A method and system for locating numbered objects based on real-time image recognition

By utilizing edge computer vision recognition technology and mobile terminal systems, the problems of low efficiency, inaccurate positioning, and network dependence in finding numbered items have been solved. This enables rapid and accurate positioning of numbered items and simultaneous searching of multiple targets, making it suitable for logistics, warehousing, and office scenarios. It improves search efficiency and accuracy and is compatible with high, medium, and low-end mobile devices.

CN122135381APending Publication Date: 2026-06-02MUTUAL HELP NETWORK TECHNOLOGY (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
MUTUAL HELP NETWORK TECHNOLOGY (SHENZHEN) CO LTD
Filing Date
2026-01-18
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Existing technologies suffer from low efficiency in finding numbered items, inaccurate positioning, poor versatility, inability to simultaneously search for multiple targets, and network dependency issues. In particular, manual searching is inefficient and has poor recognition accuracy in logistics, warehousing, and office scenarios. Furthermore, existing image recognition technologies have significant limitations, failing to achieve real-time edge recognition and cross-category versatility.

Method used

Employing edge-side computer vision recognition technology, combined with multi-method number acquisition, image frame preprocessing, millisecond-level fuzzy matching, and multi-dimensional positioning alerts, it achieves rapid and accurate positioning of numbered objects. Through the mobile terminal system's built-in number acquisition module, image acquisition module, character recognition module, and positioning alert module, it supports screenshot OCR extraction, manual input, and third-party application authorized parameter transmission, achieving real-time recognition and simultaneous positioning of multiple targets without network dependence.

Benefits of technology

It significantly improves the efficiency of finding numbered items, shortens the search time to within 1 minute, improves recognition accuracy, supports simultaneous search of multiple targets, is suitable for various scenarios, and can be used normally in environments without network access, thus reducing labor costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122135381A_ABST
    Figure CN122135381A_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for locating numbered objects based on real-time image recognition, belonging to the field of computer vision recognition and mobile terminal system application technology. It solves the problems of low efficiency, inaccurate positioning, poor versatility, and network dependence in existing numbered object retrieval methods. The method includes: acquiring one or more target numbers entered by the user; numbered objects include express parcels, warehouse goods, files, and invoices; number acquisition methods include screenshot OCR extraction, manual / paste entry, and third-party application authorization parameter transmission; acquiring real-time image frames through the mobile terminal system's camera and performing Gaussian denoising, reflection correction, and rotation correction preprocessing; calling the on-device OCR engine to recognize characters and filtering candidate numbers using a regular expression algorithm; performing millisecond-level fuzzy matching between candidate numbers and target numbers; after successful matching, performing visual annotation and audio-visual alerts; and controlling flashlights / laser accessories to achieve physical light source positioning; supporting simultaneous positioning of multiple targets, distinguished by different colors. The mobile terminal system implementing this method includes modules for number acquisition, image acquisition, character recognition, number matching, positioning alerts, and multi-target processing. This invention enables real-time edge-side identification, network-independent operation, simultaneous multi-target positioning, cross-category applicability, high search efficiency and accuracy, and is suitable for various scenarios such as logistics, warehousing, and office work.
Need to check novelty before this filing date? Find Prior Art