Available online at www.sciencedirect.com Procedia Engineering 32 (2012) 348 353 I-SEEC2011 A new technique for distance measurement of between vehicles to vehicles by plate car using image processing J. Phelawan, P. Kittisut *, N. Pornsuwancharoen Nano Photonics Research Group, Department of Electrical Engineering, Faculty of Industry and Technology, Rajamangala University of Technology Isan, Sakon Nakhon Campus, Sakon Nakhon, 47160, Thailand Elsevier use only: Received 30 September 2011; Revised 10 November 2011; Accepted 25 November 2011. Abstract We present a new technique about to measurement distance between vehicles to vehicles. The camera in front of vehicle can be capture to images the next vehicles. We have the picture process by Fast Fourier Transform (FFT). The systems can check the position of the plate car with the properties[1] of the picture. The systems can be compared with data stored in the database as reference again. When we know the distance of between vehicles to vehicles and we control the distance of break security system. 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of I-SEEC2011 Open access under CC BY-NC-ND license. Keywords: distance measurement; image processing; FFT 1. Introduction System overview Currently, Car plays more important role in daily life because it is a part of method to reach each person s destination nowadays. As a result, the amount of cars driving on everyday road is increasing with the higher risk of accident. With this concern, the researcher propose the prevention of such accident by using image processing in order to measure the distance between vehicles to vehicles by plate car. When two cars are driving so close to each other with high risk of crashing then this prevention shall make the drives recognize immediately. Figure 1 is proposed to measure the distance between the car with the car. It will take a picture of the car in front. To get the original image. And the original image onto the image. To find the location of the License Plate. Send to compare the data in the database to obtain the correct distance. 1877-7058 2012 Published by Elsevier Ltd. doi:10.1016/j.proeng.2012.01.1278 Open access under CC BY-NC-ND license.
J. Phelawan et al. / Procedia Engineering 32 (2012) 348 353 349 2. Theory and Background Image Processing Algorithm Fig. 1. Illustrate plate car locating by using image processing Digital Image Processing is using computer algorithm to process image on digital image related to image processing on calculation and color management in pixel or model of color. The color is changeable for the benefit of operation, for example, RGB benefits for human s visibility while CMYK is the real color used in computer. Dilation and Erosion Dilation [2] is the method to expand the structure of image to be larger. The aspect of expanded image will depend on structuring element scanned on image as shown in Fig. 2.
350 J. Phelawan et al. / Procedia Engineering 32 (2012) 348 353 Fig. 2. Sample of Dilation Dilation is represented by B A with the following equation:a} A] ) Bˆ A B = {z [( z by A represents image for dilation B represents structuring element Fig. 3. Erosion Erosion [2] is opposite to Dilation it is the method to decrease the size of image s structure. The aspect of reduced image will depend on structuring element scanned on image as shown in Fig. 3. Erosion is represented by A B with the following equation: A B = {z [( B ) A} z by A represents image for erosion B represents structuring element
J. Phelawan et al. / Procedia Engineering 32 (2012) 348 353 351 Opening and Closing Opening [2] is the method to make the surface of material curve or flat by cutting or removing bottleneck and extended area. The method is bring the image open through the process of erosion then bring the outcome through dilation. Segmentation Generally, segmentation is analysis or interpretation. Segmentation image is distinct region that is related to others in image. Moreover, it is also the method to combine pixels with the same qualities. Segmentation will link among low-level image processing related to management and analysis of pixels grouping specifically on or interested field. We will find some techniques of segmentation in some application programs related to detection, recognition and measurement of object in the image or detecting license plate and recognition Segmentation is deemed to be the hardest responsibilities necessary for image analysis. To complete the process, the segmentation has to be completed. Segmentation is divided into 2 types including: 1. Similarity properties used for combining pixel into the same group. 2. Discontinuity of pixel used for separating pixel out of our interested object from the background. Thresholding It is the technique used in several different image processing. Thresholding is the method of changing that stores many different values to be a new database with only 2 values, namely, determining pixel with lower value thanthreshold as one value and pixel with lower value than Threshold as another value or black and white (1, 0) The image that is processed through this process will be Binary Image and Image Threshold is deemed to be one kind of segmentation. Edge Detection Edge [6] typically might be the border of area; however, its definition here means the change in gray level when measuring. 3. Result and Discussion Programming The process of programming will receive the image from the camera mounted in front of the car then send such image to the program for transforming such image into grayscale[11] or being in the status. [0, 1] This will indicate the position of plate. However, some received images might have no plate because the image record will be taken every 2 seconds. All images will be commanded to find the position of plate. Fig.4. The process of the system
352 J. Phelawan et al. / Procedia Engineering 32 (2012) 348 353 The complete image will have plate on it then such image will be sent to module to compare with the data in database. The database has already stored the size of plates and the distances between the cars. Output will be obtained from data comparison in database then display the result of the distance between the cars. With the limited data, some images will not have the distance between the cars. Fig. 5. Flow chart over all process Fig. 6. Illustration of Program Operation
J. Phelawan et al. / Procedia Engineering 32 (2012) 348 353 353 4. Conclusion Are presented in this document is one way of measuring the movement of cars by the use of images to be processed for the desired effect. They also have defects in the clear. Or degrees of visual angle. As well as the integrity of the images in the database compared with. The obtained result will show the distance value between the cars only. The accident preventing section shall have the warning system to inform the drivers about this event by any easier warning, for example, cap light or alert sound, etc.the result of plate license recognition obtained from this project is also applied in car tracking to find the real owner of each car in order to prevent car robbery.overall, the presentation in this document is only a concept and presentation of a preliminary stage. It requires a functional test of the device, some method contents and some steps of machine working need to be improved. These problems had been solved in the experiment. Reference [1] Watcharin Khunathipapong, Sathit Sirikhul, 2007, License Plate Recognition, A Project Report for the Degree of Bachelor of Engineering, Khon Kaen University, Khon Kaen. [2] Gonzalez, R.C. and Richard E. Woods, Digital Image Processing, Second Edition,2002. [3] Borgefors, 1986 G. Borgefors, Distance transformations in digital images. Computer Vision, Graphics, and Image Processing, 34 (1986), pp. 344 371. [4] Dubey, P. Heuristic Approach for License Plate Detection, NECTEC, pp. 366-370 [5] Y. Cui, Q. Huang, Automatic license extraction from moving vehicles, 126, 129. Retrieved Dec.20, 2006 [6] Gonzalez, R.C. and Richard E. Woods, Digital Image Processing, Second Edition, 2002. [7] K.N. Plataniotis, D. Androutsos, A.N. Venetsanopoulos, Fuzzy adaptive filters for multichannel image processing Signal Processing, Volume 55, Issue 1, November 1996, pp. 93-106. [8] Amir Sedighi, Mansur Vafadust, A new and robust method for character segmentation and recognition in license plate images, Expert Systems with Applications, Volume 38, Issue 11, October 2011, pp. 13497-13504 [9] Gary G. Makowski, Kumares C. Sinha, A statistical procedure to analyze partial license plate numbers, Transportation Research, Volume 10, Issue 2, April 1976, pp. 131-132. [10] Thomas W Cusick, Computer Licence Plates, Computers & Security, Volume 20, Issue 5, 1 July 2001, pp. 392-394 [11] Andrew M Wallace, Greyscale image processing for industrial applications Image and Vision Computing, Volume 1, Issue 4, November 1983, pp. 178-188.