Fall detection method based on deep convolution neural network and long-short-term memory network
A long-short-term memory and deep convolution technology, applied in the field of artificial intelligence, can solve problems such as poor stability, insufficient video processing capabilities, and low accuracy of image sub-recognition technology, achieve high accuracy and stability, and solve real-time requirements Effect
Inactive Publication Date: 2019-01-15
浙江深眸科技有限公司
View PDF4 Cites 15 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The calculation speed is getting faster and faster, and the traditional image processing methods can no longer meet the current demand scenarios
[0003] The traditional image sub-recognition technology has low accuracy and poor stability, and has high requirements for image quality
Traditional image recognition technology is not capable of processing video and cannot guarantee real-time performance
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0021] The present invention discloses a fall detection method based on a deep convolutional neural network and a long-short-term memory network. The specific implementation of the present invention will be further described below in combination with preferred embodiments.
[0022] see attached figure 1 , figure 1 The overall flow of the fall detection method based on deep convolutional neural network and long short-term memory network is shown.
[0023] Preferably, the fall detection method based on deep convolutional neural network and long short-term memory network comprises the following steps:
[0024] Step S1: Inputting an image signal and / or video signal, and simultaneously dividing the above image signal and / or video signal into a picture frame and frames before and after the picture frame in sequence;
[0025] Step S2: extracting the picture frame in step S1 and the image features of the front and back frames through the deep convolutional neural network model;
[...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a fall detection method based on a depth convolution neural network and a long-short-term memory network, comprising the following steps: Step S1, an image signal and / or a video signal are inputted, and the image signal and / or the video signal are cut into picture frames and successive frames sequentially connected with the picture frames. Step S2: the image features of thepicture frame and the front and back frames in step S1 are extracted by the depth convolution neural network model. Step S3: the image features extracted in step S2 are converted into features related to falling behavior through a long-term and short-term memory network model. S4, cooperating with the long-term and short-term memory network model through the sliding window method to extract the time order information related to the falling behavior. The invention discloses a fall detection method based on a depth convolution neural network and a long-term and short-term memory network, whichcan real-time monitor and identify the fall behavior under a complex environment, and has high accuracy and stability.
Description
technical field [0001] The invention belongs to the technical field of artificial intelligence based on deep learning, and in particular relates to a fall detection method based on a deep convolutional neural network and a long-term and short-term memory network. Background technique [0002] With the rapid development of science and technology, the proportion of the artificial intelligence industry has increased significantly, and the infrastructure GPU server can meet the huge amount of calculation. The calculation speed is getting faster and faster, and the traditional image processing methods can no longer meet the current demand scenarios. [0003] Traditional image recognition technology has low accuracy and poor stability, and has high requirements for image quality. Traditional image recognition technology is not capable of processing video and cannot guarantee real-time performance. Contents of the invention [0004] Aiming at the state of the prior art, the pre...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04
CPCG06V20/48G06N3/045G06F18/214
Inventor 钟博煊许淞斐周礼
Owner 浙江深眸科技有限公司