CN108133216A - The charactron Recognition of Reading method that achievable decimal point based on machine vision is read - Google Patents
The charactron Recognition of Reading method that achievable decimal point based on machine vision is read Download PDFInfo
- Publication number
- CN108133216A CN108133216A CN201711166125.1A CN201711166125A CN108133216A CN 108133216 A CN108133216 A CN 108133216A CN 201711166125 A CN201711166125 A CN 201711166125A CN 108133216 A CN108133216 A CN 108133216A
- Authority
- CN
- China
- Prior art keywords
- character
- charactron
- decimal point
- image
- read
- 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
Links
Classifications
-
- 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
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- 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
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Character Input (AREA)
Abstract
The present invention relates to a kind of charactron Recognition of Reading methods that achievable decimal point based on machine vision is read.This method includes:(1) the single character picture sample database of charactron is built, the HOG features of character picture and generates SVM classifier in learning sample library;(2) charactron character picture is acquired, extracts the character zone in charactron character picture;(3) Character segmentation;(4) decimal point is read;(5) character is identified with grader.The sorting algorithm of the invention being combined using HOG features and SVM algorithm and unique decimal point read method can effectively read the charactron reading of mixed decimal point, and recognition accuracy is high and robustness is good.
Description
Technical field
The invention belongs to field of image recognition, the number of more particularly to a kind of achievable decimal point reading based on machine vision
Code pipe Recognition of Reading method.
Background technology
The every field that digital instrumentation is widely used in producing and live by the advantages that its precision height, easily use.However,
It is many times main to the reading of digital instrument at present still manually to realize, it is time-consuming by the way of manual work,
Labor intensity is big, and efficiency is low, and in some specific occasions, and safety is low, is identified with the digital instrument based on machine vision
Instead of manually reading the deficiency solved above.
Charactron identification based on machine vision is that the image of charactron is acquired with imaging sensor, is known from image automatically
Do not go out the reading of charactron.There are the charactron visual identity method more than comparison, such as look-up table, threading method, neural network at present,
But it is directed to the identification of numerical character mostly, the recognition methods of decimal point is without effective scheme.
Invention content
The technical problems to be solved by the invention be in view of the deficiencies of the prior art, provide it is a kind of based on machine vision can
Realize the charactron Recognition of Reading method that decimal point is read.The sorting algorithm that the present invention is combined using HOG features and SVM algorithm
With unique decimal point read method, the charactron reading of mixed decimal point, recognition accuracy height and robustness can be effectively read
It is good.
To achieve these goals, the present invention uses following technical scheme:
The charactron Recognition of Reading method that achievable decimal point based on machine vision is read, includes the following steps:
(1) all single character pictures of charactron for only including a character are collected, build sample database, are read in sample database
All single character sample images extract the HOG features of each character sample image, learn and generate SVM classifier;
(2) new charactron character picture is acquired, extracts the character zone in charactron character picture;
(3) Character segmentation is carried out to the character in the charactron character zone image of extraction, generates several single characters
Image;
(4) decimal point of charactron character zone is read;
(5) several single character pictures of segmentation are identified respectively with grader, existed with reference to character and decimal point
The reading of pixel coordinate position generation charactron in charactron character zone.
Further, the charactron Recognition of Reading read in the achievable decimal point of the present invention based on machine vision
In method, which is characterized in that the charactron is seven segment digital tubes.
Further, the charactron Recognition of Reading read in the achievable decimal point of the present invention based on machine vision
The step of each character sample image HOG features are extracted in method, in the step (1) is as follows:
A. to charactron character sample image preprocessing, that is, histogram equalization and median filter process are carried out;
B. binary conversion treatment is carried out to pretreated image again, obtains the binary image of charactron character sample;
C. binary image character zone is extracted, generation is only comprising character and the size character zone figure identical with character
Picture;
D. to the character zone image scaling after extraction to given size, wide a height of 32 pixel *, 64 pixels;
E. the HOG features of the character zone image after extraction scaling.
Further, the charactron Recognition of Reading read in the achievable decimal point of the present invention based on machine vision
In method, the character zone extraction in the step (2) in charactron character picture refers to that extraction only includes charactron character
Binary image, the gray value of character is 255 in binary image, and the gray value of background is 0.
Further, the charactron Recognition of Reading read in the achievable decimal point of the present invention based on machine vision
It is as follows the step of single Character segmentation in the step (3) in method:
A. the vertical range of charactron character zone image is adjusted, obtains character zone up-and-down boundary coordinate;
B. for tilted character into line character Slant Rectify;
C. single Character segmentation is carried out for all characters in region.
Further, the charactron reading read in the achievable decimal point of the present invention based on machine vision is known
In other method, in the step a, vertical range method of adjustment is:By the binary image of the charactron character of extraction to vertical
Direction projection, projection formula are:Using the line number of image as horizontal axis, with corresponding often capable gray value
Pixel number for 255 generates projection histogram as the longitudinal axis, and scanning projection histogram extracts the minimax side of histogram
Boundary's coordinate, and then obtain the up-and-down boundary of charactron character;
Wherein, SjThe summation of pixel that image pixel value for the i-th row is 255, i, j are respectively that the row and column of pixel is sat
Mark, col widths of the cols for image, P (i, j) value 0 or 1, when the gray value that coordinate is the pixel at (i, j) is 255, P
(i, j) takes 1, is otherwise 0.
Further, the charactron reading read in the achievable decimal point of the present invention based on machine vision is known
In other method, in the step b, character Slant Rectify is specially:The charactron character binaryzation image of extraction is refined, will be refined
Figure is projected to level into all directions in the range of ± 60 °, and the direction of the projection of pixel number maximum in projection histogram is selected to make
For the direction of character, then character is corrected with Shear Transform.
Further, the charactron reading read in the achievable decimal point of the present invention based on machine vision is known
In other method, in the step c, single Character segmentation step is:Character picture after correction is projected to horizontal direction, to scheme
The row number of picture generates projection histogram, sequential scan projection using the pixel number of corresponding each column as the longitudinal axis as horizontal axis
Histogram extracts the right boundary of each character successively, and the up-and-down boundary with reference to character is the region seat that can obtain each character
Mark, and then separating character.
Further, the charactron Recognition of Reading read in the achievable decimal point of the present invention based on machine vision
In method, decimal point read step is as follows in the step (4):Adjacent two characters maximum word is calculated by adjacent two character pitch
Symbol interval with reference to character height, generates a width as 2/3 times of maximum character pitch, the rectangle of a height of character height, with this rectangle
By the entire charactron character zone of sequential scan from left to right, when occurring upper 3/4 region in rectangle without pixel, 1/4 region is descended
Middle pixel number is more than a certain threshold value, then it is assumed that the position of decimal point is navigated to, otherwise without decimal point.
Further, above-mentioned threshold value is calculated according to the hem width of charactron character.
Threshold value calculation method is:
Wherein, Area is the minimum pixel area, that is, threshold value in decimal point region, and L is the pixel wide of character.
Further, the charactron Recognition of Reading read in the achievable decimal point of the present invention based on machine vision
In method, in the step (5), several single character pictures of segmentation are identified respectively with grader, with reference to character
Include the following steps with the reading of pixel coordinate position generation charactron of the decimal point in charactron character zone:
A. given size, wide a height of 32 pixel *, 64 pixels are zoomed to each character picture of segmentation;
B. the HOG features of the image after each scaling of extraction, are identified respectively with SVM classifier;
C. the pixel coordinate position in conjunction with character and decimal point in charactron character zone generates the reading of charactron
Number.
Advantageous effect of the present invention:
The method of the present invention can effectively identify the seven segment digital tubes reading of mixed decimal point, and accuracy rate is high and robustness is good.
The method of the present invention can substitute the artificial automation for realizing charactron instrument and read, and can greatly improve the efficiency of work, subtract
Few artificial cost.
Description of the drawings
Fig. 1 is the flow chart of the present invention.
Specific embodiment
To become apparent from illustrating the purpose of the present invention, scheme and advantage, below in conjunction with the accompanying drawings to embodiments of the present invention into
One step is described in detail.
