Patents
Literature
Patsnap Copilot is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Patsnap Copilot

34 results about "Shape alignment" patented technology

Centering device for valve side sleeve of converter transformer

The invention relates to a centering device for a valve side sleeve of a converter transformer. The centering device comprises a clamping device, the clamping device is connected with an adjusting device so as to adjust the top, the bottom, the left, the right, pitching and rotating positions of the sleeve, the clamping device comprises a bottom frame, the centering device comprises a first reference plate and a second reference plate, wherein the first reference plate and the second reference plate are fixedly arranged below the bottom frame from front to back, the centering device further comprises a visible light beam emitting device arranged below a wire unwinding sleeve, a through hole is formed in the center of the first reference plate, a cross-shaped alignment line is arranged in the center of the second reference plate, the through hole is formed in an extension line of a connection line of rotation centers of the pitching position and the rotation position of the sleeve, whena visible light beam emitted by the visible light beam emitting device penetrates through the through hole of the first reference plate to be aligned with the cross-shaped alignment line of the second reference plate, the sleeve and the wire unwinding sleeve are in a centered state. According to the centering device, centering of the sleeve and the wire unwinding sleeve is realized, and the alignment precision of the sleeve in the installation process is improved.
Owner:郭芬

Geometric invariant object segmentation method based on sparse independent shape component representation

InactiveCN106600603ASolve the problem of weak target segmentation abilityGuaranteed global optimal solutionImage enhancementImage analysisMethod of undetermined coefficientsLevel set
The invention discloses a geometric invariant object segmentation method based on sparse independent shape component representation. The method comprises the following steps of: firstly, carrying out log polar coordinate transformation on a training set; carrying out independent shape component extraction on a transformation result; on the basis of an independent shape component, constructing the first mapping convex shape set of an independent component space; on the basis of the extracted independent shape component, establishing a second mapping convex spare shape subset; on the basis of the first mapping convex shape set and the second mapping convex spare shape subset, constructing a sparse independent shape component representation model; combining with the above sparse representation model to construct an object segmentation model with a level set energy constraint item; adopting a gradient descent method and a soft threshold value method to solve a target function; and utilizing an evolution curve to realize image object segmentation. By use of the method, the problems of sensitive geometric deformation, poor shape representation ability and low segmentation accuracy of an existing object segmentation method based on sparse representation are overcome, and the problem that the sparse representation model has poor object segmentation ability under a situation of poor shape alignment or unknown deformation parameters can be solved.
Owner:ZHEJIANG UNIV

Three-dimensional human face sight line calculation method

The invention provides a three-dimensional human face sight line calculation method. A three-dimensional human face is accurately reconstructed, so that the sight line direction of the human face is accurately calculated. According to the method, a sight line direction calculation task is decomposed into two sub-tasks of sight line aligned three-dimensional face reconstruction and sight line contact realized through rotation. The method comprises the steps: reconstructing a basic three-dimensional face by using a three-dimensional deformation model and taking a two-dimensional reference face image as input; performing shape adjustment on the basic three-dimensional face to obtain a three-dimensional reconstructed face with an aligned shape so as to ensure that the eye region of the three-dimensional face is accurately aligned with the eye region of the input reference two-dimensional face image; and finally, acquiring the three-dimensional face with the aligned sight lines by replacing the texture of the eye region of the three-dimensional face with the aligned shape. For the sub-task of sight line contact, the invention provides a brand-new method for realizing sight line contact by rotating the three-dimensional face, and the method does not depend on excessive hypotheses and can be widely applied to the field of virtual reality.
Owner:BEIHANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products