DocumentCode
2042363
Title
Infant behavior recognition system based on pressure distribution image
Author
Harada, Tatsuya ; Saito, Akihiko ; Sato, Tomomasa ; Mori, Taketoshi
Author_Institution
Tokyo Univ., Japan
Volume
4
fYear
2000
fDate
2000
Firstpage
4082
Abstract
The authors developed a novel infant behavior recognition system based on a pressure distribution image. The system can recognize an infant´s status (quiet, moving and crying), posture, body parts´ positions and movement unrestrainedly. It can, in recognizing the behavior, cope with the infant´s rapid growth and unique physique. The algorithm of the infant behavior recognition system is summarized as follows. 1) First, the system measures the pressure distribution image with 384 pressure sensors distributed in the bed. 2) The authors propose “activity score”; this is calculated by using the measured pressure distribution image and indicates kinetic energy of the infant´s activity. Based on the activity score, the system decides the infant´s status. 3) If the infant is quiet, the system estimates the infant´s physique. 4) Based on the estimated physique, the system recognizes the infant´s posture and body part movement. Experimental results reveal that the system successfully recognizes infants´ status (quiet, moving and crying), posture, body parts position and movements
Keywords
biomechanics; biomedical equipment; computerised monitoring; paediatrics; patient monitoring; pattern recognition; pressure sensors; activity score; body part position recognition; infant behavior recognition system; infant growth; infant physique; kinetic energy; movement recognition; posture recognition; pressure distribution image; pressure sensors; Biological system modeling; Energy measurement; Head; Humans; Image recognition; Image sensors; Magnetooptic recording; Pediatrics; Pressure measurement; Sensor systems;
fLanguage
English
Publisher
ieee
Conference_Titel
Robotics and Automation, 2000. Proceedings. ICRA '00. IEEE International Conference on
Conference_Location
San Francisco, CA
ISSN
1050-4729
Print_ISBN
0-7803-5886-4
Type
conf
DOI
10.1109/ROBOT.2000.845368
Filename
845368
Link To Document