CN111476787A - Automatic reading method and device for adaptive distortion of pointer meter - Google Patents
Automatic reading method and device for adaptive distortion of pointer meter Download PDFInfo
- Publication number
- CN111476787A CN111476787A CN202010329384.7A CN202010329384A CN111476787A CN 111476787 A CN111476787 A CN 111476787A CN 202010329384 A CN202010329384 A CN 202010329384A CN 111476787 A CN111476787 A CN 111476787A
- Authority
- CN
- China
- Prior art keywords
- pointer
- reading
- image
- pointer table
- dial
- 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
- 238000000034 method Methods 0.000 title claims abstract description 57
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 47
- 238000001514 detection method Methods 0.000 claims abstract description 60
- 230000009466 transformation Effects 0.000 claims abstract description 44
- 238000013507 mapping Methods 0.000 claims abstract description 15
- 230000015654 memory Effects 0.000 claims description 20
- 239000011159 matrix material Substances 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 5
- 238000007689 inspection Methods 0.000 description 11
- 239000002609 medium Substances 0.000 description 10
- 238000012545 processing Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 239000012120 mounting media Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004309 orthophoria Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/04—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
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the application discloses an automatic reading method and device for adaptive distortion of a pointer meter. According to the technical scheme, the camera is used for shooting images, pointer table detection is carried out on the basis of the images shot by the camera, original images containing pointer table image characteristics are captured, perspective transformation is carried out on the original images on the basis of projection mapping of pixel coordinates, corresponding front-view images are obtained, and finally, the pointer table reading is calculated on the basis of the rotation angle of the pointer relative to the dial scale through positioning the dial scale and the pointer in the front-view images. By adopting the technical means, the angle distortion of the image shot by the camera can be eliminated, the influence of the shooting angle on the automatic reading of the pointer meter is reduced, the precision of the automatic reading of the pointer meter is improved, and the accuracy of the automatic reading is guaranteed.
Description
Technical Field
The embodiment of the application relates to the technical field of image processing, in particular to an automatic reading method and device for adaptive distortion of a pointer table.
Background
Pointer tables are widely used in many areas of life and industry. With the development and progress of image processing technology, the reading mode of the pointer table tends to be more intelligent. In some scenarios, automatic reading of the pointer table may be performed based on image recognition techniques.
One of the traditional methods for automatically reading a pointer table by using an image recognition technology is to adopt a convolutional neural network algorithm (CNN), and all scales are trained and recognized by extracting image features and finely classifying the image features. And the other method is to detect the circle and the pointer position through Hough transform and calculate the pointer change reading through conversion. The two reading identification modes are easily affected by image noise, and the reading precision is relatively low.
Disclosure of Invention
The embodiment of the application provides an automatic reading method and device for adaptive distortion of a pointer meter, and reading precision of the pointer meter can be improved.
In a first aspect, an embodiment of the present application provides an automatic reading method for adaptive distortion of a pointer table, including:
acquiring a camera shooting image, carrying out pointer table detection based on the camera shooting image, and intercepting an original image containing the pointer table image characteristics;
carrying out perspective transformation on the original image based on projection mapping of pixel coordinates to obtain a corresponding front-view image;
and positioning dial scales and a pointer in the front-view image, and calculating the reading of a pointer meter based on the rotation angle of the pointer relative to the dial scales.
Further, the acquiring an image shot by a camera, performing pointer table detection based on the image shot by the camera, and capturing an original image containing the image characteristics of the pointer table includes:
and inputting images shot by a camera into a pre-trained target detection model, and outputting corresponding original images through target detection.
Further, the process of constructing the target detection model includes:
collecting an image containing the image characteristics of the pointer table as a training sample;
and inputting the training sample into a ResNet-50 network for model training to obtain a corresponding target detection model.
Further, the perspective transformation of the original image based on the pixel coordinate projection mapping to obtain a corresponding front-view image includes:
positioning a first identification point of the oval dial plate in the original image, wherein the first identification point at least comprises an oval circle center and intersection points of a long axis and a short axis with the oval;
determining the first identification point to be mapped as a second identification point of the perfect circle dial plate;
calculating a perspective transformation matrix according to the first identification point and the second identification point;
and performing coordinate conversion on the pixel coordinates of the original image based on the perspective transformation matrix to obtain the front-view image.
Further, the dial scale and the pointer in the front view image are positioned, and the method comprises the following steps:
and respectively detecting dial scales and pointers in the front-view image based on a target detection algorithm.
And further, inputting the rotation angle into a preset reading calculation model, and outputting a corresponding reading of the pointer table, wherein the reading calculation model is constructed in advance according to the corresponding relation between the rotation angle and the reading of the pointer table.
Further, after the pointer meter reading is calculated based on the rotation angle of the pointer relative to the dial scale, the method further comprises the following steps:
based on a preset reading threshold value, when the reading of the pointer meter is detected to be greater than or equal to or less than the reading threshold value, outputting corresponding prompt information.
In a second aspect, the present application provides an automatic reading device for adaptive distortion of a pointer table, including:
the acquisition module is used for acquiring images shot by a camera, carrying out pointer table detection based on the images shot by the camera and intercepting original images containing the image characteristics of the pointer table;
the transformation module is used for carrying out perspective transformation on the original image based on the projection mapping of the pixel coordinates to obtain a corresponding front-view image;
and the positioning module is used for positioning the dial scales and the pointer in the front view image and calculating the reading of the pointer meter based on the rotation angle of the pointer relative to the dial scales.
In a third aspect, an embodiment of the present application provides an electronic device, including:
a memory and one or more processors;
the memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the automatic reading method of pointer table adaptive distortion as described in the first aspect.
In a fourth aspect, embodiments of the present application provide a storage medium containing computer-executable instructions for performing the automatic reading method of pointer table adaptive distortion as described in the first aspect when executed by a computer processor.
According to the embodiment of the application, the camera is used for shooting images, the pointer table detection is carried out based on the images shot by the camera, the original images containing the image characteristics of the pointer table are intercepted, the perspective transformation is carried out on the original images based on the projection mapping of pixel coordinates, the corresponding front view images are obtained, and finally the reading of the pointer table is calculated based on the rotation angle of the pointer relative to the dial scale by positioning the dial scale and the pointer in the front view images. By adopting the technical means, the angle distortion of the image shot by the camera can be eliminated, the influence of the shooting angle on the automatic reading of the pointer meter is reduced, the precision of the automatic reading of the pointer meter is improved, and the accuracy of the automatic reading is guaranteed.
Drawings
FIG. 1 is a flow chart of an automatic reading method for adaptive distortion of a pointer table according to an embodiment of the present application;
FIG. 2 is a flow chart of the construction of a target detection model in the first embodiment of the present application;
FIG. 3 is a flowchart of perspective transformation in the first embodiment of the present application;
FIG. 4 is a diagram of an original image of a pointer table in the first embodiment of the present application;
FIG. 5 is a schematic view of a front view of a pointer table in the first embodiment of the present application;
FIG. 6 is a flow chart of another automatic reading method for adaptive distortion of pointer table provided in the second embodiment of the present application;
fig. 7 is a schematic structural diagram of an automatic reading device for adaptive distortion of a pointer table according to a third embodiment of the present application;
fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, specific embodiments of the present application will be described in detail with reference to the accompanying drawings. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be further noted that, for the convenience of description, only some but not all of the relevant portions of the present application are shown in the drawings. Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the operations (or steps) as a sequential process, many of the operations can be performed in parallel, concurrently or simultaneously. In addition, the order of the operations may be re-arranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like.
The application provides an automatic reading algorithm of pointer table self-adaptation distortion, when aiming at carrying out the automatic reading of pointer table, carry out the elimination processing of angle distortion to the image to obtain the image of corresponding pointer table front visual angle. And further carrying out image recognition based on the corresponding front-view image to determine the reading of the pointer table. Therefore, high precision of automatic reading of the pointer meter is guaranteed, and influence of image noise on reading precision is reduced. For the traditional automatic reading mode using the convolutional neural network algorithm or the Hough transform algorithm, because the influence of image noise on the identification precision is not well considered, when the automatic reading of the pointer table is carried out, the automatic reading precision of the pointer table is relatively low due to the interference of the image noise. Therefore, the automatic reading method for the self-adaptive distortion of the pointer meter is provided to solve the technical problems of high interference and low precision of the automatic reading of the existing pointer meter.
The first embodiment is as follows:
fig. 1 is a flowchart of an automatic reading method for adaptive distortion of a pointer table according to an embodiment of the present application, where the automatic reading method for adaptive distortion of a pointer table provided in this embodiment may be performed by an automatic reading device for adaptive distortion of a pointer table, the automatic reading device for adaptive distortion of a pointer table may be implemented by software and/or hardware, and the automatic reading device for adaptive distortion of a pointer table may be formed by two or more physical entities or may be formed by one physical entity. Generally, the automatic reading device for adaptive distortion of the pointer table can be a computer, a server host and other processing devices.
The following description will be given taking an automatic reading device for adaptive distortion of a pointer table as an example of a main body of an automatic reading method for performing adaptive distortion of a pointer table. Referring to fig. 1, the automatic reading method for adaptive distortion of the pointer table specifically includes:
s110, acquiring images shot by a camera, carrying out pointer table detection based on the images shot by the camera, and intercepting original images containing the image characteristics of the pointer table.
Illustratively, when the pointer table is automatically read, a camera is arranged corresponding to the pointer table, and the camera shoots corresponding to the pointer table to obtain a corresponding image, which is defined as a shot image of the camera. Or the inspection robot is used for performing inspection to obtain image data of the inspection environment and taking the image data as a camera to shoot images. Furthermore, the inspection by the robot is taken as an example for description, and when the inspection by the inspection robot is performed, the inspection environment is shot in real time through a camera configured by the inspection robot, so that corresponding image data is obtained. The image data includes image data captured by the corresponding pointer table. Furthermore, based on the camera shooting image shot by the inspection robot, the processing equipment configured by the inspection robot can process the image to realize automatic reading, and the camera shooting image can be sent to a background server to be processed by the background server.
Specifically, when image processing is performed based on the acquired camera-shot image, firstly, target detection corresponding to the pointer table needs to be performed based on the camera-shot image, and an image containing the pointer table image features is intercepted from the camera-shot image according to a target detection result and defined as an original image. It should be noted that, when image acquisition is performed, in order to ensure the quality of an image sample for automatic reading, multiple images may be continuously captured corresponding to the pointer table, and one camera captured image with the highest image quality may be selected from the multiple images for further image processing. When the target (pointer table) detection is carried out based on the camera shooting image, the camera shooting image is input into a pre-trained target detection model, and a corresponding original image is output through target detection. The target detection model is constructed based on a neural network, and can be a ResNet-50 target detection model. In practical application, a corresponding network is selected for model construction according to model construction requirements. Taking the ResNet-50 target detection model as an example, referring to fig. 2, the construction process of the target detection model includes:
s1101, collecting an image containing the image characteristics of the pointer table as a training sample;
and S1102, inputting the training sample into a ResNet-50 network for model training to obtain a corresponding target detection model.
The ResNet-50 target detection model is mainly composed of convolution layers and pooling layers, naming rules of layers in the network are composed of classes of semantic numbers appearing several times in the network, for example, conv8 represents the 8 th convolution layer in the network, sampling represents an upper sampling layer in the network, output feature maps of each layer in the network represent a small resolution of the output feature maps, and the size of each layer in the network represents a small resolution of a local convolution layer, so that the convolution layers are sensitive to small target feature map learning, and the convolution layers of the target feature maps are easy to learn, and the target detection efficiency of the network is improved.
After the target detection model is built, the corresponding obtained camera shooting images are all input into the target detection model for target detection, and corresponding original images are obtained. The original image including the pointer table image feature extracted based on the object detection model may be an original image of a fixed image standard including background image data, or an original image in which only the pointer table image feature is extracted, as necessary. For the original image only intercepting the image characteristics of the pointer table, the background image noise is eliminated, and the interference of the image background on the identification precision can be reduced in the further image processing process.
And S120, carrying out perspective transformation on the original image based on the projection mapping of the pixel coordinates to obtain a corresponding front-view image.
Based on the original image obtained in step S110, when shooting is performed corresponding to the pointer table, the shooting angle of the camera cannot guarantee that shooting is performed completely corresponding to the front of the pointer table. Particularly, when the inspection robot takes images, the visual angle of the taken images is not fixed, namely, the captured original images have certain angular distortion relative to the images taken at the front visual angle of the pointer table, so that the influence of the inclination angle needs to be eliminated, the images at the front visual angle of the pointer table are provided for automatic identification of pointer reading, and the identification precision of the pointer reading is further improved.
Based on the principle of perspective transformation, the pixel coordinates of the original image are projected and mapped to obtain a corresponding front-view image. Referring to fig. 3, the process of perspective transformation of the original image includes:
s1201, positioning a first identification point of the oval dial plate in the original image, wherein the first identification point at least comprises the center of the oval circle and the intersection points of the major axis and the minor axis with the oval;
s1202, determining that the first identification point is mapped to be a second identification point of the perfect circular dial plate;
s1203, calculating a perspective transformation matrix according to the first identification point and the second identification point;
and S1204, performing coordinate conversion on the pixel coordinates of the original image based on the perspective transformation matrix to obtain the front-view image.
It can be understood that, referring to fig. 4, the dial corresponding to the original image of the pointer table is elliptical under the influence of the shooting angle. Correspondingly, the front view image of the index table obtained by perspective transformation is shown in fig. 5, and the dial thereof is in a perfect circle shape. When the perspective transformation is performed, a first identification point on the original image is determined in advance. And determining the elliptic dial plate on the original image through dial plate detection, and further calibrating the intersection points of the elliptic circle center, the long axis and the short axis of the dial plate and the ellipse. The dial plate detection can also be realized through a target detection algorithm, and the calibration of the first identification point is further carried out based on the detected elliptic dial plate to obtain the coordinate of the first identification point. Similarly, according to the projection mapping principle of the original image, the circle center and the long axis of the oval dial plate of the original image should correspond to the circle center and the radius of the perfect circle dial plate of the front view image, and second identification point calibration is carried out on the perfect circle dial plate of the front view image based on the circle center and the radius to obtain the coordinates of the second identification point, wherein the second identification point corresponds to the first identification point.
Furthermore, based on the first identification point and the second identification point, perspective transformation can be carried out to obtain an orthographic image. The essence of the Perspective Transformation is to project the image (original image) onto a new viewing plane (resulting in a front view image). The general transformation formula is as follows:
where, (u, v) is the pixel coordinates of the original image, (x ═ x '/w', y ═ y '/w') is the pixel coordinates of the orthophoria image after transformation. Further transformation yields a perspective transformation matrix as follows:
T2＝[a13a23]Tfor generating a perspective transformation of the image;
T3＝[a31a32]representing image translation.
Further obtaining the pixel coordinate expression of the front-view image after perspective transformation based on the perspective transformation matrix as follows:
based on the above formula, a perspective transformation matrix can be obtained by giving four pairs of pixel coordinates (i.e. the first identification point and the second identification point) before and after perspective transformation. Otherwise, given the perspective transformation matrix, the coordinates (x, y) of each pixel point of the front-view image after perspective transformation can be calculated based on the coordinates of each pixel point of the original image, and the conversion of the front-view image is completed.
S130, dial scales and pointers in the front view image are positioned, and reading of a pointer meter is calculated based on the rotation angle of the pointers relative to the dial scales.
Specifically, based on the determined orthographic image, automatic identification of the pointer meter reading can be performed. The dial scale and the pointer are recognized based on the front view image. And respectively detecting the dial scales and the pointers in the front-view image by adopting a target detection algorithm, and referring to the mode of detecting the pointer table based on the target detection model in the step S110, the embodiment of the application constructs the target detection model based on the dial scales and the target detection model of the pointers, and then detects and identifies the dial scales and the pointers on the front-view image. And determining the rotation angle of the pointer relative to the dial scale by calibrating the positions of the dial scale and the pointer. It can be understood that, taking the front view image of fig. 5 as an example, on the premise that the start scale ("0") and the end scale ("100") of the dial scale and the angle value corresponding to the dial scale are known, the dial scale value indicated by the current pointer can be correspondingly determined based on the rotation angle of the pointer relative to the dial scale, and then the dial scale value is extracted, so as to complete the automatic reading of the pointer meter.
Further, when automatic reading is carried out, the rotating angle is input into a preset reading calculation model, corresponding reading of the pointer table is output, and the reading calculation model is constructed in advance according to the corresponding relation between the rotating angle and the reading of the pointer table. It will be understood that each value of the angle of rotation of the pointer relative to the scale of the dial corresponds to a value of the scale of the dial. For example, the dial scale value is "0-100", the angle corresponding to the start scale is defined as "0 °", and the angle value from the start scale to the end scale of the dial scale is from "0 ° -300 °". When the rotation angle of the pointer relative to the dial plate is between 0 and 300 degrees, the corresponding dial plate scale value is between 0 and 100. The calculation model of the reading is obtained as follows:
C＝φ/3 φ＝(0，300)
wherein C is a dial scale value, phi is a rotation angle value, and the value is 0-300. It can be understood that when the rotation angle value of the pointer relative to the dial scale exceeds 300 °, the reading of the current dial scale is overproof or abnormal.
And finally, filling dial scale reading information obtained according to the automatic reading into a preset detection data table, and storing the dial scale reading information as related detection data of the current moment. The detection data table can be a temperature detection data table, a humidity detection data table, an air pressure detection data table and the like corresponding to different detection scenes. Taking temperature detection as an example, when the dial scale reading obtained by automatic reading is '25', the temperature of the current detection environment is indicated to be 25 ℃, and the dial scale reading is filled into a detection data table to indicate that the current temperature data is 25 ℃. In practical application, corresponding detection data output can be directly generated by integration according to the recognized dial scale reading and based on the preset data type and data unit information.
In addition, in one embodiment, in order to further guarantee the accuracy of automatic reading of the pointer table, the dial scale and the pointer of the original image are detected, the identification scale value on the pointer table is identified through the OCR, and based on the calibrated dial scale, the pointer and the corresponding identification scale value, it can be determined within which scale value range the scale value indicated by the current pointer falls. And further verifying the scale value determined by the automatic reading of the pointer table based on the determined scale value range. It will be appreciated that if the scale value determined by the automatic reading is not within the scale value range, then the scale value may be erroneous.
In addition, in one embodiment, whether the dial scales are calibrated accurately is verified based on the determined pixel coordinates corresponding to the center of the dial on the front-view image and the calibrated dial scales detected through the targets. As shown in fig. 5, it can be understood that, on the front view image, the distance between each pixel coordinate on the dial scale and the pixel coordinate of the circle center should be the same, and by comparing the distance between the circle center and each pixel coordinate of the dial scale, if there is a difference, the position calibration of the dial scale needs to be adjusted, so as to ensure the accurate positioning of the dial scale.
The method comprises the steps of acquiring an image shot by a camera, carrying out pointer table detection based on the image shot by the camera, intercepting an original image containing the image characteristics of the pointer table, carrying out perspective transformation on the original image based on the projection mapping of pixel coordinates to obtain a corresponding front view image, and finally calculating the reading of the pointer table based on the rotation angle of the pointer relative to the dial scale by positioning the dial scale and the pointer in the front view image. By adopting the technical means, the angle distortion of the image shot by the camera can be eliminated, the influence of the shooting angle on the automatic reading of the pointer meter is reduced, the precision of the automatic reading of the pointer meter is improved, and the accuracy of the automatic reading is guaranteed.
Example two:
on the basis of the above embodiments, fig. 6 is a flowchart of another automatic reading method for adaptive distortion of a pointer table provided in embodiment two of the present application, where the automatic reading method for adaptive distortion of a pointer table includes:
s210, acquiring an image shot by a camera, carrying out pointer table detection based on the image shot by the camera, and intercepting an original image containing the image characteristics of a pointer table;
s220, carrying out perspective transformation on the original image based on projection mapping of pixel coordinates to obtain a corresponding front-view image;
s230, positioning dial scales and a pointer in the front view image, and calculating the reading of a pointer meter based on the rotation angle of the pointer relative to the dial scales;
s240, based on a preset reading threshold value, when the reading of the pointer meter is detected to be greater than or equal to or less than the reading threshold value, outputting corresponding prompt information.
After the automatic reading of the pointer table is completed, the embodiment of the application further compares the preset reading threshold value according to the recognized pointer reading, and outputs prompt information to prompt a user when the pointer reading reaches the corresponding pointer. For example, in a temperature detection scene, a low-temperature alarm value "25 ℃ and a high-temperature alarm value" 50 ℃ are preset, and according to the result of automatic reading, if the detected temperature is lower than the low-temperature alarm value, a low-temperature alarm prompt is output, and if the detected temperature is higher than the high-temperature alarm value "50 ℃, a high-temperature alarm prompt is output. Thus, alarm prompting based on automatic reading is realized.
In addition, in one embodiment, based on historical data obtained from each automatic reading, when a certain reading approaches a reading threshold value for a period of time, prompt information is output to prompt the user of the situation. Therefore, the user is prompted to ask the user to solve the risk condition of the relevant indexes in time.
Example three:
on the basis of the foregoing embodiments, fig. 7 is a schematic structural diagram of an automatic reading device for adaptive distortion of a pointer table according to a third embodiment of the present application. Referring to fig. 7, the automatic reading device for adaptive distortion of a pointer table provided in this embodiment specifically includes: an acquisition module 31, a transformation module 32 and a positioning module 33.
The acquisition module 31 is configured to acquire an image captured by a camera, perform pointer table detection based on the image captured by the camera, and intercept an original image including features of the pointer table image;
the transformation module 32 is configured to perform perspective transformation on the original image based on projection mapping of pixel coordinates to obtain a corresponding front-view image;
the positioning module 33 is configured to position the dial scales and the pointer in the front view image, and calculate a pointer meter reading based on a rotation angle of the pointer relative to the dial scales.
The method comprises the steps of acquiring an image shot by a camera, carrying out pointer table detection based on the image shot by the camera, intercepting an original image containing the image characteristics of the pointer table, carrying out perspective transformation on the original image based on the projection mapping of pixel coordinates to obtain a corresponding front view image, and finally calculating the reading of the pointer table based on the rotation angle of the pointer relative to the dial scale by positioning the dial scale and the pointer in the front view image. By adopting the technical means, the angle distortion of the image shot by the camera can be eliminated, the influence of the shooting angle on the automatic reading of the pointer meter is reduced, the precision of the automatic reading of the pointer meter is improved, and the accuracy of the automatic reading is guaranteed.
Specifically, still include:
and the prompting module is used for outputting corresponding prompting information when the reading of the pointer meter is detected to be greater than or equal to or less than the reading threshold value based on a preset reading threshold value.
The automatic reading device for the adaptive distortion of the pointer table provided by the third embodiment of the application can be used for executing the automatic reading method for the adaptive distortion of the pointer table provided by the first embodiment and the second embodiment, and has corresponding functions and beneficial effects.
Example four:
an embodiment of the present application provides an electronic device, and with reference to fig. 8, the electronic device includes: a processor 41, a memory 42, a communication module 43, an input device 44, and an output device 45. The number of processors in the electronic device may be one or more, and the number of memories in the electronic device may be one or more. The processor, memory, communication module, input device, and output device of the electronic device may be connected by a bus or other means.
The communication module 43 is used for data transmission.
The processor 41 executes various functional applications of the device and data processing by running software programs, instructions and modules stored in the memory, namely, the automatic reading method of the pointer table adaptive distortion is realized.
The input device 44 is operable to receive input numeric or character information and to generate key signal inputs relating to user settings and function controls of the apparatus. The output device 45 may include a display device such as a display screen.
The electronic equipment provided by the embodiment I and the embodiment II can be used for executing the automatic reading method for the adaptive distortion of the pointer table, and has corresponding functions and beneficial effects.
Example five:
embodiments of the present application also provide a storage medium containing computer-executable instructions, which when executed by a computer processor, perform a method for automatically reading pointer table adaptive distortion, the method for automatically reading pointer table adaptive distortion comprising: acquiring a camera shooting image, carrying out pointer table detection based on the camera shooting image, and intercepting an original image containing the pointer table image characteristics; carrying out perspective transformation on the original image based on projection mapping of pixel coordinates to obtain a corresponding front-view image; and positioning dial scales and a pointer in the front-view image, and calculating the reading of a pointer meter based on the rotation angle of the pointer relative to the dial scales.
Storage medium-any of various types of memory devices or storage devices. The term "storage medium" is intended to include: mounting media such as CD-ROM, floppy disk, or tape devices; computer system memory or random access memory such as DRAM, DDR RAM, SRAM, EDO RAM, Lanbas (Rambus) RAM, etc.; non-volatile memory such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc. The storage medium may also include other types of memory or combinations thereof. In addition, the storage medium may be located in a first computer system in which the program is executed, or may be located in a different second computer system connected to the first computer system through a network (such as the internet). The second computer system may provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media residing in different locations, e.g., in different computer systems connected by a network. The storage medium may store program instructions (e.g., embodied as a computer program) that are executable by one or more processors.
Of course, the storage medium provided by the embodiments of the present application contains computer-executable instructions, and the computer-executable instructions are not limited to the automatic reading method for pointer table adaptive distortion described above, and may also perform related operations in the automatic reading method for pointer table adaptive distortion provided by any embodiments of the present application.
The automatic reading device, the storage medium, and the electronic device for adaptive distortion of a pointer table provided in the foregoing embodiments may perform the automatic reading method for adaptive distortion of a pointer table provided in any embodiments of the present application, and reference may be made to the automatic reading method for adaptive distortion of a pointer table provided in any embodiments of the present application without detailed technical details described in the foregoing embodiments.
The foregoing is considered as illustrative of the preferred embodiments of the invention and the technical principles employed. The present application is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present application has been described in more detail with reference to the above embodiments, the present application is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present application, and the scope of the present application is determined by the scope of the claims.
Claims (10)
1. An automatic reading method for adaptive distortion of a pointer table is characterized by comprising the following steps:
acquiring a camera shooting image, carrying out pointer table detection based on the camera shooting image, and intercepting an original image containing the pointer table image characteristics;
carrying out perspective transformation on the original image based on projection mapping of pixel coordinates to obtain a corresponding front-view image;
and positioning dial scales and a pointer in the front-view image, and calculating the reading of a pointer meter based on the rotation angle of the pointer relative to the dial scales.
2. The method for automatically reading the adaptive distortion of the pointer table as claimed in claim 1, wherein the acquiring the camera shot image, performing pointer table detection based on the camera shot image, and capturing an original image containing the characteristics of the pointer table image comprises:
and inputting images shot by a camera into a pre-trained target detection model, and outputting corresponding original images through target detection.
3. The method for automatically reading adaptive distortion of the pointer table as claimed in claim 2, wherein the construction process of the target detection model comprises:
collecting an image containing the image characteristics of the pointer table as a training sample;
and inputting the training sample into a ResNet-50 network for model training to obtain a corresponding target detection model.
4. The method for automatically reading the adaptive distortion of the pointer table as claimed in claim 1, wherein the perspective transformation of the original image based on the projection mapping of the pixel coordinates to obtain the corresponding front-view image comprises:
positioning a first identification point of the oval dial plate in the original image, wherein the first identification point at least comprises an oval circle center and intersection points of a long axis and a short axis with the oval;
determining the first identification point to be mapped as a second identification point of the perfect circle dial plate;
calculating a perspective transformation matrix according to the first identification point and the second identification point;
and performing coordinate conversion on the pixel coordinates of the original image based on the perspective transformation matrix to obtain the front-view image.
5. The method for automatically reading adaptive distortion of the pointer table as claimed in claim 1, wherein the positioning of the dial scale and the pointer in the front view image comprises:
and respectively detecting dial scales and pointers in the front-view image based on a target detection algorithm.
6. The method for automatically reading the adaptive distortion of the pointer table as claimed in claim 1, wherein calculating the reading of the pointer table based on the rotation angle of the pointer relative to the dial scale comprises:
and inputting the rotation angle into a preset reading calculation model, and outputting corresponding reading of the pointer table, wherein the reading calculation model is constructed in advance according to the corresponding relation between the rotation angle and the reading of the pointer table.
7. The method for automatically reading the adaptive distortion of the pointer table as claimed in claim 1, wherein after calculating the reading of the pointer table based on the rotation angle of the pointer relative to the dial scale, the method further comprises:
based on a preset reading threshold value, when the reading of the pointer meter is detected to be greater than or equal to or less than the reading threshold value, outputting corresponding prompt information.
8. An automatic reading device of pointer table self-adaptation distortion, characterized by comprising:
the acquisition module is used for acquiring images shot by a camera, carrying out pointer table detection based on the images shot by the camera and intercepting original images containing the image characteristics of the pointer table;
the transformation module is used for carrying out perspective transformation on the original image based on the projection mapping of the pixel coordinates to obtain a corresponding front-view image;
and the positioning module is used for positioning the dial scales and the pointer in the front view image and calculating the reading of the pointer meter based on the rotation angle of the pointer relative to the dial scales.
9. An electronic device, comprising:
a memory and one or more processors;
the memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement a method of automatic reading of adaptive distortion of a pointer table as recited in any of claims 1-7.
10. A storage medium containing computer-executable instructions for performing the method of automatic reading of adaptive distortion of a pointer table as claimed in any one of claims 1 to 7 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010329384.7A CN111476787A (en) | 2020-04-23 | 2020-04-23 | Automatic reading method and device for adaptive distortion of pointer meter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010329384.7A CN111476787A (en) | 2020-04-23 | 2020-04-23 | Automatic reading method and device for adaptive distortion of pointer meter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111476787A true CN111476787A (en) | 2020-07-31 |
Family
ID=71762861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010329384.7A Pending CN111476787A (en) | 2020-04-23 | 2020-04-23 | Automatic reading method and device for adaptive distortion of pointer meter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111476787A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037192A (en) * | 2020-08-28 | 2020-12-04 | 西安交通大学 | Method for collecting burial depth information in town gas public pipeline installation process |
| CN112287922A (en) * | 2020-10-29 | 2021-01-29 | 北京百度网讯科技有限公司 | Meter reading method, device, equipment and storage medium |
| CN112487901A (en) * | 2020-11-19 | 2021-03-12 | 国网河南省电力公司检修公司 | Monitoring method and monitoring system for running state of pressure gauge of oil-discharging nitrogen-injecting fire extinguishing device |
| CN113269193A (en) * | 2021-06-08 | 2021-08-17 | 上海高德威智能交通系统有限公司 | Pointer type meter reading method, device and storage medium |
| CN114757922A (en) * | 2022-04-19 | 2022-07-15 | 河南中原动力智能制造有限公司 | Meter reading method and device applied to pointer type pressure gauge |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303168A (en) * | 2015-10-14 | 2016-02-03 | 南京第五十五所技术开发有限公司 | Multi-view pointer type instrument identification method and device |
| CN106951900A (en) * | 2017-04-13 | 2017-07-14 | 杭州申昊科技股份有限公司 | A kind of automatic identifying method of arrester meter reading |
| CN107167169A (en) * | 2017-07-03 | 2017-09-15 | 吉林大学 | Readings of pointer type meters identification measuring method based on NI Vision Builder for Automated Inspection |
| CN110232354A (en) * | 2019-06-14 | 2019-09-13 | 四川超影科技有限公司 | A kind of pointer instrument universal identification method |
| CN110414510A (en) * | 2019-07-26 | 2019-11-05 | 华中科技大学 | A kind of readings of pointer type meters bearing calibration |
| CN110807355A (en) * | 2019-09-12 | 2020-02-18 | 天津大学 | Pointer instrument detection and reading identification method based on mobile robot |
-
2020
- 2020-04-23 CN CN202010329384.7A patent/CN111476787A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303168A (en) * | 2015-10-14 | 2016-02-03 | 南京第五十五所技术开发有限公司 | Multi-view pointer type instrument identification method and device |
| CN106951900A (en) * | 2017-04-13 | 2017-07-14 | 杭州申昊科技股份有限公司 | A kind of automatic identifying method of arrester meter reading |
| CN107167169A (en) * | 2017-07-03 | 2017-09-15 | 吉林大学 | Readings of pointer type meters identification measuring method based on NI Vision Builder for Automated Inspection |
| CN110232354A (en) * | 2019-06-14 | 2019-09-13 | 四川超影科技有限公司 | A kind of pointer instrument universal identification method |
| CN110414510A (en) * | 2019-07-26 | 2019-11-05 | 华中科技大学 | A kind of readings of pointer type meters bearing calibration |
| CN110807355A (en) * | 2019-09-12 | 2020-02-18 | 天津大学 | Pointer instrument detection and reading identification method based on mobile robot |
Non-Patent Citations (1)
| Title |
|---|
| 邢浩强 等: "变电站指针式仪表检测与识别方法", vol. 38, no. 11, pages 2813 - 2821 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037192A (en) * | 2020-08-28 | 2020-12-04 | 西安交通大学 | Method for collecting burial depth information in town gas public pipeline installation process |
| CN112287922A (en) * | 2020-10-29 | 2021-01-29 | 北京百度网讯科技有限公司 | Meter reading method, device, equipment and storage medium |
| CN112487901A (en) * | 2020-11-19 | 2021-03-12 | 国网河南省电力公司检修公司 | Monitoring method and monitoring system for running state of pressure gauge of oil-discharging nitrogen-injecting fire extinguishing device |
| CN113269193A (en) * | 2021-06-08 | 2021-08-17 | 上海高德威智能交通系统有限公司 | Pointer type meter reading method, device and storage medium |
| CN114757922A (en) * | 2022-04-19 | 2022-07-15 | 河南中原动力智能制造有限公司 | Meter reading method and device applied to pointer type pressure gauge |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111476787A (en) | Automatic reading method and device for adaptive distortion of pointer meter | |
| CN110322500B (en) | Optimization method and device for instant positioning and map construction, medium and electronic equipment | |
| CN111586360B (en) | Unmanned aerial vehicle projection method, device, equipment and storage medium | |
| CN111780764B (en) | Visual positioning method and device based on visual map | |
| CN110909693B (en) | 3D face living body detection method, device, computer equipment and storage medium | |
| CN110095752B (en) | Positioning method, apparatus, device and medium | |
| CN112307940A (en) | Model training method, human body posture detection method, device, equipment and medium | |
| WO2019196476A1 (en) | Laser sensor-based map generation | |
| CN111028205B (en) | Eye pupil positioning method and device based on binocular distance measurement | |
| CN112257676A (en) | Pointer instrument reading method and system and inspection robot | |
| CN110738703B (en) | Positioning method and device, terminal and storage medium | |
| US11049313B2 (en) | Rendering an object | |
| US11687467B2 (en) | Data sharing system and data sharing method therefor | |
| CN112634368A (en) | Method and device for generating space and OR graph model of scene target and electronic equipment | |
| CN111239684A (en) | Binocular fast distance measurement method based on YoloV3 deep learning | |
| CN109934165A (en) | A kind of joint point detecting method, device, storage medium and electronic equipment | |
| CN111353325A (en) | Key point detection model training method and device | |
| CN110007764B (en) | Gesture skeleton recognition method, device and system and storage medium | |
| CN111738319A (en) | Clustering result evaluation method and device based on large-scale samples | |
| CN111652168B (en) | Group detection method, device, equipment and storage medium based on artificial intelligence | |
| CN112580581A (en) | Target detection method and device and electronic equipment | |
| CN112102404A (en) | Object detection tracking method and device and head-mounted display equipment | |
| CN116597246A (en) | Model training method, target detection method, electronic device and storage medium | |
| CN110807375A (en) | Human head detection method, device and equipment based on depth image and storage medium | |
| WO2022183906A1 (en) | Imaging method and apparatus, device, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |