US 12,327,404 B2
Object detection apparatus using an image preprocessing artificial neural network model
Lok Won Kim, Seongnam-si (KR)
Assigned to DEEPX CO., LTD, Seongnam-si (KR)
Filed by DEEPX CO., LTD., Seongnam-si (KR)
Filed on Mar. 23, 2023, as Appl. No. 18/125,666.
Application 18/125,666 is a continuation of application No. 17/256,582, granted, now 11,636,670, previously published as PCT/KR2020/007303, filed on Jun. 4, 2020.
Claims priority of application No. 10-2019-0066240 (KR), filed on Jun. 4, 2019.
Prior Publication US 2023/0237792 A1, Jul. 27, 2023
Int. Cl. G06K 9/00 (2022.01); G06F 18/21 (2023.01); G06F 18/214 (2023.01); G06F 18/24 (2023.01); G06T 5/00 (2006.01); G06T 5/70 (2024.01); G06T 5/73 (2024.01); G06T 5/92 (2024.01); G06T 7/10 (2017.01); G06T 7/73 (2017.01); G06V 10/20 (2022.01); G06V 10/82 (2022.01); G06V 10/94 (2022.01)
CPC G06V 10/95 (2022.01) [G06F 18/2148 (2023.01); G06F 18/217 (2023.01); G06F 18/24 (2023.01); G06T 5/00 (2013.01); G06T 5/70 (2024.01); G06T 5/73 (2024.01); G06T 5/92 (2024.01); G06T 7/10 (2017.01); G06T 7/74 (2017.01); G06V 10/20 (2022.01); G06V 10/82 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20208 (2013.01)] 17 Claims
OG exemplary drawing
 
1. An apparatus comprising:
a preprocessing module that includes an image enhancement function based on a first trained artificial neural network model (ANN) and is configured to perform image enhancement processing by sequentially adjusting each of a plurality of image sub-parameters of an image input to the first trained ANN, the plurality of image sub-parameters including deblur and denoise, and
an object detection module including an object detection function based on a second trained ANN,
wherein the first trained ANN is optimized to increase a rate of object detection performed by the second trained ANN, in comparison with a case where the image enhancement function is not performed by the first trained ANN.