CN210007792U - Image recognition meter reading terminal and system, image recognition engine server and intelligent meter - Google Patents
Image recognition meter reading terminal and system, image recognition engine server and intelligent meter Download PDFInfo
- Publication number
- CN210007792U CN210007792U CN201920814166.5U CN201920814166U CN210007792U CN 210007792 U CN210007792 U CN 210007792U CN 201920814166 U CN201920814166 U CN 201920814166U CN 210007792 U CN210007792 U CN 210007792U
- Authority
- CN
- China
- Prior art keywords
- image
- image recognition
- meter reading
- reading terminal
- data platform
- 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.)
- Active
Links
Images
Landscapes
- Arrangements For Transmission Of Measured Signals (AREA)
Abstract
The image recognition meter reading terminal comprises a photographing device, an image processing device and an transceiver, wherein the photographing device is used for photographing an image of a meter, the image processing device is used for determining an edge area of the image and performing pixel compression on the edge area according to a preset compression rule to generate a compressed image, and the transceiver is used for sending the compressed image to a data platform for image restoration and application.
Description
Technical Field
The application relates to the technical field of intelligent meter reading, in particular to image recognition meter reading terminals and systems, an image recognition engine server and an intelligent meter.
Background
With the rapid development of the economy of China and the comprehensive implementation of the project policy of ' household meter', the demand of urban residents on water, electricity and gas resources is more and more large.
The inventor finds that the meter reading modes have disadvantages, and particularly in large-scale enterprises and multipoint application scenarios, the requirements of managers cannot be met in the aspects of real-time performance, accuracy and comprehensive management.
The IC card prepayment solves the problem of manual meter reading, but has lower user experience degree, and is particularly suitable for the sites of large-scale enterprises and multi-point application scenes, and the management difficulty is high.
The existing data transmission schemes are concentrated on GPRS, NB-Iot and 2G networks, the GPRS and 2G networks cannot be used intensively for communication base stations in large quantities, the communication base stations are not preset with enough simultaneous access capacity, the traffic cost is high, the power consumption is large, the battery life is short, the maintenance cost is high, but the transmission rate, the transmission data length and the network coverage situation of the communication base stations have advantages, image data can be uploaded and completed only in a sub-package mode for a plurality of times, the NB-Iot is special for the internet of things communication mode used intensively in large quantities, the traffic cost is low, the access capacity is 10 times of that of the 2G network, but the communication rate is low, the power consumption is small, the battery service time is longer, the maintenance cost is reduced in the later period, the data length of each transmission is short, the image data needs to be uploaded in sub-package mode for dozens of times, the packet loss rate determined by is uploaded for a plurality of times, the success rate of meter reading and identification of the terminal is influenced, the NB network.
In terms of image processing problems, because of the limitations of the meter reading terminal in cost, data transmission and power supply, generally does not adopt an image processing chip in the meter reading terminal, only depends on the camera to perform image processing, and a processor of the meter reading terminal converts an image into a data stream for transmission, so that the image is ensured not to be distorted, but pressure is applied to data transmission.
SUMMERY OF THE UTILITY MODEL
The embodiment of the application provides image recognition meter reading terminals, which comprise a photographing device, an image processing device and a transceiver, wherein the photographing device photographs images of meters, the image processing device determines edge regions of the images and carries out pixel compression on the edge regions according to a preset compression rule to generate compressed images, and the transceiver sends the compressed images to a data platform to carry out image restoration and application.
The aspects of the application are that the image processing device comprises a main view determining device, an edge area determining device and a compressing device, wherein the main view determining device determines the size of the main view of the image and the position of the main view in the image, the edge area determining device determines the area outside the main view in the image as the edge area based on the size and the position of the main view, and the compressing device generates a compressed image by performing pixel compression on the edge area according to a preset compression rule.
As aspects of the application, the front view determining device comprises a front view size obtaining device and a front view position determining device, wherein the front view size obtaining device obtains the size of the front view from the data platform through the transceiver, and the front view position determining device obtains the pixel distance of the edge of the image and the front view in the transverse direction and the longitudinal direction from the data platform through the transceiver and determines the position of the front view in the image.
The aspects of the application comprise a pixel array determining device and a compressor, wherein the pixel array determining device determines a pixel array formed by all pixel points in the image, and the compressor deletes parts of the pixel points in the edge area contained in the pixel array according to a preset compression rule to generate a compressed image.
As aspects of the application, the image recognition meter reading terminal further comprises a main controller, wherein the main controller controls the image processing device to send the compressed image to the transceiver, receives an image recognition result from the data platform through the transceiver, and adjusts the photographing resolution of the photographing device in response to the image recognition result, wherein the image recognition result is obtained by recognizing the compressed image by an image recognition engine server.
According to aspects of the application, the image recognition meter reading terminal further comprises a manual trigger or/and a timing trigger, the manual trigger manually triggers the photographing device to photograph the image of the meter, the timing trigger periodically triggers the photographing device to photograph the image of the meter, and the main controller sets the timing of the timing trigger according to preset parameters or parameters in a timing command from the data platform.
According to aspects of the application, the image recognition meter reading terminal further comprises a vibration sensor, the vibration sensor senses vibration of the image recognition meter reading terminal and sends alarm information to the main controller, and the main controller responds to the alarm information to control the photographing device to photograph the meter and record vibration conditions.
As aspects of this application, image identification meter reading terminal still includes information acquisition equipment, information acquisition equipment gathers ambient temperature information, ambient humidity information, and will the information provides main control unit.
The embodiment of the application further provides image recognition engine servers, which comprise a second transceiver and a restoring device, wherein the second transceiver receives a compressed image from an image recognition meter reading terminal through a data platform, and the restoring device restores part of missing pixel points of the compressed image according to a restoring rule corresponding to the preset compressing rule to obtain a restored image and applies the restored image.
As aspects of the present application, the image recognition engine server further comprises a measuring device that measures the size of the main view and the pixel distance between the image and the edge of the main view in the lateral and longitudinal directions, and the second transceiver further transmits the size of the main view and the pixel distance between the image and the edge of the main view in the lateral and longitudinal directions to the image recognition meter reading terminal through the data platform.
As aspects of the application, the image recognition engine server further comprises an identifier, the identifier identifies the definition of the restored image to obtain an image recognition result, and the second transceiver further sends the image recognition result to the image recognition meter reading terminal through the data platform.
The embodiment of the application also provides image identification meter reading systems, which comprise the image identification meter reading terminal, a data platform, the image identification engine server and a data platform server, wherein the image identification meter reading terminal shoots an image of a meter, determines an edge area of the image, performs pixel compression on the edge area according to a preset compression rule to generate a compressed image, transmits the compressed image to the data platform, the data platform receives the compressed image and transmits the compressed image to the image identification engine server, the image identification engine server receives the compressed image, restores part of pixel points missing from the compressed image according to a restoration rule corresponding to the preset compression rule to obtain a restored image, and transmits the restored image to the data platform server, and the data platform server receives the restored image, records, stores and applies the restored image.
The embodiment of the application also provides types of intelligent meters, which comprise meter meters and the image recognition meter reading terminal.
According to the technical scheme provided by the embodiment of the application, the image edge is zoomed, the data volume of data transmission is reduced under the condition that the success rate of image identification is not influenced, and the success rate in data transmission and the identification rate of image data are increased.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram illustrating a image recognition meter reading terminal according to an embodiment of the present application ;
FIG. 2 is a schematic diagram of an edge region of images provided by an embodiment of the present application ;
FIGS. 3A-3C are schematic diagrams of the image zooms provided by the embodiment of the present application;
FIG. 4 is a schematic diagram illustrating a image recognition meter reading terminal according to another embodiment of the present application;
FIG. 5 is a schematic diagram illustrating a image recognition meter reading terminal according to yet another embodiment of the present application;
FIG. 6 is a schematic diagram illustrating a kinds of image recognition meter reading terminals according to another embodiment of the present application;
FIG. 7 is a schematic diagram illustrating components of a image recognition engine servers according to an embodiment of the present application ;
FIG. 8 is a schematic diagram of image recognition engine servers according to another embodiment of the present application;
FIG. 9 is a schematic diagram illustrating a image recognition meter reading system according to an embodiment of the present application ;
FIG. 10 is a schematic diagram of a image recognition meter reading system according to another embodiment of the present application;
fig. 11 is a schematic diagram of components of types of smartwatch provided in the embodiment of of the present application.
Detailed Description
However, the embodiments and examples described below are only for illustrative purposes and are not intended to limit the application, and the present invention includes only some examples, not all examples, of , and other examples obtained by various changes in the present application by those skilled in the art are within the scope of the present application.
Fig. 1 is a schematic composition diagram of types of image recognition meter reading terminals provided in an embodiment of of the present application, including a photographing device 11, an image processing device 12, and a th transceiver 13.
The photographing device 11 photographs the image of the meter, the image processing device 12 determines the edge area of the image, and performs pixel compression on the edge area according to a predetermined compression rule to generate a compressed image, the transceiver 13 sends the compressed image to a data platform for image restoration and application.
The photographing device 11 is a main device for providing a photographed image for the image recognition meter reading terminal, and includes, but is not limited to, components such as a camera and a flash lamp, and can be triggered by timing or manual control, and after the photographing device 11 is triggered, the flash lamp is turned on, and the camera is started to photograph.
In the aspect of shooting, a camera with high pixels shoots clearly, but the data volume of an image is large, so that pressure is caused to data transmission, and a camera with low pixels can better solve the problem of transmission, but the definition of the image is not enough, so that the image can be easily identified or difficult to identify by using an image identification engine, and more importantly, in the aspect of rechecking, the resolution of rechecking personnel is not clear.
The photographing device 11 of this embodiment may select a camera supporting an image with 640 × 480 pixels at the maximum, and the size of the output pixels may be adjusted, so that the image may satisfy the requirement of the recognition rate through the verification and function requirements of the image recognition engine server.
The image processing device 12 adopts image edge compression methods, when the image processing device is installed for the first time, the photographed image is transmitted to the data platform through the transceiver 13, and then the transceiver 13 receives the image main view size and the main view position in the image determined by the image recognition engine server from the data platform, so as to determine the edge area except the main view in the image.
FIG. 2 is a schematic diagram of an edge region of images provided by an embodiment of the present application . As shown in FIG. 2, P1 is an image, P2 is a front view, x is a pixel distance from a left edge of the image to the left edge of the front view, y is a pixel distance from an upper edge of the image to the upper edge of the front view, and regions P1-P2 are edge regions.
The edge region is compressed and restored, which is mainly shown in fig. 3A, that is, in the matrix represented by (00, mn), data in even columns in the matrix is removed, but data included in the region shown by (ab, cd) is not removed. The resulting data matrix is the one shown in fig. 3B. This process belongs to the compression process of images.
The compressed image obtained after the image processing is sent to the data platform 2 through the th transceiver 13, the data platform 2 sends the compressed image to the image recognition engine server 3, and pixel reduction is performed in the image recognition engine server 3, which belongs to the image reduction process, the image recognition engine server 3 reduces the data of the even columns determined by the compressed image data according to the reduction rules corresponding to the preset compression rules, the reduction method is to copy and fill the data on the left side of the compressed image data into the even columns, the finally reduced image is as shown in fig. 3C, that is, the data of the 2i th column is the same as the data of the 2i-1 th column, and the specific image compression and reduction method is not limited to this.
The effective data volume of the image is reduced by after processing, for the th transceiver 13, the sending frequency is reduced by , so that the success rate of data transmission is improved, and the traffic usage is reduced by , so that the usage cost is reduced.
In the scheme of the image meter reading, the central area of an image is clearly required to be displayed actually and is convenient to recheck, and can be used as an area for indicating the meter number evidence, namely the area is defined as a main view P2, the image edge processed by the method, namely the area (P1-P2) can cause distortion of to a certain degree, but the distortion degree does not influence the identification of the main view P2 in the image identification engine server 3, so that the correctness of reading the meter number through image identification is ensured, and the pixel points shown on the image edge belong to other information areas of the meter, which embody the appearance and other information of the meter, the recheck of the area only confirms that the meter number does not change when the meter number is rechecked manually, and the point is considered to be the same as the point when the meter is installed, so that the meter reading system is not subjected to distortion in the use process, namely the meter reading is not influenced by the definition of the meter reading system .
According to the technical scheme provided by the embodiment, the image edge is compressed and restored, so that the data volume of data transmission is reduced, and the success rate of data transmission and the recognition rate of image data are increased under the condition that the success rate of image recognition is not influenced.
Fig. 4 is a schematic diagram of a kinds of image recognition meter reading terminals according to another embodiment of the present application, which includes a photographing device 11, an image processing device 12, and a th transceiver 13.
The photographing device 11 photographs the image of the meter, the image processing device 12 determines the edge area of the image, and performs pixel compression on the edge area according to a predetermined compression rule to generate a compressed image, the transceiver 13 sends the compressed image to a data platform for image restoration and application.
The image processing apparatus 12 includes a main view determining device 121, an edge area determining device 122, and a compressing device 123.
The main view determining device 121 determines the size of the main view of the image and the position of the main view in the image. The edge region determining device 122 determines a region other than the main view in the image as an edge region based on the size and the position of the main view. The compression device 123 performs pixel compression on the edge area according to a predetermined compression rule to generate a compressed image.
The main view determining device 121 comprises a main view size acquiring device 1211, a main view position determining device 1212, the main view size acquiring device 1211 acquires the size of the main view from the data platform through the th transceiver 13, the main view position determining device 1212 acquires the pixel distance of the edge of the image in the lateral and longitudinal directions from the data platform through the th transceiver 13, and determines the position of the main view in the image.
The compression device 123 includes a pixel array determination device 1231 and a compressor 1232, where the pixel array determination device 1231 determines a pixel array formed by all pixel points in the image, and the compressor 1232 deletes portions of the pixel points in the edge region included in the pixel array according to a predetermined compression rule to generate a compressed image.
Fig. 5 is a schematic diagram of a image recognition meter reading terminal provided in another embodiment of the present application, which includes a photographing device 11, an image processing device 12, a transceiver 13, a power supply device 14, a main controller 15, a manual trigger 16 and/or a timing trigger 17.
The photographing device 11 photographs images of a meter, the image processing device 12 determines an edge area of the image, pixel compression is carried out on the edge area according to a preset compression rule to generate a compressed image, the th transceiver 13 sends the compressed image to the data platform for image restoration and application, the main controller 15 controls the image processing device 12 to send the compressed image to the th transceiver 13, receives an image recognition result from the data platform through the th transceiver 13, and adjusts the photographing resolution of the photographing device 11 in response to the image recognition result, wherein the image recognition result is obtained by recognizing the compressed image by the image recognition engine server, the manual trigger 16 manually triggers the photographing device 11 to photograph the images of the meter, the timing trigger 17 periodically triggers the photographing device 11 to photograph the images of the meter, the main controller 15 sets the timing of the timing trigger 17 according to preset parameters or parameters in a timing command from the data platform, and the power supply device 14 supplies power to all devices of the image recognition meter reading terminal.
The image recognition meter reading terminal is in a dormant state for a long time, the manual trigger 16 is needed to start the meter reading terminal when the meter reading terminal is installed, the image is sent out through the process, the main controller 15 sets the timing trigger 17, and the image recognition meter reading terminal is triggered to work at fixed time, so that the meter reading terminal is in a long-term running state.
When an image currently photographed by the image recognition meter reading terminal needs to be acquired, the photographing device 11 of the meter reading terminal needs to be triggered through the manual trigger 16, so that the requirement that a user wants to acquire the current image can be met.
The low-power consumption sleep state of the image identification meter reading terminal is established by a PSM (Power Saving mode) power Saving mode, firstly, the selected main controller 15 and the th transceiver 13 can configure the PSM (Power Saving mode) power Saving mode, and the PSM power Saving mode is configured according to the requirements of the sleep state in the application process, namely, the data transmitted by a data platform is received only when the image identification meter reading terminal actively uploads the data, so that the aim of long-term low-power consumption work of the image identification meter reading terminal is fulfilled.
in the existing scheme is generally in 2 modes of using an external power supply and using a battery, the mode of the external power supply increases the difficulty in construction, a plurality of scenes cannot provide power supplies, but the continuous power supply of the power supply in the using process can be ensured, the mode of using the battery greatly reduces the construction difficulty, but the battery with too large capacity cannot be selected for safety, and the battery with small capacity has limitation on the service life.
The power supply device 14 in this embodiment may adopt a rechargeable battery with a moderate volume, weight and capacity, and retain an external power supply interface. The power supply module can meet the requirement that a 2G module (second-generation mobile phone communication technical specification) works for more than 3 years, or an NB-IOT module (Narrow Band Internet of Things based on honeycomb) works for more than 9 years, an external power supply interface is provided, an external power supply can be connected, or a battery is charged, and the use of various application scenes is facilitated.
As an alternative to , the image processing apparatus 12 includes a front view determining device 121, an edge area determining device 122, and a compressing device 123.
The timing trigger 17 is used for providing services for timing photographing of the image recognition meter reading terminal and comprises a clock and a timer, the clock acquires clock information from the data platform through the main controller 15 and the th transceiver 13, the time of the clock is synchronized according to the clock information, and the system performance of the clock information is guaranteed.
According to the technical scheme provided by the embodiment, the meter reading terminal is suitable for more application scenes through various power supply modes, and the application range is expanded; through the addition of the main controller and the trigger, the terminal photographing transmission is manually or automatically controlled, and the terminal photographing transmission device is suitable for mass application.
Fig. 6 is a schematic composition diagram of image recognition meter reading terminals provided in another embodiment of the present application, including a photographing device 11, an image processing device 12, a th transceiver 13, a power supply device 14, a main controller 15, a manual trigger 16 or/and a timing trigger 17, a vibration sensor 18, and an information acquisition device 19.
The photographing device 11 photographs an image of a meter, the image processing device 12 determines an edge area of the image, pixel compression is performed on the edge area to generate a compressed image, the th transceiver 13 transmits the compressed image to a data platform for image restoration and application, the main controller 15 controls the image processing device 12 to transmit the compressed image to the th transceiver 13, receives an image recognition result from the data platform through the th transceiver 13, and adjusts a photographing resolution of the photographing device 11 in response to the image recognition result, wherein the image recognition result is obtained by recognizing the compressed image by the image recognition engine server, the manual trigger 16 manually triggers the photographing device 11 to photograph the image of the meter, the timing trigger 17 regularly triggers the photographing device 11 to photograph the image of the meter, the main controller 15 sets the timing of the timing trigger 17 with a preset parameter or a parameter in a timing command from the data platform, the vibration sensor 18 senses the image recognition terminal including, but not limited to, detached, moved, and sends alarm information to the main controller 15, the main controller drives the device 11 to perform power supply, vibration information, the power supply information of the power supply equipment, the power supply information of the environmental information acquisition equipment, and the power supply information of the environmental information to the environmental information acquisition equipment 14.
The main controller 15 is a core main control of the image recognition meter reading terminal, and is used for issuing commands and receiving and processing data, and can time the timing trigger 17 according to preset parameters or parameters in a timing command given by the data platform received through the th transceiver 13 when the main controller is used for the first time, set the photographing resolution of the photographing device 11, acquire the current battery capacity, the ambient temperature and the ambient humidity information of the information acquisition device 19 after being powered on, monitor the state of the vibration sensor 18, receive the image processing information given by the image processing device 12, and send the information to the data platform 2 through the th transceiver 13.
, when the image recognition meter reading terminal is disassembled, or the meter associated with the image recognition meter reading terminal is moved, disassembled, or modified to cause the vibration of the image recognition meter reading terminal, the vibration sensor 18 will send alarm information to the main controller 15, and the transceiver 13 will send the alarm information to the data platform 2, and the data platform 2 will report the alarm information to the data platform server 4 and can display the alarm information, so that the manager can know that the image recognition meter reading terminal gives an alarm due to the vibration, the terminal or the meter associated with the terminal may be being disassembled, moved or modified, and the manager can perform corresponding processing according to the information.
Particularly, when the vibration sensor 18 is started, the vibration of the image recognition meter reading terminal is sensed, alarm information is sent to the main controller 15, and the main controller 15 controls the photographing device 11 to photograph and upload so as to record the condition when an alarm occurs.
Particularly, when the vibration sensor 18 is used for taking a picture in a delayed manner, the picture is taken in a delayed manner in order to record the situation accompanying the vibration, that is, when the vibration sensor 18 is started, the image recognition meter reading terminal and the meter only sense, and no displacement or change occurs, and the picture is taken in a delayed manner, so that the situation occurring at minutes after the vibration sensor 18 is started can be recorded.
In particular, the time setting for the th instant of time after the shock sensor 18 is activated is empirically set, and it may not be possible to record the actual situation at that time if the time is too short or too long.
As an alternative to , the image processing apparatus 12 includes a front view determining device 121, an edge area determining device 122, and a compressing device 123.
The timing trigger 17 comprises a clock and a timer, wherein the clock acquires clock information from the data platform through the main controller 15 and the th transceiver 13, and synchronizes the time of the clock according to the clock information, and the timer receives a timing instruction from the main controller 15 and sets timing according to parameters in the timing instruction.
According to the technical scheme, the vibration sensor and the information acquisition equipment are added, so that image data can be correctly transmitted when the position of the terminal changes, more temperature, humidity and battery information are acquired, and the terminal is more intelligent.
Fig. 7 is a schematic diagram of a image recognition engine servers according to an embodiment of of the present application, including a second transceiver 31 and a restoring apparatus 32.
The second transceiver 31 receives the compressed image from the image recognition meter reading terminal through the data platform 2, and the restoration device 32 restores the missing partial pixel points of the compressed image according to the restoration rule corresponding to the preset compression rule to obtain and apply the restored image.
The second transceiver 31 receives the compressed image data as shown in fig. 3B, and the restoring device 32 restores the received compressed image data to the data in the even columns that it has confirmed according to the restoring rule corresponding to the predetermined compression rule, by copying and filling the data on the left side of the compressed image data into the even columns, and finally restoring the image as shown in fig. 3C, that is, the data in the 2 i-th column is the same as the data in the 2i-1 st column. The specific method of image compression and restoration is not limited thereto.
Fig. 8 is a schematic diagram of another image recognition engine servers according to an embodiment of the present application , including a second transceiver 31, a restoring device 32, a measuring device 33, and an identifier 34.
The second transceiver 31 receives a compressed image from the image recognition meter reading terminal through the data platform 2, the reduction device 32 reduces the missing partial pixel points of the compressed image according to a reduction rule corresponding to a preset compression rule to obtain a reduced image and applies the reduced image, the measuring device 33 measures the size of the main view and the pixel distance between the image and the edge of the main view in the transverse direction and the longitudinal direction, the second transceiver 32 further sends the size of the main view and the pixel distance between the image and the edge of the main view in the transverse direction and the longitudinal direction to the image recognition meter reading terminal through the data platform 2, the recognizer 34 recognizes the definition of the reduced image to obtain an image recognition result, and the second transceiver 32 further sends the image recognition result to the image recognition meter reading terminal through the data platform 2 so that the image recognition meter reading terminal judges whether the photographing resolution needs to be changed.
Fig. 9 is a schematic composition diagram of image recognition meter reading systems provided in an embodiment of of the present application, including an image recognition meter reading terminal 1, a data platform 2, an image recognition engine server 3, and a data platform server 4.
The image recognition meter reading terminal 1 shoots an image of a meter, determines an edge area of the image, performs pixel compression on the edge area to generate a compressed image, transmits the compressed image to the data platform 2, the data platform 2 receives the compressed image, transmits the compressed image to the image recognition engine server 3, the image recognition engine server 3 receives the compressed image, restores partial missing pixel points of the compressed image according to a restoration rule corresponding to a preset compression rule to obtain a restored image, transmits the restored image to the data platform server 4, and the data platform server 4 receives the restored image, records, stores and applies the restored image.
The number of the image recognition meter reading terminals 1 is at least , and a plurality of the image recognition meter reading terminals can be provided for data intercommunication with the data platform 2.
The image recognition meter reading terminal 1 photographing device 11 can select a camera supporting the image with the maximum 640 × 480 pixels, the size of the output pixels can be adjusted, and the image can meet the requirement of the recognition rate through the verification and the function requirements of the image recognition engine.
The image recognition meter reading terminal 1 adjusts the resolution of the captured image in response to the image recognition result of the image recognition engine server 3. The strategy adopts the mode that the low-resolution image is prioritized, and the low-resolution image is replaced by the high-resolution image when the low-resolution image does not meet the requirement. Since the camera supports 640 × 480 pixels at the maximum, but considering the low resolution to affect the recognition and review, the image recognition meter reading terminal 1 can shoot 2 images at the initial stage of installation, which are 640 × 480 pixels and 320 × 240 pixels respectively, through the recognition of the image recognition engine server 3. If 320 × 240 pixels can meet the requirement, the photographing pixels of the photographing device 11 are set to 320 × 240 pixels. If 320 × 240 pixels fail to satisfy the requirement, the photographing pixels of the photographing device 11 are set to 640 × 480 pixels. In addition, the image edge sampling scaling is carried out on the image so as to reduce the size of data sent by the image and increase the transmission success rate.
Particularly, if the pixels are set to 320 × 240 pixels, if the image definition changes due to a change in the usage environment due to some factor and the image recognition engine server 3 cannot recognize the change, the image recognition engine server 3 changes the resolution sampling reset, that is, the resolution sampling reset to 640 × 480 pixels, to the image recognition meter reading terminal 1 through the data platform server 4 and the data platform 2, and then performs the photographing transmission process again.
The image recognition engine server 3 also measures the size of the main view and the pixel distance between the image and the edge of the main view in the transverse direction and the longitudinal direction based on the image transmitted by the image recognition meter reading terminal 1, and transmits the size of the main view and the pixel distance between the image and the edge of the main view in the transverse direction and the longitudinal direction to the image recognition meter reading terminal 1 through the data platform 2. The image recognition engine server 3 also recognizes the definition of the restored image to obtain an image recognition result, and sends the image recognition result to the image recognition meter reading terminal 1 through the data platform 2, so that the meter reading terminal can judge whether the photographing resolution needs to be changed.
In the process, the image identification meter reading terminal 1 not only transmits the image of this time, but also uploads the current battery capacity, the environmental temperature and the environmental humidity information of the image identification meter reading terminal 1 to the data platform 2, the data platform 2 receives corresponding data and uploads the data to the data platform server 4 according to a set protocol, and the data platform server 4 records, stores and applies the data.
According to the technical scheme provided by the embodiment, the image recognition engine is matched with the set resolution and the information storage of the data platform server, so that the success rate in data transmission, the recognition rate of image data and the intelligence of the whole system are increased.
Fig. 10 is a schematic diagram illustrating a image recognition meter reading system provided in another embodiment of the present application, including an image recognition meter reading terminal 1, a data platform 2, an image recognition engine server 3, a data platform server 4, and a display platform 5.
The image recognition meter reading terminal 1 shoots an image of a meter, determines an edge area of the image, performs pixel compression on the edge area to generate a compressed image, transmits the compressed image to the data platform 2 to perform image restoration and application, the data platform 2 receives the compressed image and transmits the compressed image to the image recognition engine server 3, the image recognition engine server 3 receives the compressed image, restores missing pixel points of the compressed image according to restoration rules corresponding to the compression rules to obtain a restored image, and transmits the restored image to the data platform server 4, the data platform server 4 receives the restored image, records, stores and applies the restored image, and the display platform 5 communicates with the data platform server 4 to display the restored image and an image recognition result.
The image recognition meter reading terminal 1 adjusts the resolution of the captured image in response to the image recognition result of the image recognition engine server 3. The specific process is the same as the above embodiment.
The image recognition engine server 3 also recognizes the image transmitted by the meter reading terminal 1 for the first time based on the image, budgets the pixel distances between the images and the edges of the main view in at least two mutually perpendicular directions and the size of the main view, and sends the budgets to the image recognition meter reading terminal 1 through the data platform server 4 and the data platform 2.
For example, when the image recognition meter reading terminal 1 finishes times of image recognition meter reading, the image shot by the image recognition meter reading terminal 1, the shooting time, the recognition result, the current battery capacity, the environmental temperature and the environmental humidity information of the image recognition meter reading terminal 1 can be seen in the display platform 5, and correspondingly, if various information is abnormal, the information is processed by the data platform server 4 and displayed in the display platform 5.
Specifically, when the image recognition meter reading terminal 1 is used for the first time, as shown in fig. 6, according to an agreed installation rule, the image recognition meter reading terminal 1 is installed on the smart meter, and the manual trigger 16 is used for waking up the image recognition meter reading terminal 1, at this time, the main controller 15 powers on the powered device 14 and powers on all the functional devices, then, the photographing device 11 performs photographing, the image processing device 12 performs image processing to obtain a compressed image, the main controller 15 receives the compressed image data of the image processing, at the same time, the information acquisition device 19 transmits the current battery power, the ambient temperature and the ambient humidity information to the main controller 15, the main controller 15 provides related information to the th transceiver 13, the th transceiver 13 transmits the main control information to the data platform 2, the th transceiver 13 obtains timing trigger setting information through the data platform 2, and simultaneously obtains the pixel distance x from the left edge of the main view to the left edge of the image, and then the pixel distance y. from the upper edge of the main view to the upper edge of the image, the th transceiver 13 transmits the timing trigger setting information to the timing trigger setting device 15, and simultaneously transmits the timing trigger setting information to the image processing device to the y 12 to perform timing recording, and then to perform timing recording.
In the above process, if the data platform 2 does not provide the timing trigger setting information, the main controller 15 performs timing setting according to the preset parameters. The initial use of the image recognition meter reading terminal 1 is completed.
After the image recognition meter reading terminal 1 is started for the first time, triggering is carried out according to timing information of the timing trigger 17, when the timing information reaches preset parameters, the main controller 15 is used for electrifying the powered device 14 and electrifying all functional devices, the photographing device 11 is used for photographing, the image processing device 12 is used for carrying out image processing to obtain a compressed image, the main controller 15 receives compressed image data of the image processing device 12 and sends the compressed image data to the th transceiver 13, or the image processing device 12 is controlled to send the compressed image to the th transceiver 13, meanwhile, the information acquisition device 19 is used for transmitting current battery power, environment temperature and environment humidity information to the main controller 15, the main controller 15 provides related information to the th transceiver 13, then the th transceiver 13 sends the information to the data platform 2, the timing trigger 17 carries out timing setting for times again according to the original timing information, and the image recognition meter reading terminal 1 waits for an image recognition meter reading process triggered by the timing trigger 17 for times.
In the above process, if the data platform 2 provides the timing trigger setting information, the th transceiver 13 will obtain the timing trigger setting information, which will reset the timing trigger 17 via the main controller 15, so far, the image recognition meter reading terminal 1 will wait for the next image recognition meter reading processes triggered by the timing trigger 17.
In the above process, if the data platform 2 gives a resolution setting command, this information will be set by the th transceiver 13 via the main controller 15 to the photographing device 11, so far, the image recognition meter reading terminal 1 will wait for the next image recognition meter reading processes triggered by the timing trigger 17.
In the above process, if the data platform 2 provides x and y information, where x is the pixel distance from the left edge of the image to the left edge of the main view, and y is the pixel distance from the upper edge of the image to the upper edge of the main view, the th transceiver 13 sets the image processing apparatus 12 via the main controller 15, so far, the image recognition meter reading terminal 1 waits for the next image recognition meter reading process times triggered by the timing trigger 17.
If the vibration sensor 18 generates an alarm, an alarm process of the image recognition meter reading terminal 1 is started, the process firstly drives the main controller 15 to electrify the powered device 14 and electrify all the functional devices, then the main controller 15 sends the collected alarm information to the data platform 2 through the transceiver 13, and finally the alarm information is displayed in the display platform 5 to remind a manager.
At this time, the main controller 15 will respond to the alarm, send the shooting timing set by the alarm to the timing trigger 17, when the timing information reaches the shooting timing, the shooting device 11 will shoot, the image processing device 12 will process the image to obtain the compressed image, the main controller 15 will receive the compressed image data processed by the image and send it to the th transceiver 13, or control the image processing device 12 to send the compressed image to the th transceiver 13. at the same time, the information acquisition device 19 will transmit the current battery power, environmental temperature, environmental humidity information to the main controller 15, the main controller 15 will provide the relevant information to the th transceiver 13, the th transceiver 13 will send the information to the data platform 2, after the data flow is circulated, finally displayed in the display platform 5, the administrator will decide the subsequent processing, until the image recognition terminal will respond to the alarm process generated by the vibration meter reading sensor 18. thereafter, the timing trigger 17 will reset to the original timing information, so far, the image recognition terminal 1 will wait for times of image recognition process triggered by the timing trigger 17 to read the meter.
According to the technical scheme provided by the embodiment, the intellectualization of the whole system is further increased by steps through information display on the display platform.
Fig. 11 is a schematic composition diagram of types of smart meters provided in an embodiment of of the present application, including a meter 6 and an image recognition meter reading terminal 1.
It should be noted that the various embodiments described above with reference to the drawings are only intended to illustrate and not to limit the scope of the application, and that modifications or equivalent substitutions made herein without departing from the spirit and scope of the application will be understood by those of ordinary skill in the art to which the present invention pertains.
Claims (13)
1, kinds of image recognition meter reading terminal, characterized by that, includes:
the photographing device is used for photographing the image of the meter;
the image processing device is used for determining the edge area of the image and carrying out pixel compression on the edge area according to a preset compression rule to generate a compressed image;
and the transceiver sends the compressed image to a data platform for image restoration and application.
2. The image recognition meter reading terminal according to claim 1, wherein the image processing device comprises:
a main view determining device that determines a size of a main view of the image and a position of the main view in the image;
an edge region determining device that determines a region other than the main view in the image as the edge region based on the size and the position of the main view;
and the compression device is used for carrying out pixel compression on the edge area according to a preset compression rule to generate a compressed image.
3. The image recognition meter reading terminal according to claim 2, wherein the front view determining device includes:
a front view size acquiring device for acquiring the size of the front view from the data platform through the th transceiver;
and the front view position determining device acquires the pixel distance of the image and the edge of the front view in the transverse direction and the longitudinal direction from the data platform through the th transceiver and determines the position of the front view in the image.
4. The image recognition meter reading terminal of claim 2, wherein the compression device comprises:
the pixel array determining device is used for determining a pixel array formed by all pixel points in the image;
and the compressor deletes parts of pixel points of the edge area contained in the pixel array according to a preset compression rule to generate the compressed image.
5. The image recognition meter reading terminal according to claim 1, further comprising:
a main controller for controlling the image processing device to transmit the compressed image to the th transceiver, and to receive the image recognition result from the data platform through the th transceiver, and to adjust the photographing resolution of the photographing device in response to the image recognition result, wherein,
the image recognition result is obtained by recognizing the compressed image by an image recognition engine server.
6. The image recognition meter reading terminal according to claim 5, further comprising:
the manual trigger is used for manually triggering the photographing device to photograph the image of the meter; or/and
and the timing trigger is used for triggering the photographing device to photograph the image of the meter at regular time, and the main controller sets the timing of the timing trigger according to preset parameters or parameters in a timing command from the data platform.
7. The image recognition meter reading terminal according to claim 5, further comprising:
the vibration sensor is used for sensing the vibration of the image recognition meter reading terminal and sending alarm information to the main controller; and the main controller responds to the alarm information to control the photographing device to photograph the image of the meter so as to record the vibration condition.
8. The image recognition meter reading terminal according to claim 5, further comprising:
and the information acquisition equipment acquires the ambient temperature information and the ambient humidity information and provides the information for the main controller.
An image recognition engine server of the type , comprising:
a second transceiver for receiving a compressed image from the image recognition meter reading terminal according to of any claim 1 to claim 8 via a data platform;
and the restoration equipment restores the missing part of the pixel points of the compressed image according to the restoration rule corresponding to the preset compression rule to obtain and apply the restored image.
10. The image recognition engine server of claim 9, further comprising:
the second transceiver also receives the image through the data platform and sends the size of the main view and the pixel distance between the image and the edge of the main view in the transverse direction and the longitudinal direction to the image identification meter reading terminal.
11. The image recognition engine server of claim 9, further comprising:
and the identifier is used for identifying the definition of the restored image to obtain an image identification result, and the second transceiver is also used for sending the image identification result to the image identification meter reading terminal through the data platform.
12, an image recognition meter reading system, comprising:
an image recognition meter reading terminal according to any of claims 1 to 8;
the data platform, which receives the compressed image from the image recognition meter reading terminal and forwards the compressed image to the image recognition engine server of any of claims 9 to 11;
the image recognition engine server receives the compressed image, restores the missing part of pixel points of the compressed image according to a restoration rule corresponding to a preset compression rule to obtain a restored image, and sends the restored image to a data platform server;
and the data platform server receives the restored image, records, stores and applies the restored image.
An intelligent watch of the type , comprising:
a meter is arranged;
an image recognition meter reading terminal according to any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920814166.5U CN210007792U (en) | 2019-05-31 | 2019-05-31 | Image recognition meter reading terminal and system, image recognition engine server and intelligent meter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920814166.5U CN210007792U (en) | 2019-05-31 | 2019-05-31 | Image recognition meter reading terminal and system, image recognition engine server and intelligent meter |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN210007792U true CN210007792U (en) | 2020-01-31 |
Family
ID=69308678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201920814166.5U Active CN210007792U (en) | 2019-05-31 | 2019-05-31 | Image recognition meter reading terminal and system, image recognition engine server and intelligent meter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN210007792U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110191281A (en) * | 2019-05-31 | 2019-08-30 | 北京安诺信科技股份有限公司 | Image recognition meter reading terminal and system, image recognition engine server, intelligent meter |
| CN111710171A (en) * | 2020-06-11 | 2020-09-25 | 北京筑梦园科技有限公司 | License plate recognition method, server and parking management system |
-
2019
- 2019-05-31 CN CN201920814166.5U patent/CN210007792U/en active Active
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110191281A (en) * | 2019-05-31 | 2019-08-30 | 北京安诺信科技股份有限公司 | Image recognition meter reading terminal and system, image recognition engine server, intelligent meter |
| CN111710171A (en) * | 2020-06-11 | 2020-09-25 | 北京筑梦园科技有限公司 | License plate recognition method, server and parking management system |
| CN111710171B (en) * | 2020-06-11 | 2022-08-30 | 北京筑梦园科技有限公司 | License plate recognition method, server and parking management system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108881834B (en) | Commercial vehicle remote camera monitoring method | |
| CN210007792U (en) | Image recognition meter reading terminal and system, image recognition engine server and intelligent meter | |
| CN208369746U (en) | Web camera and internet protocol camera system | |
| CN207441000U (en) | The wireless kilowatt meter reading-out system of timing wake-up is had the function of based on NB-IoT and visual identity | |
| CN102945365B (en) | Handheld iris identification device with multiple data transmission modes and method thereof | |
| CN106303462A (en) | A kind of water meter camera direct-reading meter method and system based on GPRS | |
| CN106921942B (en) | Equipment control method and device | |
| CN105139471A (en) | Inspection method for electrical equipment | |
| CN109683539A (en) | Equipment dynamic control device and system is monitored online in pollutant | |
| CN202759542U (en) | Remote electric energy information collection system based on image identification | |
| US20090286529A1 (en) | Wireless remote monitoring system and method therefor | |
| CN115170990A (en) | Artificial intelligent edge computing system and method for unmanned aerial vehicle airborne pod | |
| CN110475321B (en) | Power-saving optimization method and device for wireless local area network (AP) | |
| CN110191281A (en) | Image recognition meter reading terminal and system, image recognition engine server, intelligent meter | |
| CN106331623A (en) | Security protection camera, and security protection communication system and method | |
| CN101753982A (en) | 3G network intelligent security system and using method thereof | |
| CN108594471A (en) | A kind of study monitor system and method based on intelligent glasses | |
| CN114845089B (en) | Video picture transmission method and device | |
| CN215447761U (en) | Three-dimensional modeling system based on wind power unmanned aerial vehicle | |
| CN107592501A (en) | Acquisition system and method for farming activities process data | |
| CN111385532A (en) | Low-power-consumption scene analysis device and analysis method thereof | |
| CN202276420U (en) | Wireless video monitor system based on H264 | |
| TWI722338B (en) | Intelligent information reading method, intelligent reader and system thereof | |
| CN116405749B (en) | Door lock monitoring device, door lock system and implementation method for low-power-consumption continuous video recording | |
| CN110944158A (en) | Intelligent Bluetooth-NB camera meter system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |