CN108205670B - License plate recognition method and device - Google Patents
License plate recognition method and device Download PDFInfo
- Publication number
- CN108205670B CN108205670B CN201611170215.3A CN201611170215A CN108205670B CN 108205670 B CN108205670 B CN 108205670B CN 201611170215 A CN201611170215 A CN 201611170215A CN 108205670 B CN108205670 B CN 108205670B
- Authority
- CN
- China
- Prior art keywords
- character
- license plate
- area
- recognized
- recognition result
- 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Character Input (AREA)
- Character Discrimination (AREA)
Abstract
The embodiment of the application provides a license plate identification method and device. The method comprises the following steps: firstly, obtaining a license plate image area of a license plate number to be recognized, and performing character recognition on the license plate image area to obtain a first character recognition result; then, according to the successfully recognized characters in the first character recognition result, determining a character area to be recognized, which needs to be recognized for the second time, in the license plate image area, and performing character recognition on the character area to be recognized to obtain a second character recognition result; and finally, acquiring the license plate number corresponding to the license plate image area according to the first character recognition result and the second character recognition result. By applying the scheme provided by the embodiment of the application to license plate recognition, the efficiency of the license plate recognition process can be improved.
Description
Technical Field
The application relates to the technical field of intelligent traffic, in particular to a license plate recognition method and device.
Background
The license plate is the 'ID card' of the vehicle and is important information which is different from other motor vehicles. The license plate recognition technology is widely applied to scenes such as a gate, a parking lot, an electronic police and the like to acquire license plate information of vehicles in the scenes, and plays the power of an intelligent traffic algorithm in many aspects such as public security management and the like.
The license plate region is generally composed of a foreground portion (character portion) and a background portion (ground color). When a license plate image area is obtained from an actual application scene, patterns may exist in a license plate background part, and the patterns may be caused by stains such as mud spots existing on a license plate or watermarks existing in the license plate background part. For example, in the partial license plate example shown in fig. 1, irregular watermarks exist in the background part of the license plate, and the shape, color and position of the watermarks have diversity. The presence of patterns in the license plate image area will affect the license plate number recognition process for such license plate images.
In the prior art, when the license plate number of the license plate of the type is identified, one of the license plate image regions in which the license plate number needs to be identified is matched with a plurality of pre-stored license plate templates, so as to identify the license plate number, wherein the license plate templates are constructed according to the type of the license plate. The specific process is as follows: and according to the selected license plate template, performing character segmentation on characters in the license plate image area, and performing character recognition on each segmented character area. And if the character recognition is successful, the license plate template is considered to be successfully matched, and the character recognition result is determined to be the license plate number corresponding to the license plate image area. If the character recognition is unsuccessful, another license plate template is selected, and the process is repeated.
Because the types of license plates with patterns are more, in order to identify the license plates of the types, a license plate template is generally required to be constructed for each type of license plate, so that a large number of license plate templates are required to be constructed.
Under normal conditions, when the method is adopted for license plate recognition, the license plate number in the license plate image area with the pattern can be recognized. However, since a large number of license plate templates need to be matched, and the character segmentation and character recognition processes need to be completely executed once in each matching process, the efficiency of the license plate recognition process is low.
Disclosure of Invention
The embodiment of the application aims to provide a license plate recognition method and a license plate recognition device so as to improve the efficiency of a license plate recognition process. The specific technical scheme is as follows.
In order to achieve the above object, the present application discloses a license plate recognition method, including:
obtaining a license plate image area of a license plate number to be identified;
performing character recognition on the license plate image area to obtain a first character recognition result;
determining a character area to be recognized which needs to be recognized for the second time in the license plate image area according to the successfully recognized characters in the first character recognition result;
performing character recognition on the character area to be recognized to obtain a second character recognition result;
and acquiring a license plate number corresponding to the license plate image area according to the first character recognition result and the second character recognition result.
Optionally, the determining, according to the successfully recognized character in the first character recognition result, a character region to be recognized, which needs to be recognized for the second time, in the license plate image region includes:
determining a first character area corresponding to a character which is successfully recognized in the first character recognition result, and determining a second character area corresponding to a character which is not successfully recognized in the first character recognition result;
determining a second character area meeting preset conditions as a character area to be recognized, which needs to be recognized for the second time, in the license plate image area;
wherein the preset condition comprises at least one of the following conditions:
the distance between the first character areas on the two sides of the target character area is greater than a preset threshold value, and the target character area is as follows: one of the second character regions;
the size of the target character region is larger than a size threshold, and the size threshold is a threshold determined according to the average size of the first character region.
Optionally, the performing character recognition on the character region to be recognized to obtain a second character recognition result includes:
carrying out character segmentation on the character area to be recognized to obtain a target character area;
and performing character recognition on the target character area to obtain a second character recognition result.
Optionally, the performing character segmentation on the character region to be recognized to obtain a target character region includes:
determining a first character area corresponding to a character which is successfully recognized in the first character recognition result;
obtaining an average size of the first character region;
and according to the average size, performing character segmentation on the character area to be recognized to obtain a target character area.
Optionally, the performing character recognition on the license plate image region to obtain a first character recognition result includes:
performing character segmentation on the license plate image area to obtain a suspected character area;
removing non-character areas in the suspected character areas to obtain screened character areas;
and performing character recognition on the screened character area to obtain a first character recognition result.
In order to achieve the above object, the present application also discloses a license plate recognition device, the device including:
the image area obtaining module is used for obtaining a license plate image area of a license plate number to be identified;
the first character recognition module is used for carrying out character recognition on the license plate image area to obtain a first character recognition result;
the character area determining module is used for determining a character area to be recognized, which needs to be recognized for the second time, in the license plate image area according to the successfully recognized characters in the first character recognition result;
the second character recognition module is used for carrying out character recognition on the character area to be recognized to obtain a second character recognition result;
and the license plate number obtaining module is used for obtaining the license plate number corresponding to the license plate image area according to the first character recognition result and the second character recognition result.
Optionally, the character region determining module includes:
the first determining submodule is used for determining a first character area corresponding to a character which is successfully recognized in the first character recognition result, and determining a second character area corresponding to a character which is not successfully recognized in the first character recognition result;
the second determining submodule is used for determining a second character region meeting preset conditions as a character region to be recognized, which needs to be recognized for the second time, in the license plate image region;
wherein the preset condition comprises at least one of the following conditions:
the distance between the first character areas on the two sides of the target character area is greater than a preset threshold value, and the target character area is as follows: one of the second character regions;
the size of the target character region is larger than a size threshold, and the size threshold is a threshold determined according to the average size of the first character region.
Optionally, the second character recognition module includes:
the first segmentation submodule is used for carrying out character segmentation on the character area to be recognized to obtain a target character area;
and the first recognition submodule is used for carrying out character recognition on the target character area to obtain a second character recognition result.
Optionally, the first partitioning sub-module includes:
the determining unit is used for determining a first character area corresponding to a character which is successfully recognized in the first character recognition result;
an obtaining unit configured to obtain an average size of the first character region;
and the segmentation unit is used for performing character segmentation on the character area to be recognized according to the average size to obtain a target character area.
Optionally, the first character recognition module includes:
the second segmentation submodule is used for carrying out character segmentation on the license plate image area to obtain a suspected character area;
the screening submodule is used for removing a non-character area in the suspected character area to obtain a screened character area;
and the second recognition submodule is used for carrying out character recognition on the screened character area to obtain a first character recognition result.
According to the technical scheme, in the scheme provided by the embodiment of the application, firstly, character recognition is carried out on the license plate image area of the obtained license plate number to be recognized to obtain a first character recognition result, then the character area to be recognized, which needs to be subjected to secondary recognition, in the license plate image area is determined according to the successfully recognized characters in the first character recognition result, and character recognition is carried out on the character area to be recognized to obtain a second character recognition result. And finally, acquiring the license plate number corresponding to the license plate image area according to the first character recognition result and the second character recognition result.
That is to say, in the embodiment of the application, first character recognition is performed on a license plate image area, and according to characters which are successfully recognized in a result of the first character recognition, a character area which needs to be recognized for the second time in the license plate image area is determined and recognized. Therefore, according to the license plate recognition method and device, a large number of license plate templates do not need to be matched one by one, and the efficiency of the license plate recognition process can be improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the application, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
FIG. 1 is an example of a portion of a license plate in a region;
fig. 2 is a schematic flow chart of a license plate recognition method according to an embodiment of the present disclosure;
FIG. 3 is an example of a portion of a license plate of the generic type;
FIG. 4 is a schematic flow chart of step S203 in FIG. 2;
FIG. 5 is an exemplary diagram of a character region and a corresponding recognition result included in a license plate image region;
fig. 6 is a schematic structural diagram of a license plate recognition device according to an embodiment of the present disclosure.
Detailed Description
The technical solution in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It is to be understood that the described embodiments are merely a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The embodiment of the application provides a license plate recognition method and device, which are applied to electronic equipment, wherein the electronic equipment can be terminal equipment or a server and the like, and the terminal equipment can comprise a computer, a tablet personal computer, a smart phone, a vehicle data recorder and the like. By applying the technical scheme in the embodiment of the application to license plate recognition, the efficiency of the license plate recognition process can be improved. The present application will be described in detail below with reference to specific examples.
Fig. 2 is a schematic flow chart of the license plate recognition method provided in the embodiment of the present application, and the license plate recognition method is applied to an electronic device. The method comprises the following steps:
step S201: and obtaining a license plate image area of the license plate number to be identified.
The license plate image area of the license plate number to be identified can be understood as follows: and a license plate image area needing license plate identification. The license plate image area is an image area containing a license plate in the license plate image. The license plate image is an image including a license plate portion of the vehicle. In a preferred embodiment, the license plate image area may be an image area formed by the outermost frame of the license plate characters. Of course, the license plate image region may also be a region containing other image portions than the license plate characters. In general, the license plate image area may be set to a rectangular area.
For example, the image area shown in fig. 1 may be used as the license plate image area of the license plate number to be recognized.
Specifically, the license plate image area of the license plate number to be recognized may be directly obtained, or may be obtained in the following manner: and obtaining a license plate image of the license plate number to be identified, and positioning the license plate of the license plate image to obtain a license plate image area. The license plate image of the license plate number to be identified can be understood as follows: and (4) a license plate image needing license plate identification.
The license plate image of the license plate number to be recognized can be an image including a vehicle captured on a road, an image including a vehicle captured in a parking lot, and the like. Of course, the license plate image of the license plate number to be recognized may also be obtained in other manners, and the obtaining manner of the license plate image of the license plate number to be recognized is not limited in the present application.
The electronic device as the execution subject may or may not include an image capturing device inside.
When the electronic device as the execution subject includes the image capturing device therein, the electronic device may include, when obtaining the license plate image of the license plate number to be recognized: and receiving a license plate image of the license plate number to be identified, which is acquired by the image acquisition equipment.
When the electronic device as the execution subject does not include an image capturing device inside, the electronic device may be connected to an external image capturing device, and the electronic device may include, when obtaining a license plate image of a license plate number to be recognized: and acquiring a license plate image of the license plate number to be identified, which is acquired by image acquisition equipment.
The acquired license plate image of the license plate number to be identified can be acquired by the image acquisition equipment in real time, or can be not acquired in real time, but is stored after being acquired in advance by the image acquisition equipment.
Step S202: and performing character recognition on the license plate image area to obtain a first character recognition result.
Specifically, when characters in the license plate image region are recognized and a first character recognition result is obtained, the license plate image region may be first segmented by using a vertical projection method or a connected domain method, etc. to obtain each character region in the license plate image region, and then the character regions are recognized by using a preset character recognizer to obtain the first character recognition result.
The character area is an image area which may contain characters in the license plate image area. The character recognizer may include N output units, each output unit corresponding to a character. For example, the character recognizer includes 37 output units, which correspond to the following characters, respectively: 10 numbers, 26 letters and 1 "unknown".
The character segmentation result generally includes a plurality of character areas, and correspondingly, the first character recognition result generally includes a successfully recognized character and a corresponding character area, and an unsuccessfully recognized character and a corresponding character area.
Specifically, when performing character recognition on each character region, the character region may be input to the character recognizer, each output unit may output a confidence level, and a character corresponding to an output unit whose confidence level is greater than a preset threshold value is a character in the character region. At this time, the character area is considered to be successfully recognized, and the corresponding character is the character which is successfully recognized.
If the confidence of the unknown output unit is higher than the threshold value, and the confidence of other output units is lower than the threshold value, the character area is considered to be unsuccessfully recognized, and the unknown is the character which is not successfully recognized.
In addition, when the license plate image area is segmented, the feature image of the license plate image area pixel can be obtained according to a vertical projection method or a connected domain method, and the character segmentation boundary is determined from the feature image, so that the character segmentation result is obtained.
Therefore, for the license plate image with all characters having obvious and consistent characteristics and good contrast between the foreground and the background of the license plate, when the license plate region of the license plate image is segmented, a good segmentation result can be obtained generally, and then all characters in the license plate image can be identified more easily and correctly.
For example, fig. 3 is a diagram illustrating a part of a license plate of a common type, in which the characters of the license plate number part have substantially the same size and the space between adjacent characters is substantially the same. Wherein the size of the character area comprises the height and/or width of the character area. It can be seen that the characters in such license plates have distinct and consistent characteristics. In addition, the license plate shown in fig. 3 has an obvious feature that the color unity of the foreground color (i.e., the color of the character part) and the background color in the license plate region is good, the contrast between the foreground color and the background color is high, and the distinction degree is good. For example, referring to the number plate numbered 5 in fig. 3, where the foreground is a single white and the background is a single black, it can be seen that the foreground portion and the background portion in the number plate both have a single color. When the license plate image area of the license plate is subjected to character segmentation, the boundary of each character area can be distinguished well usually, a good segmentation result is obtained, and a good character recognition result can be obtained.
Certainly, in real life, the background part of the license plate image area has a pattern, and the pattern may be caused by stain on the license plate or irregular watermark in the background part of the license plate. Irregular watermarks are present in the background portion of a license plate such as that shown in fig. 1. Such a license plate background portion has poor color unicity, and the existence of patterns may cause character adhesion, thereby affecting the determination of character region boundaries, and causing a failure in accurately segmenting character regions in a license plate image region. Therefore, when character recognition is performed on the license plate image area, a recognition error may exist.
Step S203: and determining a character area to be recognized which needs to be recognized for the second time in the license plate image area according to the successfully recognized characters in the first character recognition result.
As can be seen from the foregoing description of the steps, after the first license plate recognition process, the characters in the license plate image area may or may not be completely recognized, that is, there are still unrecognized characters. Therefore, when a character area to be recognized, which needs to be recognized for the second time, in the license plate image area is determined according to the successfully recognized characters in the first character recognition result, whether the character area to be recognized, which needs to be recognized for the second time, exists in the license plate image area can be judged according to the successfully recognized characters in the first character recognition result, if the character area to be recognized exists, the character area to be recognized is determined, and if the character area does not exist, the successfully recognized characters in the first character recognition result are directly determined as the license plate number of the license plate image area.
When the character area to be recognized, which needs to be recognized for the second time, in the license plate image area is determined, the character area corresponding to the character which is not recognized successfully in the first character recognition result can be determined.
When the first character recognition result includes characters which are not successfully recognized, the character regions corresponding to the characters are not necessarily the character regions to be recognized which need to be secondarily recognized, and may be rivet regions, mud dot regions, and the like in the license plate. Therefore, it is necessary to find a character region to be subjected to secondary recognition from character regions corresponding to the characters of the first character recognition result which have not been successfully recognized.
Specifically, determining a character region to be recognized in the license plate image region, which needs to be recognized for the second time, according to the successfully recognized character in the first character recognition result may include multiple embodiments:
one is to determine a character area with abnormal size from the character areas corresponding to the characters which are not successfully recognized according to the size of the character area corresponding to the successfully recognized character, and determine the character area with abnormal size as the character area to be recognized.
It can be understood that when there is an influence of a pattern or the like on a background portion of a license plate region, there may be a possibility that characters are stuck to each other, and thus a size of a corresponding character region becomes large, and when segmenting, the characters and the background pattern are segmented together, or at least two characters are segmented together with a middle pattern. Therefore, when the character recognition is unsuccessful due to such a cause, the character area to be secondarily recognized can be determined according to whether the size of the character area is abnormal or not.
And the other is that according to the position distribution of the character areas corresponding to the successfully recognized characters in the license plate image area, the character areas to be recognized are determined from the character areas corresponding to the unsuccessfully recognized characters.
It can be understood that if there are character regions to be recognized between the successfully recognized character regions, the spacing between the successfully recognized character regions is likely to be very large, presenting an anomalous distribution characteristic. Therefore, the character region to be recognized can be determined from the abnormal distribution feature.
Of course, there are many embodiments for determining the character region to be recognized according to the successfully recognized character, which are not listed here. The embodiment of the process is not particularly limited in this embodiment.
Step S204: and performing character recognition on the character area to be recognized to obtain a second character recognition result.
It should be noted that, when identifying the character in the character region to be identified, the same process as that in step S202 may be adopted, or a process different from that in step S202 may be adopted, and details of the identification process are not described again in this embodiment.
Step S205: and acquiring a license plate number corresponding to the license plate image area according to the first character recognition result and the second character recognition result.
Specifically, since both the first character recognition result and the second character recognition result may have characters which are not successfully recognized, in order to improve the accuracy of the recognition result, the characters which are successfully recognized in the first character recognition result and the characters which are successfully recognized in the second character recognition result are usually synthesized to obtain the license plate number corresponding to the license plate image area.
More specifically, during synthesis, the characters successfully recognized in the two recognition results can be sorted according to the relative positions of the character regions corresponding to the characters, the characters are synthesized according to the sorting results, and finally the license plate number corresponding to the license plate image region is obtained.
As can be seen from the above, in this embodiment, firstly, character recognition is performed on the license plate image region of the obtained license plate number to be recognized to obtain a first character recognition result, then, according to the successfully recognized characters in the first character recognition result, the character region to be recognized, which needs to be subjected to secondary recognition, in the license plate image region is determined, and character recognition is performed on the character region to be recognized to obtain a second character recognition result. And finally, acquiring the license plate number corresponding to the license plate image area according to the first character recognition result and the second character recognition result.
That is to say, in this embodiment, first character recognition is performed on the license plate image region, and according to the character that has been successfully recognized in the result of the first character recognition, the character region that needs to be recognized for the second time in the license plate image region is determined and recognized. Therefore, when the scheme provided by the embodiment is applied to license plate recognition, a large number of license plate templates do not need to be matched one by one, and the efficiency of the license plate recognition process can be improved.
In a specific implementation manner based on the embodiment shown in fig. 2, step S203 determines a character area to be recognized in the license plate image area, which needs to be recognized for the second time, according to the successfully recognized characters in the first character recognition result, which may be performed according to a flowchart shown in fig. 4, and includes the following steps:
step S203A: and determining a first character area corresponding to a character which is successfully recognized in the first character recognition result, and determining a second character area corresponding to a character which is not successfully recognized in the first character recognition result.
Step S203B: and determining the second character area meeting the preset condition as a character area to be recognized, which needs to be recognized for the second time, in the license plate image area.
Wherein the preset condition comprises at least one of the following conditions:
the first condition is as follows: the distance between the first character areas on the two sides of the target character area is greater than a preset threshold value, and the target character area is as follows: one of the second character regions.
The preset threshold value can be determined in advance according to the characteristics in a large number of sample license plate image areas. For example, the average width of a single character region in the sample license plate image region may be used as a preset threshold.
Specifically, when the character region to be recognized is determined, first character regions on two sides of each second character region may be determined, then a distance between two pairs of the first character regions is calculated, and the second character region between two pairs of the first character regions, where the distance is greater than a preset threshold value, is determined as the character region to be recognized.
For example, fig. 5 shows a first character recognition result obtained for a license plate image region, where recognized characters and unrecognized characters in the first character recognition result are marked below the license plate image region, a character region corresponding to each character is shown by a square inside the license plate image region, and a number of each character region is listed by a number above the square. Then the character area with the number "3" may be determined as the second character area, and the left and right sides of the character area 3, and two first character areas adjacent to the left and right sides of the character area 3 may be determined as the character area 2 and the character area 4, respectively, and it is determined that the distance between the character area 2 and the character area 4 is greater than the preset threshold, then the character area 3 may be determined as the character area to be recognized that needs to be recognized for the second time. (the character of the character area numbered 5 in FIG. 5 is not shown)
Case two: the size of the target character region is larger than a size threshold, and the size threshold is a threshold determined according to the average size of the first character region. The product of the average size and the preset value can be used as the size threshold, and of course, the sum of the average size and the preset value can also be used as the size threshold. This embodiment is not particularly limited thereto.
Specifically, when the character region to be recognized is determined, the size of each second character region and the average size of all first character regions may be determined first, and the average size multiplied by a preset value is used as a size threshold. Then, the size of each second character area is compared with the size threshold, and the second character area with the size larger than the size threshold is determined as the character area to be recognized. Wherein the dimension includes at least one of a width, a height, and an aspect ratio.
Following the example in case one above, character areas 1, 2, 4, 5, 6, and 7 are all first character areas, and character area 3 is a second character area, and the average width of the first character areas can be determined. And then judging whether the width of the second character area is larger than the average width multiplied by a preset value, and if so, determining the character area 3 as a character area to be recognized.
In summary, in this embodiment, the second character region that meets the preset condition is determined as the character region that needs to be secondarily recognized in the license plate image region, and the preset condition is related to the first character region. And the first character area is the character area corresponding to the character successfully recognized in the first character recognition result, so that the character area to be recognized is determined from the second character area according to the characteristics of the first character area, and the accuracy of the determination process can be improved.
Further, in order to improve the accuracy of character recognition on the character region to be recognized, in a specific implementation manner based on the embodiment shown in fig. 2, in step S204, the character recognition on the character region to be recognized to obtain a second character recognition result may include the following sub-steps:
substep 1: and carrying out character segmentation on the character area to be recognized to obtain a target character area.
Specifically, the character region to be recognized may be subjected to character segmentation according to a vertical projection method or a connected domain method, so as to obtain a target character region.
As a specific implementation manner, in order to improve the accuracy of the character segmentation process, in substep 1, performing character segmentation on the character region to be recognized to obtain a target character region, the method may include:
determining a first character area corresponding to a successfully recognized character in the first character recognition result, obtaining an average size of the first character area, and performing character segmentation on the character area to be recognized according to the average size to obtain a target character area.
Specifically, when the character region to be recognized is subjected to character segmentation according to the average size, the average size may be used as the size of characters in the character region to be recognized, and the character region to be recognized is subjected to character segmentation by combining a vertical projection characteristic value or a connected domain characteristic value corresponding to the character region to be recognized.
It is understood that the character area to be recognized may be a character area formed by sticking characters and patterns together, or may be a character area formed by sticking characters, patterns and characters together. If the size of the character in the character area to be recognized can be obtained, the process of character segmentation on the character area to be recognized can be more accurate.
Substep 2: and performing character recognition on the target character area to obtain a second character recognition result.
In summary, in the embodiment, the electronic device as the execution subject performs character segmentation on the character region to be recognized first, and then performs character recognition on the segmented character region, where the segmented character region is a region that may contain characters, and performs character recognition on such character regions one by one, which can improve the accuracy of the character recognition process.
In addition, since the background of a part of the license plate region includes watermarks separated from the character portions, for example, the second half of the license plate numbered 1-3 in fig. 1 includes watermark patterns that are not adhered to other characters, the following embodiments may be further included in the present embodiment in order to improve the accuracy of the recognition process.
In a specific implementation manner based on the embodiment shown in fig. 2, in step S202, performing character recognition on the license plate image region to obtain a first character recognition result, the method may include the following sub-steps:
substep 1: and performing character segmentation on the license plate image area to obtain a suspected character area.
Substep 2: and removing the non-character area in the suspected character area to obtain the screened character area.
Wherein the non-character region may include a region of a single color and/or a pattern region. Specifically, when removing the non-character area in the suspected character area, the non-character area may be removed according to a pre-generated non-character recognition network, or may be removed according to a pixel feature value in the area.
When a non-character area in the suspected character area is removed according to a pre-generated non-character recognition network, the suspected character area may be input to the non-character recognition network, a confidence degree indicating that the suspected character area is the non-character area is obtained from the non-character recognition network, and whether to remove the suspected character area is determined according to the confidence degree. It is understood that the greater the confidence, the more likely the suspected character region is to be a non-character region. Therefore, suspected character areas with confidence degrees larger than a preset confidence degree threshold value can be deleted, and screened character areas can be obtained. The non-character recognition network can be trained according to pre-collected sample character images, the sample character images can comprise positive sample character images and negative sample character images, and the positive sample character images comprise non-character images.
Substep 3: and performing character recognition on the screened character area to obtain a first character recognition result.
In summary, in the embodiment, when the electronic device serving as the execution subject performs the first character recognition on the license plate image region, a "non-character recognition" process may be introduced to filter the segmented non-character regions, so as to reduce the number of character regions that are not successfully recognized, and thus improve the accuracy of the character recognition process.
Fig. 6 is a schematic structural diagram of a license plate recognition device according to an embodiment of the present application, which corresponds to the method embodiment shown in fig. 2 and is applied to an electronic device. The device comprises:
the image area obtaining module 601 is used for obtaining a license plate image area of a license plate number to be identified;
the first character recognition module 602 is configured to perform character recognition on the license plate image area to obtain a first character recognition result;
the character region determining module 603 is configured to determine a character region to be recognized, which needs to be recognized for the second time, in the license plate image region according to the successfully recognized character in the first character recognition result;
a second character recognition module 604, configured to perform character recognition on the character region to be recognized, so as to obtain a second character recognition result;
and a license plate number obtaining module 605, configured to obtain a license plate number corresponding to the license plate image region according to the first character recognition result and the second character recognition result.
In a specific implementation manner based on the embodiment shown in fig. 6, the character area determining module 603 may include:
a first determining sub-module (not shown in the figure) for determining a first character area corresponding to a character successfully recognized in the first character recognition result, and determining a second character area corresponding to a character not successfully recognized in the first character recognition result;
a second determining submodule (not shown in the figure) for determining a second character region meeting preset conditions as a character region to be recognized, which needs to be recognized for the second time, in the license plate image region;
wherein the preset condition comprises at least one of the following conditions:
the distance between the first character areas on the two sides of the target character area is greater than a preset threshold value, and the target character area is as follows: one of the second character regions;
the size of the target character region is larger than a size threshold, and the size threshold is a threshold determined according to the average size of the first character region.
In a specific implementation manner based on the embodiment shown in fig. 6, the second character recognition module 604 may include:
a first segmentation submodule (not shown in the figure) for performing character segmentation on the character region to be recognized to obtain a target character region;
and a first recognition submodule (not shown in the figure) for performing character recognition on the target character area to obtain a second character recognition result.
In a specific implementation manner based on the embodiment shown in fig. 6, the first partitioning sub-module may include:
a determining unit (not shown in the figure) for determining a first character area corresponding to a character which is successfully recognized in the first character recognition result;
an obtaining unit (not shown in the figure) for obtaining an average size of the first character area;
and a segmentation unit (not shown in the figure) for performing character segmentation on the character region to be recognized according to the average size to obtain a target character region.
In a specific implementation manner based on the embodiment shown in fig. 6, the first character recognition module 602 may include:
a second segmentation submodule (not shown in the figure) for performing character segmentation on the license plate image area to obtain a suspected character area;
a screening submodule (not shown in the figure) for removing a non-character area in the suspected character area to obtain a screened character area;
and a second recognition submodule (not shown in the figure) for performing character recognition on the screened character region to obtain a first character recognition result.
Since the device embodiment is obtained based on the method embodiment and has the same technical effect as the method, the technical effect of the device embodiment is not described herein again. For the apparatus embodiment, since it is substantially similar to the method embodiment, it is described relatively simply, and reference may be made to some descriptions of the method embodiment for relevant points.
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.
It will be understood by those skilled in the art that all or part of the steps in the above embodiments can be implemented by hardware associated with program instructions, and the program can be stored in a computer readable storage medium. The storage medium referred to herein is a ROM/RAM, a magnetic disk, an optical disk, or the like.
The above description is only for the preferred embodiment of the present application, and is not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application are included in the protection scope of the present application.
Claims (10)
1. A license plate recognition method is characterized by comprising the following steps:
obtaining a license plate image area of a license plate number to be identified;
performing character recognition on the license plate image area to obtain a first character recognition result;
determining a character area to be recognized which needs to be subjected to secondary recognition in the license plate image area according to the successfully recognized characters in the first character recognition result, wherein the character area to be recognized belongs to a character area corresponding to the unsuccessfully recognized characters in the first character recognition result;
performing character recognition on the character area to be recognized to obtain a second character recognition result;
and acquiring a license plate number corresponding to the license plate image area according to the first character recognition result and the second character recognition result.
2. The method of claim 1, wherein the determining a character region to be recognized in the license plate image region, which needs to be recognized for the second time, according to the successfully recognized characters in the first character recognition result comprises:
determining a first character area corresponding to a character which is successfully recognized in the first character recognition result, and determining a second character area corresponding to a character which is not successfully recognized in the first character recognition result;
determining a second character area meeting preset conditions as a character area to be recognized, which needs to be recognized for the second time, in the license plate image area;
wherein the preset condition comprises at least one of the following conditions:
the distance between the first character areas on the two sides of the target character area is greater than a preset threshold value, and the target character area is as follows: one of the second character regions;
the size of the target character region is larger than a size threshold, and the size threshold is a threshold determined according to the average size of the first character region.
3. The method according to claim 1, wherein the character recognition of the character area to be recognized to obtain a second character recognition result comprises:
carrying out character segmentation on the character area to be recognized to obtain a target character area;
and performing character recognition on the target character area to obtain a second character recognition result.
4. The method according to claim 3, wherein the character segmentation of the character region to be recognized to obtain a target character region comprises:
determining a first character area corresponding to a character which is successfully recognized in the first character recognition result;
obtaining an average size of the first character region;
and according to the average size, performing character segmentation on the character area to be recognized to obtain a target character area.
5. The method of claim 1, wherein the performing character recognition on the license plate image area to obtain a first character recognition result comprises:
performing character segmentation on the license plate image area to obtain a suspected character area;
removing non-character areas in the suspected character areas to obtain screened character areas;
and performing character recognition on the screened character area to obtain a first character recognition result.
6. A license plate recognition device, the device comprising:
the image area obtaining module is used for obtaining a license plate image area of a license plate number to be identified;
the first character recognition module is used for carrying out character recognition on the license plate image area to obtain a first character recognition result;
the character area determining module is used for determining a character area to be recognized, which needs to be recognized for the second time, in the license plate image area according to the successfully recognized characters in the first character recognition result, wherein the character area to be recognized belongs to a character area corresponding to the unsuccessfully recognized characters in the first character recognition result;
the second character recognition module is used for carrying out character recognition on the character area to be recognized to obtain a second character recognition result;
and the license plate number obtaining module is used for obtaining the license plate number corresponding to the license plate image area according to the first character recognition result and the second character recognition result.
7. The apparatus of claim 6, wherein the character region determination module comprises:
the first determining submodule is used for determining a first character area corresponding to a character which is successfully recognized in the first character recognition result, and determining a second character area corresponding to a character which is not successfully recognized in the first character recognition result;
the second determining submodule is used for determining a second character region meeting preset conditions as a character region to be recognized, which needs to be recognized for the second time, in the license plate image region;
wherein the preset condition comprises at least one of the following conditions:
the distance between the first character areas on the two sides of the target character area is greater than a preset threshold value, and the target character area is as follows: one of the second character regions;
the size of the target character region is larger than a size threshold, and the size threshold is a threshold determined according to the average size of the first character region.
8. The apparatus of claim 6, wherein the second character recognition module comprises:
the first segmentation submodule is used for carrying out character segmentation on the character area to be recognized to obtain a target character area;
and the first recognition submodule is used for carrying out character recognition on the target character area to obtain a second character recognition result.
9. The apparatus of claim 8, wherein the first segmentation submodule comprises:
the determining unit is used for determining a first character area corresponding to a character which is successfully recognized in the first character recognition result;
an obtaining unit configured to obtain an average size of the first character region;
and the segmentation unit is used for performing character segmentation on the character area to be recognized according to the average size to obtain a target character area.
10. The apparatus of claim 6, wherein the first character recognition module comprises:
the second segmentation submodule is used for carrying out character segmentation on the license plate image area to obtain a suspected character area;
the screening submodule is used for removing a non-character area in the suspected character area to obtain a screened character area;
and the second recognition submodule is used for carrying out character recognition on the screened character area to obtain a first character recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611170215.3A CN108205670B (en) | 2016-12-16 | 2016-12-16 | License plate recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611170215.3A CN108205670B (en) | 2016-12-16 | 2016-12-16 | License plate recognition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108205670A CN108205670A (en) | 2018-06-26 |
| CN108205670B true CN108205670B (en) | 2020-10-27 |
Family
ID=62601647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611170215.3A Active CN108205670B (en) | 2016-12-16 | 2016-12-16 | License plate recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108205670B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109034158B (en) * | 2017-06-09 | 2021-03-26 | 杭州海康威视数字技术股份有限公司 | License plate recognition method and device and computer equipment |
| CN111767909B (en) * | 2020-05-12 | 2022-02-01 | 合肥联宝信息技术有限公司 | Character recognition method and device and computer readable storage medium |
| CN113486885A (en) * | 2021-06-17 | 2021-10-08 | 杭州鸿泉物联网技术股份有限公司 | License plate recognition method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006338578A (en) * | 2005-06-06 | 2006-12-14 | Mitsubishi Electric Corp | Character recognition apparatus |
| US7182492B1 (en) * | 2003-12-22 | 2007-02-27 | Robert Louis Walter | License plate system having enhanced illumination |
| CN1996995A (en) * | 2006-12-29 | 2007-07-11 | 信息产业部电信传输研究所 | Control method for service sensing and its system |
| CN101944174A (en) * | 2009-07-08 | 2011-01-12 | 西安电子科技大学 | Identification method of characters of licence plate |
| CN103413147A (en) * | 2013-08-28 | 2013-11-27 | 庄浩洋 | Vehicle license plate recognizing method and system |
| CN105335743A (en) * | 2015-10-28 | 2016-02-17 | 重庆邮电大学 | Vehicle license plate recognition method |
-
2016
- 2016-12-16 CN CN201611170215.3A patent/CN108205670B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7182492B1 (en) * | 2003-12-22 | 2007-02-27 | Robert Louis Walter | License plate system having enhanced illumination |
| JP2006338578A (en) * | 2005-06-06 | 2006-12-14 | Mitsubishi Electric Corp | Character recognition apparatus |
| CN1996995A (en) * | 2006-12-29 | 2007-07-11 | 信息产业部电信传输研究所 | Control method for service sensing and its system |
| CN101944174A (en) * | 2009-07-08 | 2011-01-12 | 西安电子科技大学 | Identification method of characters of licence plate |
| CN103413147A (en) * | 2013-08-28 | 2013-11-27 | 庄浩洋 | Vehicle license plate recognizing method and system |
| CN105335743A (en) * | 2015-10-28 | 2016-02-17 | 重庆邮电大学 | Vehicle license plate recognition method |
Non-Patent Citations (2)
| Title |
|---|
| Vehicle license-plate recognition by image analysis;K. Miyamoto;《Proceedings IECON "91: 1991 International Conference on Industrial Electronics, Control and Instrumentation》;19911101;全文 * |
| 车牌自动识别系统的研究与设计;李坤;《中国优秀硕士学位论文全文数据库 信息科学辑》;20120515;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108205670A (en) | 2018-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108073928B (en) | License plate recognition method and device | |
| CN107545239B (en) | Fake plate detection method based on license plate recognition and vehicle characteristic matching | |
| CN108268867B (en) | License plate positioning method and device | |
| CN107729899B (en) | License plate number recognition method and device | |
| CN108229466B (en) | License plate recognition method and device | |
| CN108073926B (en) | License plate recognition method and device | |
| US9014432B2 (en) | License plate character segmentation using likelihood maximization | |
| Saleem et al. | Automatic license plate recognition using extracted features | |
| US7613328B2 (en) | Label detection | |
| CN108108734B (en) | License plate recognition method and device | |
| Paunwala et al. | A novel multiple license plate extraction technique for complex background in Indian traffic conditions | |
| Khalifa et al. | Malaysian Vehicle License Plate Recognition. | |
| CN107194393B (en) | Method and device for detecting temporary license plate | |
| CN114387591A (en) | License plate recognition method, system, equipment and storage medium | |
| CN105184291B (en) | A kind of polymorphic type detection method of license plate and system | |
| CN108073925B (en) | License plate recognition method and device | |
| CN108090484B (en) | License plate recognition method and device | |
| CN108205670B (en) | License plate recognition method and device | |
| CN109034158B (en) | License plate recognition method and device and computer equipment | |
| CN110543883A (en) | license plate recognition method based on deep learning | |
| CN110276295A (en) | Vehicle identification number detection recognition method and equipment | |
| CN110909598A (en) | Deep learning-based method for recognizing illegal traffic driving of non-motor vehicle lane | |
| Lee et al. | Adaptive local binarization method for recognition of vehicle license plates | |
| CN111353485A (en) | Seal identification method, device, equipment and medium | |
| CN112115800A (en) | Vehicle combination recognition system and method based on deep learning target detection |
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 |