CN201654804U - Character recognizer - Google Patents
Character recognizer Download PDFInfo
- Publication number
- CN201654804U CN201654804U CN2009202470504U CN200920247050U CN201654804U CN 201654804 U CN201654804 U CN 201654804U CN 2009202470504 U CN2009202470504 U CN 2009202470504U CN 200920247050 U CN200920247050 U CN 200920247050U CN 201654804 U CN201654804 U CN 201654804U
- Authority
- CN
- China
- Prior art keywords
- image
- character
- module
- interface
- recognition
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The utility model relates to a character recognizer, which comprises a hardware structure and a software structure. The hardware structure consists of an image acquisition unit, an image processing unit, a data storage unit, a liquid crystal display unit, a GPRS communication unit, a key operation unit, a power supply module and the like. The software structure consists of system scheduling, a hardware control module, a driving module, a hardware platform, a human-computer interface module, a data bank module, a recognition algorithm module and the like. The character recognizer integrates an image acquisition function, a digital image recognition function and a data transmission function based on GPRS by adopting embedded design, can automatically acquire images and rapidly and automatically recognize targets on site, and transmits recognition results to a specified server by the aid of the GPRS communication unit. The recognizer is simple in operation and convenient in use, and has wide practical value and application prospect in the technical field of mode recognition detection devices.
Description
(1) technical field:
The utility model relates to a kind of character recognizing instrument, it can discern the laser coding character under the complex background, wherein, relate to image acquisition, Flame Image Process, pattern-recognition and, belong to pattern-recognition detection type device technique field based on the wireless communication technology of GPRS (GPRS (General Packet Radio Service)).
(2) background technology:
Now, along with the development of information society, Flame Image Process and mode identification technology are in the application of all trades and professions more and more widely.The character recognition technologies development rapidly, at aspects such as personal information management, office automation, electronic publication, Internet resources, various large-scale documents and materials management database, digital library, car plate identification, bill identification, I.D. discriminatings important practical value is arranged, improve work efficiency, reduced cost of labor.
To characters in images identification, the present research object overwhelming majority is for the character under the simple background, though or entire image background complexity, the background of character zone is simple.In actual applications, the situation of text image is complicated and changeable: grain background, the background that changes, noise, font are fuzzy, character incorporates among the background or the like.For example, many laser coding characters are sprayed directly on to the external packing surface of product, and product external packaging itself has unpredictable complex pattern, makes character be in the complicated background pattern.In the face of these complicated situations, traditional simple background character recognition technologies just has significant limitation.
At present, the recognition methods of character mainly contains based on template matching method, based on the method for charcter topology with based on neural network method, these existing image-recognizing methods all more or less have certain limitation, under a kind of environment the good method of effect change a kind of environment recognition effect may be very undesirable.Some have certain versatility, the method that recognition effect is good, and often calculated amount is very big, is difficult to real-time application.
GPRS is the abbreviation of GPRS (General Packet Radio Service) (General Packet Radio Service), it is a kind of wireless, packet-switched technology based on gsm system, provide end to end, wireless IP connects, it is the technology that a high-speed data is handled, form with grouping transmits data, has significant advantage aspect a lot.
(3) summary of the invention:
1, purpose: the purpose of this utility model provides a kind of character recognizing instrument, particularly relate to the laser code-jetting character recognizer under a kind of complex background of hand-held, its adopts embedded design with image collecting function, digital picture recognition function, be integrated in one based on the data sending function of GPRS, can on-the-spot carry out image acquisition and fast automatic identification, and recognition result is sent to given server by the GPRS communication module target.
2, technical scheme: a kind of character recognizing instrument of the utility model, it is to be made of hardware configuration and software configuration two large divisions:
(1) hardware configuration:
The hardware configuration block diagram of the laser code-jetting character recognizer under the hand-held complex background is as Fig. 1.
Hardware configuration partly is made up of image acquisition units, graphics processing unit, data storage cell, liquid crystal display, GPRS communication unit, key operation unit, power module etc.
