Box recognition method, system and device based on multi-vision and storage medium

By using multi-view vision technology to identify and plan the location of container ship hatch covers in ports, the problem of high equipment cost and low accuracy in existing technologies has been solved. This enables efficient and safe self-positioning and path planning, and is suitable for vehicle-mounted equipment with limited resources.

CN115880360BActive Publication Date: 2026-06-26SHANGHAI ZHULU INTELLIGENT TECH DEV CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHANGHAI ZHULU INTELLIGENT TECH DEV CO LTD
Filing Date
2022-11-17
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

In port scenarios, hatch covers on container ships accumulate on quay cranes or yards during unloading, making it difficult to build point cloud maps and locate unmanned vehicles. Existing 3D point cloud equipment is costly and computationally complex, and monocular visual depth estimation is unreliable, affecting the accuracy of vehicle path planning.

Method used

A multi-view vision-based approach is adopted, which uses binocular cameras to acquire images from different perspectives. By using image recognition neural networks and stereo vision neural networks, the local patterns, corner positions and 3D matching information of the box are identified, and a spatial box with relative position is generated. Combined with navigation path planning, self-localization and path optimization are achieved.

Benefits of technology

It improves the alignment accuracy between container trucks and quay cranes, reduces the demand for computing resources, is suitable for vehicle-mounted edge devices with limited resources, reduces the risk of misidentification and collision, and improves driving safety and real-time performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115880360B_ABST
    Figure CN115880360B_ABST
Patent Text Reader

Abstract

The application provides a box recognition method, system and device based on multi-view vision and a storage medium, and the method comprises the following steps: obtaining a first image and a second image at different angles; inputting the first image and the second image into an image recognition neural network respectively to obtain local pattern information, corner point position, parallax information and 3D matching information representing a box; inputting the local pattern information and the corner point position into a context adjustment layer of a character detection neural network to obtain stereo information corresponding to the local pattern representing the box; and obtaining a spatial box with relative positions based on the corner point position and the stereo information. The application can enhance the algorithm recognition effect, has good real-time effect, consumes less computing resources, is convenient to deploy on a resource-limited vehicle-mounted edge device, is not prone to causing misrecognition and collision, and improves driving safety.
Need to check novelty before this filing date? Find Prior Art