Fluoroscopy examination system and method for carrying out automatic classification recognition on goods

Abstract

The present invention relates to a fluoroscopic inspection system for automatic classification and recognition of cargoes. The system includes: an image data acquiring unit, configured to perform scanning and imaging for a container by using an X-ray scanning device to acquire a scanned image; an image segmenting unit, configured to segment the scanned image into small regions each having similar gray scales and texture features; a feature extracting unit, configured to extract features of the small regions; a training unit, configured to generate a classifier according to annotated images; and a classification and recognition unit, configured to recognize the small regions by using the classifier according to the extracted features, to obtain a probability of each small region pertaining to a certain category of cargoes, and merge small regions to obtain large regions each representing a category.

Description

technical field [0001] The present invention relates to a radiation fluoroscopy imaging container safety inspection system, in particular to a method for automatic analysis and intelligent inspection of bulk container cargo, an automatic classification and identification method for bulk cargo in containers, semantic segmentation and classification of scanned images of bulk cargo, and container violations Intelligent inspection methods such as entrained cargo analysis, cargo quantity estimation, and tax calculation. Background technique [0002] Intelligent inspection is an inevitable trend in the development of container security inspection, and automatic classification and identification of goods is an important core content of intelligent inspection. Material identification technologies based on energy spectrum analysis currently include X-ray dual-energy material identification and neutron-X-ray technology. Dual energy technology can only identify four categories of orga...

AI Technical Summary

Problems solved by technology

Dual energy technology can only identify four categories of organic matter, inorganic substances, mixtures and heavy metals. Poor organic penetration, these deficiencies make it difficult to use in container cargo security

[0003] The automatic classification and recognition technology of scanned image goods based on image analysis has been slow in research for many years, and the algorithm function cannot meet the actual needs of customers.

The reason for this problem is that on the one hand, the goods are so complex that it is difficult to find effective features and rules for effective classification; on the other hand, the scanning equipment is scattered, and the customs keep the scanned images confidential, so it is difficult to obtain comprehensive image data for training classifiers On the other hand, image classification and recognition based on massive data analysis has relatively high requirements for algorithms and computing hardware, which brings certain difficulties to the research work.

Images