CN113221893A

CN113221893A - Instrument image processing method and device, computer equipment and storage medium

Info

Publication number: CN113221893A
Application number: CN202110597680.XA
Authority: CN
Inventors: 肖志明; 王兵; 王聘博; 王江
Original assignee: SHENZHEN VIATOM TECHNOLOGY CO LTD
Current assignee: SHENZHEN VIATOM TECHNOLOGY CO LTD
Priority date: 2021-05-31
Filing date: 2021-05-31
Publication date: 2021-08-06

Abstract

The application relates to a meter image processing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring an instrument image shot by a digital instrument; performing area analysis on the instrument image to obtain each interested area in the instrument image; performing feature matching on the instrument image and a standard feature image corresponding to the digital instrument to determine matching feature points in the instrument image; determining a target region of interest from each region of interest based on the distribution characteristics of the matched feature points in each region of interest; and carrying out numerical identification on the target region of interest to obtain a numerical identification result of the instrument image. By adopting the method, the processing efficiency of the reading of the digital instrument can be improved.

Description

Instrument image processing method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for processing an instrument image, a computer device, and a storage medium.

Background

With the continuous development of society, digital instruments have been widely used in industrial production and daily life, such as various medical instruments, such as various sphygmomanometers and heart rate meters, and electric meters and substation instruments, due to the advantages of high precision, convenience in reading and writing, and the like. For the data reading of the digital instrument, the data transmission can be carried out on the digital instrument with a data interface through the data interface. However, for a digital meter without a data interface, manual reading is still required, which results in inefficient reading processing of the digital meter.

Disclosure of Invention

In view of the above, it is necessary to provide a meter image processing method, apparatus, computer device and storage medium capable of improving the efficiency of digital meter reading processing.

A meter image processing method, the method comprising:

acquiring an instrument image shot by a digital instrument;

performing area analysis on the instrument image to obtain each interested area in the instrument image;

performing feature matching on the instrument image and a standard feature image corresponding to the digital instrument to determine matching feature points in the instrument image;

determining a target region of interest from each region of interest based on the distribution characteristics of the matched feature points in each region of interest;

and carrying out numerical identification on the target region of interest to obtain a numerical identification result of the instrument image.

In one embodiment, performing a region analysis on the meter image to obtain regions of interest in the meter image includes: respectively carrying out binarization processing on the channel images of the instrument image corresponding to the channels to obtain binary images of the channels; respectively extracting the contour of each channel binary image to obtain the contour of each channel binary image; and constructing regions based on the contours in the binary images of the channels to obtain regions of interest in the instrument image.

In one embodiment, the extracting the contour of each channel binary image to obtain the contour of each channel binary image includes: respectively carrying out first contour extraction on the binary images of each channel to obtain a first contour extraction result; carrying out contour deformation processing on the first contour extraction result to obtain a contour deformation result; and performing second contour extraction based on the contour deformation result to obtain the contour in the binary image of each channel.

In one embodiment, the region construction is performed based on the contour in the binary image of each channel, and each region of interest in the instrument image is obtained, including: performing region fitting based on the contour in the binary image of each channel to obtain a first region of interest; carrying out contour reconstruction on the contour in the binary image of each channel through a contour reconstruction algorithm to obtain a second region of interest; and obtaining each interested area in the instrument image according to the first interested area and the second interested area.

In one embodiment, determining the target region of interest from each region of interest based on the distribution features of the matched feature points in each region of interest includes: counting the number of feature points of the matched feature points respectively included in the region of interest; obtaining the feature point occupation ratio corresponding to each interested area according to the number of the feature points and the area of the corresponding interested area; and determining a target region of interest from the regions of interest based on the feature point ratio.

In one embodiment, the performing numerical identification on the target region of interest to obtain a numerical identification result of the instrument image includes: determining the numerical type of a numerical value to be identified in the digital instrument; performing numerical region division on the target region of interest based on the numerical type to obtain sub-regions corresponding to the numerical types respectively; and respectively carrying out numerical value identification on each subarea to obtain a numerical value identification result of the instrument image.

In one embodiment, the performing numerical identification on each sub-region respectively to obtain a numerical identification result of the instrument image includes: carrying out contour enhancement processing on each subregion to obtain contour enhancement results respectively corresponding to each subregion; extracting the contour based on each contour enhancement result to obtain a sub-region numerical contour corresponding to each sub-region; matching the sub-region numerical value outline with the numerical value mapping characteristics in the numerical value mapping table, and determining the sub-region numerical value corresponding to the sub-region numerical value outline according to the matching result; and obtaining a numerical value identification result of the instrument image according to the numerical value of the sub-region corresponding to the numerical value profile of each sub-region.

A meter image processing apparatus, the apparatus comprising:

the instrument image acquisition module is used for acquiring an instrument image shot by a digital instrument;

the interesting region determining module is used for carrying out region analysis on the instrument image to obtain each interesting region in the instrument image;

the characteristic matching module is used for carrying out characteristic matching on the instrument image and a standard characteristic image corresponding to the digital instrument to determine a matching characteristic point in the instrument image;

the target area determining module is used for determining a target interested area from each interested area based on the distribution characteristics of the matched characteristic points in each interested area;

and the numerical value identification processing module is used for carrying out numerical value identification on the target region of interest to obtain a numerical value identification result of the instrument image.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

acquiring an instrument image shot by a digital instrument;

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

acquiring an instrument image shot by a digital instrument;

According to the instrument image processing method, the instrument image processing device, the computer equipment and the storage medium, the instrument image shot by the digital instrument is subjected to region analysis to obtain each interested region in the instrument image, distribution characteristics of matching characteristic points in each interested region are determined based on characteristic matching of the instrument image and a standard characteristic image corresponding to the digital instrument, a target interested region is determined from each interested region, and numerical identification is carried out on the target interested region to obtain a numerical identification result of the instrument image. The target interesting regions are determined through the distribution characteristics of the matching characteristic points determined in the instrument image in the interesting regions, and the target interesting regions are subjected to numerical value identification, so that the regions containing numerical values are accurately determined in the instrument image for numerical value identification, the data volume of numerical value identification processing can be reduced on the premise of ensuring the pertinence of numerical value identification, and the processing efficiency of digital instrument reading is effectively improved.

Drawings

FIG. 1 is a diagram of an embodiment of a meter image processing method;

FIG. 2 is a flow diagram illustrating a method for processing a meter image in one embodiment;

FIG. 3 is a flow diagram illustrating a process for obtaining numeric identification results in one embodiment;

FIG. 4 is a meter image in one embodiment;

FIG. 5 is a channel image of the G channel in the embodiment of FIG. 4;

FIG. 6 is a channel image of the B channel in the embodiment of FIG. 4;

FIG. 7 is a channel binary image in the embodiment shown in FIG. 6;

FIG. 8 is a diagram illustrating the result of contour extraction in the embodiment shown in FIG. 7;

FIG. 9 is a diagram of a target region of interest after graying processing in one embodiment;

FIG. 10 is a diagram of the sub-regions corresponding to the systolic pressure SYS in the embodiment shown in FIG. 9;

FIG. 11 is a diagram illustrating a numerical profile of the sub-regions of the embodiment of FIG. 10;

FIG. 12 is a block diagram showing the structure of a meter image processing apparatus according to an embodiment;

FIG. 13 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The meter image processing method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The digital instrument can be a sphygmomanometer, the terminal 102 can shoot a display panel of the sphygmomanometer to obtain an instrument image corresponding to the sphygmomanometer, the terminal sends the shot instrument image to the server 104, the server 104 performs area analysis on the instrument image sent by the terminal 102 to obtain each region of interest in the instrument image, distribution characteristics of matching characteristic points in each region of interest are determined based on characteristic matching of the instrument image and a standard characteristic image corresponding to the digital instrument, a target region of interest is determined from each region of interest, numerical identification is performed on the target region of interest to obtain a numerical identification result of the instrument image, the numerical identification result is returned to the terminal 102, and the terminal 102 can display the numerical identification result in an interface. In addition, after the terminal 102 obtains the meter image by shooting for the digital meter, the meter image can be also independently processed, that is, the terminal 102 can directly perform area analysis on the meter image to obtain each interested area in the meter image, perform distribution characteristics of matching characteristic points in each interested area determined by characteristic matching based on the meter image and a standard characteristic image corresponding to the digital meter, determine a target interested area from each interested area, perform numerical identification on the target interested area, and obtain and display a numerical identification result of the meter image.

The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.

In one embodiment, as shown in fig. 2, a meter image processing method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:

step 202, acquiring a meter image obtained by shooting a digital meter.

The digital instrument is an instrument for displaying measured values by numbers, namely the digital instrument converts the measurement into digital values and displays the digital values in a digital form. The digital meter displays the measured quantity in a digital form, and the reading is intuitive and has no visual error. The digital meter generally comprises three parts of a dial and a pointer for indicating electric quantity, an electromagnetic force-based electric measuring circuit, an analog-to-digital conversion part and a digital display part. The digital instrument can be applied to different scenes so as to display various measurement data, for example, the digital instrument can be applied to a medical instrument such as a sphygmomanometer so as to directly display the blood pressure value measured by the sphygmomanometer through displaying the digital form. The specific product form of the digital instrument is determined according to the actual application scene, and the meaning of the displayed numerical value corresponds to the actual application scene. The meter image is an image obtained by shooting a digital meter, and specifically, the meter image can be obtained by shooting a display panel for displaying a measurement value in the digital meter by a user through an image sensing device, for example, a terminal by the user.

Specifically, the server acquires an instrument image obtained by shooting for the digital instrument, the instrument image is obtained by shooting specifically for a display panel of the digital instrument, for example, a user can shoot the display panel of the digital instrument through the terminal and send the shot instrument image to the server, and the server receives the instrument image uploaded by the terminal to perform numerical value identification processing for the instrument image.

And 204, performing area analysis on the instrument image to obtain each interested area in the instrument image.

The interested areas are areas which comprise the numerical values required to be identified in the instrument image, the digital instrument can display various numerical values due to different shooting ranges of the instrument image, the interested areas obtained by analyzing the instrument image in areas can be multiple, and the included ranges of the interested areas are different, so that the proper areas in the interested areas are required to be screened for numerical value identification, and accurate reading processing of the digital instrument is realized.

Specifically, after the server obtains the instrument image, the area analysis is performed on the instrument image, specifically, after the binarization processing is performed on the instrument image by the server, the contour extraction is performed based on the binarization processing result, and each region of interest in the instrument image is constructed and obtained based on the extracted contour.

And step 206, performing feature matching on the instrument image and the standard feature image corresponding to the digital instrument to determine a matching feature point in the instrument image.

The standard characteristic image is a standard image of a display panel of the digital instrument, namely the standard characteristic image corresponds to the display panel of the digital instrument, and all data of the digital instrument can be accurately and comprehensively displayed. The instrument image obtained by the user through the terminal shooting can shoot other unrelated environments besides the display panel of the digital instrument due to the shooting angle and the shooting distance, so that the instrument image needs to be intercepted, an area corresponding to the display panel of the digital instrument is intercepted from the instrument image for numerical value identification, namely, an image area matched with the standard characteristic image needs to be intercepted from the instrument image for numerical value identification. And the matched feature points are the feature points of the instrument image which is successfully matched with the standard feature image after the instrument image is matched with the standard feature image corresponding to the digital instrument.

Specifically, after the server obtains an instrument image obtained by shooting for the digital instrument, the server further queries a standard feature image corresponding to the digital instrument, where the standard feature image may be preset, for example, a display panel of the digital instrument may be shot in advance and a standard feature image corresponding to the display panel may be captured, and the standard feature image may cover the display panel of the digital instrument, including all data in the digital instrument, and not including other unrelated environment backgrounds. The server performs feature matching on the instrument image and the standard feature image, for example, the instrument image and the standard feature image are subjected to feature matching based on an image matching algorithm, specifically, a gray-scale-based matching algorithm, a feature-based matching algorithm or a relation-based matching algorithm, and a matching feature point in the instrument image is determined according to a feature matching result, wherein the matching feature point refers to a feature point in the instrument image, which is successfully matched with the standard feature image. In a specific application, the instrument image and the standard feature image corresponding to the digital instrument may be subjected to feature matching based on a Scale-invariant feature transform (SIFT) algorithm to obtain matching feature points in the instrument image.

And step 208, determining a target region of interest from each region of interest based on the distribution characteristics of the matched feature points in each region of interest.

The target interesting area is an image area corresponding to a display panel of the digital instrument in the instrument image, namely the target interesting area comprises measurement data of the digital instrument and does not carry other environment backgrounds, numerical identification is carried out on the target interesting area, and accurate reading of the digital instrument can be achieved.

Specifically, the server determines the distribution characteristics of the matching feature points in the meter image in each region of interest, such as the number of the feature points, the distribution density and the like of the matching feature points in the meter image in each region of interest. And the server screens all the interested areas based on the distribution characteristics to obtain the target interested areas. For example, the regions of interest with the largest number of matched feature points and the smallest area can be obtained by screening from the regions of interest and determined as the target regions of interest, so that the environmental background of the instrument image is removed, the digital instrument is identified based on the target regions of interest, and the accurate reading processing of the digital instrument can be realized.

And step 210, carrying out numerical identification on the target region of interest to obtain a numerical identification result of the instrument image.

The numerical value identification result refers to an identification result for identifying each numerical value displayed in the digital instrument, namely a reading result of the digital instrument. Specifically, after the server determines a target region of interest in the instrument image, the server performs numerical identification on the target region of interest, for example, the numerical feature extraction can be performed on the target region of interest, numerical matching is performed based on the extracted numerical features, and a numerical identification result of the instrument image is obtained according to the numerical matching result, so that reading processing of the numerical instrument is realized.

In the instrument image processing method, the area analysis is carried out on an instrument image obtained by shooting aiming at a digital instrument to obtain each interested area in the instrument image, the distribution characteristics of matched characteristic points in each interested area are determined by carrying out characteristic matching on the instrument image and a standard characteristic image corresponding to the digital instrument, a target interested area is determined from each interested area, and the numerical value identification result of the instrument image is obtained by carrying out numerical value identification on the target interested area. The target interesting regions are determined through the distribution characteristics of the matching characteristic points determined in the instrument image in the interesting regions, and the target interesting regions are subjected to numerical value identification, so that the regions containing numerical values are accurately determined in the instrument image for numerical value identification, the data volume of numerical value identification processing can be reduced on the premise of ensuring the pertinence of numerical value identification, and the processing efficiency of digital instrument reading is effectively improved.

Wherein a channel is a grayscale image of a color image, which consists of only one of the primary colors constituting the color image. For example, for an RGB color image, R, G and B three channels are included, respectively, each channel corresponding to a channel image. The channel image may include an image of the meter image corresponding to a different channel, and the channel image may also include an image formed by combining different channel components. For example, the channel image may include R, G channel images of three channels B, may also include channel images obtained by taking 50% components of the R channel and the G channel, respectively, and may also include channel images formed by other channel combination relations. The channel combination relationship can be set according to actual needs, for example, different channel combination relationships are set for different shooting environments of the instrument image, so that various types of channel images are obtained, and accuracy of area analysis is ensured.

The channel binary image is an image obtained by performing binarization processing on the channel image, and the channel binary image only includes black and white colors, that is, the gray value of any pixel point in the channel binary image is 0 or 255, and represents black and white respectively. The contours reflect edges in the image that are a reaction to discontinuities in local features of the image (abrupt changes in gray level, abrupt changes in color, etc.) that mark the end of one region and the beginning of another. The image can be divided into regions by the contour.

Specifically, when the instrument image is subjected to area analysis, the server performs channel separation on the instrument image to obtain channel images of the instrument image corresponding to the channels, further, the server can fuse the channel images corresponding to the individual channels according to a preset channel combination relationship to form corresponding fusion channel images, and the channel images corresponding to the instrument image are obtained according to the fusion channel images and the individual channel images. The server performs binarization processing on each channel image respectively, and specifically, the binarization processing can be performed on each channel image through a binarization algorithm, such as an OTSU (maximum inter-class variance) algorithm, a Bernsen binarization algorithm, a block analysis-based binarization algorithm or a cyclic threshold algorithm, so as to obtain each channel binary image.

Further, the server performs contour extraction on each channel binary image, for example, the contour extraction may be performed on each channel binary image by a method of hollowing out interior points, so as to obtain a contour in each channel binary image. The number of the contours in each channel binary image is related according to practical application, and a general contour comprises a plurality of contours. And the server performs region construction based on the contour in the binary image of each channel, so that a region of interest is further constructed based on the contour.

In this embodiment, the server performs binarization processing on the channel images of the instrument image corresponding to the channels, performs contour extraction on the obtained binary images of the channels, and performs region construction based on the contours in the binary images of the channels, thereby determining regions of interest in the instrument image, and accurately segmenting the instrument image to ensure the processing efficiency of the instrument image.

In one embodiment, in consideration of instrument images captured in different capturing environments, if there is a channel image suitable for instrument image processing, a mapping relationship between the capturing environment and the channel image may be pre-constructed, so that the capturing environment of the instrument image is used to determine the corresponding channel image for subsequent instrument image processing. Specifically, after obtaining the instrument image, the server may perform shooting environment analysis on the instrument image, for example, the shooting environment analysis may be performed on the instrument image through a pre-trained neural network model, so as to obtain a shooting environment corresponding to the instrument image. After the shooting environment corresponding to the instrument image is determined, the server inquires a mapping relation between the shooting environment and the channel image which are constructed in advance, and determines a target channel image corresponding to the instrument image based on the shooting environment corresponding to the instrument image and the mapping relation, so that the channel image corresponding to the instrument image to be acquired is determined, and subsequent instrument image processing is performed on the channel image. Through the shooting environment analysis of the instrument image, the target channel image is determined according to the shooting environment corresponding to the instrument image, the instrument image processing is carried out based on the target channel image, the channels can be screened according to the shooting environment corresponding to the instrument image, so that various channel images of the instrument image are prevented from being processed respectively, the data volume in the instrument image processing can be further reduced, and the processing efficiency of the instrument image is improved.

The contour deformation processing may be contour enhancement processing performed based on the first contour extraction result, for example, contour widening processing, contour narrowing processing, and the like may be sequentially performed on the first contour extraction result, so that the contour extraction effect is improved, and an accurate contour is obtained.

Specifically, when the server extracts the contours of the channel binary images, the server respectively performs first contour extraction on each channel binary image to obtain a first contour extraction result, wherein the first contour extraction is direct contour extraction based on the channel binary images, and the first contour extraction result includes contours obtained by directly performing contour extraction on the channel binary images. Further, the server performs contour deformation processing on the first contour extraction result, for example, contour widening processing and contour narrowing processing may be sequentially performed on each contour in the first contour extraction result, so as to perform contour enhancement on each contour in the first contour extraction result, and obtain a contour deformation result. And after the contour deformation result is obtained, the server performs second contour extraction based on the contour deformation result, namely, the contour extraction is continued based on the contour deformation result after the contour enhancement to obtain a second contour extraction result, and the server obtains the contour in the binary image of each channel according to the second contour extraction result, for example, the second contour extraction result can be directly used as the contour in the binary image of each channel.

In this embodiment, after the first contour extraction is performed on the channel binary image, the contour deformation processing is performed on the basis of the first contour extraction result of the first contour extraction, and the second contour extraction is further performed on the contour deformation result that is subjected to the contour deformation processing and then is subjected to the contour enhancement, so that the contour in the channel binary image is obtained, and therefore, through the two times of contour extraction and contour enhancement, the contour extraction effect can be effectively improved, and the accuracy of the contour is ensured.

The region fitting is to perform fitting processing on the contour in the binary image of each channel to fit a corresponding region, specifically, fitting may be performed as polygon fitting, so as to fit a polygon corresponding to the contour, for example, fitting a quadrangle corresponding to the contour. The region fitting can specifically perform fitting processing on the contour in the binary image of each channel through various fitting algorithms such as a polygon fitting algorithm and an iterative endpoint fitting algorithm. The contour reconstruction refers to contour reconstruction of the contour in the binary image of each channel, and specifically, the contour reconstruction of the contour in the binary image of each channel can be performed through a contour reconstruction algorithm, such as a shortest diagonal parallel contour reconstruction algorithm.

Specifically, when the server constructs a region based on the contour in each channel binary image, the server may perform region fitting based on the contour in each channel binary image to obtain a first region of interest; on the other hand, the server carries out contour reconstruction on the contour in the binary image of each channel through a contour reconstruction algorithm to obtain a second region of interest. The first interested area and the second interested area are obtained through different area construction modes, and the result of area construction is enriched. After the first region of interest and the second region of interest are obtained, the server obtains the regions of interest in the instrument image according to the first region of interest and the second region of interest, for example, the first region of interest and the second region of interest can be directly used as the regions of interest in the instrument image, and the first region of interest and the second region of interest can be further screened, so that a region construction result which does not meet the requirements of the regions of interest is filtered from the first region of interest and the second region of interest, and the regions of interest in the instrument image are obtained.

In the embodiment, the contour in the channel binary image is respectively subjected to region construction through two modes of region fitting and contour reconstruction, and each region of interest in the instrument image is obtained according to the obtained first region of interest and the second region of interest, so that the region construction result is enriched, the region construction accuracy is ensured, and the processing efficiency of the instrument image is improved.

Wherein the number of the characteristic points is the number of the matched characteristic points in the area range of the region of interest. The area of the region is the size of the area of the region of interest, and can be specifically determined according to the size of pixels of the region of interest. The feature point proportion can be obtained according to the ratio of the number of feature points of the region of interest to the corresponding area, and reflects the number of matched feature points in the unit area range of the region of interest. The higher the numerical value of the feature point ratio is, the denser the matched feature points in the region of interest are, that is, the stronger the relevance between the region of interest and the display panel of the digital instrument is, the more accurate the obtained data is when the numerical identification is performed based on the region of interest. The target interesting area is an image area corresponding to a display panel of the digital instrument in the instrument image, namely the target interesting area comprises measurement data of the digital instrument and does not carry other environment backgrounds, the target interesting area is subjected to numerical identification, and accurate reading of the digital instrument can be realized.

Specifically, when the target region of interest is screened out from each region of interest, the server performs statistics on the matching feature points in each region of interest to obtain the number of feature points of the matching feature points included in each region of interest. The server determines the area corresponding to each region of interest, for example, the area corresponding to each region of interest can be obtained according to the pixel range of each region of interest. The server obtains the feature point proportion corresponding to each region of interest according to the number of the feature points and the area of the corresponding region of interest, and specifically, the feature point proportion corresponding to the corresponding region of interest can be obtained according to the ratio of the number of the feature points to the area of the corresponding region of interest. And after the feature point proportion of each interested area is obtained, the server determines a target interested area from each interested area based on the feature point proportion. In specific implementation, the server may sort the feature point ratios, and determine the region of interest with the largest feature point ratio value as the target region of interest. In addition, the area of the target region of interest may be defined, that is, the area of the target region of interest needs to be larger than the area threshold value, so as to ensure that the range of the target region of interest can cover the display panel of the digital instrument. Specifically, the server may query a preset region area threshold, and screen each region of interest based on the feature point ratio and the region area threshold, so as to determine a target region of interest from each region of interest, for example, a region of interest with a highest feature point ratio may be determined as the target region of interest, where the region area is greater than the region area threshold.

In the embodiment, the feature point ratio is determined according to the number of the feature points of the matching feature points covered in each region of interest and the area, and each region of interest is screened based on the feature point ratio to determine the target region of interest, so that each region of interest is screened by using the matching feature points, the accuracy of the target region of interest is ensured, the accuracy of the instrument image processing is improved, and the accuracy of the digital instrument reading is ensured.

The numerical type is the type of data to be identified in the digital instrument, and different digital instruments have different numerical types. For example, for a multifunctional blood pressure meter, the measured data of systolic pressure, diastolic pressure and heart rate can be displayed at the same time, i.e. the numerical type of the numerical value to be identified in the digital meter includes systolic pressure, diastolic pressure and heart rate. In the digital instrument, different types of data are displayed in different areas, and the formats of the data displayed in the different areas may also be different, such as different font sizes, font colors, font formats, and the like. Correspondingly, in the target region of interest, different types of data are displayed in different sub-regions, and the target region of interest is subjected to numerical region division, so that the sub-regions corresponding to various data are divided, and the accuracy of numerical identification is improved for numerical identification.

Specifically, when the target region of interest is subjected to numerical identification, the server determines the numerical type of a numerical value to be identified in the digital instrument, the numerical type of the numerical value to be identified corresponds to the digital instrument, different digital instruments have different numerical types, and the numerical type is displayed in the corresponding region in a corresponding preset display mode. The server divides the target region of interest into numerical regions based on the numerical types to obtain sub-regions corresponding to the numerical types respectively, so that the association between the numerical types and the sub-regions is realized. Further, the server respectively identifies the numerical values of the sub-regions to obtain data in each sub-region, the server integrates the data in each sub-region to obtain a numerical value identification result of the instrument image, and the numerical value identification result comprises an identification result corresponding to each numerical value type in the digital instrument, so that comprehensive and accurate identification processing of the digital instrument is realized.

In the embodiment, the target region of interest is subjected to numerical region division according to the numerical type of the value to be identified in the digital instrument, and the value identification is respectively carried out on the basis of each sub-region obtained by division, so that the numerical identification is carried out on the target region of interest by combining the distribution of various numerical types in the digital instrument, and the accuracy of the numerical identification result is improved.

In one embodiment, as shown in fig. 3, the processing of obtaining the numerical recognition result, that is, performing numerical recognition on each sub-area to obtain the numerical recognition result of the meter image, includes:

and 302, performing contour enhancement processing on each sub-region to obtain contour enhancement results respectively corresponding to each sub-region.

The contour enhancement processing can improve the extraction effect of contour extraction on the sub-regions, and specifically can perform contour widening, contour narrowing and the like on the sub-regions in sequence to obtain contour enhancement results respectively corresponding to the sub-regions. In specific application, after the server obtains the sub-regions corresponding to the numerical types, the server may perform graying processing on the sub-regions, perform binarization processing based on a graying processing result, thereby obtaining a sub-region binary image, and perform contour enhancement processing according to the sub-region binary image, thereby obtaining the sub-region numerical contour corresponding to each sub-region. In other implementations, the server may also directly perform graying processing on the target region of interest after determining the target region of interest from each region of interest, perform numerical region division based on the grayed target region of interest to obtain sub-regions corresponding to each numerical type, and further perform binarization processing on each sub-region, and perform contour enhancement processing according to each sub-region after the binarization processing to obtain contour enhancement results corresponding to each sub-region.

And 304, extracting the contour based on the contour enhancement result to obtain a numerical contour of the sub-region corresponding to each sub-region.

After the contour enhancement results corresponding to the sub-regions are obtained, the server extracts the contour based on the contour enhancement results, for example, the contour extraction can be performed on the contour enhancement results by a method of hollowing out the inner points, so as to obtain the numerical contour of the sub-region corresponding to the sub-region.

And step 306, matching the sub-region numerical value outline with the numerical value mapping characteristics in the numerical value mapping table, and determining the sub-region numerical value corresponding to the sub-region numerical value outline according to the matching result.

After the numerical value profiles of the sub-regions corresponding to the sub-regions are obtained, the server obtains a numerical value mapping table, the numerical value mapping table comprises numerical value mapping characteristics, and the numerical value mapping characteristics represent profile characteristics corresponding to various numerical values. The server matches the sub-region numerical value outline with the numerical value mapping characteristics in the numerical value mapping table to obtain a matching result, and the server determines the sub-region numerical value corresponding to the sub-region numerical value outline according to the matching result. Specifically, the server may determine a value represented by the value mapping feature matched and consistent with the sub-region value profile in the value mapping table as the sub-region value corresponding to the sub-region value profile, thereby implementing the value identification processing on the sub-region.

And 308, obtaining a numerical value identification result of the instrument image according to the sub-region numerical values corresponding to the sub-region numerical value profiles.

