Back to results
US 2022/0189044 A1Published application
Method and system for visual based inspection of rotating objects
AssigneeAgency for Science, Technology and Research
InventorsWei XIONG, Wenyu CHEN, Jierong CHENG, Jia DU
Life of this patent
Estimating the term…
Abstract
This disclosure relates to method and system for visual inspection of rotating components. The method includes representing rotation cycles of a rotating component as spatial features based on video or image frames, ascertaining and/or evolving Hidden Markov Model (HMM) chains for the cycles, ascertaining a count of the rotating component in the frames and/or labelling the frames with ascertained states of the HMM chains.
Claims18
A method for visual inspection of rotating components, the method comprising:
- based on an initial subset of a plurality of observed image frames, which represents an initial rotation cycle of a rotating component, ascertaining an initial closed curve as spatial feature to represent the initial rotation cycle;
- ascertaining an initial Hidden Markov Model (HMM) chain, including a plurality of initial states located on equal curve length of the initial closed curve, for the initial set of the frames;
- for each subsequent subset of the frames, which represents each subsequent rotation cycle of the rotating component, ascertaining a subsequent HMM chain, including a plurality of subsequent states located on equal curve length of a subsequent closed curve as spatial feature, which represents the subsequent rotation cycle of the rotating component;
- and
- based on the initial HMM chain and the subsequent HMM chain of each subsequent subset of the frames, ascertaining a count of the rotating component in the frames, and/or labelling the frames with the initial states or the subsequent states.
2 more elements
A system for visual inspection of rotating components, the system comprising:
- a memory device storing a plurality of observed image frames;
- and
- a computing processor communicably coupled to the memory device and configured to:
- based on an initial subset of the frames, which represents an initial rotation cycle of a rotating component, ascertain an initial closed curve as spatial features to represent the initial rotation cycle;
- ascertain an initial Hidden Markov Model (HMM) chain, including a plurality of initial states located on equal curve length of the initial closed curve, for the initial set of the frames;
- for each subsequent subset of the frames, which represent each subsequent rotation cycle of the rotating component, ascertain a subsequent HMM chain, including a plurality of subsequent states located on equal curve length of a subsequent closed curve in spatial feature space, which represents the subsequent rotation cycle of the rotating component;
- and
- based on the initial HMM chain and the subsequent HMM chain of each subsequent subset of the frames, ascertain a count of the rotating component in the frames, and/or labelling the frames with the initial states or the subsequent states.
5 more elements
Dependencies are inferred from the claim text.