Image Analysis Apparatus with AI Logic Combination Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image analysis technologies struggle to execute appropriate reactions to events in facility security or equipment management due to difficulties in determining the urgency of events based on output results from multiple AI logics.
Innovation Solution
An image analysis apparatus and method that allow users to set a combination condition of output results from multiple AI logics, enabling the execution of appropriate reactions to events by inputting image data to selected machine learning models and specifying a degree of importance associated with the output combinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple AI logics are used for parallel image analysis, then the detection capability is improved, but it becomes difficult to determine the urgency and execute appropriate reactions to events
Solution Approach 1:
The patent introduces a combination condition setting unit as an intermediary between the multiple AI logics and the reaction execution unit. This intermediary processes the output results from multiple AI logics, applies user-defined combination conditions, and determines event urgency, thereby enabling appropriate reactions while maintaining the benefits of multiple AI logics for improved detection capability.
Solution Approach 2:
The patent implements preliminary action by allowing users to pre-define combination conditions and degrees of importance for various event scenarios before actual monitoring. When events occur, the system simply evaluates the AI logic outputs against these pre-set conditions, enabling rapid and appropriate reactions without complex real-time decision-making.
2Speed
If the system automatically determines event urgency based on AI logic outputs, then the response speed is improved, but the accuracy of determining appropriate reactions deteriorates
Solution Approach 1:
The patent implements feedback by allowing users to review and adjust the determined event urgency and associated reactions. The system provides the basis for urgency determination (combination of AI logic outputs and combination conditions) to users, who can then verify or modify the results, ensuring both fast response and high reliability in reaction appropriateness.
Solution Approach 2:
The system performs preliminary configuration of combination conditions and degree of importance mappings for different event types. This pre-programming enables the system to quickly determine appropriate reactions by simply matching current AI logic outputs against pre-defined scenarios, maintaining both speed and accuracy.
3Adaptability or versatility
If the system allows flexible configuration of combination conditions, then the adaptability to different scenarios is improved, but the system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the configuration process into distinct modules: combination condition setting unit, degree of importance setting unit, and reaction determination unit. Each module handles a specific aspect of the configuration, making the overall complex system manageable through modular design while maintaining high adaptability to different scenarios.
Solution Approach 2:
The patent implements universality by designing a standardized framework where the same combination condition setting unit and degree of importance setting unit can handle various types of events and scenarios. The system uses a unified structure for different AI logics and event types, reducing configuration complexity while maintaining versatility across multiple应用场景.
Data Source
AI summary
A server device stores therein a plurality of analysis models each of which outputs image analysis targets included in image data, accepts selection of the analysis model from among the plurality of analysis models and also accepts a degree of importance that is associated with a combination of the image analysis targets, acquires the image data, inputs the acquired image data to the analysis model for which the selection has been accepted, and executes a predetermined reaction in accordance with the degree of importance that is associated with the output combination of the image analysis targets.