CN110084232B - Method and device for recognizing Chinese characters in license plate and terminal equipment - Google Patents
Method and device for recognizing Chinese characters in license plate and terminal equipment Download PDFInfo
- Publication number
- CN110084232B CN110084232B CN201810071711.6A CN201810071711A CN110084232B CN 110084232 B CN110084232 B CN 110084232B CN 201810071711 A CN201810071711 A CN 201810071711A CN 110084232 B CN110084232 B CN 110084232B
- Authority
- CN
- China
- Prior art keywords
- license plate
- character
- chinese
- characters
- chinese character
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Abstract
The invention relates to the technical field of vehicles, in particular to a method and a device for identifying Chinese characters in a license plate and terminal equipment, wherein the method and the device for identifying the Chinese characters in the license plate are applied to the terminal equipment, and the method for identifying the Chinese characters in the license plate comprises the following steps: and receiving the license plate image, further calculating the confidence coefficient of each Chinese character in the license plate image, and setting the Chinese character as a preset target character if the confidence coefficient of the Chinese character is lower than a preset matching threshold. If the confidence of the Chinese character is lower than the preset matching threshold, the recognition result of the Chinese character is not accurate, and therefore the Chinese character is set as the preset target character so as to improve the accuracy of Chinese character recognition. It should be noted that the preset matching threshold and the preset target character are updated in real time, so that the accuracy of the chinese character in the license plate can be greatly improved by using the data as a standard.
Description
Technical Field
The invention relates to the technical field of vehicles, in particular to a method and a device for recognizing Chinese characters in a license plate and terminal equipment.
Background
In an intelligent traffic road management system, license plate recognition is a core technology, and the accuracy of license plate recognition is directly related to the effectiveness of road monitoring and law enforcement. However, for the domestic license plate, the Chinese characters with complex strokes exist, and the domestic license plate has higher recognition requirements because the recognition difficulty of the Chinese characters is relatively higher, unlike the European license plate which only has numbers and letters. Therefore, the algorithms of domestic manufacturers have province configuration items, and the provinces of default configuration can be directly assigned under the fuzzy scene and the conditions of low confidence coefficient of Chinese characters and the like. The method is not suitable for scenes with more foreign vehicles, for example, areas such as high speed, direct prefecture city and the like are not effective, and the method is extremely limited in improving the recognition accuracy of Chinese characters. Moreover, with such configuration items, the complexity of configuration is also increased. Therefore, it is very important to provide a method for improving the recognition accuracy of Chinese characters in license plates.
Disclosure of Invention
The invention aims to provide a method for identifying Chinese characters in a license plate so as to improve the accuracy of Chinese character identification.
Another object of the present invention is to provide a device for recognizing chinese characters in license plates, so as to improve the accuracy of recognition of the chinese characters.
Another object of the present invention is to provide a terminal device to improve the accuracy of chinese character recognition.
In order to achieve the above purpose, the embodiment of the present invention adopts the following technical solutions:
in a first aspect, an embodiment of the present invention provides a method for identifying a chinese character in a license plate, where the method includes: receiving a license plate image; calculating the confidence coefficient of the Chinese character in the license plate image; and if the confidence coefficient of the Chinese character is lower than a preset matching threshold, setting the Chinese character as a preset target character, wherein the preset matching threshold and the preset target character are both calculated and updated in real time.
In a second aspect, an embodiment of the present invention further provides a device for recognizing chinese characters in a license plate, where the device includes: the receiving module is used for receiving the license plate image; the calculation module is used for calculating the confidence coefficient of the Chinese character in the license plate image; and the setting module is used for setting the Chinese character as a preset target character if the confidence coefficient of the Chinese character is lower than a preset matching threshold, wherein the preset matching threshold and the preset target character are both calculated and updated in real time.
In a third aspect, an embodiment of the present invention further provides a terminal device, where the terminal device includes a memory and a processor, the memory is used to store a computer program code, and the processor is used to execute the computer program code stored in the memory to implement the method for recognizing the chinese characters in the license plate.
The embodiment of the invention provides a method and a device for identifying Chinese characters in a license plate and terminal equipment, wherein the method and the device for identifying the Chinese characters in the license plate are applied to the terminal equipment, and the method for identifying the Chinese characters in the license plate comprises the following steps: and receiving the license plate image, further calculating the confidence coefficient of each Chinese character in the license plate image, and setting the Chinese character as a preset target character if the confidence coefficient of the Chinese character is lower than a preset matching threshold. If the confidence of the Chinese character is lower than the preset matching threshold, the recognition result of the Chinese character is not accurate, and therefore the Chinese character is set as the preset target character so as to improve the accuracy of Chinese character recognition. It should be noted that the preset matching threshold and the preset target character are updated in real time, so that the accuracy of the chinese character in the license plate can be greatly improved by using the data as a standard.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 shows a schematic structural diagram of a terminal device according to an embodiment of the present invention.
Fig. 2 is a schematic flow chart illustrating a method for recognizing chinese characters in a license plate according to an embodiment of the present invention.
Fig. 3 is a flowchart illustrating a method for calculating a preset target character according to an embodiment of the present invention.
Fig. 4 is a flowchart illustrating a method for calculating a preset matching threshold according to an embodiment of the present invention.
Fig. 5 is a functional module diagram of a recognition apparatus for chinese characters in a license plate according to an embodiment of the present invention.
The figure is as follows: 100-a terminal device; 110-recognition device of Chinese characters in license plate; 120-a memory; 130-a memory controller; 140-a processor; 150-peripheral interface; 160-input-output unit; 170-an audio unit; 180-a display unit; 190-a communication unit; 111-a receiving module; 112-a calculation module; 113-selecting a module; 114-a statistics module; 115-setting module.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
The license plate recognition has important significance for road monitoring and law enforcement, and because the Chinese characters are very complex, correct recognition of the Chinese characters is difficult. In addition, in urban road monitoring, scenes and complexity, such as low visibility and interference of various ambient lights, increase the difficulty of Chinese character recognition. Therefore, the embodiment of the invention provides a method for identifying Chinese characters in a license plate, which enables a detection means to have better real-time performance and accuracy by dynamically adjusting preset parameters. The method is applied to the terminal equipment, namely, the terminal equipment analyzes the images acquired by the cameras distributed in various scenes and adjusts the preset parameters serving as the standard in real time so as to improve the accuracy of recognizing the Chinese characters in the license plate.
The terminal device may be, but is not limited to, an intelligent electronic device such as a desktop computer, and please refer to fig. 1, which is a schematic structural diagram of a terminal device 100 according to an embodiment of the present invention, where the terminal device 100 includes a recognition device 110 for chinese characters in a license plate, a memory 120, a memory controller 130, a processor 140, an external interface 150, an input/output unit 160, an audio unit 170, a display unit 180, and a communication unit 190.
The memory 120, the memory controller 130, the processor 140, the peripheral interface 150, the input/output unit 160, the audio unit 170, the display unit 180, and the communication unit 190 are electrically connected to each other directly or indirectly to implement data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The recognition device 110 for chinese characters in the license plate includes at least one software function module that can be stored in the memory 120 in the form of software or firmware (firmware) or solidified in an Operating System (OS) of the terminal device 100. The processor 140 is configured to execute an executable module stored in the memory 120, such as a software function module or a computer program included in the recognition device 110 for chinese characters in the license plate.
The Memory 120 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like. The memory 120 is configured to store a program, and the processor 140 executes the program after receiving an execution instruction, and a method executed by the terminal device 100 defined by a flow procedure disclosed in any embodiment of the invention described later may be applied to the processor 140, or implemented by the processor 140.
The processor 140 may be an integrated circuit chip having signal processing capabilities. The Processor 140 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components. The various methods, steps and logic blocks disclosed in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor 140 may be any conventional processor 140 or the like.
The peripheral interface 150 couples various input/output devices to the processor 140 as well as to the memory 120. In some embodiments, peripheral interface 150, processor 140, and memory controller 130 may be implemented in a single chip. In other examples, they may be implemented separately from the individual chips.
The input and output unit 160 is used for providing input data for a user to realize the interaction of the user with the data acquisition terminal 100. The input/output unit 160 may be, but is not limited to, a mouse, a keyboard, and the like.
The display unit 180 provides an interactive interface (e.g., a user operation interface) between the data acquisition terminal 100 and a user or is used to display image data for user reference. In this embodiment, the display unit 180 may be a liquid crystal display or a touch display. In the case of a touch display, the display can be a capacitive touch screen or a resistive touch screen, which supports single-point and multi-point touch operations. Supporting single-point and multi-point touch operations means that the touch display can sense touch operations from one or more locations on the touch display at the same time, and the sensed touch operations are sent to the processor 140 for calculation and processing.
