Finger Size Estimation Using Reference Object Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating the size of a finger for electronic ring fitting are cumbersome and inaccurate, especially in e-commerce settings where users cannot try on rings virtually.
Innovation Solution
A system and method using image processing and deep learning techniques to estimate finger size by acquiring images of a user's hand and a reference object, processing these images to recognize and isolate the hand and fingers, and determining finger dimensions based on the reference object's dimensions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional e-commerce ring purchasing is used, then users can shop conveniently online, but users cannot accurately estimate ring size without trying on physically
Solution Approach 1:
The patent uses image capture and processing to create a digital copy of the user's finger, analyzing the image to estimate finger circumference and diameter. This virtual measurement copy replaces the need for physical try-on while maintaining measurement accuracy through computer vision algorithms that calculate dimensions from the captured image and reference object comparison.
2Measurement precision
If custom finger measurement devices are developed, then measurement accuracy improves, but device complexity and cost increase
Solution Approach 1:
The patent makes the smartphone camera serve multiple functions: it acts as both the image capturing device for finger measurement and the reference object detection device. The same processor performs both image processing and dimension calculation tasks. This multi-functionality eliminates the need for dedicated measurement hardware while maintaining measurement precision through software-based image analysis.
Solution Approach 2:
The system uses the user's existing smartphone hardware (camera and processor) to perform the measurement function. The smartphone itself provides the measurement capability without requiring external specialized devices. The reference object can be any commonly available item, and the system automatically calibrates measurements using that object, making the measurement process self-sufficient.
3Measurement precision
If multiple reference objects are required for calibration, then measurement accuracy improves, but the process becomes more complex and time-consuming
Solution Approach 1:
The patent requires only a single reference object with known dimensions to be present in the image alongside the finger. This partial calibration approach (using one reference object rather than multiple) provides sufficient accuracy for ring size estimation while significantly reducing the time and complexity of the calibration process. The system calculates the scale factor from this single reference object and applies it to determine finger dimensions.
Data Source
AI summary
A system and a method for estimating size of a finger of a user. The method comprises acquiring one or more images of a hand of the user and a pre-defined reference object by an image capturing device, and recognizing the pre-defined reference object and the hand of the user, for removal of background of the one or more images. By connecting outermost points on contour of the hand, a bounding region is created by the processor. Points of convexity defects are determined on the hand for identifying each finger of the hand. A second bounding box is assigned to a finger of the hand. A size of the finger of the user is estimated based on dimension of a width of the second bounding box determined from on actual dimensions of the pre-defined reference object.