As shown in Figure 1, the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read, it utilizes receipts
The charactron character repertoire of collection with the HOG feature learnings generation SVM classifier of character picture for identifying character, and utilizes character
Size and spacing, generate rectangle template, Scan orientation decimal point, specific implementation step is as follows:
(1) all single character pictures of charactron for only including a character are collected, build sample database, are read in sample database
All single character sample images extract the HOG features of each character sample image, learn and generate SVM classifier;
The method for extracting each character sample image HOG features is:
A. to charactron character sample image preprocessing, that is, histogram equalization and median filter process are carried out;
B. again to pretreated image carry out binary conversion treatment, obtain charactron character sample binary image, two
Value method is:To image gray processing, gray level image is obtained, is then handled using big law automatic threshold.
The formula of gray processing is:
Gray (i, j)=0.299*R (i, j)+0.587*G (i, j)+0.144*B (i, j), i, j be respectively pixel row and
Row coordinate;Wherein Gray is gray value, and R, G, B are respectively three kinds of color components of red, green, blue;
C. by removing background and extraction connected region, extract binary image character zone, generation only comprising character and
The size character zone image identical with character;
D. to the character zone image scaling after extraction to given size, by image scaling to wide a height of in the present embodiment
The image of 32*64.
E. the HOG features of the character zone image after scaling are extracted, the window size that HOG features are extracted in the present embodiment is
32*64, block size 16*16, cell size 8*8, direction number 9.
(2) new charactron character picture is acquired, extracts the character zone in charactron character picture;
Character zone extraction method be:Smallest passage value is subtracted with the largest passages value of image, obtains gray-scale map, the ash
It is brighter to spend charactron region in figure, threshold process is carried out to gray-scale map, filtering, then region segmentation, obtains only comprising number
The binary image in area under control domain.
(3) Character segmentation is carried out to the character in the charactron character zone image of extraction, generates several single characters
Image;
The method of Character segmentation is:
First, the binary image of the charactron character of extraction is projected to vertical direction, projection formula is:Using the line number of image as horizontal axis, using corresponding often capable gray value as 255 pixel number as
The longitudinal axis generates projection histogram, and scanning projection histogram extracts the minimax boundary coordinate of histogram, and then obtains charactron
The up-and-down boundary of character;
Wherein, SjThe summation for the pixel that image pixel value for the i-th row is 255, i, j are respectively that the row and column of pixel is sat
Mark, col widths of the cols for image, P (i, j) value 0 or 1, when the gray value that coordinate is the pixel at (i, j) is 255, P
(i, j) takes 1, is otherwise 0.
Secondly, refine the charactron character binaryzation image of extraction, will refinement figure to level into each in the range of ± 60 °
Direction projection selects direction of the projecting direction of pixel number maximum in projection histogram as character, is then rectified with Shear Transform
Positive character.
Finally, the character picture after correction is projected to horizontal direction, using the row number of image as horizontal axis, with corresponding every
The pixel number of row generates projection histogram as the longitudinal axis, and sequential scan projection histogram extracts the left side of each character successively
Right margin can obtain the area coordinate of each character, and then separating character with reference to the up-and-down boundary of character, obtain several only
Include the bianry image of single character.
(4) decimal point of charactron character zone is read;
Decimal point read method is:Adjacent two characters maximum character pitch is calculated by adjacent two character pitch, with reference to word
Symbol height generates a width as 2/3 times of maximum character pitch, the rectangle of a height of character height, with this rectangle by suitable from left to right
Sequence scans entire charactron character zone, and when occurring upper 3/4 region in rectangle without pixel, pixel number is big in 1/4 region down
In a certain threshold value, then it is assumed that the position of decimal point is navigated to, otherwise without decimal point.
Above-mentioned threshold value is calculated according to the hem width of charactron character.
Threshold value calculation method is:
Wherein, Area is the minimum pixel area (i.e. threshold value) in decimal point region, and L is the pixel wide of character.(5) it uses and divides
Class device is respectively identified several single character pictures of segmentation, with reference to character and decimal point in charactron character zone
Pixel coordinate position generation charactron reading.Step is as follows:
A. zoom to given size to each character picture of segmentation, the pixel of 64 pixels × 32;
B. the HOG features of the image after each scaling of extraction, are identified respectively with SVM classifier;
C. the pixel coordinate position in conjunction with character and decimal point in charactron character zone generates the reading of charactron
Number.
