US 12,260,613 B2
Electronic device and method for recognizing images based on texture classification
I-Hua Chen, New Taipei (TW); Chin-Pin Kuo, New Taipei (TW); Wei-Chun Wang, New Taipei (TW); and Tzu-Chen Lin, New Taipei (TW)
Assigned to HON HAI PRECISION INDUSTRY CO., LTD., New Taipei (TW)
Filed by HON HAI PRECISION INDUSTRY CO., LTD., New Taipei (TW)
Filed on Jan. 23, 2022, as Appl. No. 17/581,934.
Claims priority of application No. 202110094532.6 (CN), filed on Jan. 23, 2021.
Prior Publication US 2022/0237893 A1, Jul. 28, 2022
Int. Cl. G06V 10/764 (2022.01); G06T 7/11 (2017.01); G06T 7/41 (2017.01); G06V 10/44 (2022.01); G06V 10/54 (2022.01); G06V 10/774 (2022.01)
CPC G06V 10/764 (2022.01) [G06T 7/11 (2017.01); G06T 7/41 (2017.01); G06V 10/449 (2022.01); G06V 10/54 (2022.01); G06V 10/774 (2022.01); G06T 2207/20021 (2013.01); G06T 2207/20081 (2013.01)] 14 Claims
OG exemplary drawing
 
1. An electronic device comprising:
at least one processor; and
a storage device coupled to the at least one processor and storing instructions for execution by the at least one processor to cause the at least one processor to:
segment each of original images showing a plurality of objects with a known category into a plurality of block images;
extract features of texture of each of the plurality of block images according to at least one Gabor filter;
determine a grayscale level co-occurrence matrix of each of the plurality of block images according to the extracted features of texture;
calculate texture feature statistics of each of the plurality of block images according to the grayscale level co-occurrence matrix, wherein the texture feature statistics comprise energy, contrast, correlation, and entropy;
generate an object recognition model by performing training using the features of texture and the texture feature statistics; and
convert an image to be recognized into a grayscale image, extract texture features and texture feature statistics of the grayscale image, recognize and classify at least one object in the image to be recognized according to the texture features and the texture feature statistics of the grayscale image by the object recognition model.