Waybill number identification method and device based on bill-of-lading scanning copy big data

Abstract

The invention discloses a waybill number identification method and device based on bill-of-lading scanning copy big data, belongs to the technical field of big data processing, wherein the device comprises a plurality of convolutional neural network (CNN) models, a voting mechanism and an image segmentation technology. The plurality of convolutional neural network (CNN) models establish a classification and identification model based on the bill-of-lading scanning copy big data, and the voting mechanism performs bill-of-lading scanning copy classification and identification. According to the invention, the plurality of models are adopted to carry out classification and identification on ship companies to which the bill-of-ladings canning copies belong, the classification result is obtainedaccording to majority voting rules, the classification and identification effect can be fully improved, an iterative updating mechanism of the identification model can be established for newly addedbill-of-ladings canning copies, and the performance improvement of the models is effectively maintained; therefore, the bill-of-lading scanned copies are accurately recognized, the bill-of-lading scanned copies are automatically renamed according to the recognition result, other services can be directly called conveniently, and the bill-of-lading scanned copies can be directly output as a file tobe archived.

Description

technical field [0001] The invention belongs to the technical field of big data processing, and in particular relates to a method and device for identifying a single number based on big data of a scanned bill of lading. Background technique [0002] With the rapid development of foreign trade export, overseas e-commerce, and cross-border logistics, sea transportation has become one of the mainstream modes of cargo transportation, showing a growing trend. As an important basis for sea transportation, the ocean bill of lading is a key element in the links of inquiry and payment. . [0003] With the development of big data and artificial intelligence technology, a special data set can be established through the big data of scanned ocean bills of lading, and the classification of scanned bills of lading and identification of tracking numbers based on deep learning recognition technology can effectively replace the traditional manual classification and entry process and improve w...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is: in order to solve the problem that the traditional operation method is generally carried out manually, which requires an operator with certain business experience to scan, classify, and enter the order number of the bill of lading. For different shipping companies, the order number may also be involved. This requires more manual intervention, and it is often difficult to quickly carry out large-scale collection when faced with a large number of scanned bills of lading, which brings more time and management costs. It is better for the manual operation of the salesman, but in the face of a large amount of scanned data of the bill of lading, the manual operation method has a large bottleneck in processing efficiency and accuracy, especially the misoperation caused by long-term repetitive work is not conducive to the bill of lading In view of the problem of effective archiving of scanned copies, a method and device for identifying single numbers based on big data of scanned bills of lading is proposed

Images