Car Number Plate Recognition - Report Example

Programming: Car Number Plate Recognition Your name Name of Assignment 19th July, 2014 Abstract Template matching in car number plate recognition is one of most important imaging technique. In this report, the issue that has been focus on is how to use Matlab while maintaining or improving the image recognition.. Some of these techniques are affected negatively by the technical problems that happen when the number plate recognition is taking place. However, these limitations or problems can be solved by segmentation techniques. Color segmentation and Hough line detection are the most efficient number plate imaging technique and is counted to be the most effective. The number plate was The text is divided in four chapters. The first, introduces the origins of digital image processing, also a little resume about the following algorithms that are needed for develop the system ANPR. The first is the initial detection of a possible number plate using edge and intensity detection to extract information from the image. The second and the third step, thresholding and normalization, are necessary to use the images in the following stages; the text of the plate is found and normalized. With the segmentation, each character of the plate is isolated for subsequent recognition. The last step reads the characters by correlation template matching, which is a simple but robust way of recognizing structured text with a small set of characters. It is evaluated the system’s speed and his error rate. Table of Contents Abstract 2 Introductions 4 Design and resolution 4 Results 5 Color segmentation 7 Template matching 9 Conclusion 10 References 12 Appendix 13 Introductions Template matching of number plate recognition of imaging is technique that can be used to accelerate number plate image acquisition. This technique can be illustrated by use of MATALB. This involves processing of number plate using matching of colors and numbers that has an asymmetric sampling of k-space. This technique is used to reduce the image acquisition time by significantly minimizing the phase encoding steps. It can also be used to reduce the noise time by moving the noise away from the center in the image window. During the image processing and filtering utilize the Fourier transform. It is common to use the Fourier transform while processing images from astronomy. But it is often very useful when dealing with structures that are repetitive in nature such as crystals. The ability of Fourier transforms to identify periodically repetitive components in a given lattice or image is what makes it a favorable method for removing irregular dirty spots or noise from images. This method helps in image compression, reconstruction, sampling, and correlation and convolution theory just to name a few. Design and resolution Location plate- the number was provided in word then converted into JPG image which was input to Matlab lab using imread function for the process of image processing. The image was resized using 480 * 640 pixels. The color of the image was maintained as yellow and black as its original. The image is saved as ‘car number plate.JPG’ in 'D:\Documents and Settings\Administrator\My Documents. Segmentation – the Otsu method was used to have a binary image to select optimal level for thresholding. Here noise is eliminated and characters and segmentation of characters. Recognition – here d template matching was used where car number plate image obtained program function. In this case each character is tested at each position and the highest probability is stored along with its position, the highest character match is selected. Results Here the results the number plated is shown in terms of color segmentation, Hough line detection and template matching. Color segmentation was done to remain with yellow color and black for the numbers as shown below. The colors were grouped using Matlab program by their pixels in the image in logical ways. This used color information, edge extraction and thresholding to able to extract and segment as shown in the extract. The following code helps to produce the original image without being processed. >> I = imread('D:\Documents and Settings\Administrator\My Documents\car number plate.JPG'); >> s=max(size(I)); figure, imshow(I); hold on for i = 1:size(s,1) rectangle('Position',s(i,:),'LineWidth',5,'LineStyle','-','EdgeColor','r'); end title('origin image'); hold off; Reduction in the Hough line detection means that few peripheral lines in the k-space are acquired; hence the image resolution, the SNR and the process time are affected. Reduction in image edges reduces the resolution as well as the processing time. However, the SNR is increased. The reduction in processing time is proportional to the reduction in noise because less phase-encodings are obtained. Images reconstructed using threshoding by eliminating color as shown. Color segmentation The figure above shows that the brightness or darkness is equal that is 50% this means that adaptive filters value separates right feature of an object or image by showing grayscale intensity. This ensures that the image background values and contract adjustments are well pronounced. The effect of adaptive filters is illustrated in the e ray above. Dark features of a bone are shown clearly while right features are pronounced. It helps to obtain a thin edge of grey level image instead of binary image. It maximizes edge information fields by having a minimum allowance for excess or missed edges. In the image above the information of the original image has been changed by minimizing edge information to produce an adaptive filtered image. When thresholding operation was used on the image, the pixels were detected and changed by the use of action filaments. The images produced and viewed using binary blocks where the grey level of the image has been turned into black. The blackness that is pronounced is as a result of binary block configurations within the edges. When a threshold function is applied to an image, filtering takes place compromising the image detail and computational complexity which turns blocks to be large thus increasing the configurations. Small configurations are minimized becoming insignificant. The filtering of the image using threshold functions helps eliminate grey parts of the image. The code that helped in separating the colors is The image produced is as follows; Template matching Character segmentation was done using blob analysis as shown by the code in the appendix. The command of ‘regionprops’ gives us the properties of groups that are connected by segregating area and BoundingBox. These codes help to locate the boundaries of characters and are used to have boundary coordinates that cropped the characters. This is done as shown below The reconstructed car number plate image had their color change as well the edges disappeared as shown above. The image had a lot of light meaning that the filtered allowed light to go through. The image was more blurred than in other cases as shown above. This means the filter reduced the sharpness of the image and the colors of the image. This filter is used in modeling optical length which can be used in blurring objects. It is also used in designing equipments which can reduce noise of the image. If the image obtained after filtering is added to the original image a clear and sharper image will be made. Conclusion The car number plate photo was filtered and new images produced had same numbers but color and characters, it can be stated that they are almost similar with the original image. This confirms that some filters are low pass filters as they allowed only high frequency features to pass through while high pass frequency features were rejected. Therefore the image is not affected very much in terms of color, inversion, opacity and saturation. Looking at the original image and the modified image side by side, comparison you will note that the arrays of the image changes are the edges not at the centre. It maintains the information of the images as the original. This is designed to reduce the brightness of the image. This kind of filter can be used to improve the quality of an image which has shadows. An image develops shadows when it is taken under too much light. It also affects the edges of the image. Looking at the contrast between the original image and the filtered image, one will note a difference in the color of the image. There is also a departure in terms of the shadow of the image. The application of this filter is when one wants to reduce the sharpness and the color of an image. It can render a color image to take black and white element. It filters is important because it is used in detecting the edge of the image. It can also be used in computing at derivative of the image that is at what rate the image change does. Most black and grey areas will be visible. In this case, the accuracy/normalization of the image is reduced and non-linear features were visible. There is a loss of information about the images edges, and when applying a high pass filter where the edges are clear but there is a big loss in information about the image contrast. References Bergenudd, E., 2006, Low Cost Real Time License Plate Recognition for a Vehicle PC’, Master’s Degree Project. Gonzalez RC & Woods, RE, 2008, ‘Digital Image Processing, Instructor's Manual’ New York: Prentice Hall. Gonzalez, RC, Woods, RE and Eddins, SL, 2009, Digital Image Processing Using MatLab’, New York: Prentice Hall. Appendix Read More