CN115131351B - Engine oil radiator detection method based on infrared image - Google Patents
Engine oil radiator detection method based on infrared image Download PDFInfo
- Publication number
- CN115131351B CN115131351B CN202211049400.2A CN202211049400A CN115131351B CN 115131351 B CN115131351 B CN 115131351B CN 202211049400 A CN202211049400 A CN 202211049400A CN 115131351 B CN115131351 B CN 115131351B
- Authority
- CN
- China
- Prior art keywords
- sliding window
- gray
- pixel point
- window corresponding
- pixel
- 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
- 239000010705 motor oil Substances 0.000 title claims abstract description 43
- 238000001514 detection method Methods 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 36
- 230000000694 effects Effects 0.000 claims abstract description 20
- 230000011218 segmentation Effects 0.000 claims abstract description 20
- 230000017525 heat dissipation Effects 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 11
- 239000003921 oil Substances 0.000 claims description 7
- 239000013598 vector Substances 0.000 description 30
- 238000012546 transfer Methods 0.000 description 8
- 238000001914 filtration Methods 0.000 description 6
- 230000005855 radiation Effects 0.000 description 5
- 102100029469 WD repeat and HMG-box DNA-binding protein 1 Human genes 0.000 description 2
- 101710097421 WD repeat and HMG-box DNA-binding protein 1 Proteins 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 230000000191 radiation effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
Abstract
The invention relates to the field of image data processing, in particular to an engine oil radiator detection method based on infrared images, which comprises the following steps: acquiring an infrared gray scale map of an engine region; performing sliding window on the gray level image, and obtaining a sliding window corresponding to each pixel point by using the variance of the gray level values of the pixel points in the sliding window corresponding to the starting point; acquiring a denoised gray scale image: when the sliding window corresponding to the pixel point is not filled, taking the gray value mean value of the pixel point in the sliding window as the gray value of the pixel point after denoising; when the sliding window corresponding to the pixel point is filled and a heat conduction direction line segment exists in the sliding window, denoising the pixel point by utilizing the variance of the gray value of the pixel point in the sliding window; when no heat conduction direction line segment exists in the sliding window, denoising the pixel points by utilizing the gray value of the normal pixel points in the sliding window; and carrying out threshold segmentation on the de-noised gray scale image, and detecting the engine oil radiator by using the obtained high-temperature region. The method is used for detecting the heat dissipation effect of the engine oil radiator, and can improve the detection accuracy.
Description
Technical Field
The invention relates to the field of image data processing, in particular to an engine oil radiator detection method based on infrared images.
Background
The engine oil radiator is also called as an oil cooler and is used for cooling the engine oil of the automobile engine. The good heat dissipation effect of the engine oil radiator plays a vital role in improving the working efficiency of equipment or devices and guaranteeing the safety of the equipment. Therefore, it is necessary to detect the heat radiation effect of the oil radiator.
At present, the method for detecting the heat dissipation effect of the engine oil radiator mainly comprises the following steps: the method comprises the steps of firstly collecting an infrared image of an automobile engine in a working state, then carrying out denoising treatment on the infrared image, carrying out threshold segmentation on the denoised infrared image to obtain a high-temperature region in the infrared image, and finally judging the heat dissipation effect of the engine oil radiator according to the high-temperature region.
However, when the heat dissipation effect of the engine oil radiator is detected at present, denoising methods of a spatial domain method and a frequency domain method are mainly adopted to denoise the infrared image, and these denoising methods perform the same denoising processing on all pixel points in the infrared image, which may cause excessive loss of edge details, make the edge of the denoised infrared image blurred, easily cause inaccurate threshold segmentation, and reduce the accuracy of the heat dissipation effect detection of the engine oil radiator.
Disclosure of Invention
The invention provides an oil radiator detection method based on infrared images, which aims to solve the problem of low accuracy of the existing oil radiator detection method.
In order to achieve the purpose, the invention adopts the following technical scheme that the method for detecting the engine oil radiator based on the infrared image comprises the following steps:
s1: acquiring an infrared gray scale map of an automobile engine region;
s2: obtaining a sliding window corresponding to each pixel point in the gray-scale image:
s201: setting a sliding window, performing sliding window traversal on the gray level image, and obtaining a sliding window corresponding to a pixel point traversed by a second sliding window by using the variance of gray levels of all pixel points in the sliding window corresponding to the starting point;
s202: obtaining a sliding window corresponding to a pixel point traversed by a third sliding window by utilizing the variance of gray values of all pixel points in the sliding window corresponding to the pixel point traversed by the second sliding window, and repeating the steps to obtain a sliding window corresponding to each pixel point in the gray map;
s3: denoising the gray scale image by using a sliding window corresponding to each pixel point in the gray scale image to obtain a denoised gray scale image:
s301: when the sliding window corresponding to each pixel point is not filled with the noise, taking the mean value of the gray values of all the pixel points in the sliding window corresponding to the pixel point as the denoised gray value of the pixel point;
s302: when the sliding window corresponding to each pixel point is filled with the full data, whether the sliding window corresponding to the pixel point contains the heat conduction direction line segment or not is judged: when the sliding window corresponding to the pixel point contains the heat conduction direction line segment, denoising the pixel point by utilizing the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point;
when no heat conduction direction line segment exists in the sliding window corresponding to the pixel point, denoising the pixel point by utilizing the gray value of a normal pixel point in the sliding window corresponding to the pixel point;
s4: carrying out threshold segmentation on the denoised gray level image to obtain a high-temperature region in the denoised gray level image;
s5: and detecting the heat dissipation effect of the engine oil radiator by using the mean value of the gray values of all the pixel points in the high-temperature area.
According to the engine oil radiator detection method based on the infrared image, a sliding window corresponding to each pixel point in the gray-scale image is obtained according to the following mode:
dividing the infrared gray-scale image of the automobile engine area into a high-temperature area and a background area by utilizing an Otsu algorithm;
calculating the variance of gray values of all pixel points in the high-temperature region and the background region respectively, and taking the maximum value of the two variances as a variance threshold;
a square sliding window is arranged, and the size of the square sliding window is
And each pixel point in the infrared gray-scale map of the automobile engine area is taken as the center point of the sliding window; the value range of h is a positive integer;
taking a pixel point at the upper left corner of the infrared gray-scale map of the automobile engine region as an initial point, and traversing the infrared gray-scale map of the automobile engine region by using a square sliding window to obtain a sliding window corresponding to the initial point;
calculating the variance of gray values of all pixel points in a sliding window corresponding to the starting point;
utilizing a variance threshold value to judge the variance of the gray values of all pixel points in a sliding window corresponding to the starting point: when the variance of the gray values of all the pixel points in the sliding window corresponding to the starting point is smaller than the variance threshold, the size of the sliding window corresponding to the starting point is enlarged to be
Obtaining a sliding window corresponding to the expanded starting point, and taking the sliding window corresponding to the expanded starting point as a sliding window corresponding to a pixel point traversed by a second sliding window; when the variance of the gray values of all the pixel points in the sliding window corresponding to the starting point is greater than or equal to the variance threshold, reducing the size of the sliding window corresponding to the starting point to be
Obtaining a sliding window corresponding to the reduced starting point, and taking the sliding window corresponding to the reduced starting point as a sliding window corresponding to a pixel point traversed by a second sliding window;
and obtaining a sliding window corresponding to the pixel point traversed by the third sliding window according to the sliding window corresponding to the pixel point traversed by the second sliding window, and so on to obtain the sliding window corresponding to each pixel point in the infrared gray-scale map of the automobile engine area.
The process of judging whether a sliding window corresponding to the pixel point contains a heat conduction direction line segment is as follows:
acquiring the gray value of a middle row pixel point in a sliding window corresponding to each pixel point, sequentially calculating the gray value difference value of two adjacent pixel points from left to right, marking the gray value difference value larger than 0 as 1, marking the gray value difference value smaller than 0 as-1, and not marking the gray value difference value equal to 0 to obtain a gray value difference value sequence of the middle row pixel points;
judging the gray value difference value sequence of the pixel points in the middle row: when the gray value difference sequence of the pixel points in the middle row is only 1, the fact that the middle row has a heat conduction direction line segment is indicated, and the heat conduction direction line segment is from left to right; when the gray value difference sequence of the pixel points in the middle row is only-1, the fact that the middle row has a heat conduction direction line segment is indicated, and the heat conduction direction line segment is from right to left; when the gray value difference value sequence of the pixel points of the middle row has 1 and-1, the middle row has no heat conduction direction line segment;
judging whether the middle column and the two diagonal lines in the sliding window corresponding to each pixel point contain the heat conduction direction line segments or not according to a method for judging whether the middle row contains the heat conduction direction line segments or not;
judging the sliding window corresponding to each pixel point according to whether the middle row, the middle column and the two diagonal lines in the sliding window corresponding to each pixel point contain heat conduction direction line segments or not: when heat conduction direction line segments exist on the middle row, the middle column and the two diagonal lines, the fact that the sliding window corresponding to the pixel point contains the heat conduction direction line segments is indicated; when no heat conduction direction line segment exists on the middle row, the middle column and the two diagonal lines, it is indicated that no heat conduction direction line segment exists in the sliding window corresponding to the pixel point.
The method for detecting the engine oil radiator based on the infrared image is characterized in that the process of denoising all pixel points in a sliding window corresponding to the pixel points by using the variance of gray values of the pixel points is as follows:
when the sliding window corresponding to the pixel point contains the heat conduction direction line segment, calculating the variance of the gray values of all pixel points in the sliding window corresponding to the pixel point;
judging the variance of the gray values of all pixel points in the sliding window corresponding to the pixel points:
when the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is smaller than the variance threshold, calculating the mean value of the gray values of all the pixel points on the heat conduction direction line segment in the sliding window corresponding to the pixel point, and taking the mean value of the gray values of all the pixel points on the heat conduction direction line segment in the sliding window corresponding to the pixel point as the denoised gray value of the pixel point;
when the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel points is greater than or equal to the variance threshold, the following operations are carried out:
acquiring the maximum value of the gray value difference of two adjacent pixel points on the heat conduction direction line segment in the sliding window corresponding to the pixel point, disconnecting the two adjacent pixel points corresponding to the maximum value of the gray value difference, and dividing the heat conduction direction line segment into two line segments;
and obtaining the segment of the heat conduction direction segment where the pixel point is located, calculating the mean value of the gray values of all the pixel points on the segment of the heat conduction direction segment where the pixel point is located, and taking the mean value of the gray values of all the pixel points on the segment of the heat conduction direction segment where the pixel point is located as the de-noised gray value of the pixel point.
According to the method for detecting the engine oil radiator based on the infrared image, the normal pixel points in the sliding window corresponding to the pixel points are obtained as follows:
calculating the gray value average value of all pixel points in the sliding window corresponding to each pixel point, and removing the pixel points with the gray values different from the gray value average value in the sliding window to obtain the residual pixel points;
acquiring eight neighborhood pixel points of each residual pixel point in a sliding window, and taking the eight neighborhood pixel points with the gray values smaller than the gray value of the residual pixel points as first pixel points;
taking the direction of the residual pixel points to the first pixel point as a vector direction, and taking the gray value difference value of the residual pixel point and the first pixel point as a vector value to obtain all first heat conduction guide quantities of each residual pixel point;
summing all the first heat conduction vector quantities of each remaining pixel point to obtain the heat conduction vector quantity of each remaining pixel point;
calculating the sum of included angles between the heat conduction vector of each residual pixel point and the heat conduction vectors of other residual pixel points;
calculating the mean value of the included angles and the sum of the heat conduction vectors of all the remaining pixel points and the heat conduction guide quantities of other remaining pixel points, and taking the mean value as the included angle and the threshold value;
and taking the residual pixel points with the sum of the included angles of the heat conduction vector and the heat conduction guide quantities of other residual pixel points being less than or equal to the included angle and the threshold value as normal pixel points in the sliding window to obtain all the normal pixel points in the sliding window.
The method for detecting the engine oil radiator based on the infrared image is characterized in that the pixel point is denoised by utilizing the gray value of the normal pixel point in the sliding window corresponding to the pixel point as follows:
when no heat conduction direction line segment exists in the sliding window corresponding to the pixel point, calculating the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point;
judging the variance of the gray values of all pixel points in the sliding window corresponding to the pixel points:
when the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is smaller than the variance threshold, taking the mean of the gray values of all the normal pixel points in the sliding window corresponding to the pixel point as the denoised gray value of the pixel point;
when the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is greater than or equal to a variance threshold, the following steps are carried out:
sorting the gray values of all normal pixel points in a sliding window corresponding to the pixel point in a mode from large to small to obtain a gray value sequence of the normal pixel points;
sequentially calculating the gray value difference value of every two adjacent pixel points in the gray value sequence of the normal pixel points to obtain the gray value difference value sequence of the normal pixel points;
disconnecting two adjacent pixels corresponding to the maximum value in the gray value difference sequence of the normal pixels, and dividing the gray value sequence of the normal pixels into a high-temperature sequence and a background sequence;
judging the number of gray values in the high-temperature sequence and the background sequence: when the number of the gray values in the high-temperature sequence is greater than or equal to that of the gray values in the background sequence, taking the average value of all the gray values in the high-temperature sequence as the denoised gray value of the pixel point; and when the number of the gray values in the high-temperature sequence is smaller than that of the gray values in the background sequence, taking the mean value of all the gray values in the background sequence as the de-noised gray value of the pixel point.
According to the method for detecting the engine oil radiator based on the infrared image, the process of detecting the heat dissipation effect of the engine oil radiator is as follows:
carrying out threshold segmentation on the denoised automobile engine region infrared gray scale image by using an Otsu algorithm to obtain a high-temperature region in the denoised automobile engine region infrared gray scale image, and taking the high-temperature region in the denoised automobile engine region infrared gray scale image as a high-temperature region to be detected;
acquiring a high-temperature area in an automobile engine area infrared gray scale image when the engine oil radiator fails according to the method for acquiring the high-temperature area to be detected, and taking the high-temperature area in the automobile engine area infrared gray scale image when the engine oil radiator fails as a contrast high-temperature area;
calculating the mean value of gray values of all pixel points in the comparison high-temperature area, and taking the mean value of the gray values as a first gray value of the comparison high-temperature area;
taking the mean value of the first gray values of all the comparison high-temperature areas as a gray value threshold, and judging the high-temperature area to be measured: and when the mean value of the gray values of all the pixel points in the high-temperature area to be detected is greater than the threshold value of the gray values, the engine oil radiator is indicated to have a fault.
According to the engine oil radiator detection method based on the infrared image, the infrared gray-scale image of the automobile engine region is obtained according to the following mode:
collecting an infrared image of an automobile engine in a working state;
performing semantic segmentation on the infrared image of the automobile engine to obtain the infrared image of the automobile engine area;
and carrying out graying processing on the infrared image of the automobile engine area to obtain the infrared grayscale image of the automobile engine area.
The invention has the beneficial effects that: the method performs sliding window traversal on the infrared gray-scale image of the automobile engine region, sequentially obtains the sliding window corresponding to each pixel point in the gray-scale image by utilizing the variance of the gray-scale values of all the pixel points in the sliding window corresponding to the starting point, prevents the window from being too large or too small by obtaining the self-adaptive filtering window of each pixel point in the infrared gray-scale image of the automobile engine region, and can effectively improve the denoising effect of the pixel points. According to the characteristics of the sliding window corresponding to each pixel point in the gray-scale image, the pixel points in the gray-scale image are divided into different types, and different denoising methods are adopted for the pixel points of the different types, so that the denoising effect of the infrared image can be improved while the edge of the infrared image is protected. According to the invention, the accuracy of threshold segmentation of the infrared image is improved by improving the denoising effect of the infrared image, so that the accuracy of detection of the heat dissipation effect of the engine oil radiator is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of an engine oil radiator detection method based on infrared images according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of S2 in an engine oil radiator detection method based on an infrared image according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of S3 in the method for detecting an oil radiator based on an infrared image according to the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
The main purposes of the invention are: the method comprises the steps of processing collected engine infrared images by using a computer vision technology, designing the size of a self-adaptive filtering window, and carrying out denoising processing on different windows in different modes to realize denoising processing of the engine infrared images. And finally, segmenting a high-temperature area in the engine infrared image by using a threshold value, and judging whether the engine oil radiator has a fault or not.
Because the acquired engine infrared image has high noise, the existing denoising method does not consider the structural information of the image, so that the details of the image are lost too much, the image is blurred too much while denoising, and the accuracy of subsequent fault detection is influenced. Therefore, the invention provides the engine oil radiator detection method based on the infrared image, which is used for improving the denoising effect of the infrared image to improve the accuracy of threshold segmentation of the infrared image, and further improving the accuracy of detection of the heat dissipation effect of the engine oil radiator.
The embodiment of the invention relates to an engine oil radiator detection method based on an infrared image, which comprises the following steps of:
s1: and acquiring an infrared gray-scale image of the automobile engine region.
In the embodiment, an infrared image of an engine region is required, denoising processing is performed on the image, a high-temperature region is divided by a threshold value, and whether the engine oil radiator has a fault is judged. Therefore, an infrared thermal imager is needed to collect the infrared image of the automobile generator, and then the characteristic information of the engine in the image is identified.
The present embodiment employs a DNN semantic segmentation approach to identify objects in segmented infrared images.
The relevant content of the DNN network is as follows:
the used data set is an automobile engine infrared image data set acquired by an infrared thermal imager under the working state.
The pixel points to be segmented are divided into 2 types, namely the label labeling process corresponding to the training set is as follows: and in the single-channel semantic label, the pixel point at the corresponding position belongs to the background class and is marked as 0, and the pixel point belonging to the engine area is marked as 1.
The task of the network is classification, so the loss function used is a cross entropy loss function.
Therefore, the infrared image of the automobile engine is processed through the DNN, and the infrared image of the automobile engine area is obtained.
And carrying out graying processing on the infrared image of the automobile engine area to obtain the infrared grayscale image of the automobile engine area.
S2: and obtaining a sliding window corresponding to each pixel point in the gray-scale image.
It should be noted that: the image edge is blurred due to the fact that the filtering window is too large in size, and the denoising effect is affected due to the fact that the window is too small in size. According to the embodiment, a filtering window with a self-adaptive size is designed through the variance of the gray value of the pixel point, then different modes are used for denoising different windows, the image edge is protected, the denoising effect is improved, and the denoising of the engine infrared image is realized. The process of obtaining the sliding window corresponding to each pixel point in the gray-scale map is shown in fig. 2.
S201: and setting a sliding window, performing sliding window traversal on the gray-scale image, and obtaining the sliding window corresponding to the pixel point traversed by the second sliding window by utilizing the variance of gray values of all pixel points in the sliding window corresponding to the starting point.
The principle of the Dajin algorithm is to divide the image into two classes, so that the intra-class variance is minimum and the inter-class variance is maximum, and the optimal segmentation threshold is obtained. Firstly, a gray level histogram of an infrared gray level image of an automobile engine region is obtained, an optimal segmentation threshold value is obtained by utilizing an Otsu algorithm, and the infrared gray level image of the automobile engine region is roughly divided into two regions, namely a high-temperature region with a gray level value larger than the segmentation threshold value and a background region with a gray level value smaller than the segmentation threshold value. Calculating the variance of the gray values of all pixel points in the high-temperature region
And the variance of the gray values of all the pixel points in the background area
Taking the maximum value of the two as a variance threshold value
。
Then designing a square sliding window with odd size and size
. Taking the upper left corner pixel point of the engine infrared image as a starting point and moving from left to rightAnd traversing pixel points from top to bottom. And if the traversal pixel points are positioned at the edge of the image, only analyzing the image pixel points contained in the window.
When the square window slides pixel by pixel, the left column of the pixel in the next window is removed and the right column is added with a new column compared with the pixel in the previous window. Therefore, when the square window slides to the similar region with uniform gray value, the window should be enlarged to better suppress noise and smooth the similar region. When the square window slides to the inter-class area with uneven gray value, namely the edge area, the window is reduced to protect the image edge.
In this embodiment, the window size h has a value range of {1,2,3,4,5}. Let the window size of the initial pixel be
And at the moment, h is 3, and the variance V of the gray values of all the pixel points in the window is calculated.
If V is less than the variance threshold
Then, it indicates that the pixel points in the region are the same type of pixel points, and the gray value is uniform, so that the window of the next traversal pixel point is enlarged, and the size is
I.e. let h' = h +1. Where h 'has a maximum value of 5, the window is no longer increased when h' = 5.
If V is greater than or equal to the variance threshold
Then, it is indicated that the pixels in the area are two types of pixels, and the gray value is not uniform, so that the window of the next traversal pixel is reduced, and the size is
Instant command
And h-1. Wherein
Is 1 when
When =1, the window is no longer reduced.
Thus, the window size of the second traversal pixel is obtained, and the variance of all the gray values of the pixels in the window is calculated.
S202: and obtaining a sliding window corresponding to the pixel point traversed by the third sliding window by utilizing the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point traversed by the second sliding window, and so on to obtain the sliding window corresponding to each pixel point in the gray map.
And obtaining the window size of the third traversal pixel point according to the method for obtaining the window size of the second traversal pixel point, and so on to obtain the window size corresponding to each pixel point in the infrared gray-scale map of the automobile engine area.
Thus, the adaptive selection of the size of the filtering window is completed.
S3: and denoising the gray scale image by using a sliding window corresponding to each pixel point in the gray scale image to obtain the denoised gray scale image.
It should be noted that the infrared image is obtained by receiving infrared radiation from the target and the scenery by using a thermal infrared imager, converting invisible radiation into a visible image through photoelectric conversion, and the brightness change of each pixel point in the image corresponds to the change of the radiation energy intensity of the target and the scenery. Therefore, the brightness of the infrared image is changed from light to dark from a high-temperature area with strong radiation energy to a low-temperature area with weak radiation energy, and the gray value of the pixel point on the gray map is changed from large to small, namely the gray map has a heat conduction direction line segment. The process of denoising the gray scale map by using the sliding window corresponding to each pixel point in the gray scale map is shown in fig. 3.
S301: and when the sliding window corresponding to each pixel point is not filled with the noise, taking the mean value of the gray values of all the pixel points in the sliding window corresponding to the pixel point as the de-noised gray value of the pixel point.
If the window with the self-adaptive size is not filled with the pixels of the engine infrared image, namely the center pixel of the window is located at the contour edge of the engine infrared image, the contour edge of the image cannot contain a high-temperature area according to the heat conduction characteristic, and therefore the gray value of the center pixel is replaced by the mean gray value of all the pixels in the window.
S302: when the sliding window corresponding to each pixel point is filled with the full, judging whether the sliding window corresponding to the pixel point contains the heat conduction direction line segment or not: when the sliding window corresponding to the pixel point contains the heat conduction direction line segment, the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is utilized to denoise the pixel point.
If the window with the self-adaptive size is completely positioned in the engine infrared image, taking any filtering window as an example. Firstly, analyzing the gray value of the adjacent pixel point of the middle row in the window, sequentially solving the gray value difference value of the two adjacent pixel points from left to right, recording the mark with the difference value as positive as 1, the mark with the difference value as negative as-1 and the mark with the difference value as 0, and obtaining the following sequence:
when the sequence has only 1, the heat conduction direction line segment of the middle row in the window is from left to right, when the sequence has only-1, the heat conduction direction line segment of the middle row in the window is from right to left, and when the sequence has 1 and-1, the heat conduction direction line segment of the middle row in the window is not heat conduction direction.
And similarly, judging that the middle column and the two diagonal lines in the window have non-heat conduction direction line segments.
If any line segment in the middle row, column and two diagonal lines in the window has a heat conduction direction line segment, it indicates that no noise point or noise point in the window is not on the heat conduction direction line segment. And then analyzing the variance V of the gray values of all the pixel points in the window:
if V is less than the variance threshold
When only one segment in the heat conduction direction exists, the mean value of the gray values of all the pixel points on the segment is calculated, and the mean value is used for replacing the gray value of the central pixel point in the window. When a plurality of heat conduction direction line segments exist, the gray value average values of all pixel points on the line segments are respectively calculated, and then the average values are obtained for replacement. The method is used for well performing smooth denoising on the intra-class area.
If V is greater than or equal to the variance threshold
And then, the window is explained to contain two types of pixel points, and the gray value of the pixel points is uneven, namely the window contains edges. When only one heat conduction direction line segment exists, the gray value difference of two adjacent pixel points is calculated in sequence according to the heat conduction direction line segment, the line segment is separated from the maximum difference, and whether the central pixel point in the window is in the front section or the rear section is judged. And calculating the mean value of the gray values of all the pixel points on the line segment where the central pixel point is located in the window so as to replace the gray value of the central pixel point in the window. When a plurality of heat conduction direction line segments exist, the mean value of the gray value is calculated respectively in the same way, and then the mean value is obtained for replacement. The method is used for denoising and protecting the inter-class edges.
When no heat conduction direction line segment exists in the sliding window corresponding to the pixel point, denoising the pixel point by utilizing the gray value of the normal pixel point in the sliding window corresponding to the pixel point.
If the middle row, column and two diagonal lines are within the window, there are no heat conducting direction line segments in these four line segments. It means that the window must contain noise points, and the noise points are the central pixel points or the noise points are on the line segment of the heat conduction direction. And then analyzing the variance V of the gray values of all the pixel points in the window:
if V is less than the variance threshold
And then, the pixels in the window are the same type of pixels, and the gray value of the pixels is relatively uniform. And traversing pixel by pixel in the window, analyzing the gray level change of the traversed pixel and 8 adjacent pixels, and only analyzing adjacent pixels in the window if the traversed pixel is at the edge of the window.
Removing obvious noise points with gray values larger or smaller than 8 neighborhood pixel points in the window, and then calculating heat conduction vectors of the rest pixel points: and taking the direction of the pixel point decreasing to the gray value in the 8 neighborhoods thereof as a vector direction and the decreasing gray value as a vector value to obtain a vector, and then taking the sum of the vectors to represent the heat conduction guiding quantity of the pixel point. Obtaining a set of heat transfer direction quantities
And m represents the number of the residual pixel points in the window, namely the number of vectors in the heat conduction vector set.
The heat conduction direction line segments of the known noise points are disordered, and the heat conduction of the normal pixel points is similar. So as to calculate the sum of the included angle between the first heat conduction guide quantity and the residual heat conduction guide quantity
Comprises the following steps:
wherein
Representation collection
The first amount of heat transfer to be conducted,
representation collection
The x-th heat conduction vector, wherein the value range of x is [2,m ]]。
Representation collection
The vector value of the first heat transfer vector,
representation collection
The vector value of the xth heat transfer vector. m represents a set
The number of medium vectors.
Representing an inverse cosine function. An included angle between the first heat conduction guide quantity and the residual heat conduction guide quantity is obtained through calculation of an arccosine function, and then the sum of the included angles of the first heat conduction guide quantity and the residual heat conduction guide quantity is obtained through summation of all the included angles, so that whether the residual pixel points corresponding to the first heat conduction guide quantity are normal pixel points or not is judged. The reason why the heat conduction vectors of the normal pixels are similar and the heat conduction direction line segments of the noise points are disordered is that the heat conduction vectors of the pixels are utilized to judge whether the pixels are normal pixels.
In the collection
Sequentially calculating the included angle between one heat transfer guide quantity and the rest heat transfer guide quantity to obtain a set
. When the heat conduction vector of the noise point is the main body, the included angle between the noise point and most heat conduction vectors is larger, so that the noise point corresponds to the heat conduction vectorSum of included angles of heat transfer guide quantity and residual heat transfer guide quantity
Is relatively large. When the heat conduction vector of the normal pixel point is the main body, the included angle between the heat conduction vector and most heat conduction vectors is smaller, so the sum of the included angles between the corresponding heat conduction vector and the residual heat conduction vector
Is smaller.
Get set
Has a mean value of
Judgment of
The corresponding pixel point is a normal pixel point. And calculating the mean value of the gray values of the normal pixel points so as to replace the gray value of the central pixel point in the window.
