US 12,137,855 B2
Mobile robot and cliff detection method thereof
Keen-Hun Leong, Penang (MY); Sai-Mun Lee, Penang (MY); and Ching-Geak Chan, Penang (MY)
Assigned to PIXART IMAGING INC., Hsin-Chu County (TW)
Filed by PIXART IMAGING INC., Hsin-Chu County (TW)
Filed on Jun. 22, 2021, as Appl. No. 17/353,818.
Claims priority of provisional application 63/090,734, filed on Oct. 13, 2020.
Prior Publication US 2022/0110498 A1, Apr. 14, 2022
Int. Cl. G06T 5/50 (2006.01); A47L 11/40 (2006.01); G05D 1/00 (2006.01)
CPC A47L 11/4011 (2013.01) [G05D 1/0223 (2013.01); G05D 1/0246 (2013.01); G06T 5/50 (2013.01); A47L 2201/04 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A detection system, comprising:
a light source, configured to be turned on within a first interval and be turned off within a second interval;
an image sensor, configured to capture multiple bright image frames within the first interval using a first shutter, and capture multiple dark image frames within the second interval using a second shutter, wherein, within the second interval, the light source is not turned on between capturing adjacent two dark image frames among the multiple dark image frames; and
a processor, configured to
calculate an average of first gray level summation and an average of first image quality of the multiple bright image frames,
calculate an average of second gray level summation and an average of second image quality of the multiple dark image frames,
calculate a gray level difference between the average of first gray level summation and the average of second gray level summation,
calculate an image quality difference between the average of first image quality and the average of second image quality,
calculate an average shutter difference between the first shutter and the second shutter, and
perform cliff identification according to the gray level difference, the image quality difference and the average shutter difference.