US 12,444,220 B2
Character segmentation method and device based on edge detection and contour detection
Chenghai Huo, Beijing (CN); and Nangeng Zhang, Beijing (CN)
Assigned to CANAAN BRIGHT SIGHT CO., LTD, Beijing (CN)
Appl. No. 17/782,683
Filed by CANAAN BRIGHT SIGHT CO., LTD, Beijing (CN)
PCT Filed Sep. 29, 2020, PCT No. PCT/CN2020/118675
§ 371(c)(1), (2) Date Jun. 6, 2022,
PCT Pub. No. WO2021/109697, PCT Pub. Date Jun. 10, 2021.
Claims priority of application No. 201911233988.5 (CN), filed on Dec. 5, 2019; and application No. 201911234826.3 (CN), filed on Dec. 5, 2019.
Prior Publication US 2023/0009564 A1, Jan. 12, 2023
Int. Cl. G06V 30/18 (2022.01); G06V 30/148 (2022.01); G06V 30/182 (2022.01)
CPC G06V 30/1801 (2022.01) [G06V 30/148 (2022.01); G06V 30/18019 (2022.01); G06V 30/182 (2022.01)] 8 Claims
OG exemplary drawing
 
1. A character segmentation method, comprising:
acquiring a character region image and converting the character region image into a grayscale image, wherein the character region image comprises at least one character frame;
adaptively adjusting parameters of an edge detection algorithm based on image quality characteristics of the grayscale image, and converting the grayscale image into an edge binary image by utilizing the edge detection algorithm;
acquiring, from the edge binary image, at least one segment for the at least one character frame by utilizing a projection approach;
determining a character contour in each of the at least one segment by utilizing a contour detection algorithm; and
determining a minimum enclosing rectangle region of the character contour in each of the at least one segment and performing character segmentation on the character region image based on the minimum enclosing rectangle region.