CN105631450A - Character identifying method and device - Google Patents
Character identifying method and device Download PDFInfo
- Publication number
- CN105631450A CN105631450A CN201511001283.2A CN201511001283A CN105631450A CN 105631450 A CN105631450 A CN 105631450A CN 201511001283 A CN201511001283 A CN 201511001283A CN 105631450 A CN105631450 A CN 105631450A
- Authority
- CN
- China
- Prior art keywords
- character
- combination
- character image
- region
- confidence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Character Discrimination (AREA)
- Character Input (AREA)
Abstract
The disclosure relates to a character identifying method and device. The method includes the steps of: performing character identification on a character image and obtaining a character identification result; determining a target character with the confidence degree less than or equal to a first preset threshold from the character identification result; cutting a character image region where the target character is, and obtaining at least two character image blocks each including a unit character; and performing character identification again on the character image region according to at least two character image blocks. The technology obtains the character image blocks which cannot be segmented further by performing careful cutting on the character image region, obtains the character combination with the high confidence degree through combination treatment on the unit character in the character image blocks, and determines the character combination with the high confidence degree as the character identification result of the character image region. The technology improves the accuracy of the character identification result.
Description
Technical field
It relates to character recognition technologies field, particularly relate to character identifying method and device.
Background technology
At present, character recognition technologies has been used in various fields. When carrying out the tasks such as such as ID card information extraction, may not be accurate for various reasons Character segmentation ground, there will be some mistakes, the radical of a front word can have been switched in next word by common mistake, such as " river " word, when cutting, it is easy to the perpendicular of the right to switch in character below, thus the accuracy causing character identification result is lower.
Summary of the invention
Disclosure embodiment provides character identifying method and device. Described technical scheme is as follows:
First aspect according to disclosure embodiment, it is provided that a kind of character identifying method, comprising:
Character image is carried out character recognition, obtains character identification result;
From described character identification result, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Being cut by the character image-region at described target character place, obtain at least two character image blocks, described in each, character image block comprises a unit character;
According to described at least two character image blocks, described character image-region is re-started character recognition.
In an embodiment, the described character image-region to described target character place cuts, and comprising:
Obtain the character arrangements direction of described target character;
Using described character arrangements direction as cutting direction, the character image-region at described target character place is cut.
In an embodiment, described in described basis, described character image-region is re-started character recognition by least two character image blocks, comprising:
Respectively described at least two character image blocks are carried out character recognition, the unit character that at least two character image blocks described in acquisition are included separately;
According to described at least two character image block putting in order in described character image-region, the unit character that described at least two character image blocks are included separately is arranged;
Described unit character after arrangement is carried out combined treatment, obtains at least two group character combinations;
From described at least two group character combinations, it is determined that go out degree of confidence and meet pre-conditioned character combination;
Described degree of confidence is met the character identification result that pre-conditioned character combination is defined as described character image-region.
In an embodiment, described degree of confidence meets pre-conditioned character combination, comprises the first character combination following or the 2nd kind of character combination:
The first character combination comprises: combination each character interior degree of confidence separately is all greater than the character combination of the 2nd predetermined threshold value; Wherein, described 2nd predetermined threshold value is more than or equal to the first predetermined threshold value;
2nd kind of character combination comprises: in described at least two group character combinations, the character combination that the interior all characters of combination degree of confidence sum separately is the highest.
In an embodiment, when described target character comprises Chinese character, described unit character comprises radical radicals by which characters are arranged in traditional Chinese dictionaries or single character.
Second aspect according to disclosure embodiment, it is provided that a kind of character recognition device, comprising:
First identification module, for character image is carried out character recognition, obtains character identification result;
Determination module, for identifying in the described character identification result that module obtains from described first, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Cutting module, the character image-region for the described target character place determined by described determination module cuts, and obtains at least two character image blocks, and described in each, character image block comprises a unit character;
2nd identification module, re-starts character recognition at least two character image blocks described in obtaining according to described cutting module to described character image-region.
In an embodiment, described cutting module comprises:
Obtain submodule block, for obtaining the character arrangements direction of described target character;
Cutting submodule block, for the described character arrangements direction that got by described acquisition submodule block as cutting direction, cuts the character image-region at described target character place.
In an embodiment, described 2nd identification module, comprising:
Recognin module, for respectively described at least two character image blocks being carried out character recognition, the unit character that at least two character image blocks described in acquisition are included separately;
Sorting sub-module, for according to described at least two character image block putting in order in described character image-region, arranging the unit character that described at least two character image blocks are included separately;
Combination submodule block, for the described unit character after described arrangement is carried out combined treatment, obtains at least two group character combinations;
First true stator modules, for from described at least two group character combinations, it is determined that go out degree of confidence and meet pre-conditioned character combination;
2nd true stator modules, for meeting the character identification result that pre-conditioned character combination is defined as described character image-region by described degree of confidence.
In an embodiment, described degree of confidence meets pre-conditioned character combination, comprises the first character combination following or the 2nd kind of character combination:
The first character combination comprises: combination each character interior degree of confidence separately is all greater than the character combination of the 2nd predetermined threshold value; Wherein, described 2nd predetermined threshold value is more than or equal to the first predetermined threshold value;
2nd kind of character combination comprises: in described at least two group character combinations, the character combination that the interior all characters of combination degree of confidence sum separately is the highest.
In an embodiment, when described target character comprises Chinese character, described unit character comprises radical radicals by which characters are arranged in traditional Chinese dictionaries or single character.
The third aspect according to disclosure embodiment, it provides a kind of character recognition device, comprising:
Treater;
The storer of instruction can be performed for storage of processor;
Wherein, described treater is configured to:
Character image is carried out character recognition, obtains character identification result;
From described character identification result, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Being cut by the character image-region at described target character place, obtain at least two character image blocks, described in each, character image block comprises a unit character;
According to described at least two character image blocks, described character image-region is re-started character recognition.
The technical scheme that embodiment of the present disclosure provides can comprise following useful effect:
The technical scheme that embodiment of the present disclosure provides, by character image-region is carried out careful cutting, the character image block that acquisition cannot be split further again, again by the combined treatment to unit character in character image block, obtain the character combination that degree of confidence is higher, character combination higher for degree of confidence is defined as the character identification result of character image-region. This technique improves the accuracy of character identification result.
Should be understood that, it is only exemplary and explanatory that above general description and details hereinafter describe, and can not limit the disclosure.
Accompanying drawing explanation
Accompanying drawing herein is by being incorporated in specification sheets and forms the part of this specification sheets, shows and meets embodiment of the present disclosure, and is used from specification sheets one and explains principle of the present disclosure.
Fig. 1 is the schema of a kind of character identifying method according to an exemplary embodiment.
Fig. 2 is the schematic diagram of a kind of character image according to an exemplary embodiment.
Fig. 3 is the schematic diagram of another kind of character image according to an exemplary embodiment.
Fig. 4 is the schema of another kind of character identifying method according to an exemplary embodiment.
Fig. 5 is the block diagram of a kind of character recognition device according to an exemplary embodiment.
Fig. 6 is the block diagram of a kind of character recognition device according to an exemplary embodiment.
Fig. 7 is the block diagram being applicable to character recognition device according to an exemplary embodiment.
Embodiment
Here exemplary embodiment being described in detail, its example representation is in the accompanying drawings. When description below relates to accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawing represents same or similar key element. Enforcement mode described in exemplary embodiment does not represent all enforcement modes consistent with the disclosure below. On the contrary, they only with as in appended claims describe in detail, the example of device that aspects more of the present disclosure are consistent and method.
Disclosure embodiment provides character recognition technologies, this technology is by carrying out careful cutting to character image-region, the character image block that acquisition cannot be split further again, again by the combined treatment to unit character in character image block, obtain the character combination that degree of confidence is higher, character combination higher for degree of confidence is defined as the character identification result of character image-region. This technique improves the accuracy of character identification result.
Fig. 1 shows according to a kind of character identifying method that disclosure embodiment provides, and the method may be used in the application program for identifying character, equipment etc., and as shown in Figure 1, the method can comprise step S101-S104:
In step S101, character image is carried out character recognition, obtain character identification result.
Wherein, step S101 can adopt existing character recognition technologies to realize. Character recognition technologies is first scanned by the character carrier having character, obtains character image, then character image is carried out character recognition, obtains character identification result.
In step s 102, from character identification result, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value.