The communication unit 190 is configured to establish a connection with a plurality of cameras through the network, thereby implementing a communication connection between the terminal device 100 and the plurality of cameras, and is configured to transmit and receive data through the network.
Fig. 2 is a schematic flow chart of a method for recognizing chinese characters in a license plate according to an embodiment of the present invention, where the method includes:
step S110, receiving a license plate image.
The terminal device 100 is connected to a plurality of cameras installed on a road, and the plurality of cameras collect license plate images of passing vehicles and then transmit the license plate images to the terminal device 100.
And step S120, calculating the confidence coefficient of the Chinese characters in the license plate image.
Specifically, since the license plate number is usually a character sequence composed of letters, numbers and chinese characters, the terminal device 100 first performs segmentation processing on the received license plate image to obtain a single character image, i.e., a single digital character image, an alphabetic character image and a chinese character image. Further, the terminal device 100 recognizes the numeric character image, the alphabetic character image, and the kanji character image by a template matching method one-to-one to obtain corresponding numeric characters, alphabetic characters, and kanji characters, that is, matches the single character image obtained by dividing with the template one by one, and if the numeric characters, the alphabetic character image, and the kanji character image are matched, the character corresponding to the template is a recognition result. The template adopts a template with the width-height ratio of 0.5 and the size ratio of 25 x 50 according to the national standard, and the Chinese characters, the letters and the numbers are respectively designed into three types of templates according to the national standard documents.
After each character in the license plate image is recognized, a confidence coefficient needs to be calculated for the recognition result of each character, the confidence coefficient is used for representing the accuracy of the recognition result, and the higher the confidence coefficient is, the more accurate the recognition result is. It should be noted that the confidence algorithms of the letters, the chinese characters and the numbers are consistent, and the confidence algorithm of the chinese characters is exemplified here. The confidence algorithm of the Chinese character is as follows:
wherein, Mod (m, n) is template data, namely a Chinese character template, the size of the template is 25 x 50, and the template is subjected to binarization processing, the foreground pixel value is 1,1 is used for representing white, the background pixel value is 0, and 0 is used for representing black; the Img (m, n) is a Chinese character obtained by segmentation, the Chinese character is subjected to scaling processing to keep the same size as a Chinese character template, namely the size of the Chinese character is 25 × 50 at the moment, and the Chinese character is also subjected to binarization processing, namely the foreground of the Chinese character is 1 and the background is 0; and further, under the condition that the sizes of the Chinese characters and the Chinese character templates are kept consistent with the image form, calculating to obtain that the Confidence coefficient of the Chinese characters is Char _ Confidence, wherein the Char _ Confidence represents the Confidence coefficient of a single Chinese character, when the Chinese characters are completely matched with the templates, the value of the Char _ Confidence is 1250, and the more the value of the Char _ Confidence is, the higher the matching degree of the Chinese characters and the templates is, and the more the recognition result is reliable. In the embodiment of the invention, the license plate is identified by a template matching method, and in addition, the license plate can be identified by a deep learning network.
Step S130, if the confidence of the Chinese character is lower than a preset matching threshold, setting the Chinese character as a preset target character, wherein the preset matching threshold and the preset target character are both calculated and updated in real time.
The Chinese characters in the license plate usually represent specific provinces, and if the Chinese characters in the license plate represent Chongqing, the provinces represented by the Chinese characters in the license plate can be external provinces or local provinces for vehicles in a certain area due to the fact that the vehicles have high circulation, so that the license plate is easy to understand and has higher possibility of representing the local provinces. The preset matching threshold is the bias that the Chinese characters in the license plate represent the province, if the confidence coefficient of the calculated Chinese characters is lower than the preset matching threshold, the Chinese characters are possibly identified wrongly, and in view of the fact that the license plate of the province is more likely to appear in the region of the province, the Chinese characters which are possibly identified wrongly are set as preset target characters, and the preset target characters are the province, such as Yu. On the contrary, if the calculated confidence coefficient of the Chinese character is larger than the preset matching threshold, the Chinese character is indicated to be correctly recognized, no matter the Chinese character is recognized as the current province or the external province, the recognition result is kept, and then the letter recognition result, the number recognition result and the Chinese character recognition result in the license plate are arranged according to the preset sequence to obtain the final license plate recognition result.
