CN111931716B - Pointer type dial plate identification method and system based on multi-dimensional characteristic analysis - Google Patents
Pointer type dial plate identification method and system based on multi-dimensional characteristic analysis Download PDFInfo
- Publication number
- CN111931716B CN111931716B CN202011001595.4A CN202011001595A CN111931716B CN 111931716 B CN111931716 B CN 111931716B CN 202011001595 A CN202011001595 A CN 202011001595A CN 111931716 B CN111931716 B CN 111931716B
- Authority
- CN
- China
- Prior art keywords
- dial plate
- pointer
- picture
- alpha
- identification
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000004458 analytical method Methods 0.000 title claims abstract description 17
- 230000008569 process Effects 0.000 claims abstract description 14
- 238000001514 detection method Methods 0.000 claims description 18
- 238000007781 pre-processing Methods 0.000 claims description 15
- 230000009467 reduction Effects 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 13
- 230000004044 response Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000003708 edge detection Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000011161 development Methods 0.000 abstract description 3
- 238000012549 training Methods 0.000 description 4
- 239000003086 colorant Substances 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/02—Recognising information on displays, dials, clocks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a pointer type dial plate identification method and system based on multi-dimensional characteristic analysis, which solve the problem that in the prior art, parameters need to be changed on a code level when other dial plates are replaced, and specific functions can be adapted only through secondary development. According to the invention, the identification library is constructed, the multidimensional characteristics are analyzed and the library is constructed, the optimal result is selected in the subsequent identification process through multidimensional characteristic matching and similarity judgment, and the dial plate identification result is rapidly determined. And the recognized result is directly output, and the unrecognized result is added into the recognition library, so that the recognition library is updated in real time, and the follow-up more accurate comparison is facilitated.
Description
Technical Field
The invention relates to the technical field of picture processing, in particular to target detection, and specifically relates to a pointer type dial plate identification method and system based on multi-dimensional feature analysis.
Background
Under the industrial conditions of pharmaceutical factories, severe environments and high-risk operations, environmental data under various environments and industrial process monitoring data need to be monitored, represented and identified, for example, pressure gauges, thermometers and the like, a pointer type dial plate design is usually adopted, but under the field environment, environmental interference such as high temperature, high light and the like exists, and a barrier exists in remote identification of the dial plate. In the prior art, when the pointer type dial plate is identified and the pointer and the circle center of the dial plate are calculated, parameters are adjusted each time, and the parameters are reset when other pointer type dial plates are replaced; the scope and the precision deviation that pointer and the lines in the centre of a circle were found out through prior art in the dial plate are great, need only can obtain more accurate scope and precision through secondary treatment, realize the function on the matching project that can not be fine on directly using prior art. In the conventional identification of a pointer type dial, parameters such as a pointer and a circle center can be identified, but when one dial is replaced, the parameters need to be changed on a code level. The specific function can be adapted only by secondary development, and the existing identification technology can not well solve the problem and has certain limitation.
Disclosure of Invention
The invention aims to provide a pointer type dial plate identification method and system based on multi-dimensional feature analysis. And the recognized result is directly output, and the unrecognized result is added into the recognition library, so that the data of the recognition library is updated, and the subsequent more accurate comparison is facilitated.
According to a first aspect of the object of the present invention, a pointer dial plate identification method based on multi-dimensional feature analysis is provided, which includes the following steps:
establishing a pointer type dial plate identification library;
establishing identification data for the identified pointer dials and storing the identification data into an identification library, wherein dial IDs are assigned to each pointer dial and a corresponding dial picture in the identification data, the dial IDs are associated with the dial pictures, the identification data further comprise multi-dimensional structured identification data for the dial pictures, the structured identification data comprise weighting weights Ai obtained by giving weighting values to the shapes, colors, scales and pointer characteristics of the pointer dials according to preset rules, and dial structure attributes and coordinates of the pointer dials, and i =1,2,3,... m, m represents the total number of the pointer dials stored in the identification library;
receiving an input dial plate picture to be identified in the pointer type dial plate identification process;
preprocessing the received dial plate picture to obtain a pure picture;
detecting the shape, color, scale and pointer of a dial in the picture by adopting a target detection algorithm on the noise-reduced pure picture, and respectively acquiring the weight values of the recognized shape, color, scale and pointer of the dial according to a preset rule;
carrying out weighting calculation on the obtained weight value to obtain a weighting weight A;
performing round training comparison on the obtained weighting weight A and the weighting weight Am in the recognition library, and sequencing the absolute value Dm of the difference value of the two to obtain the weighting weight An corresponding to the minimum absolute value and the corresponding dial plate ID and dial plate picture, wherein n is less than or equal to m;
acquiring similarity values of the dial plate pictures detected by the received pictures and the dial plate corresponding to the weighting weights An;
responding to the similarity value similarity exceeding a set similarity threshold, and outputting the weighting weight An and the corresponding dial ID as a recognition result;
in response to the fact that the similarity value similarity does not reach the set similarity threshold value, judging that a new pointer type dial plate is obtained, assigning a unique dial plate ID for the pointer type dial plate, associating the unique dial plate ID with the obtained dial plate picture and the calculated weighting weight, and storing the unique dial plate ID in a pointer type dial plate identification library; and
and executing dial plate structure attribute identification and coordinate positioning on the input dial plate picture, storing the dial plate structure attribute and the coordinate into the identification library, and associating the dial plate structure attribute and the coordinate with the corresponding dial plate ID.
According to a second aspect of the present invention, there is provided a pointer dial plate recognition system based on multidimensional feature analysis, including:
a module for creating a pointer dial identification library;
the identification data comprises a module for establishing identification data for the identified pointer type dial plates and storing the identification data in an identification library, wherein the identification data designates a dial plate ID for each pointer type dial plate and a corresponding dial plate picture, the dial plate ID is associated with the dial plate picture, the identification data also comprises multi-dimensional structured identification data for the dial plate picture, the structured identification data comprises a weighting weight Ai obtained by giving a weighting value to the shape, color, scale and pointer characteristics of the pointer type dial plate according to a preset rule, and the dial plate structure attribute and the coordinate of the pointer type dial plate, and i =1,2, 3.. m, m represents the total number of the pointer type dial plates stored in the identification library;
the module is used for receiving an input dial plate picture to be identified in the pointer type dial plate identification process;
the module is used for preprocessing the received dial plate picture to obtain a pure picture;
a module for detecting the shape, color, scale and pointer of the dial plate in the picture by adopting a target detection algorithm for the pure picture after noise reduction, and respectively obtaining a weight value for the recognized shape, color, scale and pointer of the dial plate according to a preset rule;
a module for performing weighting calculation on the obtained weight value to obtain a weighting weight A;
a module for performing round training comparison on the obtained weighting weight A and the weighting weight Am in the identification library, and sequencing An absolute value Dm of a difference value of the obtained weighting weight A and the weighting weight Am to obtain a weighting weight An corresponding to the minimum absolute value and a corresponding dial plate ID and dial plate picture, wherein n is less than or equal to m;
a module for obtaining the similarity value similarity of the dial plate picture detected by the received picture and the dial plate corresponding to the weighting weight An;
a module for outputting the weighting An and the corresponding dial ID as the identification result in response to the similarity value similarity exceeding the set similarity threshold;
a module for determining a new pointer type dial plate in response to the similarity value similarity not reaching the set similarity threshold value, assigning a unique dial plate ID for the pointer type dial plate, associating the unique dial plate ID with the obtained dial plate picture and the calculated weighting weight, and storing the unique dial plate ID in the pointer type dial plate identification library; and
and the module is used for executing dial plate structure attribute identification and coordinate positioning on the input dial plate picture, storing the dial plate structure attribute and the coordinate into the identification library and associating the dial plate structure attribute and the coordinate with the corresponding dial plate ID.
By the technical scheme, the invention has the following remarkable beneficial effects:
the method comprises the steps of shooting a picture through a camera, obtaining multi-dimensional weight values such as a graph shape, a color and a dial plate through extracting characteristic values of the picture, reducing the multi-dimensional weight values, merging scores, performing polling comparison on the merged scores and a picture library, obtaining an optimal score, calculating a final azimuth angle (namely a pointing scale value) through a pointer value and an angle, and solving the problem that parameters need to be changed on a code level when other dial plates are replaced and the problem that secondary development can be carried out to adapt to a specific function in the prior art.
According to the invention, the identification library is constructed, the multidimensional characteristics are analyzed and the library is constructed, in the subsequent identification process, the optimal result is selected through multidimensional characteristic matching and similarity judgment, and the dial plate identification result is rapidly determined. And the recognized result is directly output, and the unrecognized result is added into the recognition library, so that the data of the recognition library is updated, and the subsequent more accurate comparison is facilitated.
It should be understood that all combinations of the foregoing concepts and additional concepts described in greater detail below can be considered as part of the inventive subject matter of this disclosure unless such concepts are mutually inconsistent. In addition, all combinations of claimed subject matter are considered a part of the presently disclosed subject matter.
The foregoing and other aspects, embodiments and features of the present teachings can be more fully understood from the following description taken in conjunction with the accompanying drawings. Additional aspects of the present invention, such as features and/or advantages of exemplary embodiments, will be apparent from the description which follows, or may be learned by practice of specific embodiments in accordance with the teachings of the present invention.
Drawings
Fig. 1 is a flowchart illustrating a pointer dial plate recognition method of multi-dimensional feature analysis according to an exemplary embodiment of the present invention.
Fig. 2 is a schematic diagram illustrating a detection process of circle center coordinates according to an exemplary embodiment of the present invention.
Fig. 3 is a schematic flowchart of calculating the azimuth angle of the pointer according to an exemplary embodiment of the present invention.
Detailed Description
In order to better understand the technical content of the present invention, specific embodiments are described below with reference to the accompanying drawings.
In this disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. Embodiments of the present disclosure are not necessarily intended to include all aspects of the invention. It should be appreciated that the various concepts and embodiments described above, as well as those described in greater detail below, may be implemented in any of numerous ways, as the disclosed concepts and embodiments are not limited to any one implementation. In addition, some aspects of the present disclosure may be used alone, or in any suitable combination with other aspects of the present disclosure.
With reference to fig. 1 to 3, according to the pointer-type dial plate recognition method based on multidimensional feature analysis according to the exemplary embodiment of the present invention, a recognition library is constructed, multidimensional feature analysis and the library are constructed, in a subsequent recognition process, an optimal result is selected through multidimensional feature matching and similarity judgment, and a dial plate recognition result is determined quickly. And the recognized result is directly output, and the unrecognized result is added into the recognition library, so that the data of the recognition library is updated, and the subsequent more accurate comparison is facilitated.
The identification process of the present invention is described in more detail below with reference to fig. 1.
Step S101: and creating a pointer type dial plate identification library. In an alternative embodiment, the recognition library may be built based on an oracle database, supporting distributed data processing.
Step S102: establishing identification data for the identified pointer type dials and storing the identification data into the identification library, wherein dial plate IDs are assigned to each pointer type dial plate and a corresponding dial plate picture in the identification data, the dial plate IDs are associated with the dial plate pictures, the identification data further comprise multi-dimensional structured identification data for the dial plate pictures, the structured identification data comprise weighting weights Ai obtained by giving weighting values to the shapes, colors, scales and pointer characteristics of the pointer type dial plates according to preset rules, and dial plate structure attributes and coordinates of the pointer type dial plates, and i =1,2,3,. m, and m represents the total number of the pointer type dial plates stored in the identification library.
Wherein, in the structured identification data, the shape, color, scale and pointer characteristics of the dial plate all include at least one level of characteristics, wherein: the shape features include rectangular, triangular, square, circular and hexagonal features; the color characteristics are distinguished according to the color system rule of RGB; the scale features comprise scale values and interval features between scales; the pointer features include pointer shape, length, and width features.
In alternative embodiments, rectangular, triangular, square, circular, and hexagonal features of the shape feature are assigned, for example, a weight of center L1-100, triangle L2-150, square L3-200, rectangle L4-250, hexagon L5-300; the color features are distinguished according to the color system rule of RGB, for example, the weights of the color systems of adjacent RGB are different by 10.
In the implementation process of the invention, the distribution of the weight can be distributed according to the actual research result.
In an alternative embodiment, the dial ID may be established according to certain rules, for example, in the order of arabic natural numbers. In order to facilitate the recognition and prevent the occurrence of misjudgment, it is especially preferable to add letters before Arabic numerals, the letters are generated randomly by adopting a multi-digit mode, the digits of the letters are preferably 2-5, and the recognition and matching calculation is reduced. For example, ZSX101, SWD102, etc., are stored in the database in the order of the generated dial IDs.
In other embodiments, the end of the dial ID is also provided with a check digit, which is a number from 0 to 10 or X, randomly generated, such as ZSX 101.3. The check bit is 1 bit.
Therefore, an identification library is established for the identified pointer type dial plate, the characteristic value of the picture is extracted, the multi-dimensional weight values of the shape, the color, the dial plate and the like of the graph are obtained, and the multi-dimensional weight values are subjected to dimension reduction and merging scoring, so that the subsequent identification processing is facilitated.
Next, in the pointer dial identification process, identification processing is performed through step S103 to step S110.
Step S103: and receiving the input dial picture to be identified.
In this step, gather the scene picture through the camera, specifically include:
step 1, storing pictures into a cache library through a camera photo;
step 2, judging whether the image has double images and jitters, and discarding data with the double images or jitters;
step 3, if the pictures do not conform to each other, the camera shooting function is called again;
therefore, the high-quality picture is obtained, and the later recognition and picture processing are facilitated.
Step S104: and preprocessing the received dial picture to obtain a pure picture.
Optionally, the preprocessing includes gaussian blur noise reduction processing, specifically including:
using a Gaussian fuzzy algorithm to correct RGB three channels of bright spots in a dial picture, wherein gradient percentages are used for detection, the bright spots are detected and then corrected through median filtering, finally performing transparency mixing operation based on corrected RGB, and mixing corrected RGB according to transparency, wherein:
when Alpha =50%, the signal is transmitted,
r = r1/2 + r2/2;
g = g1/2 + g2/2;
b = b1/2 + b2/2;
when Alpha is <50%, the Alpha is expressed,
r = r1 - r1/ALPHA + r2/ALPHA;
g = g1 - g1/ALPHA + g2/ALPHA;
b = b1 - b1/ALPHA + b2/ALPHA;
when Alpha >50% of the total weight of the composition,
r = r1/ALPHA + r2 - r2/ALPHA;
g = g1/ALPHA + g2 - g2/ALPHA;
b = b1/ALPHA + b2 - b2/ALPHA;
in the foregoing formula, Alpha is the transparency of the pointer picture, r, g, b are the color values of the final output, and r1, g1, b1 are the color values of the pre-color, respectively; r2, g2 and b2 are color values of the background color, respectively.
In a preferred embodiment, the preprocessing further includes performing edge detection on the dial plate picture after the gaussian fuzzy noise reduction processing, and removing edge information to obtain an outline structure of the pointer-type dial plate.
In an alternative embodiment, the foregoing preprocessing may further include a shadow covering process, and since the brightness of the area covered by the shadow is lower than that of the background area, the determination may be performed by extracting pixel point values, and performing color complementation to remove the shadow.
Therefore, a pure picture, namely a dial picture, can be obtained.
Step S105: and detecting the shape, color, scale and pointer of the dial plate in the picture by adopting a target detection algorithm on the noise-reduced pure picture, and respectively acquiring the weight values of the recognized shape, color, scale and pointer of the dial plate according to a preset rule.
It should be understood that, in step S105, characteristic data of the dial plate picture, such as shape, color, scale, and pointer, are detected and recognized by the target detection algorithm, and a weight value is obtained for the recognized dial plate shape, color, scale, and pointer according to the same rule as that adopted for the recognition library storage.
Step S106: and carrying out weighting calculation on the obtained weight value to obtain a weighting weight A.
In this step, it is preferable to add the weight values to obtain a weight value obtained by combining the calculated weight values as a weight a.
Step S107: and performing round training comparison on the obtained weighting weight A and the weighting weight Am in the recognition library, and sequencing the absolute value Dm of the difference value of the obtained weighting weight A and the weighting weight Am to obtain the weighting weight An corresponding to the minimum absolute value and the corresponding dial plate ID and dial plate picture, wherein n is less than or equal to m.
Step S108: and acquiring the similarity value similarity of the dial plate picture detected by the received picture and the dial plate corresponding to the weighting weight An.
The similarity value similarity is obtained as follows:
similarity = obtained weighting weight a/weighting weight An corresponding to the minimum absolute value in the recognition library.
Step S109: and (3) judging the similarity:
responding to the similarity value similarity exceeding a set similarity threshold, and outputting the weighting weight An and the corresponding dial plate ID as a pointer type dial plate identification result;
and in response to the fact that the similarity value similarity does not reach the set similarity threshold value, determining that a new pointer type dial plate is obtained, assigning a unique dial plate ID for the pointer type dial plate, associating the unique dial plate ID with the obtained dial plate picture and the calculated weighting weight, and storing the unique dial plate ID into the pointer type dial plate identification library.
Therefore, on the premise that the similarity meets the set threshold, the similarity threshold can be set according to the identification strength and the accuracy requirement, and the greater the similarity threshold is, the more accurate the identification is. In this embodiment, the similarity threshold value is 95%.
Step S110: and executing dial plate structure attribute identification and coordinate positioning on the input dial plate picture, storing the dial plate structure attribute and the coordinate into the identification library, and associating the dial plate structure attribute and the coordinate with the corresponding dial plate ID.
Preferably, in combination with fig. 3, in the method, the dial plate structure attribute identification and coordinate positioning include:
locating coordinates (Cx, Cy) of a center of the circle based on the Hough circle detection;
detecting and positioning coordinates (Zx, Zy) of pointer points based on the Hough straight line; and
determining a pointer azimuth angle according to the coordinates of the circle center and the coordinates of a pointer point, wherein the preprocessed dial plate picture is divided into four quadrants which are divided into 0-90 degrees, 90-180 degrees, 180-270 degrees and 270-360 degrees; then, the coordinate of the circle center is used as a dot, the coordinate of the pointer point is used as the head of the pointer, the coordinate of the circle center is used as the tail of the pointer, and coordinate increments dx and dy are calculated, dx = Zx-Cx, and dy = Zy-Cy;
if dx, dy are not zero, first calculate a = arcatn (| dy/dx |), and then determine the pointer azimuth according to the range of dx, dy:
pointer azimuth = a when dx >0, dy > 0;
pointer azimuth =180-a when dx <0, dy > 0;
pointer azimuth =180+ a when dx <0, dy < 0; negative range of a-pi
Pointer azimuth =360-a when dx >0, dy < 0; negative range of-a
And outputting pointer azimuth angle information, and giving an alarm in real time in response to the pointer exceeding a preset scale range.
With reference to fig. 2, the process of detecting the circle center coordinate includes:
step 1, detecting each nonzero pixel point in a picture;
step 2, giving the range of the circle center;
step 3, generating a coordinate plane in a voting mode;
and 4, setting a weight to position the circle.
Generating a coordinate plane by voting according to the fact that each nonzero pixel point in the picture is possibly a point on a circle, and setting a weight to position the range of the circle; thereby determining the point at the end of the pointer in the pointer dial, i.e. the centre of the circle.
Hough line detection is the transformation of a line in picture space to a point in parameter space. If the pixels in the picture form a straight line, the method can presume that the curves corresponding to the pixel coordinate values (x, y) in the parameter space are intersected certainly, namely, a straight line passes through the center of a circle. Since the pointer length is fixed, a point on the pointer tip, i.e., pointer point coordinates, is determined.
With reference to the foregoing embodiment and fig. 1, the present invention further provides a pointer-type dial plate recognition system based on multi-dimensional feature analysis, including:
a module for creating a pointer dial identification library;
the identification data comprises a module for establishing identification data for the identified pointer type dial plates and storing the identification data in an identification library, wherein the identification data designates a dial plate ID for each pointer type dial plate and a corresponding dial plate picture, the dial plate ID is associated with the dial plate picture, the identification data also comprises multi-dimensional structured identification data for the dial plate picture, the structured identification data comprises a weighting weight Ai obtained by giving a weighting value to the shape, color, scale and pointer characteristics of the pointer type dial plate according to a preset rule, and the dial plate structure attribute and the coordinate of the pointer type dial plate, and i =1,2, 3.. m, m represents the total number of the pointer type dial plates stored in the identification library;
the module is used for receiving an input dial plate picture to be identified in the pointer type dial plate identification process;
the module is used for preprocessing the received dial plate picture to obtain a pure picture;
a module for detecting the shape, color, scale and pointer of the dial plate in the picture by adopting a target detection algorithm for the pure picture after noise reduction, and respectively obtaining a weight value for the recognized shape, color, scale and pointer of the dial plate according to a preset rule;
a module for performing weighting calculation on the obtained weight value to obtain a weighting weight A;
a module for performing round training comparison on the obtained weighting weight A and the weighting weight Am in the identification library, and sequencing An absolute value Dm of a difference value of the obtained weighting weight A and the weighting weight Am to obtain a weighting weight An corresponding to the minimum absolute value and a corresponding dial plate ID and dial plate picture, wherein n is less than or equal to m;
a module for obtaining the similarity value similarity of the dial plate picture detected by the received picture and the dial plate corresponding to the weighting weight An;
a module for outputting the weighting An and the corresponding dial ID as the identification result in response to the similarity value similarity exceeding the set similarity threshold;
a module for determining a new pointer type dial plate in response to the similarity value similarity not reaching the set similarity threshold value, assigning a unique dial plate ID for the pointer type dial plate, associating the unique dial plate ID with the obtained dial plate picture and the calculated weighting weight, and storing the unique dial plate ID in the pointer type dial plate identification library; and
and the module is used for executing dial plate structure attribute identification and coordinate positioning on the input dial plate picture, storing the dial plate structure attribute and the coordinate into the identification library and associating the dial plate structure attribute and the coordinate with the corresponding dial plate ID.
The module for preprocessing the received dial plate picture to obtain the pure picture comprises a Gaussian fuzzy noise reduction processing module and is set to perform noise reduction processing according to the following mode:
using a Gaussian fuzzy algorithm to correct RGB three channels of bright spots in a dial picture, wherein gradient percentages are used for detection, the bright spots are detected and then corrected through median filtering, finally performing transparency mixing operation based on corrected RGB, and mixing corrected RGB according to transparency, wherein:
when Alpha =50%, the signal is transmitted,
r = r1/2 + r2/2;
g = g1/2 + g2/2;
b = b1/2 + b2/2;
when Alpha is <50%, the Alpha is expressed,
r = r1 - r1/ALPHA + r2/ALPHA;
g = g1 - g1/ALPHA + g2/ALPHA;
b = b1 - b1/ALPHA + b2/ALPHA;
when Alpha >50% of the total weight of the composition,
r = r1/ALPHA + r2 - r2/ALPHA;
g = g1/ALPHA + g2 - g2/ALPHA;
b = b1/ALPHA + b2 - b2/ALPHA;
in the foregoing formula, Alpha is the transparency of the pointer picture, r, g, b are the color values of the final output, and r1, g1, b1 are the color values of the pre-color, respectively; r2, g2 and b2 are color values of the background color, respectively.
The module for preprocessing the received dial plate picture to obtain the pure picture further comprises a module for performing edge detection on the dial plate picture subjected to Gaussian fuzzy noise reduction processing, removing edge information and obtaining the profile structure of the pointer type dial plate.
The similarity value similarity is obtained as follows:
similarity = obtained weighting weight a/weighting weight An corresponding to the minimum absolute value in the recognition library.
The dial plate structure attribute identification and coordinate positioning system is used for executing dial plate structure attribute identification and coordinate positioning on an input dial plate picture, storing the dial plate structure attribute and the coordinate into the identification library, and the module associated with the corresponding dial plate ID comprises a dial plate structure attribute identification and coordinate positioning module and an associated module, wherein:
the dial structure property identification and coordinate locating module is configured to determine the dial structure property and coordinates in the following manner:
locating coordinates (Cx, Cy) of a center of the circle based on the Hough circle detection;
detecting and positioning coordinates (Zx, Zy) of pointer points based on the Hough straight line; and
determining a pointer azimuth angle according to the coordinates of the circle center and the coordinates of a pointer point, wherein the preprocessed dial plate picture is divided into four quadrants which are divided into 0-90 degrees, 90-180 degrees, 180-270 degrees and 270-360 degrees; then, the coordinate of the circle center is used as a dot, the coordinate of the pointer point is used as the head of the pointer, the coordinate of the circle center is used as the tail of the pointer, and coordinate increments dx and dy are calculated, dx = Zx-Cx, and dy = Zy-Cy;
if dx, dy are not zero, first calculate a = arcatn (| dy/dx |), and then determine the pointer azimuth according to the range of dx, dy:
pointer azimuth = a when dx >0, dy > 0;
pointer azimuth =180-a when dx <0, dy > 0;
pointer azimuth =180+ a when dx <0, dy < 0; negative range of a-pi
Pointer azimuth =360-a when dx >0, dy < 0; negative range of-a
Outputting pointer azimuth information, and giving an alarm in real time in response to exceeding a preset scale range;
and the association module is used for associating the determined coordinates (Cx, Cy) of the circle center, the pointer point coordinates (Zx, Zy) and the pointer azimuth angle to the dial plate ID.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be determined by the appended claims.
Claims (6)
1. A pointer type dial plate identification method based on multi-dimensional feature analysis is characterized by comprising the following steps:
establishing a pointer type dial plate identification library;
establishing identification data for the identified pointer dials and storing the identification data into an identification library, wherein the identification data designates a dial plate ID for each pointer dial and a corresponding dial plate picture thereof, the dial plate ID is associated with the dial plate picture, the identification data also comprises multi-dimensional structured identification data for the dial plate picture, the structured identification data comprises a weighting weight Ai obtained by giving a weight value to the shape, color, scale and pointer characteristics of the pointer dials according to a preset rule, and the dial plate structure attribute and the coordinate of the pointer dials, and i =1,2,3,. m, m represents the total number of the pointer dials stored in the identification library;
receiving an input dial plate picture to be identified in the pointer type dial plate identification process;
preprocessing the received dial plate picture to obtain a pure picture;
detecting the shape, color, scale and pointer of a dial plate in the picture by adopting a target detection algorithm on the pure picture after noise reduction, and respectively obtaining a weight value for the recognized shape, color, scale and pointer of the dial plate according to a preset rule;
carrying out weighting calculation on the obtained weight value to obtain a weighting weight A;
polling and comparing the obtained weighting weight A with the weighting weight Ai in the identification library, and sequencing the absolute value Di of the difference value of the obtained weighting weight A and the weighting weight Ai in the identification library to obtain the weighting weight An corresponding to the minimum absolute value and the corresponding dial plate ID and dial plate picture, wherein n is less than or equal to m;
acquiring similarity values of the dial plate pictures detected by the received pictures and the dial plate corresponding to the weighting weights An; the similarity value similarity is obtained as follows:
similarity = obtained weighting weight a/weighting weight An corresponding to the minimum absolute value in the recognition library;
wherein:
responding to the similarity value similarity exceeding a set similarity threshold, and outputting the weighting weight An and the corresponding dial plate ID as a pointer type dial plate identification result;
in response to the fact that the similarity value similarity does not reach the set similarity threshold value, judging that a new pointer type dial plate is obtained, assigning a unique dial plate ID for the pointer type dial plate, associating the unique dial plate ID with the obtained dial plate picture and the calculated weighting weight, and storing the unique dial plate ID in the identification library; and
executing dial plate structure attribute identification and coordinate positioning on the input dial plate picture, storing the dial plate structure attribute and the coordinate into the identification library, and associating the dial plate structure attribute and the coordinate with a corresponding dial plate ID;
wherein the dial plate structure attribute identification and coordinate positioning comprises:
locating coordinates (Cx, Cy) of a center of the circle based on the Hough circle detection;
detecting and positioning coordinates (Zx, Zy) of pointer points based on the Hough straight line; and
determining a pointer azimuth angle according to the coordinates of the circle center and the coordinates of a pointer point, wherein the preprocessed dial plate picture is divided into four quadrants which are divided into 0-90 degrees, 90-180 degrees, 180-270 degrees and 270-360 degrees; then, the coordinate of the circle center is used as a dot, the coordinate of the pointer point is used as the head of the pointer, the coordinate of the circle center is used as the tail of the pointer, and coordinate increments dx and dy are calculated, dx = Zx-Cx, and dy = Zy-Cy;
if dx, dy are not zero, first calculate a = arcatn (| dy/dx |), and then determine the pointer azimuth according to the range of dx, dy:
pointer azimuth = a when dx >0, dy > 0;
pointer azimuth =180-a when dx <0, dy > 0;
pointer azimuth =180+ a when dx <0, dy < 0; negative range of a-pi
Pointer azimuth =360-a when dx >0, dy < 0; negative range of-a
And outputting pointer azimuth angle information, and giving an alarm in real time in response to the pointer exceeding a preset scale range.
2. The method for identifying the pointer-type dial plate based on the multi-dimensional feature analysis of claim 1, wherein the preprocessing includes gaussian fuzzy noise reduction processing, and specifically includes:
using a Gaussian fuzzy algorithm to correct RGB three channels of bright spots in a dial picture, wherein gradient percentages are used for detection, the bright spots are detected and then corrected through median filtering, finally performing transparency mixing operation based on corrected RGB, and mixing corrected RGB according to transparency, wherein:
when Alpha =50%, the signal is transmitted,
r = r1/2 + r2/2;
g = g1/2 + g2/2;
b = b1/2 + b2/2;
when Alpha is <50%, the Alpha is expressed,
r = r1 - r1/ALPHA + r2/ALPHA;
g = g1 - g1/ALPHA + g2/ALPHA;
b = b1 - b1/ALPHA + b2/ALPHA;
when Alpha >50% of the total weight of the composition,
r = r1/ALPHA + r2 - r2/ALPHA;
g = g1/ALPHA + g2 - g2/ALPHA;
b = b1/ALPHA + b2 - b2/ALPHA;
wherein Alpha is the transparency of the pointer picture, r, g, b are the color values of the final output, and r1, g1, b1 are the color values of the pre-color respectively; r2, g2 and b2 are color values of the background color, respectively.
3. The method for identifying the pointer dial plate based on the multi-dimensional feature analysis as claimed in claim 1, wherein the preprocessing further includes performing edge detection on the dial plate picture after the gaussian fuzzy noise reduction processing, and removing edge information to obtain the contour structure of the pointer dial plate.
