CN210038814U - Meter automatic identification system based on deep learning - Google Patents
Meter automatic identification system based on deep learning Download PDFInfo
- Publication number
- CN210038814U CN210038814U CN201920818205.9U CN201920818205U CN210038814U CN 210038814 U CN210038814 U CN 210038814U CN 201920818205 U CN201920818205 U CN 201920818205U CN 210038814 U CN210038814 U CN 210038814U
- Authority
- CN
- China
- Prior art keywords
- camera
- identification system
- sliding
- module
- deep learning
- 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
Abstract
The utility model relates to a table meter automatic identification system based on degree of depth study. The problem of manual watch table meter of looking over can't reach unmanned tour effect is solved. The system comprises a camera module and a processing end, wherein the camera module is connected with the processing end, the camera module is arranged on a sliding module, the system also comprises an operation end for controlling the sliding module, and the operation end is connected with the sliding module. The utility model discloses can come the reading of discernment table meter in the transformer substation through the image, reach the purpose of unmanned inspection, can carry out position control simultaneously, can adopt a camera module to discern a plurality of table meter readings in the transformer substation, control simply, practiced thrift the cost simultaneously.
Description
Technical Field
The utility model belongs to the technical field of image recognition technique and specifically relates to a table meter automatic identification system based on degree of depth study is related to.
Background
The reading of present meter is looked over and is looked over by the manual work, needs the operator always or often stays in the transformer substation, has increased operator's work load, can't reach unmanned effect of patrolling. In addition, the technology of acquiring meter reading images by using cameras is also provided at present, then the cameras are fixed, only meter images at fixed positions can be acquired, the cameras cannot be adjusted, the use is not flexible, and a plurality of cameras need to be installed when a plurality of meters are arranged in a transformer substation, so that the resource waste is caused, and the cost is increased.
Disclosure of Invention
The utility model discloses solve the manual work and look over the table meter, can't reach the problem of unmanned effect of patrolling to and the camera is fixed, can only gather fixed position table meter image, use dumb, and install a plurality of cameras and cause the wasting of resources, the problem that the cost improves provides a table meter automatic identification system based on degree of depth study.
The above technical problem of the present invention can be solved by the following technical solutions: the utility model provides a table meter automatic identification system based on degree of depth study, is including the processing end that is used for gathering the camera module of table meter image, learns and discerns shooting image, and camera module is connected with the processing end, camera module sets up and carries out the sliding module that removes at its drive, and the system still includes the end of controlling that carries out control to sliding module, controls the end and links to each other with sliding module. The utility model discloses well camera module gathers the reading image of table meter, and the processing end can be for the computer or have throughput's equipment, and the processing end is including the model that can train and discern the image, and this model carries out the analysis and discerns the reading of table meter to the table meter image of gathering, and the model trains and learns through the mark image of many gathers of input, and this model can adopt the technique commonly used among the prior art, if adopt convolution neural network. The position of the camera module can be adjusted through the sliding module, the camera module cannot be fixed to collect images of a specific area, and a plurality of marks in the transformer substation can be identified through movement. The control end can be a computer or mobile equipment, the control end is connected with the sliding module through a wired or wireless network, and the control end controls the sliding module to work so as to control the moving position of the camera module. The utility model discloses can come the reading of discernment table meter in the transformer substation through the image, reach the purpose of unmanned inspection, can carry out position control simultaneously, can adopt a camera module to discern a plurality of table meter readings in the transformer substation, control simply, practiced thrift the cost simultaneously.
As a preferred scheme, the camera module comprises a shell and a camera ball movably arranged in the shell, a camera port is arranged on the camera ball, a camera is arranged on the detection port, and the camera is connected with the processing end. The ball of making a video recording can rotate in the casing in this scheme, and the user can adjust the angle of making a video recording of camera through adjusting the ball of making a video recording for collection meter reading image that can be better.
As a preferred scheme, the shell comprises a front cover and a rear seat, a first spherical concave groove is formed in the front cover, a through hole is formed in the middle of the groove, a second spherical concave groove is formed in the rear seat, the front cover and the rear seat are buckled, a circular rotating chamber is formed jointly, and the camera shooting ball is installed in the rotating chamber. This scheme can realize making a video recording the all-round wide-angle rotation of ball, and simple structure is reasonable, and convenient operation is swift, can make things convenient for nimble angle of adjusting the ball camera of making a video recording.
As a preferable scheme, a plurality of first buckle heads are arranged on the front end face of the rear seat, a plurality of buckle grooves are correspondingly arranged on the bottom face of the front cover, and the first buckle heads are inserted into and buckled in the buckle grooves to fix the front cover and the rear cover. This scheme is connected protecgulum and back seat through buckle structure, and is fixed convenient and fast.
As a preferable scheme, a window is formed at the front end of the front cover, a transparent protective cover is buckled on the window, a plurality of second buckling heads are arranged on the inner side surface of the protective cover, a ring of buckling edges are arranged on the inner side wall of the probe window, the second buckling heads are buckled on the buckling edges, and the protective cover is fixed on the window.
Preferably, the camera port comprises a section of operation hole for inserting a tool, and the shape of the operation hole is a polygon. The opening part of the camera shooting port in the scheme is composed of an operation hole, the camera is installed at the bottom of the operation hole, the operation hole is used for inserting an operation tool, the orientation angle of the detection port is adjusted by the operation tool, and the operation hole is in a polygonal shape matched with the shape of the operation tool.
As a preferred scheme, the sliding module comprises a sliding rail, a sliding groove arranged on the bottom of the rear seat and a driving mechanism for driving the camera module to move, the sliding groove is connected to the sliding rail in a matched mode, and the driving mechanism is connected with the sliding rail in a driving mode. The camera module forms a slidable structure through the cooperation of the sliding groove and the sliding rail in the scheme, and the driving mechanism can automatically drive the camera module to move on the sliding rail, so that the camera module can collect more marked images through movement.
Preferably, the driving mechanism includes a driving gear driven by a motor, the driving gear is disposed in the bottom of the sliding groove, a portion of the driving gear protrudes into the sliding groove, a tooth space is disposed on the surface of the sliding rail along the length, a driving tooth is disposed in the tooth space, and the driving gear is engaged with the driving tooth. Set up drive gear in the spout in this scheme, this drive gear is cylindric, and the slide rail surface is provided with the drive tooth, and this drive tooth is located the tooth's socket, and drive tooth top flushes with the slide rail surface. When the sliding rail is matched with the sliding groove, the driving gear is meshed with the driving teeth, the driving gear is driven to rotate by the motor, the driving gear rotates to drive the camera module to move together, and the camera module can automatically move on the sliding rail.
Therefore, the utility model has the advantages that: the reading of table meter in can coming discernment transformer substation through the image reaches the purpose of unmanned tour, can carry out position control simultaneously, can adopt a camera module to discern a plurality of table meter readings in the transformer substation, controls simply, has practiced thrift the cost simultaneously.
Drawings
FIG. 1 is a block diagram of a structure of the present invention;
fig. 2 is a schematic structural diagram of a camera module according to the present invention;
fig. 3 is a schematic structural diagram of the middle slide rail of the present invention.
The camera shooting device comprises a camera shooting module 1, a camera shooting module 2, a processing end 3, a control end 4, a sliding module 5, a shell 6, a camera shooting ball 7, a front cover 8, a rear seat 9, a first groove 10, a second groove 11, a first buckle 12, a buckle groove 13, a window 14, a protective cover 15, a second buckle 16, a buckle edge 17, a camera shooting port 18, a sliding rail 19, a sliding groove 20, a driving gear 21, a tooth groove 22 and driving teeth.
Detailed Description
The technical solution of the present invention is further specifically described below by way of examples and with reference to the accompanying drawings.
Example (b):
the embodiment is a meter automatic identification system based on deep learning, and as shown in fig. 1, the system comprises a camera module 1 for collecting meter images, a processing end 2 for learning and identifying the shot images, a sliding module 4 for driving the camera module to move, and a control end 3 for controlling the sliding module, wherein the camera module is arranged on the sliding module, the camera module is connected with the processing end, and the control end is connected with the sliding module. The camera module is equipment integrated with a camera and used for collecting reading images of the meter. The processing end can be a computer or equipment with processing capacity, the processing end comprises a model capable of training and identifying images, the model analyzes the acquired meter images and identifies the reading of the meter, the model trains and learns by inputting a plurality of acquired marked images, and the model adopts a technology commonly used in the prior art, such as a convolutional neural network. The position of the camera module can be adjusted through the sliding module, the camera module cannot be fixed to collect images of a specific area, and a plurality of marks in the transformer substation can be identified through movement. The control end can be a computer or mobile equipment, the control end is connected with the sliding module through a wired or wireless network, and the control end controls the sliding module to work so as to control the moving position of the camera module.
As shown in fig. 2, the camera module includes a housing 5 and a camera ball 6 movably disposed in the housing, a camera port 17 is disposed on the camera ball, a camera is disposed on the camera port, the camera is connected with the processing end, the camera port includes a section of operation hole for inserting a tool, and the operation hole is polygonal. The casing includes protecgulum and back seat, is provided with the first recess of spherical indent in the protecgulum, and it has the perforation to open in the middle of the recess, is provided with the second recess of spherical indent on the back seat, and the protecgulum forms a circular shape with back seat looks lock inside jointly and rotates the room, the ball of making a video recording is installed and is rotated indoor.
The front cover and the rear seat are fixed in a clamping block mode. The front end face of the rear seat is provided with a plurality of first buckling heads 11, the bottom surface of the front cover is correspondingly provided with a plurality of buckling grooves 12, and the first buckling heads are inserted into and buckled in the buckling grooves to fix the front cover and the rear cover.
The front end of the front cover is provided with a window 13, a transparent protective cover 14 is buckled on the window, a plurality of second buckling heads 15 are arranged on the inner side surface of the protective cover, a ring of buckling edges 16 are arranged on the inner side wall of the window, the second buckling heads are buckled on the buckling edges, and the protective cover is fixed on the window.
The sliding module comprises a sliding rail 18, a sliding groove 19 arranged at the bottom of the rear seat and a driving mechanism for driving the camera module to move, the sliding groove is connected to the sliding rail in a matching mode, and the driving mechanism is connected with the sliding rail in a driving mode. The guide rail comprises a base and a T-shaped track, and the sliding groove is an inverted T-shaped groove matched with the shape of the track. The drive mechanism comprises a drive gear 20 driven by a motor, the drive gear being disposed in the bottom of the slideway with part of the drive gear projecting into the slideway, as shown in figure 3, a tooth slot 21 being provided along the length of the surface of the slideway, drive teeth 22 being provided in the tooth slot, the drive gear being in mesh with the drive teeth. The motor is arranged in the rear seat, a motor rotating shaft is linked with the driving gear, and the control end controls the motor. When the camera module needs to move, a user sends an instruction to the sliding module through the control end, the motor works to drive the driving gear to rotate, and the driving gear rotates to drive the whole camera module to move on the sliding rail under the condition of being meshed with the driving gear.
When the system is used, a plurality of meter images are shot by the camera module, the reading of the meter images is cut out and then placed into the self-learning model in the processing end for training, and a model capable of identifying the meter reading is formed. After the training is finished, the camera module shoots the meter reading image to be detected, and the meter reading image is placed into the trained model to be read and identified. In addition, the control end controls the sliding module to control the position of the camera module, so that a plurality of meter images can be collected.
The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications, additions and substitutions for the specific embodiments described herein may be made by those skilled in the art without departing from the spirit of the invention or exceeding the scope of the invention as defined in the accompanying claims.
Although the terms camera module, processing side, manipulation side, sliding module, etc. are used more often herein, the possibility of using other terms is not excluded. These terms are used merely to more conveniently describe and explain the nature of the present invention; they are to be construed in a manner that is inconsistent with the spirit of the invention.
Claims (8)
1. The utility model provides a table meter automatic identification system based on deep learning which characterized in that: including the camera module that is used for gathering the table meter image, to the processing end that the image of shooing was learnt and was discerned, camera module is connected with the processing end, camera module sets up on its sliding module that removes of drive, and the system still includes the end of controlling that carries out control to sliding module, controls the end and links to each other with sliding module.
2. The deep learning based meter automatic identification system as claimed in claim 1, wherein the camera module comprises a housing and a camera ball movably arranged in the housing, a camera port is arranged on the camera ball, a camera is arranged on the camera port, and the camera is connected with the processing end.
3. The deep learning-based automatic meter identification system as claimed in claim 2, wherein the housing comprises a front cover and a rear seat, a first concave spherical recess is formed in the front cover, a through hole is formed in the middle of the recess, a second concave spherical recess is formed in the rear seat, the front cover and the rear seat are buckled to form a circular rotating chamber, and the camera ball is installed in the rotating chamber.
4. The deep learning-based automatic meter identification system as claimed in claim 3, wherein a plurality of first fasteners are provided on the front end surface of the rear seat, a plurality of corresponding fastening grooves are provided on the bottom surface of the front cover, and the first fasteners are inserted into and fastened in the fastening grooves to fix the front cover and the rear cover.
5. The deep learning-based automatic meter identification system as claimed in claim 3, wherein a window is formed at the front end of the front cover, a transparent protective cover is fastened on the window, a plurality of second fastening heads are arranged on the inner side surface of the protective cover, a ring of fastening edges are arranged on the inner side wall of the window, and the second fastening heads are fastened on the fastening edges to fix the protective cover on the window.
6. The deep learning based meter automatic identification system according to claim 2, wherein the camera port comprises a section of operation hole for inserting a tool, and the shape of the operation hole is a polygon.
7. The deep learning based meter automatic identification system as claimed in claim 3, wherein the sliding module comprises a sliding rail, a sliding groove arranged on the bottom of the rear seat, and a driving mechanism for driving the camera module to move, the sliding groove is connected to the sliding rail in a matching manner, and the driving mechanism is connected with the sliding rail in a driving manner.
8. The deep learning based automatic meter identification system as claimed in claim 7, wherein the driving mechanism comprises a driving gear driven by a motor, the driving gear is disposed in the bottom of the sliding groove, and a part of the driving gear protrudes into the sliding groove, a tooth space is disposed on the surface of the sliding rail along the length, a driving tooth is disposed in the tooth space, and the driving gear is engaged with the driving tooth.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920818205.9U CN210038814U (en) | 2019-05-31 | 2019-05-31 | Meter automatic identification system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920818205.9U CN210038814U (en) | 2019-05-31 | 2019-05-31 | Meter automatic identification system based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN210038814U true CN210038814U (en) | 2020-02-07 |
Family
ID=69346002
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201920818205.9U Active CN210038814U (en) | 2019-05-31 | 2019-05-31 | Meter automatic identification system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN210038814U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110334699A (en) * | 2019-05-31 | 2019-10-15 | 国网浙江省电力有限公司金华供电公司 | A kind of meter automatic recognition system based on deep learning |
-
2019
- 2019-05-31 CN CN201920818205.9U patent/CN210038814U/en active Active
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110334699A (en) * | 2019-05-31 | 2019-10-15 | 国网浙江省电力有限公司金华供电公司 | A kind of meter automatic recognition system based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190150895A1 (en) | Fully automatic ultrasonic scanner and scan detection method | |
| CN110091302B (en) | Intelligent robot with improved image recognition technology based on deep learning | |
| CN210038814U (en) | Meter automatic identification system based on deep learning | |
| CN208077195U (en) | A kind of face recognition device with managing caller function | |
| CN108267604A (en) | A kind of clean type alcohol content tester with image collecting function | |
| CN112747219A (en) | Artificial intelligence image recognition system and recognition method thereof | |
| CN110179411B (en) | Cleaning robot based on image recognition | |
| CN210075443U (en) | Full-automatic precision numerical control machine tool data acquisition and recording device | |
| CN218036385U (en) | Blade acquisition and 3D scanning device | |
| CN209743882U (en) | Pipeline crawling detection robot | |
| CN208079220U (en) | A kind of high light spectrum image-forming equipment | |
| CN218272647U (en) | LED lamp strip check out test set | |
| CN110334699A (en) | A kind of meter automatic recognition system based on deep learning | |
| CN112017315B (en) | Fingerprint identification device of external closed protective structure based on internet of things equipment | |
| CN109348091A (en) | A kind of mobile phone camera | |
| CN205497128U (en) | Cutter parameter measurement device based on CCD | |
| CN208418045U (en) | A kind of mobile terminal visitor's identification device based on user's portrait | |
| CN109594797B (en) | Automatic adaptation device and control method for comfortable environment of living space equipment | |
| CN214012005U (en) | Capital construction on-site key personnel identification device | |
| CN214947820U (en) | Face recognition equipment support device | |
| CN220020318U (en) | Adjustment mechanism and face recognition management and control system | |
| CN219367243U (en) | Automatic acquisition device for field investigation images | |
| CN216825016U (en) | Movable sampling device for water quality detection | |
| CN114406800B (en) | Automatic cutter wear detection device based on image recognition and use method thereof | |
| CN216667117U (en) | Non-contact sample information acquisition and management device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant |