CN112381848A - Anti-interference method of instrument visual reading on-line monitoring system in complex environment - Google Patents

Anti-interference method of instrument visual reading on-line monitoring system in complex environment Download PDF

Info

Publication number
CN112381848A
CN112381848A CN202011216139.1A CN202011216139A CN112381848A CN 112381848 A CN112381848 A CN 112381848A CN 202011216139 A CN202011216139 A CN 202011216139A CN 112381848 A CN112381848 A CN 112381848A
Authority
CN
China
Prior art keywords
image
pointer
area
target
shadow
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.)
Granted
Application number
CN202011216139.1A
Other languages
Chinese (zh)
Other versions
CN112381848B (en
Inventor
伍家成
林鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Research Institute of Zhejiang University Taizhou
Original Assignee
Research Institute of Zhejiang University Taizhou
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Research Institute of Zhejiang University Taizhou filed Critical Research Institute of Zhejiang University Taizhou
Priority to CN202011216139.1A priority Critical patent/CN112381848B/en
Publication of CN112381848A publication Critical patent/CN112381848A/en
Application granted granted Critical
Publication of CN112381848B publication Critical patent/CN112381848B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation

Abstract

The invention discloses an anti-interference method of an instrument visual reading online monitoring system in a complex environment, which comprises the following steps: receiving image data returned by a sensor, and preprocessing the image to obtain a required gray-scale image; extracting the image edge by using a canny operator to obtain an edge image; detecting image features by using Hough transform, and respectively extracting a circle representing a dial plate and a straight line representing a pointer in a target image to obtain all data of circular edges and straight line edges; screening the circular edge to obtain a target circular edge representing the dial plate; screening the straight line edges to obtain a target straight line edge representing the pointer; intercepting the image around the pointer to obtain an image of the area around the pointer; comparing the area of the pointer region with the distribution characteristics of the gray level histogram, and judging whether shadow interference exists or not to obtain a shadow detection result; and removing the interference of the shadow part to obtain an image without shadow interference.

Description

Anti-interference method of instrument visual reading on-line monitoring system in complex environment
The application patent is application number 201811607738.9 filed on 27.12.2018 and named as divisional application of an anti-interference algorithm for an instrument visual reading online monitoring system in a complex environment.
Technical Field
The invention belongs to the technical field of detection instruments, and particularly relates to an anti-interference method of an instrument visual reading online monitoring system in a complex environment.
Background
The instrument position in factory workshop is diversified, and some inlay on the instrument, some need artifical reading record data at some moment, can not wrapped up. Through actual research, more than half of instruments in places needing safety monitoring, such as pharmaceutical factories, coating workshops of chemical plants and the like, can not be monitored by using the full-wrapping type detection device, but if an open type device is used for monitoring, the instruments are influenced by various external adverse conditions, so that the detection result is deviated, and the purpose of safety production monitoring cannot be achieved.
In view of this situation, it is necessary to eliminate the interference caused by shadows and environment through the advancement of machine vision algorithms, so that the on-line monitoring meter reading does not require the greenhouse monitoring of harsh environments and light sources.
Disclosure of Invention
The invention aims to provide an anti-interference method of an instrument visual reading online monitoring system in a complex environment.
In order to solve the technical problems, the invention adopts the following technical scheme:
an anti-interference method of an instrument visual reading online monitoring system in a complex environment is characterized by comprising the following steps:
step 1, receiving image data returned by a sensor, and preprocessing the image to obtain a required gray-scale image;
step 2, extracting the image edge by using a canny operator to obtain an edge image;
step 3, detecting image characteristics by using Hough transform, and respectively extracting a circle representing a dial plate and a straight line representing a pointer in a target image to obtain all data of circular edges and straight line edges;
step 4, screening the circular edge to obtain a target circular edge representing the dial;
step 5, screening the straight line edge to obtain a target straight line edge representing the pointer, and detecting the straight line representing the pointer specifically comprises the following steps:
1) detecting all circles in the target edge image by using probability Hough transform, wherein the precision is 1 pixel, the angular precision PI/180 and the threshold value is 0.5R, R is the radius of a dial plate, and the maximum interval value of broken line segment splicing is 5 pixels;
2) screening the pointer line, and screening the rule:
a. the starting point and the end point of the line segment are positioned in the dial circle;
b. the distance L between the line segment extension line and the center of the dial is less than 0.1R;
c.
Figure BDA0002760427250000023
3) carrying out secondary screening on the obtained line segment slope by using a mean square error method to remove a noise point line segment;
4) recording the acquired pointer line segment;
5) if the pointer cannot be successfully detected, recording all data and images, and storing the data and the images in a database for further improvement of the algorithm;
step 6, intercepting the image around the pointer to obtain an image of the area around the pointer, wherein the intercepting the image around the pointer specifically comprises the following steps:
the following data are generated after the detection system successfully reads the target instrument panel reading:
target dial center coordinates (X, Y);
target dial radius R;
target pointer slope K;
a target pointer length L;
then, taking the target pointer as a reference, intercepting a section with the length of L and the width of L
Figure BDA0002760427250000022
A rectangular area with a direction K;
step 7, comparing the distribution characteristics of the area of the pointer area and the gray level histogram, and judging whether shadow interference exists or not to obtain a shadow detection result, wherein the step of comparing the distribution characteristics of the area of the pointer area and the gray level histogram specifically comprises the following steps of:
calculating the image gray histogram
Figure BDA0002760427250000031
Where n is the total number of pixels, nkIs a gray level SkProcessing the gray level histogram by using the number of pixels, smoothing by using filtering, then drawing a curve with specified precision approaching a polygon, and then carrying out concave-convex detection on the curve to obtain the wave crest and the wave trough of the gray level histogram;
the peak position corresponding to the gray value represented by the pointer is determined by an area calibration method and a color calibration method,
the area calibration method specifically comprises the following steps: the detected target dial plate is a fixed dial plate, so the system can very accurately know the area occupied by the pointer of the target dial plate and the area occupied by the background mark of the intercepted icon in each direction of the background, then the conversion ratio of the actual area and the image pixel area can be calculated according to the previously detected dial plate radius and the corresponding actual radius, and the pixel area corresponding to the area of the intercepted area pointer and the area of the background mark is set as SmCumulative histogram of image
Figure BDA0002760427250000032
L0Is a trough or 0, L1Is the next trough, and L is the last trough1＝255，
Figure BDA0002760427250000033
When S ism＝CDF(Sk) Taking the peak of the wave
Figure BDA0002760427250000034
The peak is the peak corresponding to the gray value represented by the pointer in the image;
step 8, removing the interference of the shadow part to obtain an image without shadow interference;
adopting a double-peak gathering method to filter the shadow, which comprises the following steps:
let l0Front wave trough, l, representing a shadow area1The back wave valley of the shadow region is represented by L0Front wave trough, L, representing background area1A back-wave valley representing a background area,
computing local histograms for shadow regions
Figure BDA0002760427250000035
Wherein n is
Figure BDA0002760427250000036
Total number of pixels of (1), nkIs a gray level SkNumber of pixels of (2), replacement target pixel gradation value
Figure BDA0002760427250000037
The gray value wave crest of the shadow area is zoomed by a second-order function, so that the whole wave crest gathers to the background area, and meanwhile, the wave crest of the background area gathers to the shadow area by the same algorithm, so that the shadow and the color difference of the background area are smoothed, and the purpose of eliminating the shadow is achieved;
the preprocessing of the image in step 1 includes graying and nonlinear dynamic range adjustment operations, specifically,
3) graying
Using the formula
Figure BDA0002760427250000041
Performing a graying operation on the image, wherein G (i, j) is the grayscale of the (i, j) point, R (i, j) is the red (red) of the (i, j) point, G (i, j) is the green (green) of the (i, j) point, and B (i, j) is the blue (blue) of the (i, j) point;
4) nonlinear dynamic adjustment
Processing the gray-scale image by using a formula g (i, j) ═ c × lg (1+ f (i, j)), wherein the gray scale of the original image is f (i, j), the processed image is g (i, j), and c is a gain constant and is taken according to the overall brightness of the image;
the detection of the circle representing the dial plate in the step 4 specifically comprises the following steps:
1) detecting all circles in the target edge image by using probability Hough transform;
2) taking a circle C with the largest radiusmThe radius of the circle is Rm；
3) All circle centers and C are obtainedmPhase difference<0.05RmAnd radius>0.5RmA set of circles of (d);
4) taking a circle with the smallest radius, wherein the circle is defined as a circle represented by the target dial plate, and the radius R of the circle is the inner diameter of the dial plate;
5) if the dial plate cannot be successfully detected, all data and images are recorded and stored in a database for further improvement of the algorithm.
Preferably, the color calibration method specifically comprises:
the detected target dial plate is a fixed dial plate, so the system can accurately know the pointer color and the background color of the target dial plate, the maximum wave crest certainly represents the background because the background definitely occupies the maximum proportion, and the RGB value of the ambient light reflected light can be obtained by the system according to the RGB value reflected by the background, so the RGB value corresponding to the pointer in the image is calculated, and the wave crest corresponding to the pointer can be found.
Preferably, the shadow is filtered by a binary method, specifically:
is provided with L0Front wave trough, L, representing pointer area1Representing the rear wave trough of the pointer area, binarizing the image, if so
Figure BDA0002760427250000051
Then g (i, j) is 0, otherwise g (i, j) is 255.
The invention has the following beneficial effects: because the interference of the external environment on the visual reading of the instrument panel can be filtered to the maximum extent, the online monitoring system of the instrument panel, which is invented by using the algorithm, does not need to wrap the detected meter and carry a light source, and can complete the online monitoring of the instrument by only one image sensor. Some meters which cannot be wrapped, are complex in position and large in wiring limitation can also be monitored. Most instruments in the factory workshop cannot be wrapped, and the instruments can be adapted by using the algorithm provided by the embodiment of the invention, so that the aim of full-automatic safety monitoring is fulfilled.
Drawings
FIG. 1 is a flow chart of an anti-interference method of an instrument visual reading online monitoring system in a complex environment according to an embodiment of the invention;
FIG. 2 is a diagram illustrating an image of a region around a pointer in an exemplary embodiment;
FIG. 3 is a histogram of gray levels and corresponding peak and valley plots for a specific application example;
FIG. 4 is a diagram illustrating the effect of binary processing in an exemplary embodiment;
FIG. 5 is a diagram illustrating the effect of the bimodal bunching process in an exemplary embodiment.
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 some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a flowchart of an anti-interference method for an online monitoring system of instrument visual readings in a complex environment according to an embodiment of the present invention is shown, which includes the following steps:
s1, preprocessing the image to obtain the needed gray-scale image
After the system receives the image data returned by the sensor, the image is preprocessed to obtain a required gray level image, and the system respectively carries out graying and nonlinear dynamic range adjustment operations.
1) Graying
Using the formula
Figure BDA0002760427250000061
And performing a graying operation on the image, wherein G (i, j) is the grayscale of the (i, j) point, R (i, j) is red (red) of the (i, j) point, G (i, j) is green (green) of the (i, j) point, and B (i, j) is blue (blue) of the (i, j) point.
2) Nonlinear dynamic adjustment
And processing the gray level map by using a formula g (i, j) ═ c × lg (1+ f (i, j)), wherein the gray level of the original image is f (i, j), the processed image is g (i, j), and c is a gain constant and is taken according to the overall brightness of the image.
And S2, detecting edges, and extracting the edges of the image by using a canny operator to obtain an edge image.
And S3, carrying out Hough transformation, detecting image features by using Hough transformation, and respectively extracting a circle (dial) shape and a straight line (pointer) in the target image to obtain all circular edge data and straight line edge data.
And S4, detecting the circular edge to obtain a target circular edge representing the dial.
1) Detecting all circles in an edge image of an object using a probabilistic Hough transform
2) Taking a circle C with the largest radiusmThe radius of the circle is Rm。
3) All circle centers and C are obtainedmPhase difference<0.05RmAnd radius>0.5RmSet of circles of
4) Wherein the circle with the smallest radius is defined as the circle represented by the target dial, and the radius R of the circle is the inner diameter of the dial
5) If the dial plate cannot be successfully detected, all data and images are recorded and stored in a database for further algorithm improvement
S5, detecting the straight line edge to obtain a target straight line edge representing the pointer;
1) and detecting all circles in the target edge image by using probability Hough transform, wherein the precision is 1 pixel, the angular precision PI/180, the threshold value is 0.5R (R is the radius of a dial plate), and the maximum interval value of broken line segment splicing is 5 pixels.
2) Screening the pointer line, and screening the rule:
1. the starting point and the end point of the line segment are positioned in the dial circle;
2. the distance L between the line segment extension line and the center of the dial is less than 0.1R;
3.
Figure BDA0002760427250000074
3) using a mean square error method to carry out secondary screening on the obtained line segment slope, and removing the line segment with the noise point
4) Recording the acquired pointer line segment
5) If the pointer cannot be successfully detected, all data and images are recorded and saved in a database for further improvement of the algorithm
S6, intercepting the image around the pointer to obtain the image of the area around the pointer;
the following data are generated after the detection system successfully reads the target instrument panel reading:
target dial center coordinates (X, Y);
target dial radius R;
target pointer slope K;
a target pointer length L;
then, taking the target pointer as a reference, intercepting a section with the length of L and the width of L
Figure BDA0002760427250000072
The rectangular area with the direction of K is shown in figure 2.
S7, comparing the area of the pointer region with the distribution characteristics of the gray histogram, and judging whether shadow interference exists or not to obtain a shadow detection result;
calculating the image gray histogram
Figure BDA0002760427250000073
Where n is the total number of pixels, nkIs a gray level SkNumber of pixels (2)
And processing the gray level histogram, smoothing by using filtering, then drawing a curve with specified precision approximating to a polygon, and then performing concave-convex detection on the curve to obtain the wave crest and the wave trough of the gray level histogram. Processing the example graph above yields 3 peaks, 52, 128, 227 respectively, as shown in fig. 3.
One can clearly conclude that: h1 represents a pointer, H2 represents a shadow produced by a light source, and H3 is the dashboard background.
Of course, the algorithm in our system is not visible to the naked eye, and we have developed two algorithms to determine the peak position corresponding to the gray value represented by the pointer:
1) area calibration method
The detected target dial plate is a fixed dial plate, so the system can very accurately know the area occupied by the pointer of the target dial plate and the area occupied by the background mark of the intercepted icon in each direction of the background, then the conversion ratio of the actual area and the image pixel area can be calculated according to the previously detected dial plate radius and the corresponding actual radius, and the pixel area corresponding to the area of the intercepted area pointer and the area of the background mark is set as SmCumulative histogram of image
Figure BDA0002760427250000081
L0Is a trough or 0, L1Is the next trough, and L is the last trough if1＝255，
Figure BDA0002760427250000082
When S ism＝CDF(SK) Taking the peak of the wave
Figure BDA0002760427250000083
The peak is the peak corresponding to the gray value represented by the pointer in the image.
2) Color calibration method
The detected target dial is a fixed dial, so the system can know the color of the pointer of the target dial and the color of the background very accurately. Because the background is definitely in the maximum proportion, the maximum peak necessarily represents the background, and the RGB value of the ambient light reflected light can be obtained by the RGB value system reflected by the background, so that the corresponding RGB value of the pointer in the image is calculated, and the peak corresponding to the pointer can be found.
And S8, removing the interference of the shadow part to obtain an image without shadow interference.
The system can filter the shadow after obtaining the peaks that the pointer represents inside the gray histogram. We developed two sets of algorithms to filter shadows:
1) binary method
Is provided with L0Front wave trough, L, representing pointer area1Representing the rear wave trough of the pointer area, binarizing the image, if so
Figure BDA0002760427250000084
If g (i, j) is 0, otherwise g (i, j) is 255, and the binary processing effect graph is shown in fig. 4.
2) Double peak gathering method
Let l0Front wave trough, l, representing a shadow area1The back wave valley of the shadow region is represented by L0Front wave trough, L, representing background area1Representing the back valleys of the background area.
Computing local histograms for shadow regions
Figure BDA0002760427250000085
Wherein n is
Figure BDA0002760427250000086
Total number of pixels of (1), nkIs a gray level SkNumber of pixels of (2), replacement target pixel gradation value
Figure BDA0002760427250000091
And scaling the gray value wave crest of the shadow area by a second-order function to enable the whole wave crest to gather to the background area, and meanwhile, gathering the wave crest of the background area to the shadow area by the same algorithm to smooth the shadow and the color difference of the background area, thereby achieving the purpose of eliminating the shadow.
The processing effect of the bimodal gathering process is shown in fig. 5.
It can be seen that the data obtained by our algorithm processing has completely eliminated the shadow interference caused by the non-ideal light source. The monitoring system manufactured by the algorithm can completely break away from a full-wrapping detection means with a light source, and is suitable for visual detection in various complex environments.
It is to be understood that the exemplary embodiments described herein are illustrative and not restrictive. Although one or more embodiments of the present invention have been described with reference to the accompanying drawings, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the following claims.

Claims (3)

1. An anti-interference method of an instrument visual reading online monitoring system in a complex environment is characterized by comprising the following steps:
step 1, receiving image data returned by a sensor, and preprocessing the image to obtain a required gray-scale image;
step 2, extracting the image edge by using the canny operator to obtain an edge image;
step 3, detecting image characteristics by using Hough transform, and respectively extracting a circle representing a dial plate and a straight line representing a pointer in a target image to obtain all data of circular edges and straight line edges;
step 4, screening the circular edge to obtain a target circular edge representing the dial;
step 5, screening the straight line edge to obtain a target straight line edge representing the pointer, and detecting the straight line representing the pointer specifically comprises the following steps:
1) detecting all circles in the target edge image by using probability Hough transform, wherein the precision is 1 pixel, the angular precision PI/180 and the threshold value is 0.5R, R is the radius of a dial plate, and the maximum interval value of broken line segment splicing is 5 pixels;
2) screening the pointer line, and screening the rule:
a. the starting point and the end point of the line segment are positioned in the dial circle;
b. the distance L between the line segment extension line and the center of the dial is less than 0.1R;
c.
Figure FDA0002760427240000011
3) carrying out secondary screening on the obtained line segment slope by using a mean square error method to remove a noise point line segment;
4) recording the acquired pointer line segment;
5) if the pointer cannot be successfully detected, recording all data and images, and storing the data and the images in a database for further improvement of the algorithm;
step 6, intercepting the image around the pointer to obtain an image of the area around the pointer, wherein the intercepting the image around the pointer specifically comprises the following steps:
the following data are generated after the detection system successfully reads the target instrument panel reading:
target dial center coordinates (X, Y);
target dial radius R;
target pointer slope K;
a target pointer length L;
then, taking the target pointer as a reference, intercepting a section with the length of L and the width of L
Figure FDA0002760427240000021
A rectangular area with a direction K;
step 7, comparing the distribution characteristics of the area of the pointer area and the gray level histogram, and judging whether shadow interference exists or not to obtain a shadow detection result, wherein the step of comparing the distribution characteristics of the area of the pointer area and the gray level histogram specifically comprises the following steps of:
calculating the image gray histogram
Figure FDA0002760427240000022
Where n is the total number of pixels, nkIs a gray level SkProcessing the gray level histogram by using the number of pixels, smoothing by using filtering, then drawing a curve with specified precision approaching a polygon, and then carrying out concave-convex detection on the curve to obtain the wave crest and the wave trough of the gray level histogram;
the peak position corresponding to the gray value represented by the pointer is determined by an area calibration method and a color calibration method,
the area calibration method specifically comprises the following steps: the detected target dial plate is a fixed dial plate, so the system can very accurately know the area occupied by the pointer of the target dial plate and the area occupied by the background mark of the intercepted icon in each direction of the background, then the conversion ratio of the actual area and the image pixel area can be calculated according to the detected dial plate radius and the corresponding actual radius, and the area of the corresponding intercepted area pointer + the background mark are setThe area of the pixel corresponding to the area is SmCumulative histogram of image
Figure FDA0002760427240000023
L0Is a trough or 0, L1Is the next trough, and L is the last trough1＝255，
Figure FDA0002760427240000024
When S ism＝CDF(Sk) Taking the peak of the wave
Figure FDA0002760427240000025
The peak is the peak corresponding to the gray value represented by the pointer in the image;
step 8, removing the interference of the shadow part to obtain an image without shadow interference;
adopting a double-peak gathering method to filter the shadow, which comprises the following steps:
let l0Front wave trough, l, representing a shadow area1The back wave valley of the shadow region is represented by L0Front wave trough, L, representing background area1A back-wave valley representing a background area,
computing local histograms for shadow regions
Figure FDA0002760427240000031
Wherein n is
Figure FDA0002760427240000032
Total number of pixels of (1), nkIs a gray level SkNumber of pixels of (2), replacement target pixel gradation value
Figure FDA0002760427240000033
The gray value wave crest of the shadow area is zoomed by a second-order function, so that the whole wave crest gathers to the background area, and meanwhile, the wave crest of the background area gathers to the shadow area by the same algorithm, so that the shadow and the color difference of the background area are smoothed, and the purpose of eliminating the shadow is achieved;
the preprocessing of the image in step 1 includes graying and nonlinear dynamic range adjustment operations, specifically,
1) graying
Using the formula
Figure FDA0002760427240000034
Performing a graying operation on the image, wherein G (i, j) is the grayscale of the (i, j) point, R (i, j) is the red (red) of the (i, j) point, G (i, j) is the green (green) of the (i, j) point, and B (i, j) is the blue (blue) of the (i, j) point;
2) nonlinear dynamic adjustment
Processing the gray-scale image by using a formula g (i, j) ═ c × lg (1+ f (i, j)), wherein the gray scale of the original image is f (i, j), the processed image is g (i, j), and c is a gain constant and is taken according to the overall brightness of the image;
the detection of the circle representing the dial plate in the step 4 specifically comprises the following steps:
1) detecting all circles in the target edge image by using probability Hough transform;
2) taking a circle C with the largest radiusmThe radius of the circle is Rm；
3) All circle centers and C are obtainedmPhase difference<0.05RmAnd radius>0.5RmA set of circles of (d);
4) taking a circle with the smallest radius, wherein the circle is defined as a circle represented by the target dial plate, and the radius R of the circle is the inner diameter of the dial plate;
5) if the dial plate cannot be successfully detected, all data and images are recorded and stored in a database for further improvement of the algorithm.
2. The anti-interference method for the on-line monitoring system of instrument visual reading according to claim 1, wherein the color calibration method is specifically:
the detected target dial plate is a fixed dial plate, so the system can accurately know the pointer color and the background color of the target dial plate, the maximum wave crest certainly represents the background because the background definitely occupies the maximum proportion, and the RGB value of the ambient light reflected light can be obtained by the system according to the RGB value reflected by the background, so the RGB value corresponding to the pointer in the image is calculated, and the wave crest corresponding to the pointer can be found.
3. The anti-interference method for the online monitoring system of the instrument visual reading in the complex environment according to claim 1 or 2, characterized in that the shadow is filtered by a binary method, specifically:
is provided with L0Front wave trough, L, representing pointer area1Representing the rear wave trough of the pointer area, binarizing the image, if so
Figure FDA0002760427240000041
Then g (i, j) is 0, otherwise g (i, j) is 255.
CN202011216139.1A 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading on-line monitoring system in complex environment Active CN112381848B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011216139.1A CN112381848B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading on-line monitoring system in complex environment

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811607738.9A CN110009651B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading on-line monitoring system
CN202011216139.1A CN112381848B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading on-line monitoring system in complex environment

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CN201811607738.9A Division CN110009651B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading on-line monitoring system

Publications (2)

Publication Number Publication Date
CN112381848A true CN112381848A (en) 2021-02-19
CN112381848B CN112381848B (en) 2022-06-10

Family

ID=67165313

Family Applications (3)

Application Number Title Priority Date Filing Date
CN202011217299.8A Active CN112396617B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading monitoring system
CN202011216139.1A Active CN112381848B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading on-line monitoring system in complex environment
CN201811607738.9A Active CN110009651B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading on-line monitoring system

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN202011217299.8A Active CN112396617B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading monitoring system

Family Applications After (1)

Application Number Title Priority Date Filing Date
CN201811607738.9A Active CN110009651B (en) 2018-12-27 2018-12-27 Anti-interference method of instrument visual reading on-line monitoring system

Country Status (1)

Country Link
CN (3) CN112396617B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116823808A (en) * 2023-08-23 2023-09-29 青岛豪迈电缆集团有限公司 Intelligent detection method for cable stranded wire based on machine vision
CN117798654A (en) * 2024-02-29 2024-04-02 山西漳电科学技术研究院(有限公司) Intelligent adjusting system for center of steam turbine shafting

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111209842B (en) * 2020-01-02 2023-06-30 珠海格力电器股份有限公司 Visual positioning processing method and device and robot
CN111931776B (en) * 2020-10-16 2021-04-09 江西小马机器人有限公司 Deep learning-based double-pointer instrument reading method
CN114862880B (en) * 2022-07-06 2022-09-02 山东泰恒石材有限公司 Cutting optimization method and system based on anisotropic stone

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006309405A (en) * 2005-04-27 2006-11-09 Chugoku Electric Power Co Inc:The Meter recognition system, meter recognition method, and meter recognition program
CN105260710A (en) * 2015-09-28 2016-01-20 北京石油化工学院 Water meter detecting method, apparatus, and system based on image processing

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103927507A (en) * 2013-01-12 2014-07-16 山东鲁能智能技术有限公司 Improved multi-instrument reading identification method of transformer station inspection robot
CN104333736B (en) * 2014-10-28 2017-08-29 山东大学 Unmanned substation's intelligent recognition monitoring system and method
CN107038444A (en) * 2016-02-03 2017-08-11 上海慕荣电气有限公司 A kind of image-recognizing method of pointer dial plate
CN106204614B (en) * 2016-07-21 2019-01-08 湘潭大学 A kind of workpiece appearance defects detection method based on machine vision

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006309405A (en) * 2005-04-27 2006-11-09 Chugoku Electric Power Co Inc:The Meter recognition system, meter recognition method, and meter recognition program
CN105260710A (en) * 2015-09-28 2016-01-20 北京石油化工学院 Water meter detecting method, apparatus, and system based on image processing

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
YIFAN MA等: "《A robust and high-precision automatic reading algorithm of pointer meters based on machine vision》", 《MEASURED SCIENCE AND TECHNOLOGY》 *
张文杰 等: "《基于视觉显著性的指针式仪表读数识别算法》", 《计算机辅助设计与图形学学报》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116823808A (en) * 2023-08-23 2023-09-29 青岛豪迈电缆集团有限公司 Intelligent detection method for cable stranded wire based on machine vision
CN116823808B (en) * 2023-08-23 2023-11-17 青岛豪迈电缆集团有限公司 Intelligent detection method for cable stranded wire based on machine vision
CN117798654A (en) * 2024-02-29 2024-04-02 山西漳电科学技术研究院(有限公司) Intelligent adjusting system for center of steam turbine shafting
CN117798654B (en) * 2024-02-29 2024-05-03 山西漳电科学技术研究院(有限公司) Intelligent adjusting system for center of steam turbine shafting

Also Published As

Publication number Publication date
CN112396617A (en) 2021-02-23
CN112381848B (en) 2022-06-10
CN110009651A (en) 2019-07-12
CN112396617B (en) 2022-06-10
CN110009651B (en) 2021-01-05

Similar Documents

Publication Publication Date Title
CN112381848B (en) Anti-interference method of instrument visual reading on-line monitoring system in complex environment
CN108921176B (en) Pointer instrument positioning and identifying method based on machine vision
CN103759758B (en) A kind of method for detecting position of the automobile meter pointer based on mechanical angle and scale identification
CN103308430B (en) A kind of method and device measuring thousand grain weigth
CN111968144B (en) Image edge point acquisition method and device
CN109635806B (en) Ammeter value identification method based on residual error network
CN107292310B (en) Visual positioning and automatic reading method for circular pointer type dial plate
CN116188462B (en) Noble metal quality detection method and system based on visual identification
CN115018844A (en) Plastic film quality evaluation method based on artificial intelligence
CN107038444A (en) A kind of image-recognizing method of pointer dial plate
CN114240845B (en) Light cutting method surface roughness measurement method applied to cutting workpiece
CN115661147A (en) Metering detection data identification method based on machine vision
CN112215060B (en) Hough transform-based high-precision mechanical instrument reading identification method
CN114005108A (en) Pointer instrument degree identification method based on coordinate transformation
CN116245880A (en) Electric vehicle charging pile fire risk detection method based on infrared identification
CN103340600B (en) Wave distortion processing method and processing system on monitor
CN117437223B (en) Intelligent defect detection method for high-speed board-to-board connector
CN113610041A (en) Reading identification method and device for pointer instrument
CN102637317A (en) Coin size measuring method based on vision
CN115880683B (en) Urban waterlogging ponding intelligent water level detection method based on deep learning
CN115222817B (en) Pointer positioning method of pointer type pressure gauge
CN114742849B (en) Leveling instrument distance measuring method based on image enhancement
CN115760653A (en) Image correction method, device, equipment and readable storage medium
CN116129102A (en) Water meter identification method and related device based on improved image processing algorithm
CN115471650A (en) Gas pressure instrument reading method, device, equipment and 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