CN114067305A - Full-automatic water meter positioning method - Google Patents
Full-automatic water meter positioning method Download PDFInfo
- Publication number
- CN114067305A CN114067305A CN202111375472.1A CN202111375472A CN114067305A CN 114067305 A CN114067305 A CN 114067305A CN 202111375472 A CN202111375472 A CN 202111375472A CN 114067305 A CN114067305 A CN 114067305A
- Authority
- CN
- China
- Prior art keywords
- water meter
- character wheel
- character
- area
- image
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention relates to water meter image recognition, in particular to a full-automatic water meter positioning method, which comprises the steps of detecting the type of a water meter in an acquired image by using a water meter type recognition model, calling a corresponding character wheel position detection model based on the type of the water meter to detect the position of a character wheel in the acquired image, acquiring the inclination angle and the position translation amount of the water meter based on the position of the character wheel, correcting the acquired image, and positioning in the corrected image based on the position of a prestored character corresponding to the type of the water meter; the technical scheme provided by the invention can effectively overcome the defect that the area part of the water meter print wheel in the collected image can not be accurately positioned in the prior art.
Description
Technical Field
The invention relates to water meter image recognition, in particular to a full-automatic water meter positioning method.
Background
The data reading mode of the existing intelligent water meter mainly comprises a photoelectric direct reading mode, a pulse mode, an electromagnetic mode, an ultrasonic mode, a camera shooting mode and the like, wherein the existing camera shooting remote water meter mainly shoots a character wheel part of a mechanical water meter and then remotely transmits the character wheel part to a data acquisition center for identification. The camera shooting remote water meter is a novel water meter which appears in recent years, and can shoot and upload water meter data because the water meter can solve the problem of data remote transmission, and can determine whether the water meter reading is correct or not through a manual verification mode under extreme conditions.
When the camera assembly of the camera remote water meter shoots, the shot images really need to be calculated through an algorithm and concentrated on the character wheel area part of the water meter. The existing camera shooting remote water meter system can analyze and identify a large number of useless images by algorithm, so that the identification efficiency is low, and the operation resources of an algorithm chip are greatly wasted. One of the most important reasons for this is that the area of the print wheel of the water meter in the captured image cannot be accurately located.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects in the prior art, the invention provides a full-automatic water meter positioning method which can effectively overcome the defect that the part of the acquired image, which relates to the character wheel area of the water meter, cannot be accurately positioned in the prior art.
(II) technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme:
a full-automatic water meter positioning method comprises the steps of detecting the type of a water meter in an acquired image by using a water meter type identification model, calling a corresponding character wheel position detection model based on the type of the water meter to detect the position of a character wheel in the acquired image, acquiring the inclination angle and the position translation amount of the water meter based on the position of the character wheel, correcting the acquired image, and positioning in a corrected image based on the position of a prestored character corresponding to the type of the water meter.
Preferably, the method for training the water meter type recognition model includes:
s11, collecting a large number of sample images for each type of water meter, amplifying the sample images, and performing category marking on the amplified sample images;
and S12, inputting all sample images and corresponding class labels into a DenseNet neural network model for model training, and deploying the trained model files on a cloud server.
Preferably, the print wheel position detection model is trained separately for each type of water meter, including:
s21, collecting sample pictures containing various angles for each type of water meter, manually marking the outer frame positions corresponding to the character wheels on the sample pictures, and amplifying the sample pictures;
and S22, inputting the sample picture and the corresponding marking information into a yolov5 deep network together for model training, and deploying the trained network model on a cloud server.
Preferably, the acquiring the inclination angle and the position translation amount of the water meter based on the character wheel position includes:
for the water meter with the character wheel region bottom color obviously different from the dial bottom color and the character wheel region integrally communicated, the inclination angle and the position translation amount of the water meter are obtained by adopting the following methods:
extracting a foreground image containing a character wheel area from the collected image, taking a connected domain with the largest area in the foreground image as the character wheel area, and obtaining a horizontal included angle and a central point coordinate of the character wheel area through contour analysis and straight-line segment fitting and extraction in a contour sequence so as to obtain an inclination angle and a position translation amount of the water meter.
Preferably, the acquiring the inclination angle and the position translation amount of the water meter based on the character wheel position includes:
for a water meter with a character wheel area in an independent rectangle and a character wheel area bottom color obviously different from a dial bottom color, the inclination angle and the position translation amount of the water meter are obtained by adopting the following method:
and extracting a foreground image containing a character wheel area from the acquired image, filtering the character area in the foreground image by comparing the shape and the size to obtain an independent rectangle of the character wheel area, further obtaining four corners of the whole character wheel area, and obtaining the inclination angle and the position translation amount of the water meter.
Preferably, the acquiring the inclination angle and the position translation amount of the water meter based on the character wheel position includes:
for a water meter with a rubber ring at the periphery of a character wheel area and obviously different color from the bottom color of a dial, acquiring the inclination angle and the position translation amount of the water meter by adopting the following method:
the method comprises the steps of extracting a foreground image containing a character wheel area from an acquired image, filtering out the connected domains of non-character wheel areas in the foreground image by analyzing the shape and the area of each connected domain, and performing linear fitting of an external contour on the character wheel area to further obtain the inclination angle and the position translation amount of the water meter.
Preferably, the acquiring the inclination angle and the position translation amount of the water meter based on the character wheel position includes:
for a water meter with a character wheel area being an independent rectangle, no rubber ring at the periphery and no obvious difference between the ground color of the character wheel area and the ground color of the dial plate, the inclination angle and the position translation amount of the water meter are obtained by adopting the following methods:
and extracting a foreground image containing a character wheel area from the acquired image, filtering a connected domain which does not meet the size requirement according to the character size of the water meter, determining the corresponding character wheel area by utilizing the collinear characteristic of the character centers of the first three water meters, wherein the angle of the connecting line of the character centers of the first three water meters is the inclination angle of the water meter, and the central point of the second character is used as a calibration point to calculate the position translation amount of the water meter.
Preferably, the acquiring the inclination angle and the position translation amount of the water meter based on the character wheel position includes:
for the water meter with the character wheel area being an independent rectangle, no rubber ring at the periphery, no obvious difference between the ground color of the character wheel area and the ground color of the dial plate, and the side part of the character wheel area having a black area, the inclination angle and the position translation amount of the water meter are obtained by adopting the following method:
and extracting a foreground image containing a character wheel area from the acquired image, filtering black areas on the side part of the character wheel area, water meter characters adhered to the black areas and other connected areas which do not meet the size requirement on the basis of the size information, and acquiring the inclination angle and the position translation amount of the water meter on the basis of the positions and collinear characteristics of the central points of the three middle water meter characters.
Preferably, the extracting a foreground image containing a character wheel region from the captured image includes:
s31, after detecting the water meter type in the collected image by using the water meter type identification model, calling prestored information to obtain the dial color basic value of the water meter of the type;
s32, setting a pixel threshold value, and comparing the RGB value of each pixel point in the collected image with the dial color basic value;
and S33, extracting the pixel points of which any value of the RGB values in the collected image is larger than the pixel threshold value as foreground images.
(III) advantageous effects
Compared with the prior art, the full-automatic water meter positioning method provided by the invention has the advantages that the water meter type identification model can be used for identifying the water meter type in the acquired image, the character wheel position in the acquired image is detected by calling the corresponding character wheel position detection model based on the water meter type, the inclination angle and the position translation amount can be accurately obtained by combining the character wheel position and the water meter type, the acquired image can be accurately corrected, and the accurate positioning in the corrected image is realized by means of the prestored character position corresponding to the water meter type.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a collected image and a corresponding foreground image of a first type of water meter in accordance with the present invention;
FIG. 3 is a collected image and a corresponding foreground image of a second type of water meter in accordance with the present invention;
FIG. 4 is a collected image and a corresponding foreground image of a third type of water meter in accordance with the present invention;
FIG. 5 is a collected image and a corresponding foreground image of a fourth type of water meter in accordance with the present invention;
FIG. 6 is a collected image and a corresponding foreground image of a fifth type of water meter according to the present invention;
fig. 7 is a schematic structural diagram of a DenseNet neural network model.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
A full-automatic water meter positioning method is disclosed, as shown in figure 1, a water meter type identification model is used for detecting the water meter type in an acquired image, a corresponding character wheel position detection model is called based on the water meter type to detect the character wheel position in the acquired image, the inclination angle and the position translation amount of the water meter are obtained based on the character wheel position, image correction is carried out on the acquired image, and positioning is carried out in the corrected image based on the prestored character position corresponding to the water meter type.
In order to subsequently query the corresponding pre-stored information according to the water meter type, the water meter type in the collected image needs to be obtained in advance. When the camera shooting remote water meter is installed for the first time, an image can be shot and transmitted to the cloud server, the type of the water meter in the collected image is detected and identified on the server, and then the follow-up steps are called to continue analyzing.
The training method of the water meter type recognition model comprises the following steps:
s11, collecting a large number of sample images for each type of water meter (50 sample images are collected for each type of water meter), amplifying the sample images (amplifying the sample images of each type of water meter to 500 by image rotation, translation, affine transformation and the like), and labeling the types of the amplified sample images;
and S12, inputting all sample images and corresponding class labels into a DenseNet neural network model for model training, and deploying the trained model files on a cloud server.
The shooting equipment used for collecting the sample images keeps consistent with the camera shooting remote water meter in the aspects of camera type, resolution ratio, image compression format, light intensity and the like.
Training a character wheel position detection model separately for each type of water meter, comprising:
s21, collecting sample pictures containing various angles for each type of water meter (50 sample pictures are collected for each type of water meter), manually marking the outer frame positions corresponding to the character wheels on the sample pictures, and amplifying the sample pictures (the sample pictures of each type of water meter are amplified to 500 pieces by image rotation, translation, affine transformation and other modes);
and S22, inputting the sample picture and the corresponding marking information into a yolov5 deep network together for model training, and deploying the trained network model on a cloud server.
Because the amplification mode of the sample picture is known, the amplified sample picture does not need to be marked with the outer frame position corresponding to the character wheel, and only needs to be calculated according to the outer frame position in the sample picture before amplification and the amplification mode.
After the approximate position of the character wheel area is obtained through the steps, the inclination angle and the position translation amount (including the translation amplitude in the horizontal direction and the vertical direction) of the water meter also need to be further obtained. After analyzing dial plate images of all water meters on the market, the water meters can be divided into 5 categories, and the water meters in each category can obtain the inclination angle and the position translation amount by adopting the same idea. By the method for classifying all water meters and applying the pertinence strategies respectively, automatic positioning of all types of water meters can be realized.
Firstly, as shown in fig. 2, for a water meter with a character wheel region and a dial bottom color which are obviously different and the character wheel region is integrally communicated, the inclination angle and the position translation amount of the water meter are obtained by adopting the following methods:
extracting a foreground image containing a character wheel area from the collected image, taking a connected domain with the largest area in the foreground image as the character wheel area, and obtaining a horizontal included angle and a central point coordinate of the character wheel area through contour analysis and straight-line segment fitting and extraction in a contour sequence so as to obtain an inclination angle and a position translation amount of the water meter.
Secondly, as shown in fig. 3, for a water meter with a character wheel area in an independent rectangle and a character wheel area bottom color obviously different from a dial bottom color, the inclination angle and the position translation amount of the water meter are obtained by adopting the following methods:
the method comprises the steps of extracting a foreground image containing a character wheel area from an acquired image, filtering character areas in the foreground image by comparing the shape and the size (the size of a small rectangle is known by pre-stored information), obtaining independent rectangles of the character wheel area (the independent rectangles containing 4 small rectangles are known by the pre-stored information and are influenced by the effect of an actual image, the rectangles are not completely extracted, but rectangle analysis is not influenced), further obtaining four corners of the whole character wheel area, and obtaining the inclination angle and the position translation amount of the water meter.
Thirdly, as shown in fig. 4, for a water meter with a rubber ring at the periphery of the character wheel area and obvious difference between the color of the rubber ring and the bottom color of the dial, the inclination angle and the position translation amount of the water meter are obtained by adopting the following methods:
extracting a foreground image containing a character wheel area from the collected image (a dial color basic value and a pixel threshold value of a rubber ring are extracted from prestored information corresponding to the type of water meter), analyzing the shape and the area of each connected domain (both recorded in the prestored information), filtering the connected domain of the non-character wheel area in the foreground image, and performing linear fitting of an external contour on the character wheel area to further obtain the inclination angle and the position translation amount of the water meter.
And fourthly, as shown in fig. 5, for the water meter of which the character wheel area is an independent rectangle, the periphery of which is not provided with a rubber ring, and the ground color of the character wheel area is not obviously different from the ground color of the dial, acquiring the inclination angle and the position translation amount of the water meter by adopting the following methods:
extracting a foreground image containing a character wheel area from the acquired image, filtering connected domains (such as a left H character, a right m character and various connected domains above and below) which do not meet the size requirement according to the character size of the water meter, determining the corresponding character wheel area by utilizing the collinear characteristic of the centers of the first three water meter characters, wherein the angle of the connecting line of the centers of the first three water meter characters is the inclination angle of the water meter, and the central point of the second character is used as a calibration point to calculate the position translation amount of the water meter.
Because the water meter rarely has the phenomenon that two characters are half characters at the same time (even if the situation occurs, manual intervention can be carried out, and the situation is avoided), the first three characters are all whole characters basically, and therefore the character wheel area (the size difference of the three characters can also be used as judgment information) to which the three characters belong can be determined by utilizing the collinear characteristic of the character centers of the first three water meters.
As shown in fig. 6, for a water meter in which the character wheel area is an independent rectangle, no rubber ring is arranged on the periphery, the ground color of the character wheel area is not obviously different from the ground color of the dial, and the side part of the character wheel area has a black area (the two black areas are caused by the character wheel not being wide enough to fill the space), the inclination angle and the position translation amount of the water meter are obtained by adopting the following methods:
the method comprises the steps of extracting a foreground image containing a character wheel area from a collected image, filtering black areas on the side portion of the character wheel area, water meter characters adhered to the black areas and other connected areas which do not meet size requirements (such as a spacing line between two characters) based on size information, and obtaining the inclination angle and the position translation amount of the water meter based on the positions and collinear characteristics of center points of three middle water meter characters.
In the technical scheme of this application, in the in-process of obtaining its angle of inclination, position translation volume to above-mentioned five types of water meters, draw the prospect image that contains character wheel region from gathering the image, include:
s31, after detecting the water meter type in the collected image by using the water meter type identification model, calling prestored information to obtain the dial color basic value of the water meter of the type;
s32, setting a pixel threshold value, and comparing the RGB value of each pixel point in the collected image with the dial color basic value;
and S33, extracting the pixel points of which any value of the RGB values in the collected image is larger than the pixel threshold value as foreground images.
Because the illumination condition of the collected image is consistent with that of the template image, the dial color basic value of the current collected image is consistent with that of the template image. The dial color base value refers to the color value of the water meter dial area.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (9)
1. A full-automatic water meter positioning method is characterized by comprising the following steps: the method comprises the steps of detecting the type of the water meter in an acquired image by using a water meter type identification model, calling a corresponding character wheel position detection model based on the type of the water meter to detect the position of a character wheel in the acquired image, acquiring the inclination angle and the position translation amount of the water meter based on the position of the character wheel, carrying out image correction on the acquired image, and positioning in the corrected image based on the pre-stored character position corresponding to the type of the water meter.
2. The full-automatic water meter positioning method according to claim 1, characterized in that: the training method of the water meter type recognition model comprises the following steps:
s11, collecting a large number of sample images for each type of water meter, amplifying the sample images, and performing category marking on the amplified sample images;
and S12, inputting all sample images and corresponding class labels into a DenseNet neural network model for model training, and deploying the trained model files on a cloud server.
3. The full-automatic water meter positioning method according to claim 1, characterized in that: training the print wheel position detection model separately for each type of water meter, comprising:
s21, collecting sample pictures containing various angles for each type of water meter, manually marking the outer frame positions corresponding to the character wheels on the sample pictures, and amplifying the sample pictures;
and S22, inputting the sample picture and the corresponding marking information into a yolov5 deep network together for model training, and deploying the trained network model on a cloud server.
4. The full-automatic water meter positioning method according to claim 1, characterized in that: based on angle of inclination, the position translation volume of character wheel position acquisition water gauge includes:
for the water meter with the character wheel region bottom color obviously different from the dial bottom color and the character wheel region integrally communicated, the inclination angle and the position translation amount of the water meter are obtained by adopting the following methods:
extracting a foreground image containing a character wheel area from the collected image, taking a connected domain with the largest area in the foreground image as the character wheel area, and obtaining a horizontal included angle and a central point coordinate of the character wheel area through contour analysis and straight-line segment fitting and extraction in a contour sequence so as to obtain an inclination angle and a position translation amount of the water meter.
5. The full-automatic water meter positioning method according to claim 1, characterized in that: based on angle of inclination, the position translation volume of character wheel position acquisition water gauge includes:
for a water meter with a character wheel area in an independent rectangle and a character wheel area bottom color obviously different from a dial bottom color, the inclination angle and the position translation amount of the water meter are obtained by adopting the following method:
and extracting a foreground image containing a character wheel area from the acquired image, filtering the character area in the foreground image by comparing the shape and the size to obtain an independent rectangle of the character wheel area, further obtaining four corners of the whole character wheel area, and obtaining the inclination angle and the position translation amount of the water meter.
6. The full-automatic water meter positioning method according to claim 1, characterized in that: based on angle of inclination, the position translation volume of character wheel position acquisition water gauge includes:
for a water meter with a rubber ring at the periphery of a character wheel area and obviously different color from the bottom color of a dial, acquiring the inclination angle and the position translation amount of the water meter by adopting the following method:
the method comprises the steps of extracting a foreground image containing a character wheel area from an acquired image, filtering out the connected domains of non-character wheel areas in the foreground image by analyzing the shape and the area of each connected domain, and performing linear fitting of an external contour on the character wheel area to further obtain the inclination angle and the position translation amount of the water meter.
7. The full-automatic water meter positioning method according to claim 1, characterized in that: based on angle of inclination, the position translation volume of character wheel position acquisition water gauge includes:
for a water meter with a character wheel area being an independent rectangle, no rubber ring at the periphery and no obvious difference between the ground color of the character wheel area and the ground color of the dial plate, the inclination angle and the position translation amount of the water meter are obtained by adopting the following methods:
and extracting a foreground image containing a character wheel area from the acquired image, filtering a connected domain which does not meet the size requirement according to the character size of the water meter, determining the corresponding character wheel area by utilizing the collinear characteristic of the character centers of the first three water meters, wherein the angle of the connecting line of the character centers of the first three water meters is the inclination angle of the water meter, and the central point of the second character is used as a calibration point to calculate the position translation amount of the water meter.
8. The full-automatic water meter positioning method according to claim 1, characterized in that: based on angle of inclination, the position translation volume of character wheel position acquisition water gauge includes:
for the water meter with the character wheel area being an independent rectangle, no rubber ring at the periphery, no obvious difference between the ground color of the character wheel area and the ground color of the dial plate, and the side part of the character wheel area having a black area, the inclination angle and the position translation amount of the water meter are obtained by adopting the following method:
and extracting a foreground image containing a character wheel area from the acquired image, filtering black areas on the side part of the character wheel area, water meter characters adhered to the black areas and other connected areas which do not meet the size requirement on the basis of the size information, and acquiring the inclination angle and the position translation amount of the water meter on the basis of the positions and collinear characteristics of the central points of the three middle water meter characters.
9. The full-automatic water meter positioning method according to any one of claims 4 to 8, characterized in that: the method for extracting the foreground image containing the character wheel area from the acquired image comprises the following steps:
s31, after detecting the water meter type in the collected image by using the water meter type identification model, calling prestored information to obtain the dial color basic value of the water meter of the type;
s32, setting a pixel threshold value, and comparing the RGB value of each pixel point in the collected image with the dial color basic value;
and S33, extracting the pixel points of which any value of the RGB values in the collected image is larger than the pixel threshold value as foreground images.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111375472.1A CN114067305A (en) | 2021-11-19 | 2021-11-19 | Full-automatic water meter positioning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111375472.1A CN114067305A (en) | 2021-11-19 | 2021-11-19 | Full-automatic water meter positioning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114067305A true CN114067305A (en) | 2022-02-18 |
Family
ID=80278506
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111375472.1A Pending CN114067305A (en) | 2021-11-19 | 2021-11-19 | Full-automatic water meter positioning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114067305A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115082922A (en) * | 2022-08-24 | 2022-09-20 | 济南瑞泉电子有限公司 | Water meter digital picture processing method and system based on deep learning |
-
2021
- 2021-11-19 CN CN202111375472.1A patent/CN114067305A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115082922A (en) * | 2022-08-24 | 2022-09-20 | 济南瑞泉电子有限公司 | Water meter digital picture processing method and system based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108759973B (en) | Water level measuring method | |
| CN108288264B (en) | Wide-angle camera module contamination testing method | |
| CN105424709A (en) | Fruit surface defect detection method based on image marking | |
| CN112149543B (en) | Building dust recognition system and method based on computer vision | |
| CN106370671A (en) | PCB (printed circuit board) component detection system and method based on machine vision | |
| WO2023168972A1 (en) | Linear array camera-based copper surface defect detection method and apparatus | |
| CN106682665B (en) | Seven-segment type digital display instrument number identification method based on computer vision | |
| CN103017869A (en) | Water level measuring system and method based on digital image processing | |
| CN110363798B (en) | Method for generating remote sensing image interpretation sample set | |
| CN106599890B (en) | digital instrument recognition algorithm for substation inspection robot | |
| CN109461156B (en) | Threaded sealing plug assembly detection method based on vision | |
| WO2022105420A1 (en) | Lock gate misalignment detection method and detection system | |
| CN108871185B (en) | Method, device and equipment for detecting parts and computer readable storage medium | |
| CN110414341A (en) | A kind of Blasting In The Open Mining heap measurement statistical method based on unmanned air vehicle technique | |
| CN112801947A (en) | Visual detection method for dead pixel of LED display terminal | |
| CN109544535B (en) | Peeping camera detection method and system based on optical filtering characteristics of infrared cut-off filter | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN114067305A (en) | Full-automatic water meter positioning method | |
| WO2022032675A1 (en) | Product testing method and apparatus, computer readable storage medium, and electronic device | |
| CN111652089A (en) | Water level automatic identification method and system based on image processing | |
| CN110807424A (en) | Port ship comparison method based on aerial images | |
| CN111665199A (en) | Wire and cable color detection and identification method based on machine vision | |
| CN111666880A (en) | Intelligent identification system for fire extinguisher pointer instrument | |
| CN114627461A (en) | Method and system for high-precision identification of water gauge data based on artificial intelligence | |
| CN114581760A (en) | Equipment fault detection method and system for machine room inspection |
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 |