Therefore, in the scheme, the confidence coefficient of the Chinese character of the license plate is analyzed, and if the confidence coefficient is lower than the preset matching threshold, the Chinese character is set as the province, so that the error recognition rate is reduced, and the accuracy of the integral recognition of the license plate is improved. In addition, in the invention, the preset matching threshold and the preset target character are updated in real time, so that the accuracy of judgment is further improved.
Fig. 3 is a schematic flow chart of a method for calculating a preset target character according to an embodiment of the present invention, where the method includes:
and step S210, calculating the confidence of the license plate image.
Because the license plate image is composed of a plurality of characters, the confidence coefficient of the whole license plate is added by a single character, namely the confidence coefficient of the license plate image is calculated in the following way:
wherein, Char _ Confidence is the Confidence of a single character, and Plate _ Confidence is the Confidence of the whole license Plate image.
Step S220, if the confidence coefficient of the license plate image is greater than a first threshold value and the confidence coefficient of the Chinese character in the license plate image is greater than a second threshold value, selecting the Chinese character in the license plate image as a target Chinese character.
The first threshold value is Plate _ Thres, the Plate _ Thres is a threshold value representing that the recognition result of the license Plate image is credible and effective, the second threshold value is Chn _ Thres, and the Chn _ Thres is a threshold value representing that the recognition result of the Chinese character in the license Plate image is credible and effective. If the Plate _ Confidence is larger than or equal to the Plate _ Thres and the Char _ Confidence > Chn _ Thres are met, namely the Confidence coefficient of the license Plate image is larger than a first threshold value and the Confidence coefficient of the Chinese character in the license Plate image is larger than a second threshold value, the recognition result of each Chinese character in the license Plate image is credible, the recognition result of the whole license Plate image is credible, and the Chinese character of the license Plate image is selected as the target Chinese character. It is easy to understand that since the license plate may be located in the same place or in different places, the target kanji character may include a plurality of the same kanji characters or a plurality of different kanji characters.
Step S230, selecting a first predetermined number of target kanji characters from the kanji characters of the license plate images.
That is, a certain number of target kanji characters are selected from the plurality of target kanji characters determined at step S220 as data for subsequent statistics.
Step S240, selecting the chinese character with the largest occurrence frequency in the target chinese characters as a preset target character.
Specifically, the Chinese character with the largest occurrence frequency is screened from the selected first preset number of target Chinese characters to be the preset target character.
It should be noted that the preset target character is determined in such a manner that the terminal device 100 has just started to operate, and there is no more license plate image data when the terminal device 100 has just started to operate, so that the chinese character with the largest occurrence frequency is selected from the chinese characters of a certain number of license plate images, which is the province of the user, but as the number of acquired license plate images is more and the data size is larger and larger, the data for determining the preset target character needs to be updated, so as to correct the preset target character in real time, and improve the accuracy.
In step S250, a second predetermined number of target Chinese characters are added at predetermined time intervals.
Specifically, a second predetermined number of target kanji characters is added at intervals, and the second predetermined number may be set according to the user's needs, such as 100 or 200.
In step S260, the kanji character with the largest number of times is selected from the sum of the added second predetermined number of target kanji characters and the first predetermined number of target kanji characters as a preset target character.
The preset target character is the Chinese character with the largest occurrence frequency in the first preset number of target Chinese characters as the preset target character, and in order to update the preset target character in real time, the Chinese character with the largest occurrence frequency is selected as the preset target character on the basis of adding a sample of a second preset number of target Chinese characters on the first preset number of target Chinese characters. If the first preset number of target kanji characters is 500 and the newly added second preset number of target kanji characters is 100, the preset target characters are determined according to the target kanji character amount of 500 for the first time, the preset target characters are determined according to the target kanji character amount of 600 for the second time, the preset target characters are determined according to the target kanji character amount of 700 for the third time, and so on. Because the sample size is increased in sequence and is increased on the original basis in real time, the real-time property is better reflected, and the accuracy of determining the preset target character is ensured.
Fig. 4 is a schematic flow chart of a method for calculating a preset matching threshold according to an embodiment of the present invention, where the method includes:
step S310, selecting Chinese characters in the license plate image, wherein the confidence coefficient of the Chinese characters is larger than a second threshold value, as statistical data.
The second threshold is Chn _ Thres, the Chn _ Thres is a threshold representing that the recognition result of the Chinese character in the license plate image is credible and effective, and the Chinese character with the confidence coefficient of the Chinese character being larger than the second threshold in a certain number of license plate images is further selected as statistical data.
Step S320, comparing the confidence of each chinese character in the statistical data with an initial matching threshold in sequence, and setting the chinese character whose confidence is smaller than the initial matching threshold as a preset target character.
Specifically, the initial matching threshold is a threshold set by a user in a self-defined manner and is used for representing the bias of the Chinese character to represent the province, if the confidence of the Chinese character is smaller than the initial matching threshold, it is indicated that the Chinese character is possibly recognized wrongly, the Chinese character is possibly recognized wrongly as an external province, the Chinese character is set as a preset target character, and the preset target character represents the province.
Step S330, counting the chinese character in the statistical data that is the same as the preset target character as the first data.
And counting the number of Chinese characters representing the province in the statistical data as first data, wherein the first data is represented by Num _ Local.
Step S340, counting, as second data, chinese characters in the statistical data that are different from the preset target characters.
And counting the number of Chinese character characters representing the outer province in the statistical data as second data, wherein the second data is represented by Num _ other.
And step S350, adjusting the initial matching threshold value according to the proportion of the first data and the second data to obtain a preset matching threshold value.
Specifically, the ratio of the first data to the second data is calculated, if Num _ other is 0 or Num _ Local/Num _ other is greater than or equal to 10, it indicates that many Chinese characters representing the province are present, the initial matching threshold value can be increased to obtain a preset matching threshold value, if Num _ Local/Num _ other is greater than or equal to 5 and Num _ Local/Num _ other is less than 10, it indicates that many Chinese characters representing the province are present, the initial matching threshold value can be slightly increased to obtain a preset threshold value, in other words, the degree of increasing the initial matching threshold value is determined according to the range of the ratio of the first data to the second data; if Num _ Local/Num _ other <5, it indicates that there are few Chinese characters representing the province, and the initial matching threshold is reduced to obtain the preset matching threshold.
Because the preset matching threshold is obtained by adjusting the initial matching threshold according to the currently acquired data, the data needing to be supplemented and updated is used as the basic data to recalculate the preset matching threshold along with the increase of the later data volume so as to update the preset matching threshold.
Therefore, the updating method of the preset matching threshold value comprises the following steps:
if the statistical data of the preset matching threshold is calculated to be 500 license plate images, the Chinese character corresponding to one license plate image is randomly selected from the license plate images with the set number of intervals from the statistical data to replace the Chinese character corresponding to the front license plate image in the 500 license plate images so as to obtain updated statistical data, and the updated statistical data is analyzed so as to adjust the preset matching threshold. If any one of 5 license plate images after the 500 license plate images of the statistical data is selected, because the Chinese character corresponding to the province is determined before, whether the Chinese character on the selected license plate image represents the province or the exterior province can be easily distinguished, and then the license plate image selected replaces the front license plate image in the 500 statistical data to obtain the updated statistical data.
And further analyzing the updated statistical data, namely respectively counting the Chinese characters which are the same as the preset target characters in the updated statistical data as third data, the third data representing the number of the Chinese characters of the province, the Chinese characters which are different from the preset target characters as fourth data, and the fourth data representing the number of the Chinese characters of the province, and further regulating the preset matching threshold again according to the ratio of the third data to the fourth data so as to realize the updating of the preset matching threshold. It should be noted that the set number is generally smaller, so as to achieve the effect of updating the preset matching threshold in real time.
Therefore, the method for recognizing the Chinese characters in the license plate improves the accuracy of license plate recognition by updating the preset matching threshold and the preset target characters serving as reference bases in real time. It should be noted that, when a preset target character, that is, a chinese character representing the province is used for a period of time, the chinese character in the collected license plate images is emptied, so that the chinese character is selected again as basic data according to the license plate images, and the preset target character and the preset matching threshold are determined again according to the same method, so as to implement continuous update of the whole data. During recalculation, if the determined Chinese character representing the province is the same as the previous one, the previously determined preset matching threshold value can be used, and if the determined Chinese character representing the province is different from the previous one, the preset matching threshold value is determined again on the basis of the initial matching threshold value.
Fig. 5 is a schematic diagram of functional modules of a device 110 for recognizing chinese characters in license plates according to an embodiment of the present invention, including a receiving module 111, a calculating module 112, a selecting module 113, a counting module 114, and a setting module 115.
The receiving module 111 is configured to receive a license plate image.
In the embodiment of the present invention, step S110 may be performed by the receiving module 111.
And the calculating module 112 is used for calculating the confidence coefficient of the Chinese characters in the license plate image.
