Digital camera-based three-dimensional measurement method of shoe tree and prepared shoe tree
A three-dimensional measurement and digital camera technology, applied in the field of measurement, can solve the problems of unguaranteed accuracy, inconvenient extraction, large workload, etc., and achieve the effects of low price, flexible measurement method, and simple measurement hardware.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0121] Shoe last wiring and image shooting
[0122] In this embodiment, the marked curves on the shoe tree are as follows: figure 1 shown. First, the shoe last is routed according to the characteristics that determine the geometric shape of the shoe last to form an image curve, and the target curve that needs to be reconstructed is marked on the shoe last with a colored paper tape of a certain color and width, so that the target curve is consistent with the color and brightness of the shoe last. The outer surface of the shoe last is clearly distinguished to facilitate image processing and identification. In order to reconstruct the characteristic curve of the partially lost or partially occluded image, the marked shoe last is photographed with a digital camera from different angles and multiple images are obtained.
Embodiment 2
[0124] Acquisition of characteristic curves
[0125] The color image taken in Example 1 was converted into a grayscale image, and Gaussian filtering was used to smooth the image to suppress noise and improve image quality, and then the image skeleton was extracted using the mathematical morphology thinning algorithm.
[0126] Let B=(B 1 , B 2 ) is a given structural element, and B 1 , b 2 disjoint, then for the image set X to be refined, the structural element B is used to represent the hit-and-miss transformation of the image X, and for a given image X and the structure pair sequence B={B 1 , B 2 ,L,B n}, first select the structural element B 1 Thinning the image X, and recording the thinning result as reuse the element B in the sequence of structures 2 right refinement, and the refinement result is recorded as Continue the refinement process until B n The refinement is completed, and the refinement result is The whole process can be repeated until the image no ...
Embodiment 3 3
[0151] Embodiment 3 Reconstruction of three-dimensional curve
[0152] An image curve γ(l) can be parameterized by a single variable l∈[a,b]. Suppose there are m image curves with the same name γ i (l i ), 0≤i≤m-1, each curve has its own parameter l i ∈[a i , b i ]. Assume that each curve is connected by two endpoints, and the matching of these two endpoints on m images is known. Without loss of generality, select γ 0 As the longest image curve, assume γ 0 (a 0 ) and γ i (a 0 ) corresponds, 1≤i≤m-1. Therefore, to obtain the correspondence of point columns on m curves with the same name between different images, it is actually to solve the m-1 mapping function, 1≤i≤m-1, mapping[a 0 , b 0 ] to [a i , b i ], each mapping function is a monotonically increasing function.
[0153] Referring to Wu's curve matching optimization method, after the B-spline curve is discretized, dynamic programming is used to initially match the discrete pixel points on the correspondin...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com