US 12,175,722 B2
Connected component analysis method which can reuse label
Joon Chok Lee, Penang (MY); and Zi Hao Tan, Penang (MY)
Assigned to PixArt Imaging Inc., Hsin-Chu (TW)
Filed by PixArt Imaging Inc., Hsin-Chu (TW)
Filed on Dec. 17, 2021, as Appl. No. 17/553,833.
Prior Publication US 2023/0196717 A1, Jun. 22, 2023
Int. Cl. G06V 10/44 (2022.01); G06T 7/11 (2017.01)
CPC G06V 10/457 (2022.01) [G06T 7/11 (2017.01)] 17 Claims
OG exemplary drawing
 
1. A connected component analysis method, which uses labels repeatedly, applied to an image sensor comprising a sensor array and a processing circuit, comprising:
(a) outputting a target binary image by the sensor array, wherein the binary image has M*N pixels, M>2 and N>3;
(b) defining a label pattern comprising a center label and a plurality of neighboring labels;
(c) setting a label of a current pixel of the target binary image according to a binary value of the current pixel and the neighboring pixels;
(d) setting labels of at least two of the neighboring labels according to whether the current pixel is in any one of a first row, a first column and a last column; and
(e) recording the center label to a label buffer;
wherein labels for marking pixels of the target binary image are first center labels, and then are second center labels, and are the first center labels again after the labels are the second center labels;
wherein the connected component analysis method further comprises:
performing the steps (a), (b), (c), (d) and (e) to identify at least one object in the target binary image through minimally labeling at least one object unit by the processing circuit, wherein if the number of the least one object unit is X, the number of a corresponding label is X;
wherein the minimally labeling meaning a label of the object unit is not greater than a maximum label.