3D Shape Measurement via Image Segmentation and Normal-Distance Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for composing a three-dimensional shape of a subject using depth information require significant computational resources and memory, leading to increased costs and impracticality for devices with large image sensors.
Innovation Solution
A shape measurement device and method that segment a two-dimensional image into regions, processing normal and distance information separately within each segment to compose three-dimensional shape information, reducing computational requirements by using processors and memory efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If normal information for all pixels is collectively integrated to compose three-dimensional shape information, then measurement precision is improved, but device complexity and computation cost increase significantly
Solution Approach 1:
The patent divides the image capture range into multiple segment regions and processes normal information separately for each segment region rather than collectively integrating all pixels. This segmentation approach maintains measurement precision by preserving local shape characteristics while significantly reducing computation chip requirements and memory usage, directly resolving the contradiction between accuracy and device complexity
Solution Approach 2:
The patent extracts and utilizes distance information separately to correct shape information after initial composition from normal information. By taking out the distance information correction step as a separate processing stage, the system achieves high-precision three-dimensional shape composition without requiring excessive computation chip performance, as the correction uses pre-acquired distance data rather than requiring complex real-time integration of all normal information
2Measurement precision
If a large matrix is used to process all pixel data, then measurement precision is improved, but loss of substance increases due to memory requirements
Solution Approach 1:
The patent segments the image into multiple regions and processes each segment separately, avoiding the need for a large matrix that would require substantial memory. This approach maintains measurement precision by preserving local shape details while significantly reducing memory usage, directly addressing the contradiction between accuracy and resource consumption
3Manufacturing precision
If collective integration of normal information is performed, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent implements segmentation of the image capture range into multiple segment regions, processing normal information locally for each segment rather than requiring a complex circuit to collectively integrate all pixels. This maintains manufacturing precision by preserving local shape characteristics while reducing circuit scale and device complexity
Solution Approach 2:
The patent extracts distance information and uses it separately to correct shape information after initial composition. This two-stage approach achieves high manufacturing precision without requiring excessively complex circuits, as the correction stage uses pre-acquired distance data rather than requiring complex real-time integration circuits
Data Source
AI summary
A shape measurement device composes shape information indicating a three-dimensional shape of a subject based on normal information and distance information acquired for the subject. The normal and distance information are two-dimensional information and have a pixel structure corresponding to the two-dimensional image. The device acquires the normal and distance information, sets processing-target segment regions with respect to the two-dimensional image upon composition of the shape information of the subject, composes, for each of the segment regions, first shape information based on the normal information, composes, for each of the segment regions, second shape information acquired by changing the first shape information based on the distance information, and composes combined shape information indicating the three-dimensional shape of the subject by combining the second shape information for a plurality of the segment regions.


