After Sales Quotation of Car Park License Plate Recognition System - Tigerwong

Commonly known as license plate recognition system, optical character recognition, introduction to character recognition, character recognition. The full English name is optical compact recognition, or OCR. It uses optical technology and computer technology to read and recognize printed or handwritten characters and convert them into a format that can be read by computers and people. OCR technology is a key link here. In OCR technology, printed character recognition is the most mature one, because it was first developed. As early as 1929, it was used by European and American countries to deal with a large number of newspapers and magazines, documents, documents and statements.

After more than 40 years of development and improvement, character recognition technology has become more naive and gradually realized the "electronization of information processing". However, the research on printed Chinese character recognition in China started in the late 1970s and has a development history of nearly 30 years. It can be roughly divided into the following three stages: 1. Exploration stage (on the basis of the research on the recognition of numbers, English and symbols from 1979 to 1985, since the late 1970s, researchers from a few units in China have explored the methods of Chinese character recognition, published some papers and developed a small number of analog recognition software and systems. This stage is long and has little effect, but it has bred fruitful fruits in the next stage. 2 R & D stage (from 1986 to 1988, the three years from the beginning of 1986 to the end of 1988 were the climax of the research on Chinese character recognition technology and the harvest of the research on printed Chinese character recognition technology. A total of 11 units conducted 14 evaluations on the effect of printed Chinese character recognition. These systems can achieve high indicators for sample sheet recognition: they can recognize Song typeface, imitation Song typeface, bold typeface and regular typeface, and the number of words recognized can reach 6763 at most, The recognition rate of font sizes from No. 3 to No. 5 is more than 99.5%, and the recognition speed can reach 10 14 words / s under the condition of 286 microcomputer, but the recognition rate of real text is greatly reduced, which is due to the change of printed text shape (such as text blur, stroke adhesion, broken pen, uneven black and white, poor paper quality, ink reflection, etc.) of the above system Poor adaptability and anti-interference, but the recognition system developed in the past three years has laid a foundation for the practicability of printed Chinese character recognition system, and the recognition system must go through the process from development to practicality. 3. Printed Chinese character recognition (character recognition) in the practical stage (from 1989 to now) Since its climax in 1986, the Department of electronic engineering of Tsinghua University, the intelligent center of the Institute of computing of the Chinese Academy of Sciences, Beijing Institute of information engineering, Shenyang Institute of automation and other units have developed and developed practical printed Chinese character recognition systems, especially the Tsinghua th OCR product developed by the Department of electronic engineering of Tsinghua University and the Shangshu OCR product developed by Hanwang group, It has always been at the forefront of technological development and occupies the largest market share, representing the development trend of printed Chinese character recognition technology.

At present, the research focus of printed Chinese character recognition technology has shifted from simple character recognition to automatic form recognition and input, layout analysis, layout understanding and layout recovery of mixed graphic and multilingual layout, business card recognition, Financial bill identification and ancient book identification, and presents many related identification systems, such as business card identification system, ID card identification system and "Huishi" launched by Zhong'an future technology Screen character and image recognition system, etc. the presentation of these new recognition systems marks a broad expansion of the application field of printed Chinese character recognition technology. Overview of the OCR recognition workflow of Zhongan future technology. The purpose of OCR recognition system is very simple. The image is converted to keep the graphics in the image. If there is a form, the data in the form and the characters in the image, one OCR recognition process analysis 1. Image input and preprocessing: 2. Image input: for different image formats, there are different storage formats and different compression methods.

Preprocessing: mainly includes two parts Binarization: most of the pictures taken by the camera are black images, which contain a huge amount of information. The content of the picture can be simply divided into foreground and background. In order to make the computer recognize characters faster and better, it is necessary to process the black image first, and only keep the foreground information and background information The foreground information is black and the background information is white, which is the binary image. 4. Noise removal: for different documents, the definition of dryness can be different. According to the characteristics of dryness, the license plate recognition system is called noise removal. 5. Correct inclination: because ordinary users are more casual when taking pictures of documents, the pictures taken can not be prevented In case of skew, the character recognition software is required to correct it. 6. Layout analysis: the process of dividing document pictures into paragraphs and branches is called layout analysis. Due to the diversity and complexity of actual documents, there is no fixed optimal cutting model at present. 7. Character cutting: due to the limitation of photographing conditions, it often causes character adhesion and pen breakage, so it is extremely limited To improve the performance of the recognition system, it is necessary for the character recognition software to have the character cutting function. 8. Character recognition: this research is very early.

There was template matching for a long time. Later, it focused on feature extraction. Due to the influence of text displacement, stroke thickness, broken pen, adhesion, rotation and other factors, it greatly affected the difficulty of feature extraction. 9. Layout restoration: People It is hoped that the recognized text will still be arranged like the original document pictures, and the paragraphs, positions and order will remain unchanged. This process is called layout restoration. 10. Post processing and proofreading: correct the recognition results according to the relationship of specific language context, which is post-processing. The current OCR recognition technology has been successfully applied to students for a long time All aspects of life: such as the all-in-one license plate recognition machine at the entrance and exit of the community, the bank card recognition used for bank account opening, the business card recognition tool for business card management, etc. OCR recognition technology is affecting my life.

Systematically introduce the development history of character recognition: first, character recognition is one of the branches in the field of computer vision research. This paper is a professional paper on license plate recognition system A highly informative article. It belongs to pattern recognition and artificial intelligence, an important part of computer science. Please follow the essay to see its development.