User Drawing Analysis for Mind State Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining a user's state of mind, such as questionnaires and image/video analysis, are prone to manipulation and may not accurately capture emotional states due to conscious user responses or behaviors.
Innovation Solution
A processing unit and method that analyzes shapes and colors in user-drawn images using a Machine Learning model to interpret emotional responses, converting the image into a machine-understandable format and providing recommendations based on identified emotions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If questionnaires or image/video analysis are used to determine state of mind, then automation is improved, but accuracy deteriorates due to user manipulation
Solution Approach 1:
The patent introduces drawing analysis as an intermediary method between direct user reporting (questionnaires) and physiological monitoring (image/video). The drawings serve as a mediator that captures subconscious emotional states without direct user awareness or manipulation, bridging the gap between automated detection and accurate measurement.
Solution Approach 2:
The patent replaces the mechanical/psychological system of conscious user responses (questionnaires, self-reporting) with an automated computer vision system that analyzes drawing characteristics. This substitution eliminates the need for user consciousness or effort, thereby preventing manipulation while maintaining automation.
2Measurement precision
If manual review of user-drawn art is performed, then accuracy is improved, but productivity deteriorates
Solution Approach 1:
The patent replaces the manual mechanical process of therapist review with an automated computer vision system using machine learning models. This substitution maintains the analytical depth of manual review while eliminating time constraints, thereby improving productivity without sacrificing accuracy.
Solution Approach 2:
The patent transforms the subjective artistic drawing into objective measurable parameters (shape characteristics, color distributions, spatial relationships) that can be processed automatically. This parameter transformation enables the transition from manual qualitative assessment to automated quantitative analysis, resolving the contradiction between accuracy and productivity.
3Ease of operation
If conscious user responses are analyzed, then ease of operation is improved, but reliability deteriorates due to manipulation
Solution Approach 1:
The patent uses drawing analysis as an intermediary that bypasses conscious user control. The drawing process remains simple and easy for users, but the analysis target shifts from conscious responses to subconscious emotional expressions captured in the drawing, thereby maintaining ease of operation while improving reliability.
Solution Approach 2:
The patent creates a copy of the user's emotional state through their drawing rather than directly analyzing their conscious self-report. This indirect copying method preserves the simplicity of the user's action (drawing) while eliminating the reliability issue of conscious manipulation, as the drawing reflects subconscious states beyond deliberate control.
Data Source
AI summary
Method and processing unit for identifying the state of mind of a user are described. The method for identifying state of mind of the user includes facilitating the user to provide a drawing drawn by the user. Further, the method includes receiving the provided drawing and user data while providing the image and converting the received drawing into a machine understandable format. The method further includes pre-processing the received drawing to identify one or more shapes and/or one or more colors in the received drawing. Thereafter, the method includes identifying the state of mind of the user based on the received user data, the identified one or more shapes, and/or the identified one or more colors by employing one or more Machine Learning (ML) models.