AU2020103716A4 - Training method and device of automatic identification device of pointer instrument with numbers in natural scene - Google Patents
Training method and device of automatic identification device of pointer instrument with numbers in natural scene Download PDFInfo
- Publication number
- AU2020103716A4 AU2020103716A4 AU2020103716A AU2020103716A AU2020103716A4 AU 2020103716 A4 AU2020103716 A4 AU 2020103716A4 AU 2020103716 A AU2020103716 A AU 2020103716A AU 2020103716 A AU2020103716 A AU 2020103716A AU 2020103716 A4 AU2020103716 A4 AU 2020103716A4
- Authority
- AU
- Australia
- Prior art keywords
- pointer
- information
- image
- instrument panel
- instrument
- 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.)
- Ceased
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 title claims abstract description 8
- 230000011218 segmentation Effects 0.000 abstract description 8
- 230000009466 transformation Effects 0.000 abstract description 2
- 238000013135 deep learning Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 8
- 238000000605 extraction Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 2
- 238000009776 industrial production Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement 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
- 230000001537 neural effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/16—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
Abstract
The invention discloses a training method and device for pointer type
instrument automatic reading recognition device based on deep learning.
The method includes: for the collected original image information, the
instrument panel in the image is separated from the natural background,
and the digital area in the original image is marked and extracted at the
same time; the extracted instrument panel information is used to advance
the pointer of the instrument panel Line fitting and positioning to find out
the position information of the pointer; using the obtained digital area to
identify the number on the instrument panel; using the position
information of the pointer, the numerical information of the number and
the position information of the number, the indication of the pointer type
instrument is determined. The invention can construct a complete, simple
and efficient scheme for specific domain vocabulary discovery and
classifier training.
1
Original pointer meter pictures collected s101
The image feature is extracted and the
image feature vector is obtained s102
Generate dashboard information feature s103
vector and digital area feature vector
Figure 1
Image of dashboard after segmentation s201
The binary image is obtained by color
space transformation
After binarization, the pointer fitting and
positioning are carried out for the image 03
of the instrument panel
Figure 2
1
Description
Original pointer meter pictures collected s101
The image feature is extracted and the image feature vector is obtained s102
Generate dashboard information feature s103 vector and digital area feature vector
Figure 1
Image of dashboard after segmentation s201
The binary image is obtained by color space transformation
After binarization, the pointer fitting and positioning are carried out for the image 03 of the instrument panel
Figure 2
Training method and device of automatic identification device of pointer
instrument with numbers in natural scene
The invention relates to the field of automatic reading recognition of
pointer type instrument, in particular to a training method and device for
automatic reading recognition device of pointer instrument containing
digital scale information.
With the continuous development of science and technology, the
continuous progress of society, and the continuous expansion of industrial
production scale, as an important tool for data monitoring and data
collection, instruments play an important role in production and life.
Although the detection and data transmission technology of digital
display instrument has been quite mature, in some traditional industrial
production and some complex environment, pointer instrument is still the
main measurement tool. With the information and data of industrial
production and life, pointer instrument still needs manual input of
readings, which not only consumes a lot of manpower, but also is prone
to errors due to heavy workload, which seriously affects the efficiency of
industrial production. For the pointer instrument without data
transmission, there are some problems such as high cost, complex operation and excessive waste of resources. Therefore, how to automatically read recognition of pointer meters is particularly important in industrial production and life.
At present, a lot of research has been done on the automatic reading
recognition of pointer instruments in China, and a series of achievements
have been achieved. Usually, the automatic recognition of pointer
instrument mainly includes three important parts: the separation of
instrument panel from environment background, positioning pointer and
indication recognition and automatic reading. Among them, the main
methods of instrument panel and environment background separation
include: template matching method, Ostu adaptive threshold
segmentation method. However, these two mainstream algorithms have
their limitations as image segmentation. Template matching method
limits the types of pointer instrument recognition, and can only separate
the instrument panel and background of given template, which has poor
generalization. However, Ostu adaptive threshold segmentation method
has too high requirements for illumination conditions and instrument
background, so it is unable to effectively segment instrument panel and
natural background in complex lighting conditions and natural
environment.
The purpose of the invention is to propose a training method and device
for automatic reading recognition device of pointer type instrument
containing digital scale information in natural scene, and to construct a
complete, simple and efficient scheme for pointer type instrument
identifier containing digital scale information in natural scene.
Based on the above purpose, the invention provides a method for
automatic reading recognition of pointer meter containing digital scale
information in natural scene, including:
For the original image information collected, the instrument panel is
separated from the natural background, and the digital area in the original
image is marked and extracted;
In the above operations, the information extraction of the original image
is realized by using the designed deep convolution neural network model
according to the feature vector representing the original image.
For the extracted information of the instrument panel, the pointer in the
instrument panel is fitted and positioned to find out the position
information of the pointer;
Among them, the location information of the pointer is extracted by
Hough probability line.
For the obtained digital area, the number on the instrument panel is
identified;
Among them, for the number recognition, the KNN classifier trained in
advance is used to recognize the information in the digital region.
Finally, the pointer position information, numerical value information and
digital position information obtained before are used to determine the
indication of the pointer instrument.
For the final determination of the indication, the distance method is used
to determine the indication of the pointer meter according to the
numerical information and position information of the number.
The invention also provides a training method for pointer type instrument
recognizer in natural scene, including:
Determine the feature vector of the original data, dashboard information,
digital area information and numerical information;
Using the eigenvector of original data, the feature vector of dashboard
information and the feature vector of digital area information, the
recognizer of pointer instrument with digital information in natural scene
is trained by deep convolution neural network model;
Among them, the label value of invalid information is 0, that of
dashboard information is 1, and that of digital area information is 2.
The invention also provides a pointer type instrument image recognition
reading device in natural scene, including:
The original image feature extraction module, which is used to extract the
features of the original image information and get the feature vector of
the original image.
The instrument panel and digital area recognition module, which is used
to recognize the feature vector of the original image, and the feature
vector of the original image is the feature vector of the image generated
by the feature extraction module of the original image.
The pointer positioning module of the instrument panel, which is used to
simulate and merge the pointer in the instrument panel and extract the
position information of the pointer; the information used is the feature
vector generated by the combination of the feature vector information of
the instrument panel and the original image information generated by the
instrument panel and digital area recognition module.
The digital image recognition module, which is used to identify the
digital image information extracted from the digital area and generate the
numerical information; the digital image information is the feature vector
generated by the instrument panel and the digital area recognition module
and the original image.
Digital determination module for instrument display, which is used to
determine the reading of the pointer meter. The numerical information
obtained by the numerical identification module, the digital area
information obtained by the instrument panel and digital area identification module, and the pointer position information obtained by the pointer positioning module of the instrument panel are used to determine the reading value.
In the technical scheme of the invention, the feature extraction of the
instrument panel information in the natural scene is carried out by using
the deep convolution neural network model, and then the digit
recognition, pointer positioning and reading judgment are performed
according to the extracted information. Based on the comparison and
combination of the previous pointer instrument identification design, this
scheme focuses on solving the problem that it is difficult to extract the
dashboard information in natural scenes. At the same time, it is a scheme
with high generalization, strong robustness and good universality.
Fig. 1 is a flow chart of a method for feature extraction of natural
scene instrument provided by the embodiment of the invention;
Fig. 2 is a flow chart of a method for locating the pointer of an
instrument panel provided by an embodiment of the invention;
Fig. 3 is a flow chart of a digital area numerical recognition method
provided by the embodiment of the invention;
Fig. 4 is a flow chart of a method for automatically determining the
indication of an instrument provided by an embodiment of the
invention;
Fig. 5 is a flow chart of a method for determining the automatic
reading of a pointer type instrument provided by an embodiment of
the invention
In order to make the purpose, technical scheme and advantages of the
invention clearer and clearer, the present invention is further described in
detail in combination with specific embodiments and with reference to
the attached drawings.
An embodiment of the invention is described in detail below, an example
of which is shown in the accompanying drawings, in which the same or
similar labels from beginning to end indicate the same or similar elements
or elements with the same or similar functions. The embodiments
described below with reference to the accompanying drawings are
illustrative and are intended to explain the invention only and cannot be
interpreted as a limitation of the invention.
It can be understood by those skilled in the art that the singular forms
"one" and "the" used herein may also include the plural forms, unless
specifically stated. It should be further understood that when we call an
element "connected" or "coupled" to another element, it can be directly
connected or coupled to other components, or there may be intermediate
elements. In addition, "connection" or "coupling" used herein may include wireless connection or wireless coupling. The words "and / or" as used here include all or any unit and all combinations of one or more associated listed items.
It should be noted that all expressions of "first" and "second" in the
embodiment of the present invention are used to distinguish two entities
with the same name but not the same parameters. It can be seen that
"first" and "second" are only for the convenience of expression, and
should not be understood as the limitation of the embodiment of the
invention. Subsequent embodiments will not explain them one by one.
The technical scheme of the embodiment of the invention is described in
detail below in combination with the attached drawings.
S101: original pointer meter pictures collected
Specifically, the image acquisition of the pointer instrument is carried out
to obtain the format which is convenient for the computer to read directly,
such as JPG, PNG, etc. The image feature vector read by computer is a
three-dimensional matrix (w * h * 3), where w represents the width of the
image, H represents the height of the image, and 3 represents the color
space of the image.
S102: extract image features and get image feature vector;
Specifically, the trained deep convolution neural network is used to
extract the features of the original pointer instrument image, and the
feature vector of the image is obtained.
S103: generate dashboard information feature vector and digital area
feature vector;
Specifically, the feature vector of image obtained by S102 is used to
generate the feature vector of dashboard information and digital region by
using the deep convolution network MASKR2CNN
The information feature vector of the dashboard is a two-dimensional
matrix with the same width and height as the original image. The
elements of the two-dimensional matrix are 0 or 1. Multiple connected
ones are used to represent the position information of the dashboard in the
original image, namely mask information. The representation of digital
region is a 1 * 5 eigenvector, i.e. (x, y, h, w, theta), where x and y
represent the coordinates of the center point of the digital region in the
original image. h. w represents the height and width of the digital area,
theta represents the angle between the rectangle of the digital area and the
horizontal direction.
The invention designs an optimization method for improving the model
of MASKRCNN, and uses the improved MASK2CNN model to extract
information from the collected original data.
The structure of MASKR2CNN belongs to the classic two-step
recommended target detection network. For the collected original image
information, MASKR2CNN uses Resnet101 deep convolution neural
network to extract the features of the original image. At the same time, the feature information generated from different depths is fused, and a feature pyramid FPN model is constructed to represent the image features.
On the FPN model, candidate frames with different proportions and
different aspect ratios are generated by anchor points, and the RPN model
is used to identify the candidate frames The candidate frames are
classified and scored by foreground and background, and the regression
values of candidate frames and result frames are generated to screen the
preliminarily recommended candidate frames. For the preliminarily
screened candidate frames, the scale normalization pooling operation is
performed. For the pooled characteristic information, part of the
candidate frames are classified and the coordinates of inclined frames
are generated, and some of them are used The segmentation region of the
target is generated by deconvolution. After the original image passes
through a network model of MASR2CNN, we can get the information of
the inclined frame of the digital area and the segmentation information of
the dashboard separated from the natural scene.
The embodiment of the invention provides a flow chart of a method for
positioning the pointer of an instrument panel;
S201: image of instrument panel after segmentation;
Specifically, the image information of the instrument panel is obtained by
segmenting the image of the instrument panel on the original image of the
pointer instrument panel.
