Elderly people feature series fusion gesture recognition method and device
A gesture recognition, gesture technology, applied in the field of gesture recognition
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0076] Such as figure 1 As shown, a feature series fusion gesture recognition method for the elderly is characterized in that it comprises the following steps:
[0077] Step S1, gesture image segmentation: using the color space model to convert the RGB image into a YCbCr image, and then using the ellipse model in the skin color model to segment the image to obtain the gesture part;
[0078] Step S2, gesture feature extraction: use the serial feature fusion method to serially fuse the HOG and LBP features, describe the gesture features from the two angles of edge and texture, and adopt the gesture classification and recognition based on SVM;
[0079] Step S3, Valid Gesture Recognition: Use face verification method and head pose estimation method to judge whether the recognized gesture is valid, if valid, use the gesture to identify user needs, otherwise judge it as invalid.
[0080] Specifically, step S1 includes step S11, using a monocular camera to collect images and perform...
Embodiment 2
[0137] The difference between this embodiment and Embodiment 1 is that the head pose estimation method in this embodiment is based on EPnP head pose estimation.
[0138] The EPnP algorithm is to express the three-dimensional coordinates of all feature points in the world coordinate system with the weighted sum of four virtual control point coordinates, and these four virtual control points cannot be coplanar, by solving the coordinates of the four control points in the camera coordinate system Coordinates, the conversion relationship between the coordinates can be obtained, and the posture information of the head can be further calculated according to the conversion relationship.
[0139] Record the coordinates of n feature points in the world coordinate system as The coordinates of the 4 virtual control points are The coordinates projected into the camera coordinate system become The virtual control point becomes Each feature point in the two coordinate systems is re...
Embodiment 3
[0155] An eye movement machine vision tracking device for the elderly, characterized in that it includes a gesture image segmentation module, a gesture recognition module and an effective gesture recognition module;
[0156] The gesture image segmentation module includes a video capture device, a face detection module and a gesture image segmentation submodule; the video capture device collects an image; the face detection module detects a human face in the image collected by the video capture device; the The gesture image segmentation sub-module segments the hand region in the image collected by the video capture device;
[0157] The gesture recognition module includes a gesture feature extraction module and a gesture recognition submodule; the gesture feature extraction module extracts gestures; the gesture recognition submodule compares and recognizes gestures according to preset gesture classifications and extracted gestures;
[0158] The valid gesture recognition module r...
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.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap