Method and system for navicular identification

Abstract

The invention relates to the technical field of navicular identification, in particular to a method and system for navicular identification based on interest target regions and SUSAN (Small Univalue Segment Assimilating Nucleus) edge detection. The navicular identification method comprises the following steps: S1, converting an image to be identified into a gray level image; S2, extracting an interest target region from the gray level image; S3, utilizing a SUSAN operator to perform edge detection on the interest target region to obtain edge information, and performing target contour tracing on the edge information to form an shape of the target to be identified; S4, obtaining shape context of the shape of the target to be identified; S5, matching the shape context of the shape of the target to be identified with shape context of the shape of a template, and determining whether the target to be identified is the navicular identification. The method solves the problems of identification of a navicular object on an ocean surface scene and difficulty in navicular identification for on-board cargo, and can be widely applied to monitoring of exploitation of marine resource, thereby satisfying the requirements of fishery administration.

Description

technical field [0001] The present invention relates to the technical field of ship shape recognition, in particular to a ship shape recognition method and system based on edge detection of a target area of ​​interest and SUSAN (Small Univalue Segment Assimilating Nucleus, minimum nuclear value similar area). Background technique [0002] my country is a large ocean country, and the ocean provides a steady stream of resources for my country's economic development. At the same time, the pollution and destruction of the marine environment, the over-exploitation of marine resources, the increasingly frequent maritime economic activities and the competition of various countries for marine resources Increasingly, the identification of ship-shaped targets on or near the sea is becoming more and more important. The present invention proposes a ship-shaped recognition method and system under this background, which is used to recognize ship-shaped objects in a specific scene. The spec...

AI Technical Summary

Problems solved by technology

[0024] The above technology uses visual attention and HOG feature bidirectional fusion for ship target detection. By obtaining the salient ship area, although it can eliminate background interference and reduce the target search area, thereby improving the detection efficiency, it has the following disadvantages: if there are Another non-boat-shaped target area is more significant than the ship-shaped target area, so that the final target recognition area will not contain a boat-shaped target, which will cause missed detection; and, when using SVM support vector machine training, because the boat shape is Various (such as passenger ships, cargo ships, ships, etc.), if you want to accurately recognize the shape of the ship that appears on the sea, you need to train each shape of the ship, which will require a lot of templates, and the time complexity of training will be Very high; in addition, the recognition effect of ship-shaped objects in the vehicle cargo will be very poor, and it cannot be well used in the business of monitoring illegal fishing behaviors. At the same time, the recognition effect under the condition of ship target occlusion needs to be further studied

Images