With a byte_dividing_to_bit method to store data,a bottom_to_top recurrence scheme ,a quadtree structure to represent the spatial layers,the proposed algorithm has the characteristics:loss compress and loss_less compress in one program,reading all pixel data only one time ,and gradually transporting the layer data.

This paper discusses characters segmentation and characters recognition in intelligent vehicle plate recognition system.

The new methods of the early technique of LP recognition, including LP location, LP image contrast estimation, binarization, and characters segmentation are studied and proposed.

License Plate Locating and Character Division;

The accuracy of character division is the prerequisite of correct recognition in the automatic recognition system for automobile license plate.

Various techniques of digital image processing in the license plate identification are introduced,and the key techniques such as edge detection of license plate and character division of license plate are discussed in detail.

This paper analyzes the application statement of traffic management of image auto-recognition system at home and abroad, demonstrates the advantages and disadvantages of the technique plan applied in all kinds of auto-recognition system currently, and researches the image location and character segment technique and makes a comparison between them.

Based on the advantage of FPGA in signal processing,a kind of parallel processing construction of character segment operation,which is one of the most important procedure of LPR system,and implements and optimizes the design in System Generator platform provided by Xilinx about DSP.

An algorithm combined texture-based methods with connected component methods was proposed for text segmentation in color complex image.

According to the morphic characteristics of the Chinese characters and background areas in a complex background image, we propose a fast text segmentation algorithm based on morph operation, and compare it with the LLT algorithm.