Decision-Tree Picture Search for Vague-Memory Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional picture searching methods rely on keywords, which are ineffective when users have a vague memory of the pictures they are seeking, such as remembering a person at a certain location at a certain time, lacking a good solution for such scenarios.
Innovation Solution
A picture searching method using a decision tree that acquires features of pictures, generates questions for nodes, and guides users through the tree based on their answers to find the desired pictures without relying on keywords.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If keyword-based searching is used, then searching is simple to operate, but searching accuracy deteriorates when user memory is vague
Solution Approach 1:
The patent introduces a decision tree as an intermediary between the user's vague memory and the picture database. The decision tree acts as a mediator that guides users through structured questions, translating vague recollections into precise search criteria without requiring users to formulate accurate keywords
Solution Approach 2:
The system implements feedback by asking users a series of questions based on the decision tree and adjusting the search path according to user answers. This iterative feedback process refines the search criteria dynamically, improving accuracy while maintaining ease of operation through guided interaction
2Measurement precision
If decision tree-based fuzzy searching is used, then picture searching accuracy improves for vague memories, but device complexity increases
Solution Approach 1:
The decision tree is pre-generated and prepared before the actual searching process. All possible search paths and questions are structured in advance based on picture features, so during searching, the system only needs to traverse the pre-built tree rather than constructing search logic in real-time, reducing operational complexity
Solution Approach 2:
The searching process is segmented into discrete questions and decision nodes organized in a tree structure. Each node represents a specific feature or attribute, breaking down the complex task of fuzzy picture searching into manageable, independent decision steps that are easier to process and implement
Data Source
AI summary
The present disclosure discloses a picture searching method and apparatus, an electronic device, and a computer readable storage medium, and relates to the field of artificial intelligence, such as intelligent cloud, computer vision, deep learning, or the like. The method may include: acquiring features of pictures in a picture set to be searched, generating a decision tree according to the acquired features, and configuring corresponding questions for nodes in the decision tree respectively; and if a picture searching request of a user is received, asking the user the configured question, selecting an optimal path from the decision tree in conjunction with an answer of the user, and taking the picture corresponding to a leaf node on the optimal path as a searched picture. With the solution of the present disclosure, the picture required by the user may be conveniently and accurately searched without a keyword.