Study On The Character Image Recognition System Based On Virtual Instrumentation

With the coming information times and the widely application of computer, more and more countrys pay more attention to the information industry. Contrast with the foll speed development of information technology, a problem is encountered in front of us. That is the conflict between the high speed of computer data processing, transmiting and the low speed of data input. In the past the information usually record on the paper in the form of characters image. How to transform these huge of image into character information that easily processed by couputer rapidly and reliably is an important factor for the process of societyl informational. Human usually accept information by vision and in the daily life there are 75%-85% information needed processed is image information. If wanted to realize the aim of processing character information automatically, the first thing is to realize the recognition of character image by computer.In this paper the problem of computer recognition to the character image is studied. According to the style of writing, the characters are divided into two forms, printing characters and handwriting characaters. Based on every characteristic of the two styles, system realizes the function of image pre-processing, characteristic extraction and character recognition. So the whole system is divided into four parts, image collection module, image pre-processing module, characteristic extraction module and character recognition module.The image collection hardware involves Lenove M6200 computer, which has 2.8GHZ CPU and 512M memory, NI company PCI-1411 image acquisition board, and image collection box. The box adopts black background and installs Panasonic WV-CP240/G color video camera. And the 40W loop light guarantees the collected image quality.The pre-process module involves the process of binary process, diminish noise, image enhance, character oriention, edge extraction, characters split and so on. After that a signal character with the merit of distinguished characteristic, easily processed and recognition by computer is formed. In the process of character oriention, a new methord is introduced, which detect the edges of notes and use it to establish a relative coordinates, in which the location of the characters are orientated.According to the individual needed of printing and handwriting character processing, two differents strategys are adopted by the system. For the printing characters, first the system completes the process of thining and unification and then extract the combined characteristic of the net characteristic and the intersection characteristic. Basing on these a standard characteristic database of characters is established. And for the handwriting characters, arithmetic of characteristic extraction is designed, which base on the border characteristic and the character's holes and strokes characteristic.The recognition arithmetic of printing character adopts template matching, which compares the character with the standard characteristic database and recognize thecharacter as the nearst Euclid distance standard character in the standard characteristic database. And for the handwriting character the system adopts BP neural network to realize the recognition.The software development adopts the NI Company's Virtual Instrument Labview and vision tookit IMAQ Vision. The software realizes the function of image collection, image pre-process, character extraction and character recognition. It has the merit of easily operation, friendly interface, high reliability and so on. The experiments show that the recognition accuracy of printing character is 95.5% and the recognition accuracy of handwriting character is 90.3%.