Machine vision solution automatic generation method based on large language model

Through the automatic generation system of machine vision solutions based on large language models, the problem of machine vision solutions deployment cycle and high cost is solved, and the effect of rapid deployment and reduced customer usage costs is achieved.

CN119690504AActive Publication Date: 2025-03-25SHENYANG INST OF AUTOMATION - CHINESE ACAD OF SCI
11 Cites 4 Cited by

Patent Information

Application Number
CN202411846713.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-16
Publication Date
2025-03-25
Estimated Expiration
2044-12-16

AI Technical Summary

Technical Problem

Machine vision solution providers consume too much deployment cycle and staff costs on the customer's site, and customers also have high cost of using machine vision solutions, and it is difficult for customers to adjust and maintain solutions themselves.

Method used

An automatic generation system is adopted for machine vision solutions based on large language models, including multi-agent modules, pre-implemented operator library and workflow engine. Through human language describing on-site problems, agents can automatically generate workflows and adjust parameters, reducing manual intervention and deployment time.

Benefits of technology

It effectively shortens the deployment cycle and personnel cost of machine vision solutions, reduces customer usage costs, and improves the application efficiency of machine vision solutions.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a machine vision solution automatic generation method based on a large language model. The module comprises a workflow agent, a parameter agent, an algorithm agent, a workflow engine and a pre-implementation operator library. According to the method, many problems in a traditional workflow-based machine vision solution construction process are solved to a great extent, including that workflow construction highly depends on personal experience of engineers, algorithm parameter performance is sensitive and difficult to adjust, and an application iteration debugging period is long. Besides, the method provided by the invention has multiple flexible deployment modes, the method can be deployed in carriers such as a client field industrial personal computer, a client field private cloud server, a public cloud server and the like, and a machine vision solution automatic generation task is completed. The solution deployment efficiency of a machine vision provider is greatly improved, and the personnel cost of the machine vision provider is reduced; the method is of great significance in improving the efficiency of the last kilometer of a machine vision industry chain.
Need to check novelty before this filing date? Find Prior Art