In an embodiment of the present invention, step S120 may be performed by the calculation module 112.
A selecting module 113, configured to select a first predetermined number of target chinese characters from the chinese characters of the license plate image; selecting the Chinese character with the largest occurrence frequency in the target Chinese character as a preset target character, wherein the preset matching threshold and the preset target character are both calculated and updated in real time.
In the embodiment of the present invention, steps S210 to S260 may be performed by the selecting module 113.
The statistical module 114 is configured to select, as statistical data, a chinese character in the license plate image whose confidence of the chinese character is greater than a second threshold; sequentially comparing the confidence coefficient of each Chinese character in the statistical data with an initial matching threshold value, and setting the Chinese character with the confidence coefficient smaller than the initial matching threshold value as a preset target character; counting Chinese character characters which are the same as preset target characters in the statistical data as first data; counting Chinese character characters which are different from preset target characters in the statistical data as second data; and adjusting the initial matching threshold value according to the proportion of the first data and the second data to obtain a preset matching threshold value.
In an embodiment of the present invention, steps S310 to S350 may be performed by the statistics module 114.
A setting module 115, configured to set the chinese character as a preset target character if the confidence of the chinese character is lower than a preset matching threshold.
In the embodiment of the present invention, step S130 may be performed by the setting module 115.
Since the recognition method of the Chinese characters in the license plate has been described in detail, the description is omitted here.
In summary, the method, the device and the terminal device for recognizing the chinese character in the license plate provided by the embodiment of the present invention are applied to the terminal device, and the method for recognizing the chinese character in the license plate includes: and receiving the license plate image, further calculating the confidence coefficient of each Chinese character in the license plate image, and setting the Chinese character as a preset target character if the confidence coefficient of the Chinese character is lower than a preset matching threshold. If the confidence of the Chinese character is lower than the preset matching threshold, the recognition result of the Chinese character is not accurate, and therefore the Chinese character is set as the preset target character so as to improve the accuracy of Chinese character recognition. It should be noted that the preset matching threshold and the preset target character are updated in real time, so that the accuracy of the chinese character in the license plate can be greatly improved by using the data as a standard.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (8)
1. A method for recognizing Chinese characters in a license plate is characterized by comprising the following steps:
receiving a license plate image;
calculating the confidence coefficient of the Chinese character in the license plate image;
if the confidence coefficient of the Chinese character is lower than a preset matching threshold, setting the Chinese character as a preset target character, wherein the preset matching threshold and the preset target character are both calculated and updated in real time;
the preset matching threshold is calculated in the following way:
selecting Chinese characters with the confidence degrees of the Chinese characters in the license plate image larger than a second threshold value as statistical data;
sequentially comparing the confidence coefficient of each Chinese character in the statistical data with an initial matching threshold value, and setting the Chinese character with the confidence coefficient smaller than the initial matching threshold value as a preset target character;
counting Chinese character characters which are the same as preset target characters in the statistical data as first data;
counting Chinese character characters which are different from preset target characters in the statistical data as second data;
and adjusting the initial matching threshold value according to the proportion of the first data and the second data to obtain a preset matching threshold value.
2. The method of claim 1, wherein the predetermined target character is calculated by:
selecting a first preset number of target Chinese characters from the Chinese characters of the license plate image;
and selecting the Chinese character with the largest occurrence frequency in the target Chinese character as a preset target character.
3. The method as claimed in claim 2, wherein the updating method of the preset target character comprises:
adding a second preset number of target Chinese characters at preset time intervals;
and selecting the Chinese character with the largest number of times as a preset target character from the sum of the added second preset number of target Chinese characters and the first preset number of target Chinese characters.
4. The method of claim 2 or 3, wherein prior to said selecting a first predetermined number of target kanji characters from the plurality of kanji characters of the license plate image, the method further comprises:
calculating the confidence of the license plate image;
and if the confidence coefficient of the license plate image is greater than a first threshold value and the confidence coefficient of the Chinese character in the license plate image is greater than a second threshold value, selecting the Chinese character in the license plate image as a target Chinese character.
5. The method of claim 1, wherein the preset matching threshold is updated by:
randomly selecting a Chinese character corresponding to a vehicle image from the vehicle images with a set number at intervals to replace a Chinese character corresponding to a license plate image which is in front in the statistical data, so as to obtain updated statistical data;
calculating the Chinese character which is the same as the preset target character in the updated statistical data as third data;
calculating the Chinese character which is different from the preset target character in the updated statistical data as fourth data;
and updating the preset matching threshold according to the ratio of the third data to the fourth data.
6. An apparatus for recognizing chinese characters in a license plate, the apparatus comprising:
the receiving module is used for receiving the license plate image;
the calculation module is used for calculating the confidence coefficient of the Chinese character in the license plate image;
the setting module is used for setting the Chinese character as a preset target character if the confidence coefficient of the Chinese character is lower than a preset matching threshold, and the preset matching threshold and the preset target character are both calculated and updated in real time;
the apparatus further comprises a statistics module, the statistics module further configured to:
selecting Chinese characters with the confidence degrees of the Chinese characters in the license plate image larger than a second threshold value as statistical data;
sequentially comparing the confidence coefficient of each Chinese character in the statistical data with an initial matching threshold value, and setting the Chinese character with the confidence coefficient smaller than the initial matching threshold value as a preset target character;
counting Chinese character characters which are the same as preset target characters in the statistical data as first data;
counting Chinese character characters which are different from preset target characters in the statistical data as second data;
and adjusting the initial matching threshold value according to the proportion of the first data and the second data to obtain a preset matching threshold value.
7. The apparatus of claim 6, further comprising a selection module, the selection module further configured to:
selecting a first preset number of target Chinese characters from the Chinese characters of the license plate image;
and selecting the Chinese character with the largest occurrence frequency in the target Chinese character as a preset target character.
8. A terminal device, characterized in that the terminal device comprises a memory for storing computer program codes, and a processor for executing the computer program codes stored in the memory to realize the recognition method of kanji characters in a license plate according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810071711.6A CN110084232B (en) | 2018-01-25 | 2018-01-25 | Method and device for recognizing Chinese characters in license plate and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810071711.6A CN110084232B (en) | 2018-01-25 | 2018-01-25 | Method and device for recognizing Chinese characters in license plate and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110084232A CN110084232A (en) | 2019-08-02 |
| CN110084232B true CN110084232B (en) | 2021-01-29 |
Family
ID=67411908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810071711.6A Active CN110084232B (en) | 2018-01-25 | 2018-01-25 | Method and device for recognizing Chinese characters in license plate and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110084232B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113496227A (en) * | 2020-04-08 | 2021-10-12 | 顺丰科技有限公司 | Training method and device of character recognition model, server and storage medium |
| CN111639640B (en) * | 2020-04-24 | 2023-11-14 | 深圳市金溢科技股份有限公司 | License plate recognition method, device and equipment based on artificial intelligence |
| CN113657372A (en) * | 2020-05-12 | 2021-11-16 | 中移(上海)信息通信科技有限公司 | License plate recognition method, device, equipment and computer storage medium |
| CN113723422B (en) * | 2021-09-08 | 2023-10-17 | 重庆紫光华山智安科技有限公司 | License plate information determining method, system, equipment and medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102375991A (en) * | 2010-08-24 | 2012-03-14 | 北京中星微电子有限公司 | Classifier training method and device, and character recognition method and device |
| CN103390156A (en) * | 2012-11-05 | 2013-11-13 | 深圳市捷顺科技实业股份有限公司 | License plate recognition method and device |
| CN103390173A (en) * | 2013-07-24 | 2013-11-13 | 佳都新太科技股份有限公司 | Plate number character vote algorithm based on SVM (support vector machine) confidence |
| CN104065920A (en) * | 2014-06-10 | 2014-09-24 | 北京中芯丙午媒体科技有限公司 | Vehicle monitoring and tracking method, system and server |
| CN104464302A (en) * | 2014-12-22 | 2015-03-25 | 南京中兴力维软件有限公司 | License plate recognition and intelligent error correction method and system |
| CN105117725A (en) * | 2015-07-29 | 2015-12-02 | 安徽四创电子股份有限公司 | License plate self-correcting identification method based on license plate credibility |
| CN105512660A (en) * | 2015-11-27 | 2016-04-20 | 浙江宇视科技有限公司 | License number identification method and device |
| CN106097722A (en) * | 2016-06-16 | 2016-11-09 | 广州地理研究所 | Video is utilized to carry out the system and method for trackside parking stall automatization supervision |
| CN106845478A (en) * | 2016-12-30 | 2017-06-13 | 同观科技（深圳）有限公司 | The secondary licence plate recognition method and device of a kind of character confidence level |
| CN107273896A (en) * | 2017-06-15 | 2017-10-20 | 浙江南自智能科技股份有限公司 | A kind of car plate detection recognition methods based on image recognition |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10706330B2 (en) * | 2015-10-01 | 2020-07-07 | Intellivision Technologies Corp | Methods and systems for accurately recognizing vehicle license plates |
-
2018
- 2018-01-25 CN CN201810071711.6A patent/CN110084232B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102375991A (en) * | 2010-08-24 | 2012-03-14 | 北京中星微电子有限公司 | Classifier training method and device, and character recognition method and device |
| CN103390156A (en) * | 2012-11-05 | 2013-11-13 | 深圳市捷顺科技实业股份有限公司 | License plate recognition method and device |
| CN103390173A (en) * | 2013-07-24 | 2013-11-13 | 佳都新太科技股份有限公司 | Plate number character vote algorithm based on SVM (support vector machine) confidence |
| CN104065920A (en) * | 2014-06-10 | 2014-09-24 | 北京中芯丙午媒体科技有限公司 | Vehicle monitoring and tracking method, system and server |
| CN104464302A (en) * | 2014-12-22 | 2015-03-25 | 南京中兴力维软件有限公司 | License plate recognition and intelligent error correction method and system |
| CN105117725A (en) * | 2015-07-29 | 2015-12-02 | 安徽四创电子股份有限公司 | License plate self-correcting identification method based on license plate credibility |
| CN105512660A (en) * | 2015-11-27 | 2016-04-20 | 浙江宇视科技有限公司 | License number identification method and device |
| CN106097722A (en) * | 2016-06-16 | 2016-11-09 | 广州地理研究所 | Video is utilized to carry out the system and method for trackside parking stall automatization supervision |
| CN106845478A (en) * | 2016-12-30 | 2017-06-13 | 同观科技（深圳）有限公司 | The secondary licence plate recognition method and device of a kind of character confidence level |
| CN107273896A (en) * | 2017-06-15 | 2017-10-20 | 浙江南自智能科技股份有限公司 | A kind of car plate detection recognition methods based on image recognition |
Non-Patent Citations (1)
| Title |
|---|
| 一种基于模糊模板匹配的车牌汉字识别方法;吴炜等;《信息技术与网络安全》;20051130(第11期);第57-59页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110084232A (en) | 2019-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110084232B (en) | Method and device for recognizing Chinese characters in license plate and terminal equipment | |
| WO2021184727A1 (en) | Data abnormality detection method and apparatus, electronic device and storage medium | |
| CN109948497B (en) | Object detection method and device and electronic equipment | |
| CN110276342B (en) | License plate identification method and system | |
| KR20210110823A (en) | Image recognition method, training method of recognition model, and related devices and devices | |
| WO2019134298A1 (en) | Image processing method, appapatus, electronic device and storage medium | |
| CN112561080B (en) | Sample screening method, sample screening device and terminal equipment | |
| KR101912748B1 (en) | Scalable Feature Descriptor Extraction and Matching method and system | |
| CN109697468B (en) | Sample image labeling method and device and storage medium | |
| CN108986125B (en) | Object edge extraction method and device and electronic equipment | |
| CN112329847A (en) | Abnormity detection method and device, electronic equipment and storage medium | |
| CN111931864B (en) | Method and system for multiple optimization of target detector based on vertex distance and cross-over ratio | |
| CN111783767B (en) | Character recognition method, character recognition device, electronic equipment and storage medium | |
| CN112200081A (en) | Abnormal behavior identification method and device, electronic equipment and storage medium | |
| CN111783665A (en) | Action recognition method and device, storage medium and electronic equipment | |
| WO2021189827A1 (en) | Method and apparatus for recognizing blurred image, and device and computer-readable storage medium | |
| CN108986137B (en) | Human body tracking method, device and equipment | |
| CN111666907B (en) | Method, device and server for identifying object information in video | |
| US20180181804A1 (en) | Data normalization for handwriting recognition | |
| CN111402579A (en) | Road congestion degree prediction method, electronic device and readable storage medium | |
| CN112766218A (en) | Cross-domain pedestrian re-identification method and device based on asymmetric joint teaching network | |
| WO2019120025A1 (en) | Photograph adjustment method and apparatus, storage medium and electronic device | |
| CN111368632A (en) | Signature identification method and device | |
| CN110889437B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN117218622A (en) | Road condition detection method, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |