A New Algorithm for Character Segmentation of License Plate

1、A New Algorithm for Character Segmentation of License PlateYungang ZhangDept. of Automation, Tsinghua University The Institute of Information Processing Beijing 100084, China AbstractCharacter segmentation is an important step in License Plate Recognition (LPR) system. Ther

2、e are many dif-ficulties in this step, such as the influence of image noise, plate frame, rivet, the space mark, and so on. This paper presents a new algorithm for character seg-mentation, using Hough transformation and the prior knowledge in horizontal and vertical segmentation re-spectively. Furth

3、ermore, a new object enhancement technique is used for image preprocessing. The experi-ment results show a good performance of this new seg-mentation algorithm.1 IntroductionThere are many useful applications for a LPR (License Plate Recognition) system. The LPR algorithm con-sists of three steps: l

4、icense plate locating, character segmentation and character recognition. This paper presents a new algorithm for character segmentation.There are many factors that cause the character seg-mentation task dicult, such as image noise, plate frame, rivet, space mark, plate rotation and illumina-tion var

5、iance. Our algorithm uses Hough transforma-tion and the prior knowledge in horizontal and vertical segmentation respectively and overcomes the dicul-ties mentioned above.Compared with the method of image binarization 2, this algorithm uses the information of intensity and avoids the abruption and co

6、nglutination of characters that are the drawbacks of image binarization. And because of using Hough transformation and the prior knowledge, the segmentation is more accurate and ro-bust than the simple projection method15.Changshui ZhangDept. of Automation, Tsinghua UniversityThe Institute of Inform

7、ation ProcessingBeijing 100084,C2 AlgorithmThe algorithm has three steps: preprocessing, horizon-tal segmentation and vertical segmentation.2.1 PreprocessingPreprocessing is very important for the good perfor-mance of character segmentation. Our preprocessing consists of s

8、ize normalization, determination of plate kind and object enhancement:2.1.1 Size normalization: The size of the plate images is an important factor for the accuracy of character segmentation. All the license plate images are normalized to 160*40 in pixel. The experiments show that this scale is fit

9、for character segmentation.2.1.2 Determination of plate kind: There are three kinds of Chinese license plates: black char-acters on a yellow background, white characters on a blue background and white characters on a black back-ground. The gray scale images are of two kinds: black characters on a wh

10、ite background and white charac-ters on a black background. The ratios of number of white pixels to that of black pixels are quite dierent in these two kinds of gray scale images. So the kind of a plate image can be determined by histogram analysis. Our posterior segmentation algorithm deals with pl

11、ate images of white characters on a black background. So in preprocessing the kind of the plate image is deter-mined and if a plate image is white characters on a black background its color will be reversed.2.1.3 Object enhancement: The quality of plate images varies much in dierent capture condi-ti

12、ons. Illumination variance and noise make it dicult for character segmentation. Then some image enhance-ment should be adopted to improve the quality of im-ages. As we all know, the image enhancement methods of histogram equalization and gray level scaling have some side eects. They may have the noi

13、se enhanced as well. For character segmentation, only the character pixels need to be enhanced and the background pixelsshould be weakened at the same time. In fact, a license plate image contains about 20% character pixels3. So these 20% character pixels need to be enhanced and the rest pixels need

14、 to be weakened. It is called ob-ject enhancement. The object enhancement algorithm consists of two steps. Firstly, gray level of all pixels is scaled into the range of 0 to 100 and compared with the original range 0 to 255, the character pixels and the background pixels are both weakened. Secondly,

15、 sorting all pixels by gray level in descending order and multiply the gray level of the top 20% pixels by 2.55. Then most characters pixels are enhanced while back-ground pixels keep weakened. Fig. 1 shows the result of object enhancement. It can be seen from Fig. 1 that after object enhancement th

16、e contrast of peaks and valleys of the projection is more significant than the original.Figure 1: Object enhancement2.2 Horizontal segmentation2.2.1 Hough transformation: The Hough transformation can be used to detect lines in an image6. For each pixel in image space (x0, y0), us-ing transformation,

17、r = x cos + y sin We get a curve r = x0 cos +y0 sin in the parameter space (, r). Suppose that there are n points in the image space. After translating them to the parameterspace, we obtain n curves in the parameter space. If these curves cross the same point (0, r0), then the n points in the image

