Unlock instant, AI-driven research and patent intelligence for your innovation.

Life searching method and system based on multi-sensor fusion

A multi-sensor fusion and sensor technology, applied in the life search method and system field based on multi-sensor fusion, can solve the problems of large error in life search results, misjudgment of life scenes, blurred edges, etc., to improve robustness and accuracy , the effect of improving the accuracy rate

Active Publication Date: 2019-07-12
SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI
View PDF22 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Infrared imaging uses an infrared video sensor to image the human body, thereby detecting human targets through a series of methods such as background segmentation and human target detection in the portrait area. Low, lack of color texture information, halo effect around the target, blurred edges, easy to misjudge the current life scene
[0006] It can be seen that the traditional life search uses a single sensor for life detection, which has great limitations. The output life search results have large errors and poor stability. The use process needs to rely on the manual operation of the staff

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Life searching method and system based on multi-sensor fusion
  • Life searching method and system based on multi-sensor fusion
  • Life searching method and system based on multi-sensor fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] Please also refer to Figure 1 to Figure 6 , the following describes the multi-sensor fusion-based life searching method of this embodiment.

[0062] The life search method mainly includes the following steps:

[0063] Step 101: setting up several sensor modules to acquire the sensor data sequence group collected by each sensor module at a set frequency;

[0064] Step 102: Perform neural network-based preprocessing on the sensor data sequence group to form a preliminary probability output of life detection corresponding to the sensor module;

[0065] Step 103: Each preliminary probability output is used as an input of the deep neural network, and the several preliminary probability outputs of the several sensor modules are spliced ​​and fused through the deep neural network, and the final probability value of the existence of life is output.

[0066] Such as image 3 As shown, the method is described below taking infrared video sensors, acoustic wave sensors, and ult...

Embodiment 2

[0125] please refer again figure 2 as well as Figure 9 , this embodiment relates to a life search system based on multi-sensor fusion.

[0126] The system includes a controller 10 and several sensor modules. The plurality of sensor modules include an ultra-wideband radar sensor, an infrared video sensor, and an acoustic wave sensor.

[0127] The controller 10 includes an ultra-wideband radar sensor output probability output sequence 21 , an infrared video sensor output probability output sequence 22 and an acoustic wave sensor output probability output sequence 23 for storing data.

[0128] The controller 10 also includes several preprocessing modules based on neural networks that are set corresponding to the sensor modules, such as the first preprocessing module 31 that is set corresponding to the ultra-wideband radar sensor, and the second preprocessing module that is set corresponding to the infrared video sensor. 32, and a third preprocessing module 33 corresponding t...

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

No PUM Login to View More

Abstract

The invention discloses a life searching method and system based on multi-sensor fusion. The method comprises the following steps of: arranging a plurality of sensor modules, and collecting a sensor data sequence group collected by each sensor module at a set frequency; performing preprocessing of the sensor data sequence group based on a neural network to form life detection preliminary probability outputs corresponding to the sensor modules; and taking each preliminary probability output as one path of input of the deep neural network, splicing and fusing the plurality of preliminary probability outputs of the plurality of sensor modules through the deep neural network, and outputting a final probability value of life existence. The life searching method and system are based on the plurality of sensor modules to collect various environmental information, and employ the deep neural network for fusing and splicing life detection data under different environmental information so that the robustness and the accuracy of life detection are improved in a stage.

Description

technical field [0001] The invention relates to the technical field of geological disaster prevention and robot control, in particular to a life search method and system based on multi-sensor fusion. Background technique [0002] Studies have shown that mainland China has entered a new seismically active period and is always facing the threat of strong earthquakes. The state attaches great importance to the use of scientific and technological means to improve the ability to respond to public emergencies. For example, how to implement scientific and technological rescue, improve the search efficiency of survivors in the earthquake ruins, and reduce injuries have become important problems for on-site rescue after disasters. [0003] In actual on-site rescue, natural disasters such as earthquakes will cause a large number of houses to collapse. How to determine the location of living bodies in the ruins is a major problem in disaster rescue. [0004] In the existing technology...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G01V11/00G01V8/10G01V3/12G01V1/00G01S13/88G01S7/41G01S7/292G01S7/35G06K9/62G06K9/00G06N3/04G06N3/08
CPCG01V11/00G01V8/10G01V3/12G01V1/001G01S13/888G01S7/417G01S7/2923G01S7/354G06N3/08G06N3/044G06N3/045G06F2218/06G06F2218/12G06F18/2411G06F18/214
Inventor 周翊民殷力
Owner SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI