Sign language recognition method based on convolutional neural network
A convolutional neural network and recognition method technology, applied in biological neural network models, neural architecture, character and pattern recognition, etc., can solve the problem of low efficiency of tasks with large amounts of data, avoid sensitivity to light, overcome artificial The design features are more complex and the effect of automatic gesture recognition
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0081] Step 1: Use the Kinect depth camera to collect sign language images. When collecting data, the Kinect depth camera is about 1 meter away from the person and about 1.2 meters away from the ground. The sign language image database of this embodiment contains 30 categories of sign language actions with different semantics, each category contains 1000 frames of images, recorded by 5 people respectively, and each data set contains 30*1000=30000 frames of static gesture images.
[0082] By using the Image AcquisitionToolbox toolbox in MATLA to acquire images, import the sign language image database obtained from the Kinec depth camera into MATLAB and save it locally in .jpg format. Randomly divide the data in the color sign language dataset and deep sign language dataset into training set, cross-validation set and test set. The training set and cross-validation set are used to train the convolutional neural network model and evaluate the performance of the recognition model. T...
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