US 11,694,086 B2
Systems and methods for detecting region of interset in image
Yaozong Gao, Shanghai (CN); Yanbo Chen, Shanghai (CN); Jiyong Wang, Shanghai (CN); Weiyang Xie, Shanghai (CN); and Yiqiang Zhan, Shanghai (CN)
Assigned to SHANGHAI UNITED IMAGING INTELLIGENCE CO., LTD., Shanghai (CN)
Filed by SHANGHAI UNITED IMAGING INTELLIGENCE CO., LTD., Shanghai (CN)
Filed on Nov. 22, 2021, as Appl. No. 17/455,923.
Application 17/455,923 is a continuation of application No. 16/600,960, filed on Oct. 14, 2019, granted, now 11,188,773.
Claims priority of application No. 201811417230.2 (CN), filed on Nov. 26, 2018.
Prior Publication US 2022/0083804 A1, Mar. 17, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 10/25 (2022.01); G06N 3/045 (2023.01); G06N 3/084 (2023.01); G06N 3/08 (2023.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 10/82 (2022.01)
CPC G06N 3/084 (2013.01) [G06N 3/045 (2023.01); G06N 3/08 (2013.01); G06V 10/25 (2022.01); G06V 10/764 (2022.01); G06V 10/7747 (2022.01); G06V 10/82 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A region of interest (ROI) detection system, comprising:
at least one storage device including a set of instructions; and
at least one processor configured to communicate with the at least one storage device, wherein when executing the set of instructions, the at least one processor is configured to direct the system to perform operations including:
acquiring a target image and an ROI detection model; and
performing ROI detection on the target image by applying the ROI detection model to the target image, the ROI detection model including a multi-scale trained model configured to perform the ROI detection at a plurality of resolution levels, wherein
the multi-scale trained model includes a plurality of sequentially connected layers having different detection accuracies regarding ROIs of different sizes, and
each of the plurality of layers corresponds to one of the plurality of resolution levels and is configured to generate an ROI detection result at the corresponding resolution level.