And obtaining a sub-region numerical value corresponding to each sub-region numerical value outline, namely obtaining a numerical value identification result of each sub-region, and obtaining a numerical value identification result of the instrument image by the server according to the sub-region numerical value corresponding to each sub-region numerical value outline. For example, the server may associate the sub-region numerical value corresponding to each sub-region numerical value profile with the corresponding numerical value type, so as to obtain a numerical value identification result of the meter image.

In this embodiment, contour extraction is performed after contour enhancement processing is performed on each sub-region, so that the contour extraction effect of each sub-region can be ensured, the obtained numerical value contour of each sub-region is matched with the numerical value mapping feature in the numerical value mapping table, so as to obtain the numerical value of each sub-region, and the numerical value recognition result of the instrument image is obtained by integrating the numerical values of the sub-regions, so that accurate recognition of each type of data in the instrument image is realized.

In one embodiment, a meter image processing method is provided, which is illustrated by applying the method to the server in fig. 1 as an example. In this embodiment, as shown in fig. 4, the digital instrument is a sphygmomanometer, three types of measurement data, namely systolic pressure, diastolic pressure and pulse rate, are displayed on a display panel of the sphygmomanometer, and the three types of measurement data displayed on the sphygmomanometer need to be identified.

Specifically, the user performs blood pressure measurement through the sphygmomanometer, and after the sphygmomanometer displays measurement data, the user photographs the sphygmomanometer through the terminal to obtain an instrument image corresponding to the sphygmomanometer. The terminal sends the instrument image to the server, and after the server receives the instrument image, the server extracts the gray-scale map of each channel of the RGB instrument image and the gray-scale map of 50% of each RG component, and increases the contrast. As shown in fig. 5, the instrument image shown in fig. 4 corresponds to a channel image of a G channel; as shown in fig. 6, the meter image shown in fig. 4 corresponds to a channel image of the B channel. And after the channel images are obtained, the server carries out binarization processing on the channel images to obtain binary images corresponding to the channel images. As shown in fig. 7, the channel binary image is obtained by performing binarization processing on the channel image of the B channel shown in fig. 6. And after the binary image of each channel is obtained, the server extracts the contour of the binary image of each channel to obtain the contour in the binary image of each channel. As shown in fig. 8, the contour extraction result obtained after contour extraction is performed on the channel binary image shown in fig. 7 is obtained. And the server constructs regions based on the contours in the binary images of the channels to obtain regions of interest in the instrument images. Specifically, on one hand, the server performs region fitting based on the contour in the binary image of each channel to obtain a first region of interest; on the other hand, the server carries out contour reconstruction on the contour in the binary image of each channel through a contour reconstruction algorithm to obtain a second region of interest; further, the server obtains each region of interest in the meter image according to the first region of interest and the second region of interest, where each region of interest in this embodiment is a quadrilateral region.

Further, the server performs feature matching on the instrument image and the standard feature image corresponding to the digital instrument, and determines matching feature points in the instrument image. Specifically, the server can perform feature matching on the instrument image through an SIFT algorithm to obtain matching feature points in the instrument image. After the matching feature points are obtained, the server counts the number of feature points of the matching feature points included in the interested areas respectively, obtains the feature point occupation ratio corresponding to each interested area according to the ratio of the number of the feature points to the area of the corresponding interested area, and determines the interested area with the largest feature point occupation ratio as the target interested area from each interested area. And after the target interesting region is obtained, the server extracts and intercepts the target interesting region from the instrument image, and performs graying processing on the target interesting region to obtain the grayed target interesting region. As shown in fig. 9, which is an embodiment, the object region of interest after the graying process. And the server performs numerical region division on the target region of interest based on the numerical type of the numerical value to be identified in the digital instrument to obtain sub-regions corresponding to the numerical types respectively. Specifically, the server may perform numerical region division on the target region of interest according to three numerical types, such as systolic pressure, diastolic pressure, pulse rate, of the numerical value to be identified in the digital instrument, separate foreground and background scenes in the target region of interest, and perform clustering processing to obtain circumscribed rectangles corresponding to various numerical types, that is, sub-regions corresponding to each numerical type respectively. As shown in fig. 10, the sub-regions corresponding to the systolic pressure SYS obtained by dividing the target region of interest shown in fig. 9 are shown. Further, the server performs contour enhancement processing on each sub-region, and performs contour extraction based on each obtained contour enhancement result to obtain a sub-region numerical contour corresponding to each sub-region. As shown in fig. 11, a sub-region numerical contour is obtained by contour extraction for the sub-region corresponding to the systolic pressure SYS shown in fig. 10.

And after the numerical value contour of each sub-region is obtained, the server matches the numerical value contour of the sub-region with the numerical value mapping characteristics in the numerical value mapping table, and determines to obtain the numerical value of the sub-region corresponding to the numerical value contour of the sub-region according to the matching result, so as to obtain the numerical value identification result of the instrument image. Specifically, the brightness of 8 sections of nixie tubes in the numerical value profile of each sub-region is judged, corresponding numbers are calculated according to the brightness, and the corresponding numbers can be obtained through inquiring a mapping table from the display condition of the nixie tubes to the numbers, so that accurate and efficient reading of the digital instrument is realized.

It should be understood that although the steps in the flowcharts of fig. 2 and 3 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2 and 3 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.

In one embodiment, as shown in fig. 12, there is provided a meter image processing apparatus 1200 including: a meter image acquisition module 1202, a region of interest determination module 1204, a feature matching module 1206, a target region determination module 1208, and a numerical identification processing module 1210, wherein:

a meter image acquisition module 1202, configured to acquire a meter image obtained by shooting a digital meter;

an interested region determining module 1204, configured to perform region analysis on the instrument image to obtain each interested region in the instrument image;

the feature matching module 1206 is used for performing feature matching on the instrument image and a standard feature image corresponding to the digital instrument to determine a matching feature point in the instrument image;

a target region determining module 1208, configured to determine a target region of interest from each region of interest based on the distribution features of the matching feature points in each region of interest;

and the numerical value identification processing module 1210 is used for performing numerical value identification on the target region of interest to obtain a numerical value identification result of the instrument image.

In one embodiment, the region of interest determining module 1204 includes a binarization processing module, a contour extraction module, and a region construction module; wherein: the binarization processing module is used for respectively carrying out binarization processing on the channel images of the channels corresponding to the instrument image to obtain binary images of the channels; the contour extraction module is used for respectively extracting contours of the binary images of the channels to obtain the contours in the binary images of the channels; and the region construction module is used for constructing regions based on the contours in the binary images of the channels to obtain the regions of interest in the instrument image.

In one embodiment, the contour extraction module comprises a first extraction module, a deformation processing module and a second extraction module; wherein: the first extraction module is used for respectively carrying out first contour extraction on the binary images of each channel to obtain a first contour extraction result; the deformation processing module is used for carrying out contour deformation processing on the first contour extraction result to obtain a contour deformation result; and the second extraction module is used for carrying out second contour extraction based on the contour deformation result to obtain the contour in the binary image of each channel.

In one embodiment, the region construction module comprises a region fitting module, a contour reconstruction module, and a region of interest acquisition module; wherein: the region fitting module is used for performing region fitting on the basis of the contour in the binary image of each channel to obtain a first region of interest; the contour reconstruction module is used for carrying out contour reconstruction on the contour in the binary image of each channel through a contour reconstruction algorithm to obtain a second region of interest; and the interested region obtaining module is used for obtaining each interested region in the instrument image according to the first interested region and the second interested region.

In one embodiment, the target region determination module 1208 includes a feature point statistics module, a proportion determination module, and a target region determination module; wherein: the characteristic point counting module is used for counting the number of characteristic points of the matched characteristic points respectively included in the region of interest; the proportion determining module is used for obtaining the proportion of the characteristic points corresponding to each interested area according to the number of the characteristic points and the area of the corresponding interested area; and the target area determining module is used for determining a target interested area from each interested area based on the feature point ratio.

In one embodiment, the numerical identification processing module 1210 includes a numerical type determination module, a numerical region division module, and a sub-region identification module; wherein: the numerical type determining module is used for determining the numerical type of the numerical value to be identified in the digital instrument; the numerical region division module is used for carrying out numerical region division on the target interesting region based on the numerical type to obtain sub-regions respectively corresponding to the numerical types; and the sub-region identification module is used for respectively carrying out numerical value identification on each sub-region to obtain a numerical value identification result of the instrument image.

In one embodiment, the sub-region identification module comprises an enhancement processing module, a sub-region outline module, a feature matching module and a numerical synthesis module; wherein: the enhancement processing module is used for carrying out contour enhancement processing on each subregion to obtain contour enhancement results respectively corresponding to each subregion; the sub-region outline module is used for extracting outlines based on the outline enhancement results to obtain sub-region numerical value outlines corresponding to the sub-regions; the characteristic matching module is used for matching the numerical value profile of the sub-region with the numerical value mapping characteristics in the numerical value mapping table and determining the numerical value of the sub-region corresponding to the numerical value profile of the sub-region according to the matching result; and the numerical value synthesis module is used for obtaining the numerical value identification result of the instrument image according to the sub-region numerical values corresponding to the sub-region numerical value profiles.

For specific limitations of the instrument image processing device, reference may be made to the above limitations of the instrument image processing method, which are not described herein again. The respective modules in the above-described meter image processing apparatus may be wholly or partially implemented by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 13. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a meter image processing method.

Those skilled in the art will appreciate that the architecture shown in fig. 13 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

acquiring an instrument image shot by a digital instrument;

In one embodiment, the processor, when executing the computer program, further performs the steps of: respectively carrying out binarization processing on the channel images of the instrument image corresponding to the channels to obtain binary images of the channels; respectively extracting the contour of each channel binary image to obtain the contour of each channel binary image; and constructing regions based on the contours in the binary images of the channels to obtain regions of interest in the instrument image.

In one embodiment, the processor, when executing the computer program, further performs the steps of: respectively carrying out first contour extraction on the binary images of each channel to obtain a first contour extraction result; carrying out contour deformation processing on the first contour extraction result to obtain a contour deformation result; and performing second contour extraction based on the contour deformation result to obtain the contour in the binary image of each channel.

In one embodiment, the processor, when executing the computer program, further performs the steps of: performing region fitting based on the contour in the binary image of each channel to obtain a first region of interest; carrying out contour reconstruction on the contour in the binary image of each channel through a contour reconstruction algorithm to obtain a second region of interest; and obtaining each interested area in the instrument image according to the first interested area and the second interested area.

In one embodiment, the processor, when executing the computer program, further performs the steps of: counting the number of feature points of the matched feature points respectively included in the region of interest; obtaining the feature point occupation ratio corresponding to each interested area according to the number of the feature points and the area of the corresponding interested area; and determining a target region of interest from the regions of interest based on the feature point ratio.

In one embodiment, the processor, when executing the computer program, further performs the steps of: determining the numerical type of a numerical value to be identified in the digital instrument; performing numerical region division on the target region of interest based on the numerical type to obtain sub-regions corresponding to the numerical types respectively; and respectively carrying out numerical value identification on each subarea to obtain a numerical value identification result of the instrument image.

In one embodiment, the processor, when executing the computer program, further performs the steps of: carrying out contour enhancement processing on each subregion to obtain contour enhancement results respectively corresponding to each subregion; extracting the contour based on each contour enhancement result to obtain a sub-region numerical contour corresponding to each sub-region; matching the sub-region numerical value outline with the numerical value mapping characteristics in the numerical value mapping table, and determining the sub-region numerical value corresponding to the sub-region numerical value outline according to the matching result; and obtaining a numerical value identification result of the instrument image according to the numerical value of the sub-region corresponding to the numerical value profile of each sub-region.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

acquiring an instrument image shot by a digital instrument;

In one embodiment, the computer program when executed by the processor further performs the steps of: respectively carrying out binarization processing on the channel images of the instrument image corresponding to the channels to obtain binary images of the channels; respectively extracting the contour of each channel binary image to obtain the contour of each channel binary image; and constructing regions based on the contours in the binary images of the channels to obtain regions of interest in the instrument image.

In one embodiment, the computer program when executed by the processor further performs the steps of: respectively carrying out first contour extraction on the binary images of each channel to obtain a first contour extraction result; carrying out contour deformation processing on the first contour extraction result to obtain a contour deformation result; and performing second contour extraction based on the contour deformation result to obtain the contour in the binary image of each channel.

In one embodiment, the computer program when executed by the processor further performs the steps of: performing region fitting based on the contour in the binary image of each channel to obtain a first region of interest; carrying out contour reconstruction on the contour in the binary image of each channel through a contour reconstruction algorithm to obtain a second region of interest; and obtaining each interested area in the instrument image according to the first interested area and the second interested area.

In one embodiment, the computer program when executed by the processor further performs the steps of: counting the number of feature points of the matched feature points respectively included in the region of interest; obtaining the feature point occupation ratio corresponding to each interested area according to the number of the feature points and the area of the corresponding interested area; and determining a target region of interest from the regions of interest based on the feature point ratio.

In one embodiment, the computer program when executed by the processor further performs the steps of: determining the numerical type of a numerical value to be identified in the digital instrument; performing numerical region division on the target region of interest based on the numerical type to obtain sub-regions corresponding to the numerical types respectively; and respectively carrying out numerical value identification on each subarea to obtain a numerical value identification result of the instrument image.

In one embodiment, the computer program when executed by the processor further performs the steps of: carrying out contour enhancement processing on each subregion to obtain contour enhancement results respectively corresponding to each subregion; extracting the contour based on each contour enhancement result to obtain a sub-region numerical contour corresponding to each sub-region; matching the sub-region numerical value outline with the numerical value mapping characteristics in the numerical value mapping table, and determining the sub-region numerical value corresponding to the sub-region numerical value outline according to the matching result; and obtaining a numerical value identification result of the instrument image according to the numerical value of the sub-region corresponding to the numerical value profile of each sub-region.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A meter image processing method, characterized in that the method comprises:

acquiring an instrument image shot by a digital instrument;

performing feature matching on the instrument image and a standard feature image corresponding to the digital instrument to determine a matching feature point in the instrument image;

2. The method of claim 1, wherein the performing a region analysis on the meter image to obtain regions of interest in the meter image comprises:

respectively carrying out binarization processing on the channel images of the instrument image corresponding to the channels to obtain binary images of the channels;

respectively extracting the contour of each channel binary image to obtain the contour of each channel binary image;

and constructing regions based on the contours in the channel binary images to obtain regions of interest in the instrument image.

3. The method according to claim 2, wherein the performing contour extraction on each channel binary image to obtain a contour in each channel binary image comprises:

respectively carrying out first contour extraction on each channel binary image to obtain a first contour extraction result;

carrying out contour deformation processing on the first contour extraction result to obtain a contour deformation result;

and performing second contour extraction based on the contour deformation result to obtain a contour in each channel binary image.

4. The method of claim 2, wherein the region construction based on the contour in each channel binary image to obtain each region of interest in the instrument image comprises:

performing region fitting on the basis of the contour in each channel binary image to obtain a first region of interest;

carrying out contour reconstruction on the contour in each channel binary image through a contour reconstruction algorithm to obtain a second region of interest;

and obtaining each interested area in the instrument image according to the first interested area and the second interested area.

5. The method according to claim 1, wherein the determining a target region of interest from each of the regions of interest based on the distribution feature of the matched feature points in each of the regions of interest comprises:

counting the number of feature points of the matching feature points each included in the region of interest;

obtaining the feature point occupation ratio corresponding to each interested area according to the number of the feature points and the area of the corresponding interested area;

and determining a target region of interest from each region of interest based on the feature point occupation ratio.

6. The method according to any one of claims 1 to 5, wherein the performing numerical identification on the target region of interest to obtain a numerical identification result of the meter image comprises:

determining the numerical type of the numerical value to be identified in the digital instrument;

performing numerical region division on the target region of interest based on the numerical type to obtain sub-regions corresponding to the numerical types respectively;

and respectively carrying out numerical value identification on each sub-region to obtain a numerical value identification result of the instrument image.

7. The method according to claim 6, wherein the performing numerical identification on each of the sub-regions to obtain a numerical identification result of the instrument image comprises:

performing contour enhancement processing on each sub-region to obtain contour enhancement results corresponding to each sub-region;

extracting the contour based on each contour enhancement result to obtain a numerical contour of each sub-region corresponding to each sub-region;

matching the sub-region numerical value outline with numerical value mapping characteristics in a numerical value mapping table, and determining a sub-region numerical value corresponding to the sub-region numerical value outline according to a matching result;

and obtaining a numerical value identification result of the instrument image according to the sub-region numerical values corresponding to the sub-region numerical value profiles.

8. A meter image processing apparatus, characterized in that the apparatus comprises:

a target region determining module, configured to determine a target region of interest from each of the regions of interest based on a distribution feature of the matching feature points in each of the regions of interest;

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.