Mi, C.; Cao, L.; Zhang, Z.; Feng, Y.; Yao, L., and Wu, Y., 2020. A port container code recognition algorithm under natural conditions. In: Yang, Y.; Mi, C.; Zhao, L., and Lam, S. (eds.), Global Topics and New Trends in Coastal Research: Port, Coastal and Ocean Engineering. Journal of Coastal Research, Special Issue No. 103, pp. 822–829. Coconut Creek (Florida), ISSN 0749-0208.

Automatic container code recognition is very important for modern container intelligent management system. Under natural conditions, aiming at the problems of uneven illumination, tilt and deflection of container number in port container code recognition. A new differential edge detection algorithm is used to realize binary segmentation of uneven illumination container number image, and then the problem of accurate location of container number deflection is solved effectively by the improved least square method, then use gradient descent projection based character correction and segmentation algorithm to correct and segment tilt container number; BP neural network to recognize the segmented characters. Finally, experiments are carried out on the images taken under different conditions. The comprehensive recognition rate is 96.8%, the localization rate is 2.4% higher than the traditional method, and the comprehensive recognition rate is 6.5% higher than yolov3 algorithm, which meets the real-time requirements.