If V is greater than or equal to the variance threshold
And then, the window is explained to contain two types of pixel points, and the gray value of the pixel points is uneven, namely the window contains edges. Obtaining normal pixel points in the window, sequencing the gray values of the normal pixel points from large to small to obtain a gray value sequence
Where y represents the number of normal pixels within the window. Then, gray value difference values of two adjacent pixel points in the sequence are sequentially calculated from left to right, and a gray value difference value sequence G = &isobtained
Get the maximum value of the sequence
. According to
Corresponding adjacent two pixel point sequence
High-temp sequence divided into large grey values
Great and background sequence with small gray value
}。
The known high-temperature area is a circular connected area, when the central pixel point of the window belongs to the high-temperature area, more than half of the window is the pixel point of the high-temperature area, and when the central pixel point of the window belongs to the background area, more than half of the window is the pixel point of the background area.
Therefore, the replacement value H of the gray value of the central pixel point in the window is:
wherein q denotes a sequence of gray values
And dividing the window into two types of segmentation serial numbers, wherein y represents the number of normal pixel points in the window.
Representing a sequence of grey values
The ith gray scale value of (1). The meaning here is: when the number of the gray values in the high-temperature sequence is more than or equal to that in the background sequence, taking the average value of all the gray values in the high-temperature sequence as the imageThe gray value of the denoised prime points; and when the number of the gray values in the high-temperature sequence is smaller than that of the gray values in the background sequence, taking the average value of all the gray values in the background sequence as the denoised gray value of the pixel point. The central pixel point is denoised according to the gray value distribution condition of the pixels in the neighborhood of the central pixel point, and the denoising effect is better.
Therefore, the replacement of the gray value of the central pixel point in the window is completed. And completing denoising processing of the infrared image of the engine.
S4: and carrying out threshold segmentation on the denoised grey-scale image to obtain a high-temperature region in the denoised grey-scale image.
Acquiring a denoised engine infrared image, solving an optimal segmentation threshold by using an Otsu algorithm, obtaining a high-temperature area according to pixel points of which the gray values are greater than the optimal segmentation threshold, and calculating a gray value mean value P of the high-temperature area.
S5: and detecting the heat dissipation effect of the engine oil radiator by using the mean value of the gray values of all the pixel points in the high-temperature area.
Taking an infrared image of the automobile engine in a working state that 20 engine oil radiators have faults, obtaining a mean value P' of the gray values of a high-temperature region according to a method for obtaining the high-temperature region in a de-noised gray image, and taking the mean value of the group of data as
When the mean value P of the gray value of the high-temperature area in the de-noised gray image is larger than the threshold value of the gray value, taking the mean value P as the threshold value of the gray value
And judging that the heat radiation effect of the engine oil radiator is poor, and indicating that the engine oil radiator has a fault.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (3)
1. An engine oil radiator detection method based on infrared images is characterized by comprising the following steps:
s1: acquiring an infrared gray scale map of an automobile engine region;
s2: obtaining a sliding window corresponding to each pixel point in the gray-scale image:
s201: setting a sliding window, performing sliding window traversal on the gray-scale image, and obtaining a sliding window corresponding to a pixel point traversed by a second sliding window by using the variance of gray values of all pixel points in the sliding window corresponding to the starting point;
s202: obtaining a sliding window corresponding to a pixel point traversed by a third sliding window by utilizing the variance of gray values of all pixel points in the sliding window corresponding to the pixel point traversed by the second sliding window, and so on to obtain a sliding window corresponding to each pixel point in the gray map;
the sliding window corresponding to each pixel point in the gray-scale map is obtained according to the following mode:
dividing the infrared gray-scale image of the automobile engine area into a high-temperature area and a background area by utilizing an Otsu algorithm;
respectively calculating the variances of the gray values of all the pixel points in the high-temperature region and the background region, and taking the maximum value of the two variances as a variance threshold;
a square sliding window is arranged, and the size of the square sliding window is
And each pixel point in the infrared gray level image of the automobile engine area is taken as the center point of the sliding window; the value range of h is a positive integer;
taking a pixel point at the upper left corner of the infrared gray scale image of the automobile engine region as an initial point, and performing sliding window traversal on the infrared gray scale image of the automobile engine region by using a square sliding window to obtain a sliding window corresponding to the initial point;
calculating the variance of gray values of all pixel points in a sliding window corresponding to the starting point;
utilizing a variance threshold value to judge the variance of the gray values of all pixel points in a sliding window corresponding to the starting point: the gray value of all pixel points in the sliding window corresponding to the starting pointWhen the difference is less than the variance threshold value, the size of the sliding window corresponding to the starting point is enlarged to
Obtaining a sliding window corresponding to the expanded starting point, and taking the sliding window corresponding to the expanded starting point as a sliding window corresponding to a pixel point traversed by a second sliding window; when the variance of the gray values of all the pixel points in the sliding window corresponding to the starting point is greater than or equal to the variance threshold, reducing the size of the sliding window corresponding to the starting point to be
Obtaining a sliding window corresponding to the reduced starting point, and taking the sliding window corresponding to the reduced starting point as a sliding window corresponding to a pixel point traversed by a second sliding window;
obtaining a sliding window corresponding to a pixel point traversed by a third sliding window according to the sliding window corresponding to the pixel point traversed by the second sliding window, and so on to obtain a sliding window corresponding to each pixel point in the infrared gray-scale map of the automobile engine area;
s3: denoising the gray scale image by using a sliding window corresponding to each pixel point in the gray scale image to obtain a denoised gray scale image:
s301: when the sliding window corresponding to each pixel point is not filled with the noise, taking the mean value of the gray values of all the pixel points in the sliding window corresponding to the pixel point as the denoised gray value of the pixel point;
s302: when the sliding window corresponding to each pixel point is filled with the full, judging whether the sliding window corresponding to the pixel point contains the heat conduction direction line segment or not: when the sliding window corresponding to the pixel point contains the heat conduction direction line segment, denoising the pixel point by utilizing the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point;
when no heat conduction direction line segment exists in the sliding window corresponding to the pixel point, denoising the pixel point by utilizing the gray value of a normal pixel point in the sliding window corresponding to the pixel point;
the process of judging whether the sliding window corresponding to the pixel point contains the heat conduction direction line segment is as follows:
acquiring the gray value of a middle row pixel point in a sliding window corresponding to each pixel point, sequentially calculating the gray value difference value of two adjacent pixel points from left to right, marking the gray value difference value larger than 0 as 1, marking the gray value difference value smaller than 0 as-1, and not marking the gray value difference value equal to 0 to obtain a gray value difference value sequence of the middle row pixel points;
judging the gray value difference value sequence of the pixel points in the middle row: when the gray value difference sequence of the pixel points in the middle row is only 1, the fact that the middle row has a heat conduction direction line segment is indicated, and the heat conduction direction line segment is from left to right; when the gray value difference sequence of the pixel points in the middle row is only-1, the fact that the middle row has a heat conduction direction line segment is indicated, and the heat conduction direction line segment is from right to left; when the gray value difference value sequence of the pixel points of the middle row has 1 or-1, the middle row has no heat conduction direction line segment;
judging whether the middle column and the two diagonal lines in the sliding window corresponding to each pixel point contain the heat conduction direction line segments or not according to a method for judging whether the middle row contains the heat conduction direction line segments or not;
judging the sliding window corresponding to each pixel point according to whether the middle row, the middle column and the two diagonal lines in the sliding window corresponding to each pixel point contain heat conduction direction line segments: when heat conduction direction line segments exist on the middle row, the middle column and the two diagonal lines, the fact that the sliding window corresponding to the pixel point contains the heat conduction direction line segments is indicated; when no heat conduction direction line segment exists on the middle row, the middle column and the two diagonal lines, the fact that no heat conduction direction line segment exists in the sliding window corresponding to the pixel point is indicated;
the process of denoising the pixel point by using the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is specifically as follows: when the sliding window corresponding to the pixel point contains the heat conduction direction line segment, calculating the variance of the gray values of all pixel points in the sliding window corresponding to the pixel point;
judging the variance of the gray values of all pixel points in the sliding window corresponding to the pixel points: when the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is smaller than a variance threshold, calculating the mean value of the gray values of all the pixel points on the heat conduction direction line segment in the sliding window corresponding to the pixel point, and taking the mean value of the gray values of all the pixel points on the heat conduction direction line segment in the sliding window corresponding to the pixel point as the de-noised gray value of the pixel point;
when the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is greater than or equal to the variance threshold, the following operations are carried out: acquiring the maximum value of the gray value difference value of two adjacent pixel points on the heat conduction direction line segment in the sliding window corresponding to the pixel point, disconnecting the two adjacent pixel points corresponding to the maximum value of the gray value difference value, and dividing the heat conduction direction line segment into two line segments;
acquiring a heat conduction direction line segment subsection where the pixel point is located, calculating the mean value of gray values of all the pixel points on the heat conduction direction line segment subsection where the pixel point is located, and taking the mean value of the gray values of all the pixel points on the heat conduction direction line segment subsection where the pixel point is located as the de-noised gray value of the pixel point;
the process of denoising the pixel point by using the gray value of the normal pixel point in the sliding window corresponding to the pixel point is as follows: when no heat conduction direction line segment exists in the sliding window corresponding to the pixel point, calculating the variance of all pixel point gray values in the sliding window corresponding to the pixel point;
judging the variance of the gray values of all pixel points in the sliding window corresponding to the pixel points: when the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is smaller than the variance threshold, taking the mean of the gray values of all the normal pixel points in the sliding window corresponding to the pixel point as the denoised gray value of the pixel point;
when the variance of the gray values of all the pixel points in the sliding window corresponding to the pixel point is greater than or equal to a variance threshold, the following steps are carried out: sorting the gray values of all normal pixel points in a sliding window corresponding to the pixel point in a mode from large to small to obtain a gray value sequence of the normal pixel points; sequentially calculating the gray value difference value of every two adjacent pixel points in the gray value sequence of the normal pixel points to obtain the gray value difference value sequence of the normal pixel points; disconnecting two adjacent pixels corresponding to the maximum value in the gray value difference sequence of the normal pixels, and dividing the gray value sequence of the normal pixels into a high-temperature sequence and a background sequence;
judging the number of gray values in the high-temperature sequence and the background sequence: when the number of the gray values in the high-temperature sequence is greater than or equal to that of the gray values in the background sequence, taking the average value of all the gray values in the high-temperature sequence as the denoised gray value of the pixel point; when the number of the gray values in the high-temperature sequence is smaller than that of the gray values in the background sequence, taking the average value of all the gray values in the background sequence as the denoised gray value of the pixel point;
s4: carrying out threshold segmentation on the denoised gray level image to obtain a high-temperature region in the denoised gray level image;
s5: and detecting the heat dissipation effect of the engine oil radiator by using the mean value of the gray values of all the pixel points in the high-temperature area.
2. The method for detecting the oil radiator based on the infrared image as claimed in claim 1, wherein the process for detecting the heat dissipation effect of the oil radiator is as follows:
carrying out threshold segmentation on the denoised automobile engine region infrared gray-scale image by utilizing an Otsu algorithm to obtain a high-temperature region in the denoised automobile engine region infrared gray-scale image, and taking the high-temperature region in the denoised automobile engine region infrared gray-scale image as a high-temperature region to be detected;
acquiring a high-temperature area in an infrared gray-scale image of an automobile engine area when the engine oil radiator fails according to the method for acquiring the high-temperature area to be detected, and taking the high-temperature area in the infrared gray-scale image of the automobile engine area when the engine oil radiator fails as a comparison high-temperature area;
calculating the mean value of gray values of all pixel points in the comparison high-temperature area, and taking the mean value of the gray values as a first gray value of the comparison high-temperature area;
taking the mean value of the first gray values of all the comparison high-temperature areas as a gray value threshold, and judging the high-temperature area to be measured: and when the mean value of the gray values of all the pixel points in the high-temperature area to be detected is greater than the threshold value of the gray values, the engine oil radiator is indicated to have a fault.
3. The engine oil radiator detection method based on the infrared image as claimed in claim 1, wherein the infrared gray-scale map of the automobile engine region is obtained as follows:
collecting an infrared image of an automobile engine in a working state;
performing semantic segmentation on the infrared image of the automobile engine to obtain the infrared image of the automobile engine area;
and carrying out graying processing on the infrared image of the automobile engine area to obtain the infrared grayscale image of the automobile engine area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211049400.2A CN115131351B (en) | 2022-08-31 | 2022-08-31 | Engine oil radiator detection method based on infrared image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211049400.2A CN115131351B (en) | 2022-08-31 | 2022-08-31 | Engine oil radiator detection method based on infrared image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115131351A CN115131351A (en) | 2022-09-30 |
| CN115131351B true CN115131351B (en) | 2022-11-18 |
Family
ID=83386998
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211049400.2A Active CN115131351B (en) | 2022-08-31 | 2022-08-31 | Engine oil radiator detection method based on infrared image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115131351B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330758B (en) * | 2022-10-11 | 2023-12-22 | 青岛信和钢结构有限公司 | Welding quality detection method based on denoising treatment |
| CN115471503B (en) * | 2022-11-03 | 2023-04-07 | 江西捷锐机电设备有限公司 | Equipment abnormity detection method for numerical control ingot splitting machine |
| CN116883412B (en) * | 2023-09-08 | 2023-11-17 | 浙江中骏石墨烯科技有限公司 | Graphene far infrared electric heating equipment fault detection method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113129326A (en) * | 2021-06-17 | 2021-07-16 | 苏州高视半导体技术有限公司 | Self-adaptive threshold segmentation method of gray level image |
| CN114972370A (en) * | 2022-02-08 | 2022-08-30 | 黄河水利职业技术学院 | Remote sensing image self-adaptive segmentation method for neural network reasoning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335947B (en) * | 2014-05-26 | 2019-03-01 | 富士通株式会社 | Image de-noising method and image denoising device |
| CN113313641B (en) * | 2021-04-28 | 2022-05-03 | 北京理工大学 | CT image denoising method with self-adaptive median filtering |
| CN114445387A (en) * | 2022-01-29 | 2022-05-06 | 泗阳富艺木业股份有限公司 | Fiberboard quality classification method based on machine vision |
-
2022
- 2022-08-31 CN CN202211049400.2A patent/CN115131351B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113129326A (en) * | 2021-06-17 | 2021-07-16 | 苏州高视半导体技术有限公司 | Self-adaptive threshold segmentation method of gray level image |
| CN114972370A (en) * | 2022-02-08 | 2022-08-30 | 黄河水利职业技术学院 | Remote sensing image self-adaptive segmentation method for neural network reasoning |
Non-Patent Citations (2)
| Title |
|---|
| Removal Of Low-DENSITY Impulse Noise From Grayscale Images Using New Modified Hybrid MedianFilTerWith Various Sliding Window Sizes;R. Sravani，and etc;《2018 3rd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)》;20200227;第1750-1754页 * |
| 基于偏微分中值滤波的巡检图像去噪研究;黄晶晶等;《光电技术应用》;20191031;第42-47页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115131351A (en) | 2022-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115131351B (en) | Engine oil radiator detection method based on infrared image | |
| CN107045634B (en) | Text positioning method based on maximum stable extremum region and stroke width | |
| CN111260616A (en) | Insulator crack detection method based on Canny operator two-dimensional threshold segmentation optimization | |
| CN110782477A (en) | Moving target rapid detection method based on sequence image and computer vision system | |
| CN114972326A (en) | Defective product identification method for heat-shrinkable tube expanding process | |
| CN109850518B (en) | Real-time mining adhesive tape early warning tearing detection method based on infrared image | |
| CN111598897B (en) | Infrared image segmentation method based on Otsu and improved Bernsen | |
| CN111539980B (en) | Multi-target tracking method based on visible light | |
| CN113780110A (en) | Method and device for detecting weak and small targets in image sequence in real time | |
| CN112529853A (en) | Method and device for detecting damage of netting of underwater aquaculture net cage | |
| WO2024016632A1 (en) | Bright spot location method, bright spot location apparatus, electronic device and storage medium | |
| CN110633705A (en) | Low-illumination imaging license plate recognition method and device | |
| CN116205939A (en) | Line extraction method, line extraction apparatus, and computer storage medium | |
| CN111223050A (en) | Real-time image edge detection algorithm | |
| CN115578629A (en) | Highway guardrail extraction method | |
| CN110807348A (en) | Method for removing interference lines in document image based on greedy algorithm | |
| CN113505811A (en) | Machine vision imaging method for hub production | |
| CN112949389A (en) | Haze image target detection method based on improved target detection network | |
| CN112528895A (en) | Computing method for alteration detection of financial bill image ticket number | |
| CN112287831A (en) | Following robot multi-target identification system and method based on coded thermal infrared mark | |
| CN113487496B (en) | Image denoising method, system and device based on pixel type inference | |
| Chehdi et al. | A blind system to identify and filter degradations affecting an image | |
| Soumya et al. | Enhancement and segmentation of historical records | |
| JP4230960B2 (en) | Image processing apparatus, image processing method, and image processing program | |
| CN116630205B (en) | Image processing method, intelligent terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Oil Radiator Detection Method Based on Infrared Image Effective date of registration: 20230626 Granted publication date: 20221118 Pledgee: Industrial and Commercial Bank of China Limited Weishan sub branch Pledgor: Weishan Hongjie Machinery Co.,Ltd. Registration number: Y2023980045492 |