Although above-mentioned be described the specific implementation of the present invention with reference to attached drawing, not to the scope of the present invention
Limitation, for those skilled in the art under the premise of creative achievement is not needed to, the various modifications made should all be considered as belonging to this
The protection domain of invention.
Claims (10)
1. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read, which is characterized in that including following
Step:
(1) all single character pictures of charactron for only including a character are collected, build sample database, reads in sample database and owns
Single character sample image, extract the HOG features of each character sample image, learn and generate SVM classifier;
(2) new charactron character picture is acquired, extracts the character zone in charactron character picture;
(3) Character segmentation is carried out to the character in the charactron character zone image of extraction, generates several single character pictures;
(4) decimal point of charactron character zone is read;
(5) several single character pictures of segmentation are identified respectively with grader, with reference to character and decimal point in number
The reading of pixel coordinate position generation charactron in pipe character zone.
2. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as described in claim 1,
It is characterized in that, the charactron is seven segment digital tubes.
3. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as described in claim 1,
The step of being characterized in that, each character sample image HOG features are extracted in the step (1) is as follows:
A. to charactron character sample image preprocessing, that is, histogram equalization and median filter process are carried out;
B. binary conversion treatment is carried out to pretreated image again, obtains the binary image of charactron character sample;
C. binary image character zone is extracted, generation is only comprising character and the size character zone image identical with character;
D. to the character zone image scaling after extraction to given size;
E. the HOG features of the character zone image after extraction scaling.
4. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as described in claim 1,
It is characterized in that, the image after character zone in the step (2) in extraction charactron character picture is only includes charactron word
The binary image of symbol, the gray value of character is 255 in binary image, and the gray value of background is 0.
5. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as described in claim 1,
It is characterized in that, it is as follows the step of Character segmentation in the step (3):
A. the vertical range of charactron character zone image is adjusted, obtains character zone up-and-down boundary coordinate;
B. for tilted character into line character Slant Rectify;
C. single Character segmentation is carried out for all characters in region.
6. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as claimed in claim 5,
It is characterized in that, in the step a, vertical range is adjusted specially:By the binary image of the charactron character of extraction to
Vertical direction projects, and projection formula is:Using the line number of image as horizontal axis, with corresponding often capable ash
The pixel number that angle value is 255 generates projection histogram as the longitudinal axis, and scanning projection histogram extracts the maximum of histogram most
Small boundary coordinate, and then obtain the up-and-down boundary of charactron character;
Wherein, SjThe summation of pixel that image pixel value for the i-th row is 255, i, j are respectively the row and column coordinate of pixel,
Col widths of the cols for image, P (i, j) value 0 or 1, when the gray value that coordinate is the pixel at (i, j) is 255, P (i, j)
1 is taken, is otherwise 0.
7. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as claimed in claim 5,
It is characterized in that, in the step b, character Slant Rectify is specially:The charactron character binaryzation image of extraction is refined, will be refined
Figure is projected to level into all directions in the range of ± 60 °, and the direction of the projection of pixel number maximum in projection histogram is selected to make
For the direction of character, then character is corrected with Shear Transform.
8. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as claimed in claim 5,
It is characterized in that, in the step c, single Character segmentation step is:Character picture after correction is projected to horizontal direction, to scheme
The row number of picture generates projection histogram, sequential scan projection using the pixel number of corresponding each column as the longitudinal axis as horizontal axis
Histogram extracts the right boundary of each character successively, and the up-and-down boundary with reference to character is the region seat that can obtain each character
Mark, and then separating character.
9. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as described in claim 1,
It is characterized in that, decimal point read step is as follows in the step (4):It is maximum that adjacent two character is calculated by adjacent two character pitch
Character pitch with reference to character height, generates a width as 2/3 times of maximum character pitch, the rectangle of a height of character height, with this square
Shape, when occurring upper 3/4 region in rectangle without pixel, descends 1/4th area by the entire charactron character zone of sequential scan from left to right
Pixel number is more than a certain threshold value in domain, then it is assumed that the position of decimal point is navigated to, otherwise without decimal point.
10. the charactron Recognition of Reading method that the achievable decimal point based on machine vision is read as described in claim 1,
It is characterized in that, in the step (5), several single character pictures of segmentation is identified respectively with grader, with reference to word
The reading of the pixel coordinate position generation charactron of symbol and decimal point in charactron character zone includes the following steps:
A. given size is zoomed to each character picture of segmentation;
B. the HOG features of the image after each scaling of extraction, are identified respectively with SVM classifier;
C. the pixel coordinate position in conjunction with character and decimal point in charactron character zone generates the reading of charactron.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711166125.1A CN108133216B (en) | 2017-11-21 | 2017-11-21 | Nixie tube reading identification method capable of realizing decimal point reading based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711166125.1A CN108133216B (en) | 2017-11-21 | 2017-11-21 | Nixie tube reading identification method capable of realizing decimal point reading based on machine vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108133216A true CN108133216A (en) | 2018-06-08 |
CN108133216B CN108133216B (en) | 2021-10-12 |
Family
ID=62388761
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711166125.1A Active CN108133216B (en) | 2017-11-21 | 2017-11-21 | Nixie tube reading identification method capable of realizing decimal point reading based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108133216B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109034160A (en) * | 2018-07-06 | 2018-12-18 | 江苏迪伦智能科技有限公司 | A kind of mixed decimal point digital instrument automatic identifying method based on convolutional neural networks |
CN110059693A (en) * | 2019-04-18 | 2019-07-26 | 华北电力大学(保定) | A kind of digital instrument Recognition of Reading system based on wireless sensor network |
CN110084241A (en) * | 2019-05-05 | 2019-08-02 | 山东大学 | A kind of ammeter automatic reading method based on image recognition |
CN110443220A (en) * | 2019-08-13 | 2019-11-12 | 树根互联技术有限公司 | Digital table image-recognizing method, device, electronic equipment and storage medium |
CN110688996A (en) * | 2019-09-23 | 2020-01-14 | 天津大学 | Embedded automatic ruler reading device and method based on visual sensing |
CN111222507A (en) * | 2020-01-10 | 2020-06-02 | 随锐科技集团股份有限公司 | Automatic identification method of digital meter reading and computer readable storage medium |
CN112348026A (en) * | 2020-11-08 | 2021-02-09 | 北京工业大学 | Magnetic hard disk sequence code identification method based on machine vision |
CN113159027A (en) * | 2021-04-13 | 2021-07-23 | 杭州电子科技大学 | Seven-segment type digital display instrument identification method based on minimum external rectangle variant |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002158877A (en) * | 2000-11-21 | 2002-05-31 | Ricoh Co Ltd | Image data processing method and image data processor |
CN103984930A (en) * | 2014-05-21 | 2014-08-13 | 南京航空航天大学 | Digital meter recognition system and method based on vision |
CN105335745A (en) * | 2015-11-27 | 2016-02-17 | 小米科技有限责任公司 | Recognition method, device and equipment for numbers in images |
CN106529537A (en) * | 2016-11-22 | 2017-03-22 | 亿嘉和科技股份有限公司 | Digital meter reading image recognition method |
CN106682665A (en) * | 2016-12-27 | 2017-05-17 | 陕西科技大学 | Digital recognition method for seven-segment digital indicator |
-
2017
- 2017-11-21 CN CN201711166125.1A patent/CN108133216B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002158877A (en) * | 2000-11-21 | 2002-05-31 | Ricoh Co Ltd | Image data processing method and image data processor |
CN103984930A (en) * | 2014-05-21 | 2014-08-13 | 南京航空航天大学 | Digital meter recognition system and method based on vision |
CN105335745A (en) * | 2015-11-27 | 2016-02-17 | 小米科技有限责任公司 | Recognition method, device and equipment for numbers in images |
CN106529537A (en) * | 2016-11-22 | 2017-03-22 | 亿嘉和科技股份有限公司 | Digital meter reading image recognition method |
CN106682665A (en) * | 2016-12-27 | 2017-05-17 | 陕西科技大学 | Digital recognition method for seven-segment digital indicator |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109034160A (en) * | 2018-07-06 | 2018-12-18 | 江苏迪伦智能科技有限公司 | A kind of mixed decimal point digital instrument automatic identifying method based on convolutional neural networks |
