Emotion recognition method and system based on mobile intelligent terminal sensor equipment
A technology of mobile intelligent terminal and sensor equipment, which is applied in the fields of emotion recognition and psychological diagnosis and treatment, and can solve the problems of lack of research results and applications.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0074] Embodiment 1, the mobile smart terminal used is a smart phone, and the sensor device on the smart phone is used to collect raw sensor data.
[0075] Please refer to figure 1 , in the mobile smart terminal of the user, the embodiment of the present invention provides a data collection method, comprising the following steps:
[0076] Step 1. Receive the emotion recognition task from the user and start the data collection service.
[0077] Step 2, obtaining the corresponding sensor management object corresponding to the data collection.
[0078] Step 3: Register the listener for the sensor device with the smartphone, start the monitoring service, monitor the data of each sensor in real time, record its real-time changes, and obtain all the sensor data required within the time period.
[0079] Step 4: Send the sensor information data obtained through monitoring to the server, and log off the sensor monitor after the data collection is completed.
[0080] Step 5, receivin...
Embodiment 2
[0082] Embodiment 2, on the server side, the embodiment of the present invention provides an emotion recognition method, comprising the following steps:
[0083] Step 1. Perform data processing, data cleaning and preprocessing for the raw sensor data collected in real time by each sensor on the user's mobile smart phone.
[0084] Step 11, filtering illegal data, adding a code verification function in the background of the server, filtering illegal data, such as null values, data that obviously exceeds the sensor data range, and eliminating them.
[0085] Step 12, the data is subjected to a median filtering operation. In general, data collected from hardware requires filtering operations. There are two main filtering methods: mean filtering and median filtering. Among them, mean filtering is realized based on calculating the average value of the neighborhood, so the edge information and feature information in the waveform image will be blurred. The median filter can highligh...
Embodiment 2
[0088] In step 2 of Embodiment 2, the sensors include motion sensors, light sensors, GPS sensors or / and network sensors. The methods for extracting features from various sensor data are as follows:
[0089] Acceleration sensors and gyroscopes belong to motion sensors; for motion sensors, their data are divided into three dimensions of x, y, and z, corresponding to the real three-dimensional space, which are drawn into curved waveform graphs, and their peak and valley values are counted as For the violent value of the fluctuation, count the number of peaks and troughs as the number of fluctuations, and count the time occupied by the peaks and troughs as the fluctuation time, so as to obtain the eigenvector corresponding to the motion sensor data.
[0090] For the light sensor, the following features are extracted from the light sensor data as the corresponding feature vectors: According to the light sensor data, the usage environment information of the mobile smart terminal a...
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