Gesture Recognition Device Hand Orientation Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gesture recognition systems face errors in determining the left or right hand of a user based on smartphone posture, leading to incorrect hand recognition.
Innovation Solution
A gesture recognition device with a digital camera having an imaging range changing mechanism, which adjusts to capture a proper face position image and determines hand orientation using machine learning models to accurately identify whether the hand is the right or left hand.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If hand determination is made based on smartphone posture, then gesture recognition can be performed, but erroneous determination of left or right hand occurs
Solution Approach 1:
The patent replaces the mechanical/posture-based hand determination method with a machine learning-based image analysis method. The processor uses a trained model to analyze the captured image and determine whether the hand is the left or right hand, eliminating the need to infer hand type from smartphone holding posture and thereby eliminating determination errors.
2Device complexity
If only right hand gesture dictionary is prepared, then recognition simplicity is maintained, but left hand gestures cannot be recognized
Solution Approach 1:
The patent applies asymmetry by performing mirror-image inversion processing on the captured image when the hand is determined to be the left hand. This transforms the asymmetric left hand image into a symmetric form that matches the right hand gesture dictionary, allowing the existing right-hand-only dictionary to recognize both left and right hand gestures without requiring separate dictionaries.
Solution Approach 2:
The patent uses image inversion as a solution: when a left hand is detected, the image is inverted horizontally to convert it into an equivalent right hand image. This inversion approach allows the system to use a single right hand gesture dictionary for both left and right hand recognition, maintaining simplicity while achieving versatility.
3Adaptability or versatility
If face position is not centered, then imaging flexibility is maintained, but hand position relative to face cannot be accurately determined
Solution Approach 1:
The patent applies preliminary action by first adjusting the captured image so that the face position is centered before performing hand determination. This preprocessing step establishes a consistent reference frame, allowing the system to accurately determine hand position relative to the face using fixed positional relationships, thereby improving measurement precision while maintaining imaging flexibility through subsequent image processing.
Data Source
AI summary
A gesture recognition device that recognizes, based on an image obtained by imaging a person to be imaged with a digital camera having an imaging range changing mechanism for changing an imaging range, a gesture represented by a hand of the person to be imaged includes at least one processor. The at least one processor controls, in a case where a position of a face of the person to be imaged in the image deviates from a predetermined set position, an operation of the imaging range changing mechanism to cause the digital camera to capture a proper face position image with the position of the face as the set position, and determines whether the hand detected from the proper face position image is a right hand or a left hand based on a positional relationship between a position of the detected hand and the set position.