18、space are on a line. So we can find lines in the image space by searching the cross points in the parameter space.2.2.2 Horizontal segmentation using Hough transformation: For plate images with large rotation, it is dicult to obtain horizontal segment lines by horizontal projection analysis. How-eve

19、r, for a single character, rotation has little eect on its horizontal projection. It is easier to analyze the horizontal projection of a single character and find the horizontal segment lines. So the horizontal segmentation algorithm is as follows:1) Find valleys of the vertical projection and then

20、ver-tically divide the plate image into many blocks. The division will not be very accurate because of the influ-ence of frame and rivet.2) Find the horizontal segmentation line for each block by analyzing the horizontal projection of the block. We call the horizontal segmentation line for a single

21、block a subsection line.3) Use Hough transformation on the midpoints of all subsection lines to eliminate the incorrect subsection lines and combine the correct subsection lines into a whole line.This method has a number of advantages. First, Hough transformation utilizing a vote strategy 6 and the

22、in-correct subsection lines are the minority, so the incor-rect subsection lines can be eliminated. For example, the horizontal segment lines of the block with rivet are often incorrect and can be eliminated by Hough trans-formation. On the contrary, the linear fitting method is more sensitive to th

23、e incorrect subsection lines. Sec-ond, it is a local projection method, which can weaken the influence of background, illumination variance and the rotation of plate. Third, it avoids the rotation cor-rection of images. In fact, rotation correction can cause distortion of image and make the characte

24、r recognition more dicult.Fig. 2 shows some results of horizontal segmentation. The white lines denote the horizontal segmentation po-sitions. There are images with rotation, background noise, illumination variance, rivet and plate frame influ-ence in the figure. The results show that the horizontal

25、 segmentation algorithm has a good performance.2.3 Vertical segmentationThe vertical segmentation algorithm is based on projec-tion analysis, constrained by the prior knowledge. As(a) Images with rotaiotn(b) Images with noise(c) Images with illumination variance(d) Images with rivet and plate frameF

26、igure 2: Horizontal segmentationwe know, the size of license plate is 440*140(mm), each character is 45*90(mm), and the interval between char-acters is 12(mm). And there is a big interval (34mm) between the first two characters and the last five char-acters. This information is used as prior knowled

27、ge. And by using the prior knowledge, the segmentation becomes more accurate. The vertical segmentation al-gorithm consists of four steps, as follows:1) Find candidates for vertical segmentation lines. We assigned a candidate for each valley of the vertical pro-jection.2) Estimate the size of the pl

28、ate and each character by using the position information of the horizontal seg-mentation lines and the candidates.Figure 3: Vertical segmentation3) Estimate the position of the left and right borders of the big interval, using the prior knowledge of character size. The variance of the pixels gray le

29、vel along a seg-mentation line should be small, because a segmentation line should be located in the interval of the plate and the pixels it crosses are background pixels with similar gray level. On the contrary, when it crosses a character, the gray level variance will be much larger. Based on this

30、 fact, the vertical segmentation lines (the left and right borders) for the big interval can be retrieved by searching around the estimated positions and finding the best segmentation lines with the minimum variance from the candidates.4) The other vertical segmentation lines can be located in the s

31、ame way.Fig. 3 shows some results of vertical segmentation. It can be seen that the vertical segmentation algorithm can exclude the influence of the space mark and the plate frame satisfactorily.3 Experiment resultsA database containing 697 license plate images is used to test the algorithm. Experim

32、ents show that the al-gorithm has good performance on character segmenta-tion, and can deal with images with disturb of noise, plate frame, rivet, space mark, rotation and illumina-tion variance. Fig. 4 shows some results.(a) Normal images(b) Images with noise(c) Images with illumination variance(d)

33、 Images with rotation(e) Plate images of black characters on a white background4 Conclusion and future workThe algorithm presented in this paper can segment the characters in license plate images accurately. The pre-processing can improve the accuracy of the segmenta-tion. The algorithm for horizontal segmentation, using Hough Transformation, can solve the problem of rivet, rotation, and illumination variance. The prior knowl-edge constrained vertical segmentation algorithm can restrain the influence of plate frame and space mark.There are still some further researches to