Power module is responsible for this identifier various piece power supply; Key operation unit and liquid crystal display are man-machine interfaces, are responsible for the input of operational order and result's output and show; Image acquisition units is responsible for images acquired; The GPRS communication unit is responsible for the wireless transmission of recognition result.Whole process is: the user gives an order by key operation unit, processor receives order, control image acquisition units acquisition of image data, store by data storage cell, carry out image recognition then, the result that will discern sends to given server by the GPRS communication unit at last, simultaneously the result is stored in the data storage cell, is convenient to inquire about in the future.
Described image acquisition units is a high integration, high-resolution CMOS sensing chip OV9650, its inner integrated sequential circuit, analog signal processing circuit, digital signal processing circuit; It has Serial Camera ControlBus (SCCB) interface of standard, can realize various figure image intensifyings and control function by this interface, as automatic exposure, automatic gain, balance Control Scheme etc., and control image color, saturation degree, gamma correction, sharpening, camera lens calibration, noise and white pixel deletion etc.
Described graphics processing unit is to adopt Samsung ARM9 S3C2440 processor, and it is based on 32 embedded microprocessors of ARM920T kernel, and maximum operation frequency can reach 533MHz; It has the MMU memory management unit can support embedded OSs such as Windows CE, Linux, have resource on the abundant sheet, as LCD, dma controller, and interface such as USB, Ethernet, UART, camera, touch-screen.
Described data storage cell is the K9F1208 chip (64M * 8bit), be used for Boot Loader, OS kernel mirror image, file system and the user application of stocking system that adopts a slice Samsung; (4M * 16bit * 4Banks) constitutes the SDRAM (Synchronous Dynamic Random Access Memory) of 64MB, is used to load Windows CE operating system and runs application to adopt the HY57V561620 chip of two Samsungs.
Described liquid crystal display is 3.5 cun TFT touch-screen LQ035Q7DH01 that select Sharp for use; Wherein, S3C2440 provides LCD interface, and its control signal, frame synchronizing signal, line synchronizing signal, pixel clock signal, data output enable signal and data-signal are linked to each other with the LCD respective signal; The back 4 passage A of S3C2440 are linked to each other with the touch screen interface of LCD, realize the collection of contact position.
Described GPRS communication unit is to adopt the sim300 module, it has three frequency GSM (global system for mobile communications) and GPRS functions, embedded ICP/IP protocol is utilized the transparent transmission of GPRS network Platform Implementation data message, passes through serial communication between GPRS and the processor.
Described key operation unit is individual 4 * 4 matrix keyboard, and it links to each other by 8 universal I with controller, is used for the operation to whole identifier.
Described power-supply system is to adopt outside 5V power supply, utilizes two LM1117-33 respectively the 5V of outside input to be converted to 2 road 3.3V, controller and peripheral circuit device is powered the influence that independent power supply has at utmost avoided peripheral circuit to bring to controller separately.Because S3C2440 needs 3.3V and two kinds of voltages of 1.3V, so utilize a slice LM1117-ADJ that 3.3V is changed into 1.3V again.Power-supply system provides 5V, 3.3V and 1.3V voltage for the various piece of whole identifier.
(2) software configuration:
The software architecture synoptic diagram of this hand-held laser code-jetting character recognizer is as Fig. 2.
Software configuration partly is made up of system call, hardware controls module, driver module, hardware platform, human-computer interface module, database module, recognizer module etc.System call is responsible for the task scheduling of whole software system, at first, to come on the image data acquiring by driver module control hardware platform by the hardware controls module, through the recognizer module, obtain recognition result, then by human-computer interface module with its demonstration, the result is stored by data memory module, the control GPRS communication unit by the hardware controls module sends to specified server with recognition result at last.
Described system call is the core of whole software system, and it is responsible for the task scheduling of whole software system, the sending and control all and finished by system call of the operation task of other module.
The hardware controls part that described hardware controls module is a total system, its control camera, GPRS communication unit, real time clock (RTC) and keyboard.The control of camera drives based on the camera stream interface of independent research, and the basic procedure of operation as shown in Figure 3.The GPRS communication unit adopts the control of RS232 serial ports; The control of real time clock (RTC) is to adopt the built-in RTC interface of S3C2440, and the control of keyboard is by 8 universal I, adopts the mode of inquiry, the matrix keyboard of control 4 * 4.
Described driver module is the interface between hardware controls module and the hardware platform, and the hardware controls module is to finish control to hardware platform by driver module.
Described hardware platform is meant the hardware components of total system, comprises camera, GPRS communication module, button etc.
Described human-computer interface module realizes that user and identifier carry out man-machine interaction, it is developed based on MFC dialog box mode, realized login interface, main interface such as main interface, database master interface, database editing interface, the state exchange between each interface is as shown in Figure 5.After the beginning, at first enter into log-in interface, can enter into main interface after correctly landing, can enter into database master interface from main interface, can enter into the editing interface of database by editor and interpolation, withdraw from the database editing interface and can get back to database master interface, withdraw from database master interface and can get back to system master interface, withdraw from log-in interface and main interface point and can close whole man-machine interface.
Described database module is used for the recognition result storage, and the database EDB design that it adopts Windows CE to carry realizes.Its idiographic flow as shown in Figure 4.At first, first carry database volume; Open the database volume then; If database is then created in failure again, open database again, if success is then proceeded database manipulation, as searched database, write record, read record etc.; If no longer need to carry out database manipulation, then closing database unloads the database volume at last.
The software section main body that described recognizer module is this character recognizing instrument, it comprises: image pre-service, character zone location, binaryzation, inclination rectification, Character segmentation, feature extraction, preliminary identification, secondary identification etc.
This image pre-service (include image gray processing, remove noise and grey level stretching) is to handle for the ease of follow-up rapid image, needs earlier view data to be changed, and makes coloured image become 256 grades of gray-scale maps.Inevitably contain noise in the image, we adopt medium filtering that image is carried out pre-service.In order to strengthen the contrast of background area and character zone, image is carried out grey level stretching.
This character zone location is that the coding character zone under the complex background is located out,, because original image is a coloured image, if only handle, can lose a lot of useful informations in gray level image here, causes and the character zone accurate in locating can't be come out.Therefore, the design will start with simultaneously from gray level image and two targets of coloured image, on the one hand, carry out character locating in gray level image, use the method for texture analysis; In coloured image, position on the other hand, use the method for color cluster, the domain analysis of homochromy UNICOM.After analyzing end by above method, object information is merged, (character zone the ratio of width to height is between 6: 1 to 7: 1 to add prior imformation, the coding character color is a black), all these information are merged, finally obtain the result of decision, promptly obtain character zone.
This binaryzation is that the gray level image in the character zone is carried out binary conversion treatment, and we adopt maximum variance between clusters.Be inter-class variance and the method for class internal variance ratio, self-adaptation is calculated gray threshold, thinks the target area less than the zone of this threshold value, thinks the background area greater than the zone of this threshold value.
It is because the problem of camera shooting angle may be brought the little inclination of some angles that this inclination is corrected, and the result of character recognition is affected, and we have carried out self-adaptation to bianry image and have tilted to correct.Promptly obtain the angle of inclination of character zone, adopt affined transformation to correct according to the angle of inclination of character zone at last with the hough conversion.
This Character segmentation is to adopt regional area minimum value method to cut apart, and character is carried out vertical projection, and minimum place is cut apart in the vertical projection value of character adhesion scope.
Character feature is promptly extracted in this feature extraction.The feature extraction mode that native system adopts statistical nature and architectural feature to combine is in the hope of obtaining classifying quality preferably.Character picture is divided into 4 * 4 grid, adds up the ratio that black picture element in each zone accounts for this zone total pixel number, form 16 dimension statistical natures; Character picture respectively along level and vertical direction projection, is asked the peak of projection, form 2 dimension architectural features, two kinds of characteristics combination are formed 18 dimensional features as character feature.
Should preliminary identification be to adopt the support vector machine sorting algorithm, letter and number was set up the SVM model of cognition respectively.With the digit recognition is example, and model can be described as: given training set is input as 18 dimension sample characteristics inputs, be output as 10 kinds of possible outputs,, obtain decision function by sample training, make to draw corresponding output result for test sample book, with this as preliminary recognition result.
This secondary identification is on preliminary base of recognition, and character is discerned once more.Test is found, for some similar characters, is difficult to once discern successfully, need discern once more.Mainly some similar characters are discerned once more in conjunction with the specific structural features of character, such as Aspect Ratio, regional closure etc., with this recognition result as final recognition result.After secondary identification, can effectively raising character identification rate.
The recognizer flow process at first, is carried out pre-service to gathering the image that comes as shown in Figure 6, comprises image gray processing, removes noise, grey level stretching; Carry out the character zone location then, the zone location at character place is come out; Next character zone is carried out binary conversion treatment, obtain bianry image; Carry out slant correction then; Image after proofreading and correct is carried out Character segmentation, whole character zone is divided into single character; Each character is carried out feature extraction; The preliminary identification of advanced line character; At last by character like the phase-splitting of secondary cog region; Each character is carried out feature extraction, tentatively identification and secondary identification, get recognition result to the end.
3, beneficial effect:
A kind of character recognizing instrument of the utility model, its advantage and beneficial effect are as follows:
1. anti-preferably electromagnetic interference (EMI) circuit is adopted in the main electronic modular unit design of equipment, gets rid of the influence of electromagnetic interference (EMI);
2. the mode that adopts coloured image identification and gray level image identification to combine improves the character identification rate under the complex background;
3. adopt recognizer efficiently, shorten recognition time greatly;
4. recognition result can be sent to given server automatically by GPRS, improve work efficiency.
(4) description of drawings:
Fig. 1 the utility model hardware configuration synoptic diagram
The embedded character recognizing instrument software architecture of Fig. 2 synoptic diagram
Fig. 3 camera control flow synoptic diagram
Fig. 4 EDB database manipulation schematic flow sheet
Fig. 5 user interface state conversion synoptic diagram
Fig. 6 recognizer schematic flow sheet
Symbol description is as follows among the figure:
1 graphics processing unit; 2 liquid crystal displays; 3 key operation unit;
4 image acquisition units; The 5GPRS communication unit; 6 data storage cells;
7 power modules; 8 system calls; 9 recognizer modules; 10 hardware controls modules;
11 driver modules; 12 hardware platforms; 13 human-computer interface module; 14 database module.
(5) embodiment
A kind of character recognizing instrument of the utility model, it is to be made of hardware configuration and software configuration two large divisions.
(1) hardware configuration:
The hardware configuration block diagram of the laser code-jetting character recognizer under the hand-held complex background is as Fig. 1.
Hardware configuration partly is made up of image acquisition units 4, graphics processing unit 1, data storage cell 6, liquid crystal display 2, GPRS communication unit 5, key operation unit 3, power module 7 etc.Power module 7 is responsible for this identifier various piece power supply; Key operation unit 3 and liquid crystal display 2 are man-machine interfaces, are responsible for the input of operational order and result's output and show; Image acquisition units 4 is responsible for images acquired; GPRS communication unit 5 is responsible for the wireless transmission of recognition result.Whole process is: the user gives an order by key operation unit 3, processor receives order, control image acquisition units 4 acquisition of image data, be put in the data storage cell 6, carry out image recognition then, the result that will discern sends to given server by GPRS communication unit 5 at last, the result is stored in the data storage cell 6 that carries simultaneously.
Described image acquisition units 4 is a high integration, high-resolution CMOS sensing chip OV9650, its inner integrated sequential circuit, analog signal processing circuit, digital signal processing circuit; It has Serial Camera ControlBus (SCCB) interface of standard, can realize various figure image intensifyings and control function by this interface, as automatic exposure, automatic gain, balance Control Scheme etc., and control image color, saturation degree, gamma correction, sharpening, camera lens calibration, noise and white pixel deletion etc.
Described graphics processing unit 1 is to adopt Samsung ARM9S3C2440 processor, and it is based on 32 embedded microprocessors of ARM920T kernel, and maximum operation frequency can reach 533MHz; It has the MMU memory management unit can support embedded OSs such as Windows CE, Linux, have resource on the abundant sheet, as LCD, dma controller, and interface such as USB, Ethernet, UART, camera, touch-screen.
Described data storage cell 6 is the K9F1208 chip (64M * 8bit), be used for Boot Loader, OS kernel mirror image, file system and the user application of stocking system that adopt a slice Samsung; (4M * 16bit * 4Banks) constitutes the SDRAM (Synchronous Dynamic Random Access Memory) of 64MB, is used to load Windows CE operating system and runs application to adopt the HY57V561620 chip of two Samsungs.
Described liquid crystal display 2 is 3.5 cun TFT touch-screen LQ035Q7DH01 that select Sharp for use; Wherein, S3C2440 provides LCD interface, and its control signal, frame synchronizing signal, line synchronizing signal, pixel clock signal, data output enable signal and data-signal are linked to each other with the LCD respective signal; The back 4 passage A of S3C2440 are linked to each other with the touch screen interface of LCD, realize the collection of contact position.
Described GPRS communication unit 5 is to adopt the sim300 module, it has three frequency GSM (global system for mobile communications) and GPRS functions, embedded ICP/IP protocol is utilized the transparent transmission of GPRS network Platform Implementation data message, passes through serial communication between GPRS and the processor.
Described key operation unit 3 is matrix keyboards of individual 4 * 4, and it links to each other by 8 universal I with controller, is used for the operation to whole identifier.
Described power module 7 is to adopt outside 5V power supply, utilize two LM1117-33 respectively the 5V of outside input to be converted to 2 road 3.3V, controller and peripheral circuit device are powered the influence that independent power supply has at utmost avoided peripheral circuit to bring to controller separately.Because S3C2440 needs 3.3V and two kinds of voltages of 1.3V, so utilize a slice LM1117-ADJ that 3.3V is changed into 1.3V again.Power module 7 provides 5V, 3.3V and 1.3V voltage for the various piece of whole identifier.
(2) software configuration:
The software architecture synoptic diagram of this hand-held laser code-jetting character recognizer is as Fig. 2.
Software configuration partly is made up of system call 8, hardware controls module 10, driver module 11, hardware platform 12, human-computer interface module 13, database module 14, recognizer module 9 etc.System call 8 is responsible for the task scheduling of whole software system, at first, to come on the image data acquiring by driver module 11 control hardware platforms 12 by hardware controls module 10, through recognizer module 9, obtain recognition result, then by human-computer interface module 13 with its demonstration, by database module 14 with result storage, by hardware controls module 10 control GPRS communication units 5 recognition result is sent to specified server at last.
Described system call 8 is meant the core of whole software system, and it is responsible for the task scheduling of whole software system, the sending and control all and finished by system call of the operation task of other module.
The hardware controls part that described hardware controls module 10 is total systems, its control image acquisition units 4, GPRS communication unit 5, real time clock (RTC) and key operation unit 3.The control of image acquisition units 4 drives based on the camera stream interface of independent research, and the basic procedure of operation as shown in Figure 3.GPRS communication unit 5 adopts RS232 serial communication mode; The control of real time clock (RTC) is to adopt the built-in RTC interface of S3C2440, and the control of key operation unit 3 is by 8 universal I, adopts the mode of inquiry, the key operation unit 3 of control 4 * 4.
Described driver module 11 is the interfaces between hardware controls module 10 and the hardware platform 12, and hardware controls module 10 is to finish control to hardware platform 12 by driver module 11.
Described hardware platform 12 is meant the hardware components of total system, comprises camera, GPRS communication unit 5, key operation unit 3 etc.
Described database module 14 is used for the recognition result storage, and the database EDB design that it adopts Windows CE to carry realizes.Its idiographic flow as shown in Figure 4.At first, first carry database volume; Open the database volume then; If database is then created in failure again, open database again, if success is then proceeded database manipulation, as searched database, write record, read record etc.; If no longer need to carry out database manipulation, then closing database unloads the database volume at last.
Described human-computer interface module 13 realizes that user and identifier carry out man-machine interaction, it is developed based on MFC dialog box mode, realized login interface, main interface such as main interface, database master interface, database editing interface, the state exchange between each interface is as shown in Figure 5.After the beginning, at first enter into log-in interface, can enter into main interface after correctly landing, can enter into database master interface from main interface, can enter into the editing interface of database by editor and interpolation, withdraw from the database editing interface and can get back to database master interface, withdraw from database master interface and can get back to system master interface, withdraw from log-in interface and main interface point and can close whole man-machine interface.
The software section main body that described recognizer module 9 is these character recognizing instruments, it comprises: image pre-service, character zone location, binaryzation, inclination rectification, Character segmentation, feature extraction, preliminary identification, secondary identification etc.
This image pre-service (include image gray processing, remove noise and grey level stretching) is to handle for the ease of follow-up rapid image, needs earlier view data to be changed, and makes coloured image become 256 grades of gray-scale maps.Inevitably contain noise in the image, we adopt medium filtering that image is carried out pre-service.In order to strengthen the contrast of background area and character zone, image is carried out grey level stretching.
This character zone location is that the coding character zone under the complex background is located out,, because original image is a coloured image, if only handle, can lose a lot of useful informations in gray level image here, causes and the character zone accurate in locating can't be come out.Therefore, the design will start with simultaneously from gray level image and two targets of coloured image, on the one hand, carry out character locating in gray level image, use the method for texture analysis; In coloured image, position on the other hand, use the method for color cluster, the domain analysis of homochromy UNICOM.After analyzing end by above method, object information is merged, (character zone the ratio of width to height is between 6: 1 to 7: 1 to add prior imformation, the coding character color is a black), all these information are merged, finally obtain the result of decision, promptly obtain character zone.
This binaryzation is that the gray level image in the character zone is carried out binary conversion treatment, and we adopt maximum variance between clusters.Be inter-class variance and the method for class internal variance ratio, self-adaptation is calculated gray threshold, thinks the target area less than the zone of this threshold value, thinks the background area greater than the zone of this threshold value.
It is because the problem of camera shooting angle may be brought the little inclination of some angles that this inclination is corrected, and the result of character recognition is affected, and we have carried out self-adaptation to bianry image and have tilted to correct.Promptly obtain the angle of inclination of character zone, adopt affined transformation to correct according to the angle of inclination of character zone at last with the hough conversion.
This Character segmentation is to adopt regional area minimum value method to cut apart, and character is carried out vertical projection, and minimum place is cut apart in the vertical projection value of character adhesion scope.
Character feature is promptly extracted in this feature extraction.The feature extraction mode that native system adopts statistical nature and architectural feature to combine is in the hope of obtaining classifying quality preferably.Character picture is divided into 4 * 4 grid, adds up the ratio that black picture element in each zone accounts for this zone total pixel number, form 16 dimension statistical natures; Character picture respectively along level and vertical direction projection, is asked the peak of projection, form 2 dimension architectural features, two kinds of characteristics combination are formed 18 dimensional features as character feature.
Should preliminary identification be to adopt the support vector machine sorting algorithm, letter and number was set up the SVM model of cognition respectively.With the digit recognition is example, and model can be described as: given training set is input as 18 dimension sample characteristics inputs, be output as 10 kinds of possible outputs,, obtain decision function by sample training, make to draw corresponding output result for test sample book, with this as preliminary recognition result.
This secondary identification is on preliminary base of recognition, and character is discerned once more.Test is found, for some similar characters, is difficult to once discern successfully, need discern once more.Mainly some similar characters are discerned once more in conjunction with the specific structural features of character, such as Aspect Ratio, regional closure etc., with this recognition result as final recognition result.After secondary identification, can effectively raising character identification rate.
The recognizer flow process at first, is carried out pre-service to gathering the image that comes as shown in Figure 6, comprises image gray processing, removes noise, grey level stretching; Carry out the character zone location then, the zone location at character place is come out; Next character zone is carried out binary conversion treatment, obtain bianry image; Carry out slant correction then; Image after proofreading and correct is carried out Character segmentation, whole character zone is divided into single character; Each character is carried out feature extraction; The preliminary identification of advanced line character; At last by character like the phase-splitting of secondary cog region; Each character is carried out feature extraction, tentatively identification and secondary identification, get recognition result to the end.
When identifier is worked, at first manually control button, the laser coding character picture that will have complex background by image acquisition units 4 is gathered up; Then the image that collects is carried out pre-service, remove noise; The method that adopts coloured image and gray level image to combine then locatees out with character zone; And then, image is carried out the binaryzation operation; The character integral body that tilts is carried out slant correction; Then whole character zone is divided into a plurality of independent characters; Each independent character is carried out feature extraction, carry out the preliminary identification of character; At last, character similar, that obscure is easily carried out secondary identification, obtain final recognition result.
Obtain after the recognition result by recognizer, by serial ports, control GPRS communication unit 5 sends to specified server with recognition result, finishes the repertoire of whole character recognizing instrument.
Claims (1)
1. character recognizing instrument is characterized in that it comprises:
Image acquisition units, graphics processing unit, data storage cell, liquid crystal display, GPRS communication unit, key operation unit and power module are formed;
Power module is responsible for this identifier various piece power supply; Key operation unit and liquid crystal display are man-machine interfaces, are responsible for the input of operational order and result's output and show; Image acquisition units is responsible for images acquired; The GPRS communication unit is responsible for the wireless transmission of recognition result, whole process is: the user gives an order by key operation unit, processor receives order, control image acquisition units acquisition of image data, store by data storage cell, carry out image recognition then, the result that will discern sends to given server by the GPRS communication unit at last, simultaneously the result is stored in the data storage cell, be convenient to inquire about in the future;
Described image acquisition units is a high integration, high-resolution CMOS sensing chip OV9650, its inner integrated sequential circuit, analog signal processing circuit, digital signal processing circuit; The Serial Camera Control Bus that it has standard is the SCCB interface, can realize various figure image intensifyings and control function by this interface, i.e. automatic exposure, automatic gain, balance Control Scheme and control image color, saturation degree, gamma correction, sharpening, camera lens calibration, noise and white pixel deletion;
Described graphics processing unit is to adopt Samsung ARM9 S3C2440 processor;
Described data storage cell is to adopt the K9F1208 chip of a slice Samsung and the HY57V561620 chip of two Samsungs;
Described liquid crystal display is selected 3.5 cun TFT touch-screen LQ035Q7DH01 of Sharp for use; Described GPRS communication unit adopts the sim300 module;
Described key operation unit is individual 4 * 4 matrix keyboard;
Described power module is to adopt outside 5V power supply.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009202470504U CN201654804U (en) | 2009-11-17 | 2009-11-17 | Character recognizer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009202470504U CN201654804U (en) | 2009-11-17 | 2009-11-17 | Character recognizer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201654804U true CN201654804U (en) | 2010-11-24 |
Family
ID=43120047
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009202470504U Expired - Fee Related CN201654804U (en) | 2009-11-17 | 2009-11-17 | Character recognizer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201654804U (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102819619A (en) * | 2011-06-10 | 2012-12-12 | 上海市电力公司 | Device for simulating streams of people through BIM (building information modeling) |
| CN104484867A (en) * | 2014-12-30 | 2015-04-01 | 小米科技有限责任公司 | Picture processing method and device |
| CN105654083A (en) * | 2015-12-23 | 2016-06-08 | 天津天地伟业数码科技有限公司 | Wide-range license plate inclination angle rapid calculation method |
| CN104598903B (en) * | 2015-02-04 | 2018-11-06 | 浙江科技学院 | The adaptive character recognition device of object temperature and method |
| CN111582262A (en) * | 2020-05-07 | 2020-08-25 | 京源中科科技股份有限公司 | Segment type liquid crystal picture content identification method, device, equipment and storage medium |
| CN113780038A (en) * | 2020-06-10 | 2021-12-10 | 深信服科技股份有限公司 | Picture auditing method and device, computing equipment and storage medium |
-
2009
- 2009-11-17 CN CN2009202470504U patent/CN201654804U/en not_active Expired - Fee Related
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102819619A (en) * | 2011-06-10 | 2012-12-12 | 上海市电力公司 | Device for simulating streams of people through BIM (building information modeling) |
| CN104484867A (en) * | 2014-12-30 | 2015-04-01 | 小米科技有限责任公司 | Picture processing method and device |
| CN104598903B (en) * | 2015-02-04 | 2018-11-06 | 浙江科技学院 | The adaptive character recognition device of object temperature and method |
| CN105654083A (en) * | 2015-12-23 | 2016-06-08 | 天津天地伟业数码科技有限公司 | Wide-range license plate inclination angle rapid calculation method |
| CN105654083B (en) * | 2015-12-23 | 2019-05-03 | 天津天地伟业物联网技术有限公司 | A kind of license plate sloped angle quick calculation method of fixed range |
| CN111582262A (en) * | 2020-05-07 | 2020-08-25 | 京源中科科技股份有限公司 | Segment type liquid crystal picture content identification method, device, equipment and storage medium |
| CN113780038A (en) * | 2020-06-10 | 2021-12-10 | 深信服科技股份有限公司 | Picture auditing method and device, computing equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101706875B (en) | Hand-held laser code-jetting character recognizer under complex background | |
| CN201654804U (en) | Character recognizer | |
| CN106650780A (en) | Data processing method, device, classifier training method and system | |
| CN110458095A (en) | A kind of recognition methods, control method, device and the electronic equipment of effective gesture | |
| CN101853071A (en) | Gesture identification method and system based on visual sense | |
| CN104102912A (en) | Sub-item metering energy consumption data collection device based on video image identification and method thereof | |
| CN113705554A (en) | Training method, device and equipment of image recognition model and storage medium | |
| CN103488415A (en) | Handwritten text and graphic recognition system based on paper medium and recognition method | |
| CN110866900A (en) | Water body color identification method and device | |
| CN103105924A (en) | Man-machine interaction method and device | |
| CN109241962A (en) | A kind of character identifying method and device | |
| CN101192124B (en) | System and method for automatic distinguishing and processing for touch screen input information | |
| Chen et al. | Multi-objective solid waste classification and identification model based on transfer learning method | |
| CN101893960B (en) | Direction vector-based character recognition method and device | |
| CN109492573A (en) | A kind of pointer read method and device | |
| Zhang et al. | License plate recognition model based on CNN+ LSTM+ CTC | |
| Sharma et al. | Automatic number plate recognition and parking management | |
| Gao et al. | Chinese character components segmentation method based on faster RCNN | |
| Bao et al. | A new approach to hand tracking and gesture recognition by a new feature type and HMM | |
| CN101901348A (en) | Normalization based handwriting identifying method and identifying device | |
| CN102541347A (en) | Automatic identification system and method of handwriting Chinese character | |
| CN108681736A (en) | A kind of embedded vehicle license plate identifying system based on ARM microprocessor | |
| Liu et al. | Reagent color recognition model for Android platform based on opencv and machine learning | |
| CN113673527A (en) | License plate recognition method and system | |
| Fazheng et al. | Research on location of chinese handwritten signature based on efficientdet |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20101124 Termination date: 20121117 |