MBCNN-based ship target fine-grained classification system and method

Abstract

The invention discloses a ship target fine-grained classification system and method based on an MBCNN. The method comprises the steps of obtaining a ship image of a visible light remote sensing image, inputting the ship image into a pre-trained ship target detection model, and outputting position coordinate information of all ships through network forward reasoning; cutting the visible light remote sensing ship image according to the position coordinate information of the ship to obtain a ship image; preprocessing the ship image to obtain a normalized ship image; constructing a ship target fine-grained classification model, and performing pre-training; and inputting the ship image into a pre-trained ship target fine-grained classification model, and outputting a category label to which the ship belongs through network forward reasoning. The method has the characteristics of high accuracy and high recognition efficiency, and solves some calculation defects of the traditional BCNN and problems existing in fine-grained classification.

Description

technical field [0001] The invention relates to the field of computer image classification, in particular to an MBCNN-based ship target fine-grained classification system and method. Background technique [0002] Most of the current classification techniques are mainly aimed at the classification of general objects, focusing on the specific category without taking into account the subtle differences between objects of the same type. Compared with general-purpose image classification, fine-grained image classification needs to judge more fine-grained image categories. Maritime ship target recognition requires fine-grained screening. In terms of civilian use, it can be used to monitor and manage the work of ships, which can improve the efficiency of offshore operations and ensure safety. [0003] An article on fine-grained classification in ICCV2015 Bilinear CNNs for Fine-grained Visual Recognition The feasibility and advantages of bilinear network in fine-grained classif...

AI Technical Summary

Problems solved by technology

[0005]The purpose of this invention is to address the shortcomings of the existing fine-grained classification technology for ship targets, especially for the presence of multiple ships and fine-grained areas in visible light remote sensing images at sea Intra-class differences are greater than inter-class differences, and image quality is blurred, which leads to poor fine-grained recognition of ship targets and slow recognition efficiency. A MBCNN-based ship target fine-grained classification system and method are provided.

Images