation and control method and system for intelligent camera based on artificial intelligence |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2365209C (en) * | 2001-05-09 | 2012-03-20 | National Research Council Of Canada | Method for micro-fabricating a pixelless infrared imaging device |
| DE102018218726A1 (en) * | 2018-10-31 | 2020-04-30 | Robert Bosch Gmbh | Method for operating a thermal imaging camera and thermal imaging camera |
| WO2021217643A1 (en) * | 2020-04-30 | 2021-11-04 | 深圳市大疆创新科技有限公司 | Method and device for infrared image processing, and movable platform |
| CN113573608B (en) * | 2020-05-09 | 2022-06-14 | 深圳汝原科技有限公司 | Drying apparatus |
| CN115060374A (en) * | 2022-06-10 | 2022-09-16 | 汉得利（常州）电子股份有限公司 | Infrared thermopile sensor capable of improving absorption efficiency and MEMS process manufacturing method thereof |
| CN115439474B (en) * | 2022-11-07 | 2023-01-24 | 山东天意机械股份有限公司 | Rapid positioning method for power equipment fault |
| CN115601630B (en) * | 2022-12-15 | 2023-03-28 | 山东天意装配式建筑装备研究院有限公司 | Stain recognition method for automatic wallboard mold cleaning machine |
-
2023
- 2023-02-16 CN CN202310120232.XA patent/CN115830106B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1489557A2 (en) * | 2003-06-10 | 2004-12-22 | hema electronic GmbH | Method for adaptive edge detection |
| CN115633259A (en) * | 2022-11-15 | 2023-01-20 | 深圳市泰迅数码有限公司 | Automatic regulation and control method and system for intelligent camera based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115830106A (en) | 2023-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11221107B2 (en) | Method for leakage detection of underground pipeline corridor based on dynamic infrared thermal image processing | |
| CN110458839B (en) | Effective wire and cable monitoring system | |
| CN104966304B (en) | Multi-target detection tracking based on Kalman filtering and nonparametric background model | |
| CN106296670B (en) | A kind of Edge detection of infrared image based on the watershed Retinex--Canny operator | |
| CN115830106B (en) | Auxiliary positioning method for electrified cleaning of equipment in machine room | |
| Zakaria et al. | Object shape recognition in image for machine vision application | |
| CN113902641B (en) | Data center hot zone judging method and system based on infrared image | |
| CN108765406A (en) | A kind of snow mountain detection method based on infrared remote sensing image | |
| CN111898486A (en) | Method and device for detecting abnormity of monitoring picture and storage medium | |
| CN113609901A (en) | Power transmission and transformation equipment fault monitoring method and system | |
| CN110223344A (en) | A kind of infrared small target detection method based on morphology and vision noticing mechanism | |
| CN110490872A (en) | The foreign matter detecting method and system of process equipment | |
| CN117575953B (en) | Detail enhancement method for high-resolution forestry remote sensing image | |
| Murari et al. | Algorithms for the automatic identification of MARFEs and UFOs in JET database of visible camera videos | |
| CN109815784A (en) | A kind of intelligent method for classifying based on thermal infrared imager, system and storage medium | |
| Saranyaraj | Image de-noising and edge segmentation using bilateral filtering and gabor-cut for edge representation of a breast tumor | |
| WO2024016632A1 (en) | Bright spot location method, bright spot location apparatus, electronic device and storage medium | |
| CN112037137A (en) | Method and device for eliminating fuzzy region of insulator disc surface edge in infrared image | |
| Luzi et al. | Automatic evaluation of vertex structural defects on the anode surface of a low-light-level image intensifier based on proposed individual image processing strategies | |
| Pathak et al. | Evaluation of effect of pre-processing techniques in solar panel fault detection | |
| CN116862918B (en) | Real-time detection method for condensation inside ring main unit based on artificial intelligence | |
| Sasikala et al. | An adaptive edge detecting method for satellite imagery based on canny edge algorithm | |
| Zhang et al. | A novel method for low illumination image | |
| Sadic et al. | Enhancement Technique of Infrared Images | |
| Kurniastuti et al. | Segmentation of Finger Nails Image based on Image Processing methods |
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 |