US 12,106,098 B2
Semiconductor device
Hyun Pil Kim, Seoul (KR); Hyun Woo Sim, Seoul (KR); and Seong Woo Ahn, Yongin-Si (KR)
Assigned to Samsung Electronics Co., Ltd., (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Mar. 31, 2023, as Appl. No. 18/129,119.
Application 18/129,119 is a division of application No. 17/216,323, filed on Mar. 29, 2021, granted, now 11,645,072.
Application 15/905,979 is a division of application No. 15/717,989, filed on Sep. 28, 2017, granted, now 10,649,771, issued on May 12, 2020.
Application 17/216,323 is a continuation of application No. 16/520,761, filed on Jul. 24, 2019, granted, now 10,990,388, issued on Apr. 27, 2021.
Application 16/520,761 is a continuation of application No. 15/905,979, filed on Feb. 27, 2018, granted, now 10,409,593, issued on Sep. 10, 2019.
Claims priority of application No. 10-2017-0042125 (KR), filed on Mar. 31, 2017.
Prior Publication US 2023/0236832 A1, Jul. 27, 2023
Int. Cl. G06F 9/30 (2018.01); G06F 1/3287 (2019.01); G06T 1/20 (2006.01)
CPC G06F 9/3001 (2013.01) [G06F 1/3287 (2013.01); G06F 9/30101 (2013.01); G06T 1/20 (2013.01); G06T 2207/20024 (2013.01); G06T 2207/20164 (2013.01)] 5 Claims
OG exemplary drawing
 
1. An image processing method of a semiconductor device, wherein the semiconductor device comprises a first processor comprising a first register, and a second processor comprising a second register, wherein the first register is shared with the second processor, and the second register is shared with the first processor, the method comprising:
inputting an image for corner detection to the first processor;
calculating derivative values by the first processor;
calculating a derivative product with respect to every pixel from the derivative values by the first processor;
calculating a sum of squared difference using the derivative product by the first processor;
defining a key point matrix by the second processor;
calculating a response function by the second processor; and
detecting a key point by performing a non-maximum suppression calculation by the first processor.