CN110191281A - Image recognition meter reading terminal and system, image recognition engine server, intelligent meter - Google Patents

Image recognition meter reading terminal and system, image recognition engine server, intelligent meter Download PDF

Info

Publication number
CN110191281A
CN110191281A CN201910470446.3A CN201910470446A CN110191281A CN 110191281 A CN110191281 A CN 110191281A CN 201910470446 A CN201910470446 A CN 201910470446A CN 110191281 A CN110191281 A CN 110191281A
Authority
CN
China
Prior art keywords
image
meter reading
image recognition
compression
reading terminal
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
Application number
CN201910470446.3A
Other languages
Chinese (zh)
Inventor
赵剑开
戴敦
蔡江涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuxi Zhuanxinzhizhi Technology Co ltd
Original Assignee
Beijing Annuoxin Polytron Technologies Inc
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Annuoxin Polytron Technologies Inc filed Critical Beijing Annuoxin Polytron Technologies Inc
Priority to CN201910470446.3A priority Critical patent/CN110191281A/en
Publication of CN110191281A publication Critical patent/CN110191281A/en
Pending legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • H04N23/65Control of camera operation in relation to power supply
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • H04N23/66Remote control of cameras or camera parts, e.g. by remote control devices

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Arrangements For Transmission Of Measured Signals (AREA)

Abstract

The application provides image recognition meter reading terminal and system, image recognition engine server, intelligent meter.Described image identifies that meter reading terminal includes: camera arrangement, shoots the image of meter;Image processing apparatus determines the fringe region of described image, carries out pixel compression to the fringe region according to predetermined reduced rule and generates compression image;First transceiver sends the compression image to data platform and carries out image restoring and application.

Description

Image recognition meter reading terminal and system, image recognition engine server, intelligent meter
Technical field
This application involves smart meter reading technical fields, and in particular to a kind of image recognition meter reading terminal and system, image are known Other engine server, intelligent meter.
Background technique
Implemented in full with the rapid development of our country's economy with " Every household has an ammeter " engineering policy, town dweller to water, electricity, The demand of gas resource is increasing.The meter reading method generallyd use at present has manual metering, remote automatic meter reading, or uses IC The mode of card pre-payment.
Inventors have found that all there is drawback in these meter reading methods, especially in some large enterprises, multipoint application scene, Its demand that some managers are unable to satisfy in terms of real-time, accuracy and integrated management.
Manual metering low efficiency, it is at high cost, it is easy error.IC card pre-payment solves the problems, such as manual metering, but user's body Degree of testing is lower, and particular for large enterprise, the scene of multipoint application scene, management difficulty is big.Remote automatic meter reading is most at present It using the modes such as carrier wave, micropower be wireless, is difficult open to big by water, electricity, gas resource management enterprise's unified monitoring and management Type enterprise, multipoint application scene scene.
The scheme of data transmission at present concentrates on GPRS, NB-Iot, 2G network these types.GPRS and 2G network is for communication For base station, can not largely intensive use, access capability while communication base station Non-precondition is enough, and flow cost is high, Power consumption is big, and battery life is lower, and maintenance cost is high, but occupies in its transmission rate, transmission data length and network coverage situation Advantage, image data can subpackage can upload completion several times;And NB-Iot is a kind of specifically for a large amount of intensive uses Internet of Things communication mode, flow cost is low, and access capability is 10 times of 2G network, but its traffic rate is low, small power consumption, electricity Pond is longer using the time, and the later period reduces maintenance cost, and the data length transmitted every time is short, needs image data subpackage tens times It uploads, and repeatedly uploads certain packet loss, influence the success rate that meter reading terminal completes identification, NB network is currently in development Stage, also not up to all standing, respective regions do not have NB signal.
On image processing problem, because the limitation of the transmission of meter reading terminal cost itself, data and power supply aspect, is generally being copied Picture processing chip is not used in meter terminal, is only relied upon camera itself and is carried out image procossing, the processor of meter reading terminal will Image is converted into data flow and is transmitted, and guarantees that image is undistorted, but has pressure to data transmission.
Summary of the invention
The embodiment of the present application provides a kind of image recognition meter reading terminal, including camera arrangement, image processing apparatus, first Transceiver, the image of the camera arrangement shooting meter;Described image processing unit determines the fringe region of described image, according to Predetermined reduced rule carries out pixel compression to the fringe region and generates compression image;The first transceiver sends the compression Image carries out image restoring and application to data platform.
As the one aspect of the application, described image processing unit includes main view determining module, fringe region determination Module, compression module, size that the main view determining module determines the main view of described image and the main view are described Position in image;Size and the position of the fringe region determining module based on the main view, determine described image Described in region except main view be the fringe region;The compression module is according to predetermined reduced rule to the marginal zone Domain carries out pixel compression and generates compression image.
As the one aspect of the application, the main view determining module includes main view dimension acquisition module, main view Position determination module, the main view dimension acquisition module obtain the institute from the data platform by the first transceiver State the size of main view;The main view position determination module is obtained by the first transceiver and comes from the data platform , described image and the main view in the pixel distance at horizontal and vertical edge, and determine the main view in the figure Position as in.
As the one aspect of the application, the compression module includes pixel array determination unit, compression unit, the picture Pixel array determination unit determines the pixel array that all pixels point forms in described image;The compression unit is according to predetermined compression Rule deletes the one part of pixel point for the fringe region for including in the pixel array, to generate compression image.
As the one aspect of the application, described image identifies that meter reading terminal further includes main control module, the main control module It controls described image processing unit and the compression image is sent to the first transceiver, also connect by the first transceiver The image recognition result from the data platform is received, the bat of the camera arrangement is adjusted in response to described image recognition result According to resolution ratio, wherein described image recognition result is identified the compression image by image recognition engine server and obtained.
As the one aspect of the application, described image identifies that meter reading terminal further includes manual trigger module or/and timing Trigger module, the manual trigger module trigger the image of the camera arrangement shooting meter manually;The timing trigger module Camera arrangement described in clocked flip shoots the image of meter, and the main control module is with parameter preset or from the data platform The timing of timing trigger module described in parameter setting in timer command.
As the one aspect of the application, it further includes vibration-sensing module, the vibration that described image, which identifies meter reading terminal, Sensing module incudes the vibration of described image identification meter reading terminal, and Xiang Suoshu main control module sends a warning message;The master control mould Block controls the camera arrangement in response to the warning information and carries out taking pictures for the meter, records shock conditions.
As the one aspect of the application, described image identifies that meter reading terminal further includes information acquisition module, the information Acquisition module acquires ambient temperature information, ambient humidity information, and the information is supplied to the main control module.
The embodiment of the present application also provides a kind of image recognition engine server, including second transceiver, recovery module, described Second transceiver receives the compression image from image recognition meter reading terminal by data platform;The recovery module according to institute The corresponding also meta-rule of predetermined reduced rule is stated, the one part of pixel point of the missing of the compression image is restored and is gone back Original image is simultaneously applied.
As the one aspect of the application, described image identifies that engine server further includes measurement module, the measurement mould Block measures the size and described image and pixel distance of the main view at horizontal and vertical edge of the main view, institute State that second transceiver also sends the size of the main view by the data platform and described image and the main view exist The pixel distance at horizontal and vertical edge identifies meter reading terminal to described image.
As the one aspect of the application, described image identifies that engine server further includes identification module, the identification mould The clarity that original image is gone back described in block identification obtains image recognition result, and the second transceiver also passes through the data platform and sends out Described image recognition result is sent to identify meter reading terminal to described image.
The embodiment of the present application also provides a kind of image recognition kilowatt meter reading-out system, whole including image recognition meter reading described above End, data platform, image recognition engine server described above, data platform server, described image identify meter reading terminal The image for shooting meter, determines the fringe region of described image, carries out pixel to the fringe region according to predetermined reduced rule Compression generates compression image, transmits the compression image to data platform;The data platform receives the compression image, transmission The compression image identifies engine server to described image;Described image identifies that engine server receives the compression image, According to also meta-rule corresponding with predetermined reduced rule, the one part of pixel point of the missing of compression image is restored and is gone back Original image goes back original image to data platform server described in transmission;Original image is gone back in the reception of data platform server, record, store, Using going back original image.
The embodiment of the present application also provides a kind of intelligent meter, including meter, image recognition meter reading terminal described above.
The technical solution that embodiments herein proposes, zooms in and out image border, is identified as function not influencing image In the case where rate, reduce the data volume of data transmission, increases the discrimination of the success rate and image data in data transmission.
Detailed description of the invention
In order to more clearly explain the technical solutions in the embodiments of the present application, make required in being described below to embodiment Attached drawing is briefly described, it should be apparent that, the drawings in the following description are only some examples of the present application, for For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other Attached drawing.
Fig. 1 is a kind of composition schematic diagram for image recognition meter reading terminal that one embodiment of the application provides;
Fig. 2 is a kind of schematic diagram for image edge area that one embodiment of the application provides;
Fig. 3 A-3C is a kind of schematic diagram for image scaling that one embodiment of the application provides;
Fig. 4 is a kind of composition schematic diagram for image recognition meter reading terminal that another embodiment of the application provides;
Fig. 5 is a kind of composition schematic diagram for image recognition meter reading terminal that the another embodiment of the application provides;
Fig. 6 is a kind of composition schematic diagram for image recognition meter reading terminal that the application another embodiment provides;
Fig. 7 is a kind of composition schematic diagram for image recognition engine server that one embodiment of the application provides;
Fig. 8 is a kind of composition schematic diagram for image recognition engine server that another embodiment of the application provides;
Fig. 9 is a kind of composition schematic diagram for image recognition kilowatt meter reading-out system that one embodiment of the application provides;
Figure 10 is a kind of composition schematic diagram for image recognition kilowatt meter reading-out system that another embodiment of the application provides;
Figure 11 is a kind of composition schematic diagram for intelligent meter that one embodiment of the application provides.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below with reference to drawings and examples, The specific embodiment progress of technical scheme in further detail, is clearly illustrated.However, specific implementations described below Mode and embodiment are for illustrative purposes only, rather than the limitation to the application.It only contains the implementation of the application a part Example, instead of all the embodiments, the other embodiments that those skilled in the art obtain the various change of the application all belong to In the range of the application protection.
Fig. 1 is a kind of composition schematic diagram for image recognition meter reading terminal that one embodiment of the application provides, including dress of taking pictures Set 11, image processing apparatus 12, first transceiver 13.
The image of the shooting meter of camera arrangement 11.Image processing apparatus 12 determines the fringe region of image, according to pre- level pressure Contraction ga(u)ge then carries out pixel compression to fringe region and generates compression image.First transceiver 13 send compression image to data platform into Row image restoring and application.
Camera arrangement 11 is to provide the main modular for image of taking pictures for image recognition meter reading terminal, including but not limited to camera shooting The components such as head, flash lamp, it can carry out control triggering by timing or manually, after camera arrangement 11 is triggered, dodge opening Light lamp, and start camera, it takes pictures.
On the problem of taking pictures, the high camera of pixel is taken pictures clearly, but the data volume of image is big, is caused to data transmission Pressure, and the low camera of pixel can preferably solve the problems, such as transmission, but the clarity of image is inadequate, be easy to cause use Image recognition engine can not identify or identify difficulty, it is often more important that in terms of review, reviewing officer also differentiates unclear.
The camera arrangement 11 of the present embodiment can choose the camera for supporting highest 640*480 pixel image, can adjust Whole output pixel size, by image recognition engine server verifying and function in terms of demand, such image can expire The requirement of sufficient discrimination.
Image processing apparatus 12 uses a kind of image border compression method, when installing and using for the first time, passes through the first transmitting-receiving Then the image of the transmission shooting of device 13 receives knowing by image from data platform by first transceiver 13 to data platform The position of the size and main view for the image main view that other engine server determines in the picture determines therefrom that in image except main view Fringe region except figure.In the later period in use, if image recognition meter reading terminal installation condition does not change, main view size and The position of main view in the picture is also constant.
Fig. 2 is a kind of schematic diagram for image edge area that one embodiment of the application provides.As shown in Fig. 2, P1 is image, P2 is main view.X is pixel distance of the image left edge to main view left edge, and y is image top edge to main view top edge Pixel distance.The region of P1-P2 is fringe region.
Fringe region is compressed and is restored, is mainly manifested in as shown in Figure 3A, i.e., the matrix represented in (00, mn) In, the data of even column in matrix are rejected, but do not reject the data that region shown in (ab, cd) includes.Obtained data Matrix is content shown in Fig. 3 B.This process belongs to the compression process of image.
The compression image obtained after above-mentioned image procossing is sent to data platform 2 by first transceiver 13, number It is sent to image recognition engine server 3 by image is compressed according to platform 2, pixel will be carried out in image recognition engine server 3 Reduction, this process belong to the reduction process of image.The compressing image data that image recognition engine server 3 will receive, according to Predetermined reduced rule corresponding also meta-rule, the data of the even column really lost to it restore, and the method for reduction is by its left side The data of side are replicated and are filled into even column, and the image finally restored is as shown in Figure 3 C, i.e. the data and the of 2i column The data of 2i-1 column are identical.Specific compression of images and the method for reduction are not limited thereto.
By processing, the valid data amount of image will reduce one third.For first transceiver 13, the frequency of transmission Rate will also reduce one third, to improve the success rate of data transmission, the use of flow also reduces one third, to subtract Use cost is lacked.
This method is based on application scenarios, firstly, image recognition meter reading terminal does not taking cost into account include with image The processing unit of Processing Algorithm does not have complicated image sampling function.Secondly, the operation that complicated image sampling function needs Energy consumption can be allowed to increase, so simple image processing method can only be used.And in image meter reading scheme, the center of image itself Domain, which is actually needed, clearly to be showed, and review is facilitated, and can be used as the region of meter registration evidence, i.e., is defined as this region Main view P2, by this method treated image border, i.e. the region of (P1-P2) can cause a degree of distortion, but first First the degree of its distortion has no effect on identification of the main view P2 of diagram in image recognition engine server 3, ensure that meter The correctness that registration is read by image recognition, secondly, pixel shown in image border belongs to meter other information region, body Show the information such as the dial plate appearance of meter, when manual review only confirms that it does not change to the review in this region, it is believed that with It is same meter when installation, so reducing clarity caused by its distortion is approved in range in industrial application, i.e., not Influence image recognition meter reading terminal use process in systems.
The technical solution that the present embodiment proposes is not influencing image recognition by being compressed and being restored to image border In the case where success rate, reduce the data volume of data transmission, increases the knowledge of the success rate and image data in data transmission Not rate.
Fig. 4 is a kind of composition schematic diagram for image recognition meter reading terminal that another embodiment of the application provides, including takes pictures Device 11, image processing apparatus 12, first transceiver 13.
The image of the shooting meter of camera arrangement 11.Image processing apparatus 12 determines the fringe region of image, according to pre- level pressure Contraction ga(u)ge then carries out pixel compression to fringe region and generates compression image.First transceiver 13 send compression image to data platform into Row image restoring and application.
Image processing apparatus 12 includes main view determining module 121, fringe region determining module 122, compression module 123.
Main view determining module 121 determines the position of the size and main view of the main view of image in the picture.Marginal zone Size and location of the domain determining module 122 based on main view determines that the region in image except main view is fringe region.Compression Module 123 carries out pixel compression according to predetermined reduced rule to fringe region and generates compression image.
Main view determining module 121 includes main view dimension acquisition module 1211, main view position determination module 1212.It is main View dimensions obtain the size that module 1211 obtains the main view from data platform by first transceiver 13.Main view position Determining module 1212 obtains from data platform, image and main view at horizontal and vertical edge by first transceiver 13 Pixel distance, and determine main view position in the picture.
Compression module 123 includes pixel array determination unit 1231, compression unit 1232, pixel array determination unit 1231 Determine the pixel array that all pixels point forms in image.Compression unit 1232 deletes pixel array according to predetermined reduced rule In include fringe region one part of pixel point, to generate compression image.
Fig. 5 is a kind of composition schematic diagram for image recognition meter reading terminal that the another embodiment of the application provides, including takes pictures Device 11, image processing apparatus 12, first transceiver 13, power supply module 14, main control module 15, manual trigger module 16 or/and Timing trigger module 17.
The image of the shooting meter of camera arrangement 11.Image processing apparatus 12 determines the fringe region of image, according to pre- level pressure Contraction ga(u)ge then carries out pixel compression to fringe region and generates compression image.First transceiver 13 send compression image to data platform into Row image restoring and application.Main control module 15 controls image processing apparatus 12 and is sent to first transceiver 13 for image is compressed, and The image recognition result from data platform is received by first transceiver 13, dress of taking pictures is adjusted in response to image recognition result Set 11 resolution ratio of taking pictures, wherein image recognition result is identified compression image by image recognition engine server and obtained.Manually Trigger module 16 triggers the image that camera arrangement 11 shoots meter manually;17 clocked flip camera arrangement 11 of timing trigger module is clapped Take the photograph the image of meter, the parameter setting clocked flip in timer command of the main control module 15 with parameter preset or from data platform The timing of module 17.Power supply module 14 gives each module for power supply of image recognition meter reading terminal.
Since image recognition meter reading terminal is chronically at dormant state, need in meter reading terminal installation using triggering manually Module 16 carries out meter reading terminal starting, sends image by the above process, and allow main control module 15 to clocked flip Module 17 is set, clocked flip image recognition meter reading terminal work, so that meter reading terminal be allowed to be in longtime running state.Eventually The working condition at end is, after clocked flip setting, i.e., by according to the setting of clocked flip, every timing cycle starting is primary, in addition to When starting, other times are dormant state.
When needing to obtain the image that image recognition meter reading terminal is currently taken pictures, then need by manual trigger module 16 into The camera arrangement 11 of row triggering meter reading terminal can satisfy user in this way and want to obtain the demand of present image.
The dormant state of the low-power consumption of image recognition meter reading terminal is by PSM (Power Saving Mode) battery saving mode It is established, selected main control module 15 and first transceiver 13 can configure PSM (Power Saving Mode) power saving first Mode in application process configures it according to the demand of above-mentioned dormant state, that is to say, that only in image recognition When meter reading terminal active upload data, the data of data platform transmission are just received, to reach image recognition meter reading terminal The purpose of long-term low-power consumption work.
It is usually using external power supply and to use 2 kinds of modes of battery in existing scheme.The mode of external power supply is in construction Difficulty is increased, and many scenes can not provide power supply, but can guarantee the continued power of power supply in use process;Use electricity The mode in pond allows difficulty of construction to reduce significantly, but for the battery that security consideration cannot select capacity too big, the small battery of capacity There is limitation for service life.
The moderate rechargeable battery of volume, weight, capacity can be used in power supply module 14 in the present embodiment, and retains outer Portion's power supply interface.2G mould group (second generation mobile communication technical specification) work 3 years or more or NB-IOT mould group can be met (Narrow Band Internet of Things is based on cellular narrowband Internet of Things) work 9 years or more, and provide external Power supply interface, can be with external power supply, or charges the battery, and facilitates the use of plurality of application scenes.
Wherein, alternatively, image processing apparatus 12 includes main view determining module 121, marginal zone Domain determining module 122, compression module 123.It is specific same as the previously described embodiments.
Timing trigger module 17 is to take pictures to provide service for image recognition meter reading terminal timing, includes clock and timing Device, clock obtains clock information to data platform by main control module 15 and first transceiver 13, when synchronous according to clock information The time of clock ensure that the uniformity of clock information.Timer receives the time-of-the-day order from main control module 15, timer internal It is timed, is taken pictures when timing terminates that camera arrangement 11 will be triggered, after taking pictures according to the parameter in time-of-the-day order Timing trigger module 17 will be timed again according to the timing of agreement.
The technical solution that the present embodiment proposes adapts to more application scenarios by meter reading terminal by a variety of powering modes, Extend application range;By the addition of main control module and trigger module, so that terminal is taken pictures, transmission is realized manually or automatically Control is suitble to high volume applications.
Fig. 6 is a kind of composition schematic diagram for image recognition meter reading terminal that the application another embodiment provides, including takes pictures Device 11, image processing apparatus 12, first transceiver 13, power supply module 14, main control module 15, manual trigger module 16 or/and Timing trigger module 17, vibration-sensing module 18, information acquisition module 19.
The image of the shooting meter of camera arrangement 11.Image processing apparatus 12 determines the fringe region of image, to fringe region It carries out pixel compression and generates compression image.First transceiver 13 sends compression image and carries out image restoring to data platform and answer With.Main control module 15 controls image processing apparatus 12 and is sent to first transceiver 13 for image is compressed, and passes through first transceiver 13 receive the image recognition result from data platform, and the resolution of taking pictures of camera arrangement 11 is adjusted in response to image recognition result Rate, wherein image recognition result is identified compression image by image recognition engine server and obtained.Manual trigger module 16 is manually Trigger the image that camera arrangement 11 shoots meter;17 clocked flip camera arrangement 11 of timing trigger module shoots the image of meter, The timing of parameter setting timing trigger module 17 in timer command of the main control module 15 with parameter preset or from data platform. 18 sensed image of vibration-sensing module identifies that meter reading terminal includes but is not limited to vibration caused by being disassembled, move, reequiping, to master Control module 15 sends a warning message;Main control module carries out taking pictures for meter in response to warning information driving camera arrangement 11, records Shock conditions.Information acquisition module 19 acquires ambient temperature information, ambient humidity information, the battery information of power supply module 14, announcement Alert information, and provide information to main control module 15.Power supply module 14 gives each module for power supply of image recognition meter reading terminal.
Main control module 15 is the core master control of image recognition meter reading terminal, the receiving area issued with data for order Reason, can for the first time use when according to preset parameter or by first transceiver 13 receive data platform to timing order Parameter in order is timed timing trigger module 17.Can carry out camera arrangement 11 taking pictures the setting of resolution ratio.It can be with Acquisition of the information acquisition module 19 to current battery charge, environment temperature, ambient humidity information is got after the power-up.It can supervise Control the state of vibration-sensing module 18.It can receive the information for the image procossing that image processing apparatus 12 provides, and pass through the One transceiver 13 is sent to data platform 2.
Vibration-sensing module 18 is the safeguard measure of image recognition meter reading terminal, mainly based on shock sensor Module.In general, when image recognition meter reading terminal be disassembled or image recognition meter reading terminal associated by meter moved, torn open It unloads, reequip lamp and cause the vibration of image recognition meter reading terminal, then vibration-sensing module 18 can be allowed to send to main control module 15 and accused Alert information, and being sent a warning message from first transceiver 13 to data platform 2, data platform 2 then report and alarm information to data Platform Server 4, and warning information can be shown, finally allow administrative staff to understand that image recognition meter reading terminal because of shake It moves and alerts, the meter of terminal or terminal association may be disassembled, moves or be transformed, and administrative staff can be according to letter Breath does corresponding processing.
Particularly, when vibration-sensing module 18 starts, sensed image identifies the vibration of meter reading terminal, to main control module 15 It sends a warning message, main control module 15 controls camera arrangement 11 and carries out upload of taking pictures, to record situation when alerting.
Particularly, when vibration-sensing module 18 is taken pictures using delay, it is to record vibration generation that delay, which is taken pictures, The situation of adjoint generation afterwards, that is, when vibration-sensing module 18 starts, image recognition meter reading terminal and meter are thoughts It answers, there are no being subjected to displacement or changing, delay is taken pictures, and is able to record a period of time after the starting of vibration-sensing module 18, is occurred Situation.
Particularly, vibration-sensing module 18 start after take pictures for a period of time behavior time set, be rule of thumb It is set, this time is too short or too long possibly can not all record truth at that time.
Wherein, alternatively, image processing apparatus 12 includes main view determining module 121, marginal zone Domain determining module 122, compression module 123.It is specific same as the previously described embodiments.
Timing trigger module 17 includes clock, timer.Clock passes through main control module 15 and first transceiver 13 to data Platform obtains clock information, according to the time of clock information synchronised clock.Timer receives the timing from main control module 15 and refers to It enables, according to the parameter setting timing in time-of-the-day order.
The technical solution that the present embodiment proposes, by the addition of vibration-sensing module and information acquisition module, so that terminal Image data can be transmitted correctly when change in location, and acquire more temperature, humidity, battery information, make terminal more intelligence Energyization.
A kind of composition schematic diagram for image recognition engine server that Fig. 7 provides for one embodiment of the application, including second Transceiver 31, recovery module 32.
Second transceiver 31 receives the compression image from image recognition meter reading terminal by data platform 2.Recovery module 32, according to also meta-rule corresponding with predetermined reduced rule, are restored to obtain to the one part of pixel point of the missing of compression image Also original image and applied.
Second transceiver 31 receives the number of compressed images that compressing image data is as shown in Figure 3B, and recovery module 32 will receive According to according to also meta-rule corresponding with predetermined reduced rule, the data of the even column really lost to it are restored, the method for reduction It is that the data on the left of it are replicated and be filled into even column, finally goes back the number that original image is as shown in Figure 3 C, i.e. 2i is arranged It is identical according to the data arranged with 2i-1.Specific compression of images and the method for reduction are not limited thereto.
Fig. 8 is the composition schematic diagram for another image recognition engine server that one embodiment of the application provides, including the Two transceivers 31, recovery module 32, measurement module 33, identification module 34.
Second transceiver 31 receives the compression image from image recognition meter reading terminal by data platform 2.Recovery module 32, according to also meta-rule corresponding with predetermined reduced rule, are restored to obtain to the one part of pixel point of the missing of compression image Also original image and applied.The size and image and main view of the measurement main view of measurement module 33 are on horizontal and vertical side The pixel distance of edge, second transceiver 32 is also by the size and image and main view of the transmission main view of data platform 2 in cross To the pixel distance with longitudinal edge to image recognition meter reading terminal.Identification module 34 identifies that the clarity for going back original image obtains Image recognition result, second transceiver 32 also send image recognition result to image recognition meter reading terminal by data platform 2, with Just image recognition meter reading terminal judges whether to need to change resolution ratio of taking pictures.
Fig. 9 is a kind of composition schematic diagram for image recognition kilowatt meter reading-out system that one embodiment of the application provides, including image is known Other meter reading terminal 1, data platform 2, image recognition engine server 3, data platform server 4.
Image recognition meter reading terminal 1 shoots the image of meter, determines the fringe region of image, carries out pixel to fringe region Compression generates compression image, transmission compression image to data platform 2.Data platform 2 receive compression image, transmission compression image to Image recognition engine server 3.Image recognition engine server 3 receives compression image, according to corresponding with predetermined reduced rule Also meta-rule is restored to obtain and goes back original image to the one part of pixel point of the missing of compression image, and transmission goes back original image to number According to Platform Server 4.Original image is gone back in the reception of data platform server 4, record, storage, using going back original image.
Image recognition meter reading terminal 1 at least one, can have many.Data interchange is carried out with data platform 2.
1 camera arrangement 11 of image recognition meter reading terminal can choose the camera for supporting highest 640*480 pixel image, Adjustable output pixel size, by image recognition engine verifying and function in terms of demand, such image can expire The requirement of sufficient discrimination.
Image recognition meter reading terminal 1 in response to image recognition engine server 3 image recognition result and adjust acquisition figure The resolution ratio of picture.Its strategy uses, and the image of preferential low resolution replaces with high score again when low-resolution image is undesirable Resolution image.Since camera highest supports the image of 640*480 pixel, but in view of the too low influence identification of resolution ratio and again Core, image recognition meter reading terminal 1 can clap 2 images at installation initial stage, be 640*480 pixel and 320*240 pixel respectively, pass through The identification of image recognition engine server 3.If 320*240 pixel can satisfy requirement, the picture of taking pictures of camera arrangement 11 is set Element is 320*240 pixel.If 320*240 pixel is unable to satisfy requirement, the pixel of taking pictures that camera arrangement 11 is arranged is 640*480 Pixel.Image border sampling scale separately is done to image again, to reduce the data volume size of image transmission, increases transmission success rate.
Particularly, if having been set to 320*240 pixel, when use environment causes image clear due to certain factors change Clear degree variation, to image recognition engine server 3 can not be identified, then image recognition engine server 3 will pass through data Platform Server 4, data platform 2 send resolution ratio sampling to image recognition meter reading terminal 1 and reset, that is, are changed to 640* 480 pixels, then re-start transmission flow of taking pictures.
The image that image recognition engine server 3 is also transmitted based on image recognition meter reading terminal 1, measures the size of main view And image and main view be in the pixel distance at horizontal and vertical edge, by data platform 2 send the size of main view with And image and main view horizontal and vertical edge pixel distance to image recognition meter reading terminal 1.Image recognition engine clothes Business device 3 also identifies that the clarity for going back original image obtains image recognition result, sends image recognition result to figure by data platform 2 As identification meter reading terminal 1, so that meter reading terminal judges whether to need to change resolution ratio of taking pictures.
In above process, image of the image recognition meter reading terminal 1 in addition to transmitting this, can also be whole by image recognition meter reading The current battery charge at end 1, environment temperature, ambient humidity information are uploaded to data platform 2, and data platform 2 receives corresponding number According to, data can be uploaded to data platform server 4 according to given protocol, recorded by data platform server 4, store and Using.
The technical solution that the present embodiment proposes, is equipped with resolution ratio and data Platform Server by image recognition engine Information storage, increase data transmission in success rate and image data discrimination and whole system intelligence.
Figure 10 is a kind of composition schematic diagram for image recognition kilowatt meter reading-out system that another embodiment of the application provides, including image Identify meter reading terminal 1, data platform 2, image recognition engine server 3, data platform server 4, display platform 5.
Image recognition meter reading terminal 1 shoots the image of meter, determines the fringe region of image, carries out pixel to fringe region Compression generates compression image, and transmission compression image carries out image restoring and application to data platform 2.Data platform 2 receives compression Image, transmission compression image to image recognition engine server 3.Image recognition engine server 3 receive compression image, according to The corresponding also meta-rule of reduced rule is restored to obtain to the one part of pixel point of the missing of compression image goes back original image, transmits Also original image is to data platform server 4.Original image is gone back in the reception of data platform server 4, record, storage, using going back original image. Display platform 5 is communicated with data platform server 4, and original image and image recognition result are gone back in displaying.
Image recognition meter reading terminal 1 in response to image recognition engine server 3 image recognition result and adjust acquisition figure The resolution ratio of picture.The same above-described embodiment of detailed process.
The image that image recognition engine server 3 is also transmitted based on image recognition meter reading terminal 1 for the first time, budget image and master The pixel distance at edge of the view at least two mutually perpendicular directions and the size of main view, are taken by data platform Business device 4 is sent to image recognition meter reading terminal 1 by data platform 2.
Image recognition meter reading citing is completed with image recognition meter reading terminal 1, then it can be seen that figure in display platform 5 As this image for being clapped of identification meter reading terminal 1, the time taken pictures, identification as a result, and working as there are also image recognition meter reading terminal 1 Preceding battery capacity, environment temperature, ambient humidity information.Correspondingly, will be taken by data platform if having exception for various information Business device 4 is handled, and is shown in display platform 5.
Specifically, work as image recognition meter reading terminal 1 when first used, as shown in fig. 6, according to the installation rule of agreement, Image recognition meter reading terminal 1 is installed on intelligent meter, and image recognition meter reading terminal 1 is called out using manual trigger module 16 It wakes up, main control module 15 will be powered module 14 and power at this time, and power on to all functional modules.Thereafter, camera arrangement 11 take pictures.Image processing apparatus 12 carries out image procossing and obtains compression image.Main control module 15 receives the pressure of image procossing Compressed image data.Meanwhile current battery charge, environment temperature, ambient humidity information are passed to master control by information acquisition module 19 Relevant information is supplied to first transceiver 13 by module 15, main control module 15.First transceiver 13 sends information to data again Platform 2.First transceiver 13 will obtain clocked flip set information by data platform 2, while obtain the left side of main view Edge the pixel distance x to image left side edge, the pixel distance y of the top edges of main view to image top edges.Thereafter, One transceiver 13 sends clocked flip set information to main control module 15, at the same by x, y pass to image processing apparatus 12 into Row record.Thereafter, main control module 15 will be timed setting to timing trigger module 17.
In the above process, if data platform 2 does not provide clocked flip set information, then will by main control module 15 according to Parameter preset is timed setting.So far the first use of image recognition meter reading terminal 1 is completed.
After image recognition meter reading terminal 1 is opened for the first time, it will be triggered according to the timing information of timing trigger module 17, when When its timing information reaches parameter preset, main control module 15 will be powered module 14 and power on, and carry out to all functional modules It powers on.Camera arrangement 11 is taken pictures, and image processing apparatus 12 carries out image procossing and obtains compression image, and main control module 15 receives The compressing image data of image processing apparatus 12 is simultaneously sent to first transceiver 13, or control image processing apparatus 12 schemes compression As being sent to first transceiver 13.Meanwhile information acquisition module 19 is by current battery charge, environment temperature, ambient humidity information Pass to main control module 15.Relevant information is supplied to first transceiver 13 by main control module 15.Thereafter, first transceiver 13 will be believed Breath is sent to data platform 2.Timing trigger module 17 carries out timing setting next time according to original timing information again.So far scheme The image recognition meter reading process for triggering waiting by timing trigger module 17 next time as identification meter reading terminal 1.
In the above process, if data platform 2 provides clocked flip set information, then first transceiver 13 will obtain timing Set information is triggered, this information will reset timing trigger module 17 via main control module 15.So far image recognition meter reading terminal 1 The image recognition meter reading process that waiting is triggered by timing trigger module 17 next time.
In the above process, if data platform 2 provides resolution ratio setting command, this information will be by first transceiver 13 via master Control module 15 sets camera arrangement 11.So far image recognition meter reading terminal 1 will wait next time by timing trigger module The image recognition meter reading process of 17 triggerings.
In the above process, if data platform 2 provides x, y information, x be image left edge to main view left edge pixel away from From y is image top edge to the pixel distance of main view top edge, and information will be right via main control module 15 by first transceiver 13 Image processing apparatus 12 is set.So far image recognition meter reading terminal 1 is triggered waiting by timing trigger module 17 next time Image recognition meter reading process.
If vibration-sensing module 18 generates alarm, the alarm processing of image recognition meter reading terminal 1 is started, the process is first First driving main control module 15 will be powered module 14 and power on, and power on to all functional modules.Then main control module 15 Collected warning information is sent information into data platform 2 by first transceiver 13, is finally shown in display platform 5 Out, to remind manager.
And at this point, main control module 15 will respond above-mentioned alarm, sends to timing trigger module 17 and be arranged because of above-mentioned alarm Shooting timing, when its timing information reaches the shooting timing, camera arrangement 11 is taken pictures, image processing apparatus 12 carry out Image procossing obtains compression image, and main control module 15 receives the compressing image data of image procossing and is sent to first transceiver 13, or control image processing apparatus 12 and be sent to first transceiver 13 for image is compressed.Meanwhile information acquisition module 19 will be current Battery capacity, environment temperature, ambient humidity information pass to main control module 15, and relevant information is supplied to first by main control module 15 Transceiver 13, first transceiver 13 send information to data platform 2 and finally open up in display platform 5 after data flow circulation It shows and, subsequent processing is determined by administrator.So far meter reading terminal completes response because vibration-sensing module 18 generates the mistake of alarm Journey.Thereafter, timing trigger module 17 resets to original timing information.So far image recognition meter reading terminal 1 will wait next time The image recognition meter reading process triggered by timing trigger module 17.
The technical solution that the present embodiment proposes further increases whole system by the information displaying in display platform Intelligence.
Figure 11 is a kind of composition schematic diagram for intelligent meter that one embodiment of the application provides, including meter 6, image recognition are copied Meter terminal 1.
It should be noted that each embodiment above by reference to described in attached drawing is only to illustrate the application rather than limits this The range of application, those skilled in the art should understand that, it is right under the premise of not departing from spirit and scope The modification or equivalent replacement that the application carries out, should all cover within the scope of the present application.In addition, signified unless the context Outside, the word occurred in the singular includes plural form, and vice versa.In addition, unless stated otherwise, then any embodiment All or part of in combination with any other embodiment all or part of come using.

Claims (13)

1. a kind of image recognition meter reading terminal, comprising:
Camera arrangement shoots the image of meter;
Image processing apparatus determines the fringe region of described image, carries out picture to the fringe region according to predetermined reduced rule Element compression generates compression image;
First transceiver sends the compression image to data platform and carries out image restoring and application.
2. image recognition meter reading terminal according to claim 1, wherein described image processing unit includes:
Main view determining module determines the position of the size and the main view of the main view of described image in described image;
Fringe region determining module, size and the position based on the main view, determines main view described in described image Except region be the fringe region;
Compression module carries out pixel compression to the fringe region according to predetermined reduced rule and generates compression image.
3. image recognition meter reading terminal according to claim 2, wherein the main view determining module includes:
Main view dimension acquisition module obtains the ruler of the main view from the data platform by the first transceiver It is very little;
Main view position determination module obtains from the data platform, described image and institute by the first transceiver Main view is stated in the pixel distance at horizontal and vertical edge, and determines position of the main view in described image.
4. image recognition meter reading terminal according to claim 2, wherein the compression module includes:
Pixel array determination unit determines the pixel array that all pixels point forms in described image;
Compression unit deletes the one part of pixel for the fringe region for including in the pixel array according to predetermined reduced rule Point, to generate the compression image.
5. image recognition meter reading terminal according to claim 1, further includes:
Main control module controls described image processing unit for the compression image and is sent to the first transceiver, and passes through institute It states first transceiver and receives the image recognition result from the data platform, adjust institute in response to described image recognition result State the resolution ratio of taking pictures of camera arrangement, wherein
Described image recognition result is identified the compression image by image recognition engine server and is obtained.
6. image recognition meter reading terminal according to claim 5, further includes:
Manual trigger module triggers the image of the camera arrangement shooting meter manually;Or/and timing trigger module, timing are touched The image for shooting meter of sending out camera arrangement described, timing life of the main control module with parameter preset or from the data platform The timing of timing trigger module described in parameter setting in order.
7. image recognition meter reading terminal according to claim 5, further includes:
Vibration-sensing module, the vibration of induction described image identification meter reading terminal, Xiang Suoshu main control module send a warning message;Institute It states main control module and controls the image taking that the camera arrangement carries out the meter in response to the warning information, to record vibration Situation.
8. image recognition meter reading terminal according to claim 5, further includes:
Information acquisition module acquires ambient temperature information, ambient humidity information, and the information is supplied to the master control mould Block.
9. a kind of image recognition engine server, comprising:
Second transceiver is received by data platform and identifies meter reading from any one of claim 1 to claim 8 described image The compression image of terminal;
Recovery module, according to also meta-rule corresponding with the predetermined reduced rule, to one of the missing of the compression image Point pixel, which is restored to obtain, to be gone back original image and is applied.
10. image recognition engine server according to claim 9, further includes:
Measurement module, the size for measuring the main view and described image and the main view are at horizontal and vertical edge Pixel distance, the second transceiver also send size and described image and the institute of the main view by the data platform It states pixel distance of the main view at horizontal and vertical edge and identifies meter reading terminal to described image.
11. image recognition engine server according to claim 9, further includes:
Identification module, the clarity that original image is gone back described in identification obtain image recognition result, and the second transceiver also passes through institute It states data platform and sends described image recognition result to described image identification meter reading terminal.
12. a kind of image recognition kilowatt meter reading-out system, comprising:
Claim 1 to any one of claim 8 described image identifies meter reading terminal;
The data platform, receive from described image identification meter reading terminal the compression image be forwarded to claim 9 to Any one of claim 11 described image identifies engine server;
Described image identifies engine server, receives the compression image, goes back meta-rule according to corresponding with predetermined reduced rule, The one part of pixel point of the missing of the compression image is restored to obtain and goes back original image, original image is gone back described in transmission to data Platform Server;
The data platform server, original image is gone back described in reception, and record, storage go back original image using described.
13. a kind of intelligent meter, comprising:
Meter;
Claim 1 to any one of claim 8 described image identifies meter reading terminal.
CN201910470446.3A 2019-05-31 2019-05-31 Image recognition meter reading terminal and system, image recognition engine server, intelligent meter Pending CN110191281A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910470446.3A CN110191281A (en) 2019-05-31 2019-05-31 Image recognition meter reading terminal and system, image recognition engine server, intelligent meter

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910470446.3A CN110191281A (en) 2019-05-31 2019-05-31 Image recognition meter reading terminal and system, image recognition engine server, intelligent meter

Publications (1)

Publication Number Publication Date
CN110191281A true CN110191281A (en) 2019-08-30

Family

ID=67719563

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910470446.3A Pending CN110191281A (en) 2019-05-31 2019-05-31 Image recognition meter reading terminal and system, image recognition engine server, intelligent meter

Country Status (1)

Country Link
CN (1) CN110191281A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112435440A (en) * 2020-10-30 2021-03-02 成都蓉众和智能科技有限公司 Non-contact type indoor personnel falling identification method based on Internet of things platform
CN112712465A (en) * 2020-12-31 2021-04-27 四川长虹网络科技有限责任公司 Method and system for optimizing communication data volume of photographing type meter reading terminal

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005036493A1 (en) * 2003-10-15 2005-04-21 M. Tech Korea Co., Ltd. Remote meter-reading system using in wire and wireless
US20060067583A1 (en) * 2004-09-24 2006-03-30 Fuji Photo Film Co., Ltd. Image compression apparatus, and image compression program storage medium
KR20080096263A (en) * 2007-04-27 2008-10-30 금오공과대학교 산학협력단 System for processing digit image, and method thereof
CN103716505A (en) * 2012-09-28 2014-04-09 北京蒙恬科技有限公司 Image recognition system and method
CN210007792U (en) * 2019-05-31 2020-01-31 北京安诺信科技股份有限公司 Image recognition meter reading terminal and system, image recognition engine server and intelligent meter

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005036493A1 (en) * 2003-10-15 2005-04-21 M. Tech Korea Co., Ltd. Remote meter-reading system using in wire and wireless
US20060067583A1 (en) * 2004-09-24 2006-03-30 Fuji Photo Film Co., Ltd. Image compression apparatus, and image compression program storage medium
KR20080096263A (en) * 2007-04-27 2008-10-30 금오공과대학교 산학협력단 System for processing digit image, and method thereof
CN103716505A (en) * 2012-09-28 2014-04-09 北京蒙恬科技有限公司 Image recognition system and method
CN210007792U (en) * 2019-05-31 2020-01-31 北京安诺信科技股份有限公司 Image recognition meter reading terminal and system, image recognition engine server and intelligent meter

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112435440A (en) * 2020-10-30 2021-03-02 成都蓉众和智能科技有限公司 Non-contact type indoor personnel falling identification method based on Internet of things platform
CN112435440B (en) * 2020-10-30 2022-08-09 成都蓉众和智能科技有限公司 Non-contact type indoor personnel falling identification method based on Internet of things platform
CN112712465A (en) * 2020-12-31 2021-04-27 四川长虹网络科技有限责任公司 Method and system for optimizing communication data volume of photographing type meter reading terminal
CN112712465B (en) * 2020-12-31 2023-08-04 四川长虹网络科技有限责任公司 Method and system for optimizing communication data volume of photographing meter reading terminal

Similar Documents

Publication Publication Date Title
Gupta Fingerprint based student attendance system using GSM
CN106303462A (en) A kind of water meter camera direct-reading meter method and system based on GPRS
CN109272598A (en) Substation's routine cruising inspection system and its method based on bluetooth iBeacon intelligent terminal
CN101320053A (en) Wireless data and image double-acquisition anti-electricity theft monitoring system and method thereof
CN202422195U (en) Comprehensive application mobile terminal for city management and community service
CN110191281A (en) Image recognition meter reading terminal and system, image recognition engine server, intelligent meter
CN201413534Y (en) Digitized transformation patrolling device
CN109683539A (en) Equipment dynamic control device and system is monitored online in pollutant
CN206042375U (en) Management system is patrolled and examined to intelligence based on BLE
CN108541013A (en) Information processing method, device, mobile terminal and computer readable storage medium
CN106447873A (en) 5G base station on-line management system based on Internet of Things
CN210007792U (en) Image recognition meter reading terminal and system, image recognition engine server and intelligent meter
CN109727329A (en) A kind of assets automatic tour inspection system
CN101662796A (en) Method and device for scanning wireless local area network (LAN) channels
CN104243683A (en) Data processing method, terminal, user identification card and data processing system
CN206961192U (en) A kind of automatic attendance checking device based on LoRa Internet of Things transmission technologys
CN208335287U (en) A kind of system of registering based on bluetooth and NB-IOT
CN107770490A (en) A kind of LED advertisements identification monitoring system
CN112929442B (en) Intelligent gas management system
CN205959254U (en) Intelligence system of patrolling and examining
CN206193901U (en) Case -involving property management and control of policeman and information sharing system
CN107195174A (en) Acquisition terminal on-site calibrator
CN207573526U (en) A kind of Electric Power Patrol terminal and Electric Power Patrol system
CN203799487U (en) Power information testing device based on wireless communication
CN103985233A (en) Wireless intelligent digital instrument acquisition remote transmitting apparatus

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20240411

Address after: G7-401, China Sensor Network International Innovation Park, 200 Linghu Avenue, Xinwu District, Wuxi City, Jiangsu Province, 214000

Applicant after: WUXI ZHUANXINZHIZHI TECHNOLOGY Co.,Ltd.

Country or region after: China

Address before: 100142 first floor, building 14, No. 45, Beiwa Road, Haidian District, Beijing

Applicant before: Beijing Anoxin Technology Co.,Ltd.

Country or region before: China