S202: transform the image color space to get binary image;
Specifically, the image information of instrument panel is binarized based
on Ostu method, and the binary image information of dashboard is
obtained.
S203: fitting and positioning the pointer on the image of the instrument
panel after binarization;
Specifically, using the binary image of the instrument panel, the
probability Hough transform method is used to locate the pointer of the
instrument panel, and the location information (x1, Y, X2, Y2) of the
pointer is the two endpoint coordinates of the pointer segment.
The embodiment of the invention provides a flow chart of a method for
numerical recognition of a digital region;
S301: digital region image after segmentation;
Specifically, using the digital region information obtained before, the
digital region is segmented on the original image of the pointer
instrument, and the image information of several different digital regions
is obtained.
S302: change the color space of digital region image to get binary image;
Specifically, the image information of digital region is binarized based on
Ostu method, and the binary image information of digital region is
obtained.
S303: to recognize the value of the digital region image after binarization;
Specifically, using the image information of binary digital region
obtained before, the digital region is segmented by contour tracking
method, which makes the digital area become a single character
connected form. Then the size of the segmented image is normalized and
input into the pre trained KNN for 11 classification of single digit and
random information The whole recognition of digital area is realized. Get
the indication of the last number are{ Sa, Sb , Sn}, such as 20, 10,
523 and so on.
The embodiment of the invention provides a method flow chart for
automatic determination of meter indication;
S401: determine the distance between the digital area and the pointer;
Specifically, by using the position feature vector (x, y, h, w, theta) of
the digital region and the position information of the pointer (x1 , y,
x 2 ,Y2), the distance information between the digital region and the
pointer is obtained {L 1 , L 2 ,--- ,L,}
S402: determine the indication of pointer instrument
Specifically, according to the obtained distance information, the
minimum two values La and Lb are selected, and the corresponding
digital area recognition scores Sa and Sb are obtained. According to the
following formula, the indication s of the pointer instrument is obtained:
S = Sa + La *(Sb - Sa) La+Lb
The embodiment of the invention provides a flow chart of a method for
determining the automatic reading of a pointer type instrument;
The detection and extraction module 501 of instrument panel and
instrument number is used to separate the instrument panel and natural
background in the collected original picture letter, and mark and extract
the digital area in the original picture.
The digital recognition module 502 of the instrument panel is used to fit
and locate the pointer in the instrument panel by using the extracted
instrument panel information, and find out the position information of the
pointer.
The number recognition module 503 of the instrument panel is used to
recognize the number on the instrument panel by using the obtained
digital area.
The indicator judging module 504 of the pointer type instrument is used
to obtain the position information of the pointer, the numerical value
information of the number and the position information of the number, so
as to determine the indication of the pointer instrument.
In the technical scheme of the invention, firstly, the feature extraction of
instrument panel information in natural scene is carried out by using the
designed deep convolution neural network model, and then digital
recognition, pointer positioning and reading judgment are performed
according to the extracted information. On the basis of comparison and combination of the previous pointer instrument recognition design, on the one hand, it solves the problem that the information of the instrument panel is difficult to extract in the natural scene, and on the other hand, it solves the problem of slant digit recognition on the instrument panel. It is a scheme with high generalization, strong robustness and good versatility.
It will be understood by those skilled in the art that the invention includes
apparatus for performing one or more of the operations described in the
present application. These devices may be specially designed and
manufactured for the desired purpose or may include known devices in
general-purpose computers. These devices have computer programs
stored in them, which are selectively activated or reconstructed. Such a
computer program may be stored in a device (e.g., computer) readable
medium, or in any type of medium suitable for storing electronic
instructions and separately coupled to the bus. The computer-readable
medium includes, but is not limited to, any type of disk (including soft
disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM
(read only memory) Read memory, RAM (random access memory),
EPROM (erasable programmable read only memory), EEPROM
(electrically erasable programmable read only memory), flash memory,
magnetic card or light card. That is, a readable medium includes any
medium in which information is stored or transmitted by a device (E. G.,
a computer) in a form that can be read. It can be understood by those skilled in the art that each frame in these block diagrams and / or block diagrams and / or flow diagrams and combinations of boxes in these structure diagrams and / or block diagrams and / or flow diagrams can be realized by computer program instructions. It can be understood by those skilled in the art that these computer program instructions can be provided to a general-purpose computer, a professional computer or a processor of other programmable data processing methods to implement the structure diagram and / or frame disclosed by the invention through a computer or a processor of other programmable data processing methods
The scheme specified in the box or boxes of a graph and / or flow
diagram.
Those skilled in the art can understand that the various operations,
methods, steps, measures and schemes in the process discussed in the
invention can be alternated, modified, combined or deleted. Further, other
steps, measures and schemes with various operations, methods and
processes discussed in the present invention can also be alternated,
changed, rearranged, decomposed, combined or deleted. Furthermore, the
steps, measures and schemes in various operations, methods and
processes disclosed in the prior art can also be alternated, modified,
rearranged, decomposed, combined or deleted. Those of ordinary skill in
the art should understand that the discussion of any of the above
embodiments is only exemplary and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples; Under the idea of the invention, the technical features in the above embodiments or different embodiments can also be combined, and the steps can be implemented in any order, and there are many other changes in different aspects of the invention as described above, which are not provided in details for the sake of simplicity. Therefore, the
Any omission, modification, equivalent replacement, improvement, etc.
shall be included in the protection scope of the invention.
Claims (3)
1. A training method for automatic reading recognition device of pointer
type instrument containing digital scale information in natural scene,
characterized in that it comprises the following steps: constructs a
complete, simple and efficient scheme of pointer type instrument
identifier containing digital scale information in natural scene.
2. The method according to claim 1, characterized in that, the feature
vector, dashboard information eigenvector, digital area information
eigenvector and numerical information of the original data are
determined.
3. The method according to claim 1, characterized in that, the digital
recognition of the invention adopts the KNN classifier trained in advance
to recognize the digital area.
Figure 2 Figure 1
Figure 4 Figure 3
Figure 5
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2020103716A AU2020103716A4 (en) | 2020-11-27 | 2020-11-27 | Training method and device of automatic identification device of pointer instrument with numbers in natural scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2020103716A AU2020103716A4 (en) | 2020-11-27 | 2020-11-27 | Training method and device of automatic identification device of pointer instrument with numbers in natural scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| AU2020103716A4 true AU2020103716A4 (en) | 2021-02-11 |
Family
ID=74502219
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| AU2020103716A Ceased AU2020103716A4 (en) | 2020-11-27 | 2020-11-27 | Training method and device of automatic identification device of pointer instrument with numbers in natural scene |
Country Status (1)
| Country | Link |
|---|---|
| AU (1) | AU2020103716A4 (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112906694A (en) * | 2021-03-25 | 2021-06-04 | 中国长江三峡集团有限公司 | Reading correction system and method for inclined pointer instrument image of transformer substation |
| CN113065556A (en) * | 2021-04-16 | 2021-07-02 | 西安交通大学 | Digital instrument positioning and digital identification method and device and computer equipment |
| CN113326787A (en) * | 2021-06-02 | 2021-08-31 | 武汉理工大学 | Automatic identification method, system and equipment for reading of pointer instrument |
| CN113361539A (en) * | 2021-05-21 | 2021-09-07 | 煤炭科学技术研究院有限公司 | Instrument reading method and device of underground inspection robot and electronic equipment |
| CN116310285A (en) * | 2023-02-16 | 2023-06-23 | 武汉科技大学 | Automatic pointer instrument reading method and system based on deep learning |
| 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 |
-
2020
- 2020-11-27 AU AU2020103716A patent/AU2020103716A4/en not_active Ceased
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112906694A (en) * | 2021-03-25 | 2021-06-04 | 中国长江三峡集团有限公司 | Reading correction system and method for inclined pointer instrument image of transformer substation |
| CN112906694B (en) * | 2021-03-25 | 2022-10-04 | 中国长江三峡集团有限公司 | Reading correction system and method for transformer substation inclined pointer instrument image |
| CN113065556A (en) * | 2021-04-16 | 2021-07-02 | 西安交通大学 | Digital instrument positioning and digital identification method and device and computer equipment |
| CN113065556B (en) * | 2021-04-16 | 2024-04-02 | 西安交通大学 | Digital instrument positioning and digital identifying method and device and computer equipment |
| CN113361539A (en) * | 2021-05-21 | 2021-09-07 | 煤炭科学技术研究院有限公司 | Instrument reading method and device of underground inspection robot and electronic equipment |
| CN113326787A (en) * | 2021-06-02 | 2021-08-31 | 武汉理工大学 | Automatic identification method, system and equipment for reading of pointer instrument |
| CN113326787B (en) * | 2021-06-02 | 2023-09-01 | 武汉理工大学 | Automatic identification method, system and equipment for pointer instrument reading |
| CN116310285A (en) * | 2023-02-16 | 2023-06-23 | 武汉科技大学 | Automatic pointer instrument reading method and system based on deep learning |
| CN116310285B (en) * | 2023-02-16 | 2024-02-27 | 科大集智技术湖北有限公司 | Automatic pointer instrument reading method and system based on deep learning |
| 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 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2020103716A4 (en) | Training method and device of automatic identification device of pointer instrument with numbers in natural scene | |
| CN112906694A (en) | Reading correction system and method for inclined pointer instrument image of transformer substation | |
| WO2021109775A1 (en) | Methods and devices for generating training sample, training model and recognizing character | |
| CN110378225A (en) | The recognition methods of pointer instrument automatic reading and identification device | |
| CN113449594A (en) | Multilayer network combined remote sensing image ground semantic segmentation and area calculation method | |
| CN112949338A (en) | Two-dimensional bar code accurate positioning method combining deep learning and Hough transformation | |
| CN110956100A (en) | High-precision map generation method and device, electronic equipment and storage medium | |
| CN111854758A (en) | Indoor navigation map conversion method and system based on building CAD (computer-aided design) drawing | |
| CN109993154A (en) | The lithium sulfur type instrument intelligent identification Method of substation's simple pointer formula | |
| EP4105600A2 (en) | Method for automatically producing map data, related apparatus and computer program product | |
| CN110659637A (en) | Electric energy meter number and label automatic identification method combining deep neural network and SIFT features | |
| CN112434706B (en) | High-precision traffic element target extraction method based on image point cloud fusion | |
| CN116188756A (en) | Instrument angle correction and indication recognition method based on deep learning | |
| CN110516527B (en) | Visual SLAM loop detection improvement method based on instance segmentation | |
| CN115841669A (en) | Pointer instrument detection and reading identification method based on deep learning technology | |
| CN109508714B (en) | Low-cost multi-channel real-time digital instrument panel visual identification method and system | |
| CN104200226A (en) | Particle filtering target tracking method based on machine learning | |
| CN116704490B (en) | License plate recognition method, license plate recognition device and computer equipment | |
| CN111553345B (en) | Method for realizing meter pointer reading identification processing based on Mask RCNN and orthogonal linear regression | |
| CN109492573A (en) | A kind of pointer read method and device | |
| CN116758045B (en) | Surface defect detection method and system for semiconductor light-emitting diode | |
| CN113034511A (en) | Rural building identification algorithm based on high-resolution remote sensing image and deep learning | |
| CN115169375B (en) | AR and gun ball linkage-based high-level material visualization method | |
| CN116188755A (en) | Instrument angle correction and reading recognition device based on deep learning | |
| CN109145912A (en) | A kind of digital instrument reading automatic identifying method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FGI | Letters patent sealed or granted (innovation patent) | ||
| MK22 | Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry |