License plate identification method and system thereof

Abstract

A license plate identification method is provided, including the following steps of: obtaining a to-be-processed image; obtaining a plurality of feature maps including target features through a feature map extraction module; obtaining at least one region including the target feature in each feature map and giving each frame of each feature map scores corresponding to the target features through a target location extraction module; classifying each frame in each feature map according to the scores through a target candidate classification module and retaining at least one region that corresponds to character features; and obtaining a license plate identification result according to the region that corresponds to the character feature through a voting/statistics module.

1 . A license plate identification method, comprising steps of:
 obtaining a to-be-processed image;   obtaining a plurality of feature maps including target features through a feature map extraction module;   obtaining at least one region including the target feature in each feature map and giving each frame of each feature map scores corresponding to the target features through a target location extraction module;   classifying each frame in each feature map according to the scores through a target candidate classification module and retaining at least one region that corresponds to character features; and   obtaining a license plate identification result according to the region that corresponds to the character feature through a voting/statistics module.   
     
     
         2 . The license plate identification method as claimed in  claim 1 , wherein the target features include the character features corresponding to different spatial frequencies, features of a license plate appearance, background features, or vehicle features. 
     
     
         3 . The license plate identification method as claimed in  claim 1 , further comprising a step of:
 extracting one frame every predetermined number of pixels on the feature map by clustering or a custom size and giving each frame of each feature map the scores that correspond to the target features through the target location extraction module.   
     
     
         4 . The license plate identification method as claimed in  claim 1 , further comprising a step of:
 obtaining at least one target feature point on the feature map through a simple classifier, circling a plurality of regions located near the target feature points by using frames with different sizes, and giving each frame on the feature map the scores that correspond to the target features through the feature map extraction module.   
     
     
         5 . The license plate identification method as claimed in  claim 1 , further comprising a step of:
 retaining the target features whose scores are largest and exceed a predetermined value by means of non-maximum value suppression and through the target candidate classification module.   
     
     
         6 . The license plate identification method as claimed in  claim 1 , further comprising steps of:
 receiving license plate identification results;   dividing the plurality of license plate identification results into at least two groups according to a license plate grouping rule through the voting/statistics module;   voting for each sub-identification result in each of the groups through the voting/statistics module; and   when, for each group, there is one sub-identification result having a voting score that is higher than a threshold value, generating a final license plate identification result according the sub-identification results through the voting/statistics module.   
     
     
         7 . The license plate identification method as claimed in  claim 1 , further comprising a step of:
 updating the feature map extraction module according to the license plate identification result.   
     
     
         8 . The license plate identification method as claimed in  claim 1 , further comprising following steps of:
 receiving a raw image;   comparing the raw image with a historical background image to determine an amount of image change; and   determining whether the amount of image change is greater than a predetermined value,   wherein when the amount of image change is greater than the predetermined value, the to-be-processed image comprising all of the characters is generated.   
     
     
         9 . The license plate identification method as claimed in  claim 1 , further comprising following steps of:
 receiving a raw image;   obtaining a vehicle front image or a vehicle rear image from the raw image through a vehicle front image capturing module or a vehicle rear image capturing module by using a first image feature and a first classifier;   obtaining at least one character block from the vehicle front image or the vehicle rear image through a license plate character region detection model by using a second image feature and a second classifier;   determining a magnification according to the number of character blocks; and   obtaining the to-be-processed image comprising all of the characters according to the magnification through the license plate character region detection model.   
     
     
         10 . The license plate identification method as claimed in  claim 1 , further comprising steps of:
 receiving license plate identification results;   assigning a weight to each license plate identification result according to a time sequence of all the license plate identification results; and   generating the final license plate identification result according to the license plate identification results that are assigned with the weights.   
     
     
         11 . A license plate identification system, comprising:
 an image capturing unit configured to capture at least one raw image; and   a processor configured to:   receive the raw image from the image capturing unit;   obtain a to-be-processed image according to the raw image;   obtain a plurality of feature maps including target features through a feature map extraction module;   obtain at least one region including the target feature in each feature map and give each frame of each feature map scores corresponding to the target features through a target location extraction module;   classify each frame in each feature map according to the scores through a target candidate classification module and retain at least one region that corresponds to character features; and   obtain the license plate identification result according to the region that corresponds to the character feature through a voting/statistics module.   
     
     
         12 . The license plate identification system as claimed in  claim 11 , wherein the target features include the character features corresponding to different spatial frequencies, features of a license plate appearance, background features, or vehicle features. 
     
     
         13 . The license plate identification system as claimed in  claim 11 , wherein the processor is further configured to:
 extract one frame every predetermined number of pixels on the feature map by clustering or a custom size and give each frame of each feature map the scores that correspond to the target features through the target location extraction module.   
     
     
         14 . The license plate identification system as claimed in  claim 11 , wherein the processor is further configured to:
 obtain at least one target feature point on the feature map through a simple classifier, circle a plurality of regions located near the target feature points by using frames with different sizes, and give each frame on the feature map the scores that correspond to the target features through the feature map extraction module.   
     
     
         15 . The license plate identification system as claimed in  claim 11 , wherein the processor is further configured to:
 retain the target features whose scores are largest and exceed a predetermined value by means of non-maximum value suppression and through the target candidate classification module.   
     
     
         16 . The license plate identification system as claimed in  claim 11 , wherein the processor is further configured to:
 receive license plate identification results;   divide the plurality of license plate identification results into at least two groups according to a license plate grouping rule through the voting/statistics module;   vote for each sub-identification result in each of the groups through the voting/statistics module; and   when, for each group, there is one sub-identification result having a voting score that is higher than a threshold value, generate a final license plate identification result according the sub-identification results through the voting/statistics module.   
     
     
         17 . The license plate identification system as claimed in  claim 11 , wherein the processor is further configured to:
 update the feature map extraction module according to the license plate identification result.   
     
     
         18 . The license plate identification system as claimed in  claim 11 , wherein the processor is configured to:
 compare the raw image with a historical background image to determine an amount of image change; and   determine whether the amount of image change is greater than a predetermined value,   wherein when the amount of image change is greater than the predetermined value, the to-be-processed image comprising all of the characters is generated.   
     
     
         19 . The license plate identification system as claimed in  claim 11 , wherein the processing unit is further configured to:
 obtain a vehicle front image or a vehicle rear image from the raw image through a vehicle front image capturing module or a vehicle rear image capturing module by using a first image feature and a first classifier;   obtain at least one character block from the vehicle front image or the vehicle rear image through a license plate character region detection model by using a second image feature and a second classifier;   determine a magnification according to the number of character blocks; and   obtain the to-be-processed image including all of the characters according to the magnification through the license plate character region detection model.   
     
     
         20 . The license plate identification system as claimed in  claim 11 , wherein the processing unit is further configured to:
 receive license plate identification results;   assign a weight to each license plate identification result according to a time sequence of all the license plate identification results; and   generate the final license plate identification result according to the license plate identification results that are assigned with the weights.