US 10,891,845 B2
Mouth and nose occluded detecting method and system thereof
Chuan-Yu Chang, Yunlin County (TW); and Fu-Jen Tsai, New Taipei (TW)
Assigned to NATIONAL YUNLIN UNIVERSITY OF SCIENCE AND TECHNOLOGY, Yunlin County (TW)
Filed by NATIONAL YUNLIN UNIVERSITY OF SCIENCE AND TECHNOLOGY, Yunlin County (TW)
Filed on Nov. 28, 2018, as Appl. No. 16/202,108.
Prior Publication US 2020/0168071 A1, May 28, 2020
Int. Cl. G08B 21/04 (2006.01); A61B 5/00 (2006.01); G06T 7/00 (2017.01); G06K 9/00 (2006.01); G16H 30/20 (2018.01); G06T 7/90 (2017.01); G06K 9/62 (2006.01)
CPC G08B 21/0476 (2013.01) [A61B 5/0077 (2013.01); G06K 9/00248 (2013.01); G06K 9/00255 (2013.01); G06K 9/6217 (2013.01); G06T 7/0014 (2013.01); G06T 7/90 (2017.01); G16H 30/20 (2018.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30004 (2013.01); G06T 2207/30201 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A mouth and nose occluded detecting method, comprising:
a detecting step, comprising:
a facial detecting step, wherein an image is captured by an image capturing device, and a facial portion image is obtained from the image according to a facial detection;
an image extracting step, wherein a mouth portion is extracted from the facial portion image according to an image extraction so as to obtain a mouth portion image; and
an occluded determining step, wherein the mouth portion image is entered into an occluding convolutional neural network so as to produce a determining result, and the determining result is an occluding state or a normal state; and
a warning step, wherein a warning is provided according to the determining result, when the determining result is the normal state, the detecting step is performed; when the determining result is the occluding state, the warning is provided;
wherein the occluding convolutional neural network comprises a softmax layer, the softmax layer comprises at least one image state, the mouth portion image, at least one image state parameter, at least one image state probability and an image state probability set, the softmax layer is corresponded by:

OG Complex Work Unit Drawing
wherein y(i) is the at least one image state, k is a number of the image state, x(i) is the mouth portion image, θ12 . . . , θK are the image state parameters, p(y(i)=k|x(i):θ) is the image state probability, hθ(x(i)) is the image state probability set and T means transpose matrix.