Number Plate Detection system is very important in present scenario and is used to find the lost vehicles, to monitor and regulate the traffic. Number plate extraction for vehicle tracking has been developed, but each has its advantages and disadvantages. It is assumed that this system worked on images which have been captured in different climatic conditions. An automated system for extracting text is developed using JAVA in which, if we would like to extract and verify text from an image then we use OCR (Optical Character Recognition) technology. OCR software extracts all the information from the image into easily editable text format.
OCR is a system of converting scanned printed/handwritten image files into its machine readable text format. OCR software works by analyzing a document and comparing it with fonts stored in its database and/or by noting features typical to characters. There is good number of free OCR software tools. As our preferred program is Java, we can use one of the Java OCR libraries to extract text from an image.
We used ASPRISE OCR-java library. This extracted text is then used to find the path of the vehicle, i.e.., whenever the police man gives the vehicle number can get the details of the vehicle and the path of the vehicle in which it travelled. A sensor here senses the vehicle while it is approaching the traffic signal and also activates the digital camera so that the camera captures the image and vehicle number is extracted from the captured number plate. This extracted text can be stored in a text file for further verification.
Here, conversion from image to text is important because direct image comparisons may produce wrong results as images are captured in different climatic conditions. We compare this extracted number with other numbers which are captured at different traffic signals.
By doing so, we get the details of the vehicle like the path in which the vehicle travelled. We are proposing a real time application which recognizes license plates from vehicles to track the vehicle path. It is very difficult to identify the lost vehicle and also the vehicles which violate traffic rules. It is not possible to track the vehicle, by the user because they may not be able to identify the number from the moving vehicle. Therefore, it is necessary to capture the number plate of the vehicle and use this captured number to track the path of the vehicle. In order to track the path we need to extract number from the captured image, using JAVA OCR-libraries.
The digital camera is used to capture the image. A sensor is used to detect the vehicle approach and activates the digital camera as to stand ready to capture the image. The captured image is then passed for identification or recognition purposes.
After the image acquisition we have to perform Image digitization. Digitizing is the way of storing images in a form which is suitable for transmission and computer processing, which is nothing but storing in the form of bits.
Firstly Thinning will be applied to the plate in order to get the single pixels of each character that lies in the number plate image. Character segmentation is usually slicing the extracted image according to individual characters that lie on the plate image. Horizontal segmentation eliminates the unnecessary top and bottom parts of the license plate image. Vertical segmentation helps in separating each character on the license plate picture.
OCR software extracts all the information from the image into easily editable text format. Optical character recognition (OCR) is a system of converting scanned printed/handwritten image files into its machine readable text format. It is commonly used for passport documents, bank statements, computerized receipts, business cards etc.., It is a familiar technique of digitizing printed texts so that it can be electronically edited, searched, stored easily.OCR is a field of research in pattern recognition, artificial intelligence and computer vision.
After the Optical character recognition is completed the acquired information is then stored in to the text file.
With the obtained results, we track the path of the vehicle in which it has travelled by comparing the vehicle details at different traffic signals.