| CPC G06T 7/0014 (2013.01) [A61B 5/02042 (2013.01); G06T 7/90 (2017.01); G16H 30/40 (2018.01); G06T 2207/10024 (2013.01); G06T 2207/10068 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30092 (2013.01); G06T 2207/30096 (2013.01); G06T 2207/30104 (2013.01)] | 10 Claims |

|
1. A method for detect tissue hemorrhage with image analysis, comprising steps of:
producing a plurality of hyperspectral image information by a host according to a plurality of reference images;
extracting an input image to said host by an image extraction unit;
transforming said input image by said host according to said plurality of hyperspectral image information to produce a hyperspectral input image;
producing an input image spectrum by said host according to said hyperspectral input image;
performing a feature operation on said input image spectrum by said host according to a preset cell band corresponding to a surface cell of a small intestine for generating a plurality of corresponding feature bands;
performing at least one convolution operation on said plurality of feature bands by said host according to a plurality of kernels for producing a convolution result;
matching and comparing said convolution result with at least one hyperspectral sample band of at least one hyperspectral sample spectrum of at least one hyperspectral sample image by said host for producing at least one comparison result; and
judging said input image to be a hemorrhage image on the surface layer of said small intestine when said at least one comparison result shows equality between said convolution result and said at least one hyperspectral sample band.
|