Wherein, the degree of confidence of each character refers to the accuracy of each character in character identification result, it is possible to adopt the method for the existing degree of confidence determining the character identified to determine. First predetermined threshold value can pre-set, if the span of degree of confidence is [0,1], then the first predetermined threshold value should be greater than 0 and be less than 1, such as, can be 0.5,0.4; Concrete value is how many, can set as required.
In step s 103, being cut by the character image-region at target character place, obtain at least two character image blocks, each character image block comprises a unit character. Unit character does not comprise neither radical radicals by which characters are arranged in traditional Chinese dictionaries, does not form again the text structure of independent Chinese character.
In an embodiment, target character can be the different spoken and written languages such as Chinese character, Korean, Japanese, English. When target character is Chinese character, the character image-region at target character place may comprise one or more unit character, unit character refers to the minimum unit of composition Chinese character, can be radical radicals by which characters are arranged in traditional Chinese dictionaries or single character, wherein, single character refers to the Chinese character that can not split out radical radicals by which characters are arranged in traditional Chinese dictionaries again, such as " my god ", " cutter " etc.; And the Chinese character combined by radical radicals by which characters are arranged in traditional Chinese dictionaries, single character, radical radicals by which characters are arranged in traditional Chinese dictionaries can be split out, in Chinese literature, this kind of word is combinde rqdical character usually, such as " body ", " phase " etc.
Visible, this step is that the character image-region to target character place carries out more meticulous cutting, obtains the character image block that cannot cut further again.
In an embodiment, step S103 can be embodied as: the character arrangements direction obtaining target character; Using character arrangements direction as cutting direction, the character image-region at target character place is cut.
Citing below is described: such as, after a character image is carried out Text region, obtaining its Word message is " we to be drawn functional block diagram to the other side ", calculate the degree of confidence of each character, assume that the respective degree of confidence of its 6th character " former ", the 7th character " reason ", the 8th these three character of character " figure " is all less than the first predetermined threshold value, then the character image-region at these three character places is cut. As shown in Figure 2, label 20 is referred to as character image, if these three words from left to right arrange in character image-region 20, then cutting direction is also from left to right, and cutting result is " former ", " king ", " inner ", " wood ", " rectifying ", " figure ". Again such as, after a character image is carried out Text region, obtaining its Word message is " place placing star ", calculate the degree of confidence of each character, assume that the respective degree of confidence of its 2nd character " putting ", the 3rd character " star ", the 4th these three character of character " star " is all less than the first predetermined threshold value, then the character image-region at these three character places is cut. As shown in Figure 3, label 30 is referred to as character image, if these three words arrange from top to bottom in character image-region 30, then cutting direction is also from top to bottom, and cutting result is " four ", " directly ", " day ", " life ", " day ", " life ".
In step S104, according at least two character image blocks, character image-region is re-started character recognition.
In an embodiment, as shown in Figure 4, step S104 can be embodied as following step S401-S405:
In step S401, respectively at least two character image blocks are carried out character recognition, obtain the unit character that at least two character image blocks are included separately.
In step S402, according at least two character image block putting in order in character image-region, the unit character that at least two character image blocks are included separately is arranged.
In step S403, the unit character after arrangement is carried out combined treatment, obtain at least two group character combinations.
In step s 404, from least two group character combinations, it is determined that go out degree of confidence and meet pre-conditioned character combination.
Wherein, degree of confidence meets pre-conditioned character combination, refers to any one character combination following, can also be the character combination of other types certainly, all within disclosure protection domain:
The first character combination: combination each character interior degree of confidence separately is all greater than the character combination of the 2nd predetermined threshold value; Wherein, the 2nd predetermined threshold value is more than or equal to the first predetermined threshold value. That is, the recognition result determined after the character image-region at target character place again being identified, its degree of confidence is higher than the degree of confidence of target character so that again identify meaningful.
2nd kind of character combination: in above-mentioned at least two group character combinations, the character combination that the interior all characters of combination degree of confidence sum separately is the highest. That is, using the again recognition result of character combination the highest for total degree of confidence as the character image-region at target character place. This kind be recognition result again, can think roughly that the average confidence of each character in character combination is higher, thus also can improve again the accuracy of recognition result.
When reality implements such scheme, if it is determined that do not go out the first character combination, that is, if not having a kind of character combination to be that the degree of confidence of each character is all greater than the 2nd predetermined threshold value, then the 2nd kind of character combination can be met pre-conditioned character combination as above-mentioned degree of confidence. That is, in prioritizing selection above-mentioned first, character combination is that above-mentioned degree of confidence meets pre-conditioned character combination. This is owing to character group credit union in first makes again recognition result more accurate.
In step S405, above-mentioned degree of confidence is met the character identification result that pre-conditioned character combination is defined as character image-region.
In an embodiment, utilize the character identification result of this character image-region, replace in abovementioned steps S101 the target character in the character identification result obtained, thus the final character identification result that can make character image is more accurate.
For above-mentioned steps S401-S405, continue to adopt citing shown in accompanying drawing 2 to explain: the cutting result of the character image-region at the place of target character shown in Fig. 2 is " former ", " king ", " inner ", " wood ", " rectifying ", " figure ", to each the unit character in this cutting result, carry out combination possible arbitrarily successively along order from left to right, form following character combination:
Calculate in above-mentioned six groups of character combinations respectively, often organizing the degree of confidence of each character in character combination, wherein, first group of character combination " principle frame " is the first character combination above-mentioned or the 2nd kind of character combination, therefore, can using the character identification result of this character combination as character image-region.
The aforesaid method that corresponding disclosure embodiment provides, as shown in Figure 5, disclosure embodiment provides a kind of character recognition device, and this device comprises:
First identification module 51, is configured to character image is carried out character recognition, obtains character identification result;
Determination module 52, is configured to from the character identification result that the first identification module 51 obtains, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Cutting module 53, the character image-region at target character place being configured to be determined by determination module 52 cuts, and obtains at least two character image blocks, and each character image block comprises a unit character;
2nd identification module 54, is configured at least two character image blocks according to cutting module 53 obtains and character image-region is re-started character recognition.
In an embodiment, as shown in Figure 6, cutting module 53 can comprise:
Obtain submodule block 61, it is configured to obtain the character arrangements direction of target character;
Cutting submodule block 62, the character image-region at target character place, as cutting direction, is cut by the character arrangements direction being configured to get acquisition submodule block.
In an embodiment, more as shown in Figure 6, the 2nd identification module 54 can comprise:
Recognin module 63, is configured to respectively at least two character image blocks be carried out character recognition, obtains the unit character that at least two character image blocks are included separately;
Sorting sub-module 64, is configured to according at least two character image block putting in order in character image-region, is arranged by the unit character that at least two character image blocks are included separately;
Combination submodule block 65, is configured to the unit character after arrangement is carried out combined treatment, obtains at least two group character combinations;
First true stator modules 66, is configured to from least two group character combinations, it is determined that go out degree of confidence and meet pre-conditioned character combination;
2nd true stator modules 67, is configured to degree of confidence meets the character identification result that pre-conditioned character combination is defined as character image-region.
In an embodiment, above-mentioned degree of confidence meets pre-conditioned character combination, comprises the first character combination following or the 2nd kind of character combination:
The first character combination comprises: combination each character interior degree of confidence separately is all greater than the character combination of the 2nd predetermined threshold value; Wherein, the 2nd predetermined threshold value is more than or equal to the first predetermined threshold value;
2nd kind of character combination comprises: at least two group character combinations, the character combination that the interior all characters of combination degree of confidence sum separately is the highest.
In an embodiment, when target character comprises Chinese character, unit character comprises radical radicals by which characters are arranged in traditional Chinese dictionaries or single character.
The aforesaid method that corresponding disclosure embodiment provides, disclosure embodiment provides a kind of character recognition device, comprising:
Treater;
The storer of instruction can be performed for storage of processor;
Wherein, treater is configured to:
Character image is carried out character recognition, obtains character identification result;
From character identification result, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Being cut by the character image-region at target character place, obtain at least two character image blocks, each character image block comprises a unit character;
According at least two character image blocks, character image-region is re-started character recognition.
The technical scheme that embodiment of the present disclosure provides can comprise following useful effect:
Technique scheme, by character image-region is carried out careful cutting, the character image block that acquisition cannot be split further again, again by the combined treatment to unit character in character image block, obtain the character combination that degree of confidence is higher, character combination higher for degree of confidence is defined as the character identification result of character image-region. This technique improves the accuracy of character identification result.
Fig. 7 is a kind of block diagram for character recognition device 800 according to an exemplary embodiment. Such as, device 800 can be mobile telephone, computer, digital broadcast terminal, messaging devices, game console, tablet device, medical facilities, body-building equipment, personal digital assistant etc.
With reference to Fig. 7, device 800 can comprise following one or more assembly: processing components 802, storer 804, power supply module 806, multimedia groupware 808, audio-frequency assembly 810, the interface 812 of I/O (I/O), sensor module 814, and communications component 816.
The overall operation of the usual control device 800 of processing components 802, such as with display, the operation that telephone call, data corresponding, camera operation and recording operation are associated. Processing element 802 can comprise one or more treater 820 to perform instruction, to complete all or part of step of above-mentioned method. In addition, processing components 802 can comprise one or more module, and what be convenient between processing components 802 and other assemblies is mutual. Such as, processing element 802 can comprise multi-media module, mutual with what facilitate between multimedia groupware 808 and processing components 802.
Storer 804 is configured to store various types of data to be supported in the operation of equipment 800. The example of these data comprises for any application program of operation on device 800 or the instruction of method, contact data, telephone book data, message, picture, video etc. Storer 804 can be realized by the volatibility of any type or non-volatile memory device or their combination, such as static RAM (SRAM), electrically erasable read-only storage (EEPROM), erasable programmable read-only storage (EPROM), programmable read only memory (PROM), read-only storage (ROM), magneticstorage, flash device, disk or CD.
The various assembly that electric power assembly 806 is device 800 provides electric power. Electric power assembly 806 can comprise power-supply management system, one or more power supply, and other generate, manage and distribute, with for device 800, the assembly that electric power is associated.
The screen that an output interface is provided that multimedia groupware 808 is included between described device 800 and user. In certain embodiments, screen can comprise liquid-crystal display (LCD) and touch panel (TP). If screen comprises touch panel, screen may be implemented as touch-screen, to receive the input signal from user. Touch panel comprises one or more touch sensing device with the gesture on sensing touch, slip and touch panel. Described touch sensing device can the border of not only sensing touch or sliding action, but also the detection time length relevant to described touch or slide and pressure. In certain embodiments, multimedia groupware 808 comprises a front-facing camera and/or rearmounted camera. When equipment 800 is in operator scheme, during such as screening-mode or video pattern, front-facing camera and/or rearmounted camera can receive outside multi-medium data. Each front-facing camera and rearmounted camera can be a fixing optical lens system or have focal length and optical zoom ability.
Audio-frequency assembly 810 is configured to export and/or input audio signal. Such as, audio-frequency assembly 810 comprises a microphone (MIC), and when device 800 is in operator scheme, during such as calling pattern, record pattern and speech recognition pattern, microphone is configured to receive external audio signal. The sound signal received can be stored in storer 804 further or be sent via communications component 816. In certain embodiments, audio-frequency assembly 810 also comprises a loud speaker, for output audio signal.
I/O interface 812 is for providing interface between processing components 802 and peripheral interface module, and above-mentioned peripheral interface module can be keyboard, some striking wheel, button etc. These buttons can include but not limited to: home button, volume button, startup button and locking button.
Sensor module 814 comprises one or more sensor, for providing the state estimation of all respects for device 800. Such as, sensor module 814 can detect the opening/closing state of equipment 800, the relative location of assembly, such as described assembly is indicating meter and the keypad of device 800, the position that sensor module 814 can also detect device 800 or device 800 1 assemblies changes, the presence or absence that user contacts with device 800, the temperature variation of device 800 orientation or acceleration/deceleration and device 800. Sensor module 814 can comprise close to sensor, be configured to without any physical contact time detection near the existence of object. Sensor module 814 can also comprise optical sensor, such as CMOS or ccd image sensor, for using in imaging applications. In certain embodiments, this sensor module 814 can also comprise acceleration transducer, gyro sensor, Magnetic Sensor, pressure transmitter or temperature sensor.
Communications component 816 is configured to be convenient to the communication of wired or wireless mode between device 800 and other equipment. Device 800 can access the wireless network based on communication standard, such as WiFi, 2G or 3G, or their combination. In an exemplary embodiment, communication component 816 receives the broadcast signal from outside broadcasting management systems or broadcast related information via broadcast channel. In an exemplary embodiment, described communication component 816 also comprises near-field communication (NFC) module, to promote short distance communication. Such as, can based on RF identification (RFID) technology in NFC module, Infrared Data Association (IrDA) technology, ultra broadband (UWB) technology, bluetooth (BT) technology and other technologies realize.