CN109034160B (en) * | 2018-07-06 | 2019-07-12 | 江苏迪伦智能科技有限公司 | A kind of mixed decimal point digital instrument automatic identifying method based on convolutional neural networks |
CN110059693A (en) * | 2019-04-18 | 2019-07-26 | 华北电力大学(保定) | A kind of digital instrument Recognition of Reading system based on wireless sensor network |
CN110084241A (en) * | 2019-05-05 | 2019-08-02 | 山东大学 | A kind of ammeter automatic reading method based on image recognition |
CN110084241B (en) * | 2019-05-05 | 2023-05-30 | 山东大学 | Automatic ammeter reading method based on image recognition |
CN110443220A (en) * | 2019-08-13 | 2019-11-12 | 树根互联技术有限公司 | Digital table image-recognizing method, device, electronic equipment and storage medium |
CN110688996A (en) * | 2019-09-23 | 2020-01-14 | 天津大学 | Embedded automatic ruler reading device and method based on visual sensing |
CN111222507A (en) * | 2020-01-10 | 2020-06-02 | 随锐科技集团股份有限公司 | Automatic identification method of digital meter reading and computer readable storage medium |
CN111222507B (en) * | 2020-01-10 | 2024-03-19 | 随锐科技集团股份有限公司 | Automatic identification method for digital meter reading and computer readable storage medium |
CN112348026A (en) * | 2020-11-08 | 2021-02-09 | 北京工业大学 | Magnetic hard disk sequence code identification method based on machine vision |
CN113159027A (en) * | 2021-04-13 | 2021-07-23 | 杭州电子科技大学 | Seven-segment type digital display instrument identification method based on minimum external rectangle variant |
CN113159027B (en) * | 2021-04-13 | 2024-02-09 | 杭州电子科技大学 | Seven-segment digital display instrument identification method based on minimum external rectangular variant |
Also Published As
Publication number | Publication date |
---|---|
CN108133216B (en) | 2021-10-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108133216A (en) | The charactron Recognition of Reading method that achievable decimal point based on machine vision is read | |
CN105046196B (en) | Front truck information of vehicles structuring output method based on concatenated convolutional neutral net | |
CN104794421B (en) | A kind of positioning of QR codes and recognition methods | |
CN105046252B (en) | A kind of RMB prefix code recognition methods | |
CN106097254B (en) | A kind of scanning document image method for correcting error | |
CN104751142B (en) | A kind of natural scene Method for text detection based on stroke feature | |
CN106023151B (en) | Tongue object detection method under a kind of open environment | |
CN106169080B (en) | A kind of combustion gas index automatic identifying method based on image | |
CN110084241B (en) | Automatic ammeter reading method based on image recognition | |
CN104598907B (en) | Lteral data extracting method in a kind of image based on stroke width figure | |
CN102750540A (en) | Morphological filtering enhancement-based maximally stable extremal region (MSER) video text detection method | |
CN108537782A (en) | A method of building images match based on contours extract with merge | |
CN106407983A (en) | Image body identification, correction and registration method | |
CN110276279B (en) | Method for detecting arbitrary-shape scene text based on image segmentation | |
CN107767379A (en) | Pcb board marks print quality inspection method | |
CN110598566A (en) | Image processing method, device, terminal and computer readable storage medium | |
CN112434699A (en) | Automatic extraction and intelligent scoring system for handwritten Chinese characters or components and strokes | |
CN111539330A (en) | Transformer substation digital display instrument identification method based on double-SVM multi-classifier | |
CN107122775A (en) | A kind of Android mobile phone identity card character identifying method of feature based matching | |
CN108171157A (en) | The human eye detection algorithm being combined based on multiple dimensioned localized mass LBP histogram features with Co-HOG features | |
CN110929562A (en) | Answer sheet identification method based on improved Hough transformation | |
CN109741273A (en) | A kind of mobile phone photograph low-quality images automatically process and methods of marking | |
CN111401364B (en) | License plate positioning algorithm based on combination of color features and template matching | |
CN115588208A (en) | Full-line table structure identification method based on digital image processing technology | |
CN105654140B (en) | The positioning of rail tank car license number and recognition methods towards complex industrial environment |
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 |