4. A pointer dial plate recognition system based on multi-dimensional feature analysis, comprising:
a module for creating a pointer dial identification library;
the identification data comprises a module for establishing identification data for the identified pointer type dial plates and storing the identification data in an identification library, wherein the identification data designates a dial plate ID for each pointer type dial plate and a corresponding dial plate picture, the dial plate ID is associated with the dial plate picture, the identification data also comprises multi-dimensional structured identification data for the dial plate picture, the structured identification data comprises a weighting weight Ai obtained by giving a weighting value to the shape, color, scale and pointer characteristics of the pointer type dial plate according to a preset rule, and the dial plate structure attribute and the coordinate of the pointer type dial plate, and i =1,2, 3.. m, m represents the total number of the pointer type dial plates stored in the identification library;
the module is used for receiving an input dial plate picture to be identified in the pointer type dial plate identification process;
the module is used for preprocessing the received dial plate picture to obtain a pure picture;
a module for detecting the shape, color, scale and pointer of the dial plate in the picture by adopting a target detection algorithm for the pure picture after noise reduction, and respectively obtaining a weight value for the recognized shape, color, scale and pointer of the dial plate according to a preset rule;
a module for performing weighting calculation on the obtained weight value to obtain a weighting weight A;
a module for polling and comparing the obtained weighting A with the weighting Ai in the identification library, and sequencing the absolute value Di of the difference between the two, so as to obtain the weighting An corresponding to the minimum absolute value and the corresponding dial ID and dial picture, wherein n is less than or equal to m;
a module for obtaining the similarity value similarity of the dial plate picture detected by the received picture and the dial plate corresponding to the weighting weight An; the similarity value similarity is obtained in the following manner:
similarity = obtained weighting weight a/weighting weight An corresponding to the minimum absolute value in the recognition library;
a module for outputting the weighting An and the corresponding dial ID as the identification result in response to the similarity value similarity exceeding the set similarity threshold;
a module for determining a new pointer type dial plate in response to the similarity value similarity not reaching the set similarity threshold value, assigning a unique dial plate ID for the pointer type dial plate, associating the unique dial plate ID with the obtained dial plate picture and the calculated weighting weight, and storing the unique dial plate ID in the pointer type dial plate identification library; and
a module used for executing dial plate structure attribute identification and coordinate positioning to the input dial plate picture, storing the dial plate structure attribute and the coordinate into the identification library and associating with the corresponding dial plate ID;
the dial plate structure attribute identification and coordinate positioning system is used for executing dial plate structure attribute identification and coordinate positioning on an input dial plate picture, storing the dial plate structure attribute and the coordinate into the identification library, and the module associated with the corresponding dial plate ID comprises a dial plate structure attribute identification and coordinate positioning module and an associated module, wherein:
the dial structure property identification and coordinate locating module is configured to determine the dial structure property and coordinates in the following manner:
locating coordinates (Cx, Cy) of a center of the circle based on the Hough circle detection;
detecting and positioning coordinates (Zx, Zy) of pointer points based on the Hough straight line; and
determining a pointer azimuth angle according to the coordinates of the circle center and the coordinates of a pointer point, wherein the preprocessed dial plate picture is divided into four quadrants which are divided into 0-90 degrees, 90-180 degrees, 180-270 degrees and 270-360 degrees; then, the coordinate of the circle center is used as a dot, the coordinate of the pointer point is used as the head of the pointer, the coordinate of the circle center is used as the tail of the pointer, and coordinate increments dx and dy are calculated, dx = Zx-Cx, and dy = Zy-Cy;
if dx, dy are not zero, first calculate a = arcatn (| dy/dx |), and then determine the pointer azimuth according to the range of dx, dy:
pointer azimuth = a when dx >0, dy > 0;
pointer azimuth =180-a when dx <0, dy > 0;
pointer azimuth =180+ a when dx <0, dy < 0; negative range of a-pi
Pointer azimuth =360-a when dx >0, dy < 0; negative range of-a
Outputting pointer azimuth information, and giving an alarm in real time in response to exceeding a preset scale range;
and the association module is used for associating the determined coordinates (Cx, Cy) of the circle center, the pointer point coordinates (Zx, Zy) and the pointer azimuth angle to the dial plate ID.
5. The system for identifying a dial plate with pointers based on multi-dimensional feature analysis of claim 4, wherein the module for preprocessing the received dial plate picture to obtain a pure picture comprises a Gaussian blur noise reduction module configured to perform noise reduction according to the following method:
using a Gaussian fuzzy algorithm to correct RGB three channels of bright spots in a dial picture, wherein gradient percentages are used for detection, the bright spots are detected and then corrected through median filtering, finally performing transparency mixing operation based on corrected RGB, and mixing corrected RGB according to transparency, wherein:
when Alpha =50%, the signal is transmitted,
r = r1/2 + r2/2;
g = g1/2 + g2/2;
b = b1/2 + b2/2;
when Alpha is <50%, the Alpha is expressed,
r = r1 - r1/ALPHA + r2/ALPHA;
g = g1 - g1/ALPHA + g2/ALPHA;
b = b1 - b1/ALPHA + b2/ALPHA;
when Alpha >50% of the total weight of the composition,
r = r1/ALPHA + r2 - r2/ALPHA;
g = g1/ALPHA + g2 - g2/ALPHA;
b = b1/ALPHA + b2 - b2/ALPHA;
wherein Alpha is the transparency of the pointer picture, r, g, b are the color values of the final output, and r1, g1, b1 are the color values of the pre-color respectively; r2, g2 and b2 are color values of the background color, respectively.
6. The system for identifying a dial plate based on multi-dimensional feature analysis of claim 5, wherein the module for preprocessing the received dial plate picture to obtain a pure picture further comprises a module for performing edge detection on the dial plate picture after Gaussian blur noise reduction processing, and removing edge information to obtain a contour structure of the dial plate.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011001595.4A CN111931716B (en) | 2020-09-22 | 2020-09-22 | Pointer type dial plate identification method and system based on multi-dimensional characteristic analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011001595.4A CN111931716B (en) | 2020-09-22 | 2020-09-22 | Pointer type dial plate identification method and system based on multi-dimensional characteristic analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111931716A CN111931716A (en) | 2020-11-13 |
CN111931716B true CN111931716B (en) | 2021-01-19 |
Family
ID=73333664
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011001595.4A Active CN111931716B (en) | 2020-09-22 | 2020-09-22 | Pointer type dial plate identification method and system based on multi-dimensional characteristic analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111931716B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108460327A (en) * | 2018-01-12 | 2018-08-28 | 河南大学 | A kind of readings of pointer type meters automatic identifying method based on image procossing |
CN110580480A (en) * | 2019-07-21 | 2019-12-17 | 天津大学 | Meter reading identification method based on image processing |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8965762B2 (en) * | 2007-02-16 | 2015-02-24 | Industrial Technology Research Institute | Bimodal emotion recognition method and system utilizing a support vector machine |
CN104751187B (en) * | 2015-04-14 | 2019-04-16 | 山西科达自控股份有限公司 | Meter reading automatic distinguishing method for image |
CN108256524B (en) * | 2018-01-24 | 2021-12-17 | 国家电网公司 | Automatic reading method of multi-dial pointer instrument |
-
2020
- 2020-09-22 CN CN202011001595.4A patent/CN111931716B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108460327A (en) * | 2018-01-12 | 2018-08-28 | 河南大学 | A kind of readings of pointer type meters automatic identifying method based on image procossing |
CN110580480A (en) * | 2019-07-21 | 2019-12-17 | 天津大学 | Meter reading identification method based on image processing |
Also Published As
Publication number | Publication date |
---|---|
CN111931716A (en) | 2020-11-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11403839B2 (en) | Commodity detection terminal, commodity detection method, system, computer device, and computer readable medium | |
CN111307727B (en) | Water body water color abnormity identification method and device based on time sequence remote sensing image | |
CN110766095B (en) | Defect detection method based on image gray level features | |
CN108491498B (en) | Bayonet image target searching method based on multi-feature detection | |
CN106529559A (en) | Pointer-type circular multi-dashboard real-time reading identification method | |
CN111398176B (en) | Water body water color abnormity remote sensing identification method and device based on pixel scale characteristics | |
CN111563896B (en) | Image processing method for detecting abnormality of overhead line system | |
CN110197185B (en) | Method and system for monitoring space under bridge based on scale invariant feature transform algorithm | |
CN116309557B (en) | Method for detecting fracture of track shoe of excavator | |
CN117689655B (en) | Metal button surface defect detection method based on computer vision | |
CN114627089A (en) | Defect identification method, defect identification device, computer equipment and computer readable storage medium | |
CN112164055A (en) | Photovoltaic cell color difference classification method based on color segmentation | |
CN111931716B (en) | Pointer type dial plate identification method and system based on multi-dimensional characteristic analysis | |
CN112036391B (en) | Electronic inspection method and system based on shale gas field production station | |
Yi et al. | A clustering-based algorithm for automatic detection of automobile dashboard | |
CN112461345B (en) | Truck scale rolling line out-of-bounds detection method based on LSD (least squares distortion) linear detection algorithm | |
CN110378425B (en) | Intelligent image comparison method and system | |
CN115222652A (en) | Method for identifying, counting and centering end faces of bundled steel bars and memory thereof | |
CN112288747A (en) | Intelligent detection method and device for steel billets | |
CN113239944A (en) | Image feature extraction method and device, electronic equipment and medium | |
CN112541471A (en) | Shielded target identification method based on multi-feature fusion | |
CN117115275B (en) | Distortion parameter determination method and device and computer equipment | |
CN114299037B (en) | Quality evaluation method and device for object detection result, electronic equipment and computer readable storage medium | |
CN117115172B (en) | Donkey-hide gelatin quality detection method and system based on machine vision | |
CN113591564B (en) | Scene abnormal state detection method |
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 | ||
CP01 | Change in the name or title of a patent holder |
Address after: Building 10, No.19, ningshuang Road, Yuhuatai District, Nanjing City, Jiangsu Province, 210000 Patentee after: Jiahuan Technology Co.,Ltd. Address before: Building 10, No.19, ningshuang Road, Yuhuatai District, Nanjing City, Jiangsu Province, 210000 Patentee before: Nanjing Jiahuan Technology Co., Ltd |
|
CP01 | Change in the name or title of a patent holder |