In the exemplary embodiment, device 800 can be realized by one or more application specific unicircuit (ASIC), digital signal processor (DSP), digital signal processing appts (DSPD), programmable logic device part (PLD), field-programmable gate array (FPGA), controller, microcontroller, microprocessor or other electronic components, for performing aforesaid method.
In the exemplary embodiment, additionally providing a kind of non-transitory computer-readable recording medium comprising instruction, such as, comprise the storer 804 of instruction, above-mentioned instruction can perform aforesaid method by the treater 820 of device 800. Such as, described non-transitory computer-readable recording medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and optical data storage equipment etc.
A kind of non-transitory computer-readable recording medium, when the instruction in described storage media is performed by the treater of mobile terminal so that mobile terminal can perform a kind of character identifying method, and described method comprises:
Character image is carried out character recognition, obtains character identification result;
From described character identification result, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Being cut by the character image-region at described target character place, obtain at least two character image blocks, described in each, character image block comprises a unit character;
According to described at least two character image blocks, described character image-region is re-started character recognition.
In an embodiment, the described character image-region to described target character place cuts, and comprising:
Obtain the character arrangements direction of described target character;
Using described character arrangements direction as cutting direction, the character image-region at described target character place is cut.
In an embodiment, described in described basis, described character image-region is re-started character recognition by least two character image blocks, comprising:
Respectively described at least two character image blocks are carried out character recognition, the unit character that at least two character image blocks described in acquisition are included separately;
According to described at least two character image block putting in order in described character image-region, the unit character that described at least two character image blocks are included separately is arranged;
Described unit character after arrangement is carried out combined treatment, obtains at least two group character combinations;
From described at least two group character combinations, it is determined that go out degree of confidence and meet pre-conditioned character combination;
Described degree of confidence is met the character identification result that pre-conditioned character combination is defined as described character image-region.
In an embodiment, described degree of confidence meets pre-conditioned character combination, comprises the first character combination following or the 2nd kind of character combination:
The first character combination comprises: combination each character interior degree of confidence separately is all greater than the character combination of the 2nd predetermined threshold value; Wherein, described 2nd predetermined threshold value is more than or equal to the first predetermined threshold value;
2nd kind of character combination comprises: in described at least two group character combinations, the character combination that the interior all characters of combination degree of confidence sum separately is the highest.
In an embodiment, when described target character comprises Chinese character, described unit character comprises radical radicals by which characters are arranged in traditional Chinese dictionaries or single character.
Those skilled in the art, after considering specification sheets and putting into practice disclosed herein disclosing, will easily expect other embodiment of the present disclosure. The application is intended to contain any modification of the present disclosure, purposes or adaptations, and these modification, purposes or adaptations are followed general principle of the present disclosure and comprised the unexposed common practise in the art of the disclosure or conventional techniques means. Specification sheets and embodiment are only regarded as exemplary, and true scope of the present disclosure and spirit are pointed out by claim below.
Should be understood that, the disclosure is not limited to accurate structure described above and illustrated in the accompanying drawings, and can carry out various amendment and change not departing from its scope. The scope of the present disclosure is only limited by appended claim.
Claims (11)
1. a character identifying method, it is characterised in that, comprising:
Character image is carried out character recognition, obtains character identification result;
From described character identification result, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Being cut by the character image-region at described target character place, obtain at least two character image blocks, described in each, character image block comprises a unit character;
According to described at least two character image blocks, described character image-region is re-started character recognition.
2. the method for claim 1, it is characterised in that, the described character image-region to described target character place cuts, and comprising:
Obtain the character arrangements direction of described target character;
Using described character arrangements direction as cutting direction, the character image-region at described target character place is cut.
3. the method for claim 1, it is characterised in that, described in described basis, described character image-region is re-started character recognition by least two character image blocks, comprising:
Respectively described at least two character image blocks are carried out character recognition, the unit character that at least two character image blocks described in acquisition are included separately;
According to described at least two character image block putting in order in described character image-region, the unit character that described at least two character image blocks are included separately is arranged;
Described unit character after arrangement is carried out combined treatment, obtains at least two group character combinations;
From described at least two group character combinations, it is determined that go out degree of confidence and meet pre-conditioned character combination;
Described degree of confidence is met the character identification result that pre-conditioned character combination is defined as described character image-region.
4. method as claimed in claim 3, it is characterised in that, described degree of confidence meets pre-conditioned character combination, comprises the first character combination following or the 2nd kind of character combination:
The first character combination comprises: combination each character interior degree of confidence separately is all greater than the character combination of the 2nd predetermined threshold value; Wherein, described 2nd predetermined threshold value is more than or equal to the first predetermined threshold value;
2nd kind of character combination comprises: in described at least two group character combinations, the character combination that the interior all characters of combination degree of confidence sum separately is the highest.
5. the method for claim 1, it is characterised in that,
When described target character comprises Chinese character, described unit character comprises radical radicals by which characters are arranged in traditional Chinese dictionaries or single character.
6. a character recognition device, it is characterised in that, comprising:
First identification module, for character image is carried out character recognition, obtains character identification result;
Determination module, for identifying in the described character identification result that module obtains from described first, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Cutting module, the character image-region for the described target character place determined by described determination module cuts, and obtains at least two character image blocks, and described in each, character image block comprises a unit character;
2nd identification module, re-starts character recognition at least two character image blocks described in obtaining according to described cutting module to described character image-region.
7. device as claimed in claim 6, it is characterised in that, described cutting module comprises:
Obtain submodule block, for obtaining the character arrangements direction of described target character;
Cutting submodule block, for the described character arrangements direction that got by described acquisition submodule block as cutting direction, cuts the character image-region at described target character place.
8. device as claimed in claim 6, it is characterised in that, described 2nd identification module, comprising:
Recognin module, for respectively described at least two character image blocks being carried out character recognition, the unit character that at least two character image blocks described in acquisition are included separately;
Sorting sub-module, for according to described at least two character image block putting in order in described character image-region, arranging the unit character that described at least two character image blocks are included separately;
Combination submodule block, for the described unit character after described arrangement is carried out combined treatment, obtains at least two group character combinations;
First true stator modules, for from described at least two group character combinations, it is determined that go out degree of confidence and meet pre-conditioned character combination;
2nd true stator modules, for meeting the character identification result that pre-conditioned character combination is defined as described character image-region by described degree of confidence.
9. device as claimed in claim 8, it is characterised in that, described degree of confidence meets pre-conditioned character combination, comprises the first character combination following or the 2nd kind of character combination:
The first character combination comprises: combination each character interior degree of confidence separately is all greater than the character combination of the 2nd predetermined threshold value; Wherein, described 2nd predetermined threshold value is more than or equal to the first predetermined threshold value;
2nd kind of character combination comprises: in described at least two group character combinations, the character combination that the interior all characters of combination degree of confidence sum separately is the highest.
10. device as claimed in claim 6, it is characterised in that,
When described target character comprises Chinese character, described unit character comprises radical radicals by which characters are arranged in traditional Chinese dictionaries or single character.
11. 1 kinds of character recognition devices, it is characterised in that, comprising:
Treater;
The storer of instruction can be performed for storage of processor;
Wherein, described treater is configured to:
Character image is carried out character recognition, obtains character identification result;
From described character identification result, it is determined that go out the target character that degree of confidence is less than or equals the first predetermined threshold value;
Being cut by the character image-region at described target character place, obtain at least two character image blocks, described in each, character image block comprises a unit character;
According to described at least two character image blocks, described character image-region is re-started character recognition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511001283.2A CN105631450A (en) | 2015-12-28 | 2015-12-28 | Character identifying method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511001283.2A CN105631450A (en) | 2015-12-28 | 2015-12-28 | Character identifying method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105631450A true CN105631450A (en) | 2016-06-01 |
Family
ID=56046363
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511001283.2A Pending CN105631450A (en) | 2015-12-28 | 2015-12-28 | Character identifying method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105631450A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106203425A (en) * | 2016-07-01 | 2016-12-07 | 北京旷视科技有限公司 | Character identifying method and device |
| CN110334711A (en) * | 2019-06-04 | 2019-10-15 | 新华三技术有限公司 | Character match method and device |
| CN110555441A (en) * | 2019-09-10 | 2019-12-10 | 杭州橙鹰数据技术有限公司 | character recognition method and device |
| CN111630521A (en) * | 2018-02-28 | 2020-09-04 | 佳能欧洲股份有限公司 | Image processing method and image processing system |
| CN111652225A (en) * | 2020-04-29 | 2020-09-11 | 浙江省北大信息技术高等研究院 | Non-invasive camera reading method and system based on deep learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101251892A (en) * | 2008-03-07 | 2008-08-27 | 北大方正集团有限公司 | Method and apparatus for cutting character |
| US20090324081A1 (en) * | 2008-06-24 | 2009-12-31 | Samsung Electronics Co., Ltd. | Method and apparatus for recognizing character in character recognizing apparatus |
| CN101650779A (en) * | 2008-08-11 | 2010-02-17 | 欧姆龙株式会社 | Character recognition device, program and method |
| CN103870799A (en) * | 2012-12-17 | 2014-06-18 | 北京千橡网景科技发展有限公司 | Character direction judging method and device |
| CN104951741A (en) * | 2014-03-31 | 2015-09-30 | 阿里巴巴集团控股有限公司 | Character recognition method and device thereof |
-
2015
- 2015-12-28 CN CN201511001283.2A patent/CN105631450A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101251892A (en) * | 2008-03-07 | 2008-08-27 | 北大方正集团有限公司 | Method and apparatus for cutting character |
| US20090324081A1 (en) * | 2008-06-24 | 2009-12-31 | Samsung Electronics Co., Ltd. | Method and apparatus for recognizing character in character recognizing apparatus |
| CN101650779A (en) * | 2008-08-11 | 2010-02-17 | 欧姆龙株式会社 | Character recognition device, program and method |
| CN103870799A (en) * | 2012-12-17 | 2014-06-18 | 北京千橡网景科技发展有限公司 | Character direction judging method and device |
| CN104951741A (en) * | 2014-03-31 | 2015-09-30 | 阿里巴巴集团控股有限公司 | Character recognition method and device thereof |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106203425A (en) * | 2016-07-01 | 2016-12-07 | 北京旷视科技有限公司 | Character identifying method and device |
| CN106203425B (en) * | 2016-07-01 | 2020-02-04 | 北京旷视科技有限公司 | Character recognition method and device |
| CN111630521A (en) * | 2018-02-28 | 2020-09-04 | 佳能欧洲股份有限公司 | Image processing method and image processing system |
| CN110334711A (en) * | 2019-06-04 | 2019-10-15 | 新华三技术有限公司 | Character match method and device |
| CN110555441A (en) * | 2019-09-10 | 2019-12-10 | 杭州橙鹰数据技术有限公司 | character recognition method and device |
| CN111652225A (en) * | 2020-04-29 | 2020-09-11 | 浙江省北大信息技术高等研究院 | Non-invasive camera reading method and system based on deep learning |
| CN111652225B (en) * | 2020-04-29 | 2024-02-27 | 杭州未名信科科技有限公司 | Non-invasive camera shooting and reading method and system based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105528606A (en) | Region identification method and device | |
| CN107145800A (en) | Method for protecting privacy and device, terminal and storage medium | |
| CN105138259A (en) | Operation execution method and operation execution device | |
| CN104021350A (en) | Privacy-information hiding method and device | |
| CN105631450A (en) | Character identifying method and device | |
| CN105159871A (en) | Text information detection method and apparatus | |
| CN105068976A (en) | Ticket information exhibition method and device | |
| CN104965704A (en) | Information display method and apparatus | |
| CN105447109A (en) | Key word searching method and apparatus | |
| CN104461236A (en) | Method and device for displaying application icons | |
| CN106485660A (en) | The Zoom method of electronic chart and device | |
| CN105511777A (en) | Session display method and device of touch display screen | |
| CN105205153A (en) | File storage method and device | |
| KR102063764B1 (en) | Mobile terminal and control method thereof | |
| CN104516951A (en) | Page display method and apparatus and electronic device | |
| CN105224644A (en) | Information classification approach and device | |
| CN105678296A (en) | Method and apparatus for determining angle of inclination of characters | |
| CN104216969A (en) | Reading marking method and device | |
| CN104317851A (en) | Word prompt method and device | |
| CN106921958A (en) | The method and apparatus for quitting the subscription of business | |
| CN104715007A (en) | User identification method and device | |
| CN103995844B (en) | Information search method and device | |
| CN105677215A (en) | Application control method and apparatus | |
| CN105488074A (en) | Photo clustering method and device | |
| KR20140054586A (en) | Mobile terminal set |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160601 |