Research And Implementation Of Container Number Identification Based On Video

With the development of information and computer technology, the container number identification system has got unprecedented progressing. After bar code identification system, a video identification system is present. The video identification system has better performance and more advantages than bar code identification system, so that the world's container number identification systems almost chose video identification systems. Bar code identification system is almost out of the stage of history.Although the video-based container number identification system has many advantages, it also has some disadvantages. For example, the use of the system is affected by fog, rain, oil, dust, concavo-convex creases of container's surface and other factors. So that the container number characters will distort or miss. These situations will affect the identification quality. To resolve the problem, many researchers join together to improve the quality of the system. In order to reduce the influence of the container's surface concavo-convex creases, depth correction algorithm is designed.Depth correction algorithm is designed by using computer vision, digital image processing technology and the relevant prior knowledge, which is practical. The purpose of the algorithm is to correct the number character, which is on the crease slope. It is based on the knowledge of container standards and doses the affine transformation by using the vertical projection to determine the location and width of the concave-convex crease. The piecewise linear function to enhance the image, median filter, geometric correction, Hough transform are finished in the image preprocessing. With the prior knowledge of the container number of the region, canny edge detection operator, dilation and erosion are used to find the number region. Vertical projection is used to segment the character and decide the width. After the normalization and standardization, the module matching is chosen to recognize character.The thesis improves algorithm of the video identification system, and there is also a simulation about it under Matlab platform. The simulation makes the algorithm product and gives the expected result.