Container code identification method and device

Abstract

The invention relates to the technical field of image processing, in particular to a container code recognition method and device, and the method comprises the steps: carrying out the image collection of a container, and inputting the collected image into a neural network prediction model for processing, thereby obtaining a target box image. And searching the contour in the target box image, determining a minimum enclosing rectangle for the contour, and when the area of the minimum enclosing rectangle is greater than an area threshold value, determining that the region is a coding region image. The method comprises the following steps: encoding a plurality of character images, carrying out image processing on the encoded area image to obtain independent single character images arranged in sequence, identifying through a neural network prediction model, determining prediction probabilities and indexes of the character images, selecting the index corresponding to the maximum prediction probability in the prediction probabilities, and determining the specific encoding content of the character images. And code identification of the container is completed. According to the scheme, the recognition accuracy and the recognition speed are greatly ensured.

Description

technical field [0001] The present invention relates to the technical field of image processing, and in particular, to a container code identification method and device. Background technique [0002] With the continuous development of economic globalization, global trade exchanges are becoming more and more frequent. As a loading unit in transportation, containers are more and more favored by the transportation industry due to their high strength, fast loading and unloading, and large volume. The container throughput of major ports is increasing year by year. In order to improve the rapid customs clearance of containers, it is necessary to change the traditional manual registration of container numbers. Therefore, the automatic identification technology of container number plays a very important role in the modern intelligent management of the transportation industry. [0003] Many scholars in the industry have conducted research on container identification algorithms, and...

AI Technical Summary

Problems solved by technology

[0003] Many scholars in the industry have conducted research on container identification algorithms, and proposed algorithms such as template matching algorithm, support vector machine, and faster CNN to identify container codes, but these algorithms have low real-time performance. These existing identification systems for container numbers, The time from inputting the container image to the successful identification of the final container number is still at the second level

For a real-time system, it still cannot meet the requirements of millisecond level. At the same time, the robustness of the algorithm is poor, and it is greatly affected by the outside world. Once the illumination changes or the characters are incomplete, the recognition rate will be greatly reduced.

Images