-
公开(公告)号:US20210074017A1
公开(公告)日:2021-03-11
申请号:US16950293
申请日:2020-11-17
Applicant: SZ DJI TECHNOLOGY CO., LTD.
Inventor: Shaojie SHEN , Xianliang WU , Liansheng CHEN
IPC: G06T7/73 , B64C39/02 , G06K9/00 , G06K9/46 , G06K9/62 , A63H30/04 , A63H27/00 , B64D47/08 , G01V8/10
Abstract: A method for controlling a movable object includes acquiring one or more images captured by an imaging device borne by the movable object and at least partially blocked by an obstructive object attached to the movable object, detecting at least portion of the obstructive object projected in the one or more images, and processing the one or more images with the projected obstructive object to assist an operation of the movable object.
-
公开(公告)号:US20190287213A1
公开(公告)日:2019-09-19
申请号:US16432378
申请日:2019-06-05
Applicant: SZ DJI TECHNOLOGY CO., LTD.
Inventor: Rui XU , Ketan TANG , Shaojie SHEN , Qi CHEN
Abstract: An image rectification method includes determining one or more reference points in a raw image captured using a lens for a plurality of target pixels in a target image, obtaining a subsection of the raw image based on the one or more reference points, and calculating a target pixel value for each of the plurality of target pixels based on one or more pixel values of one or more pixels in the subsection of the raw image.
-
13.
公开(公告)号:US20190188879A1
公开(公告)日:2019-06-20
申请号:US16285373
申请日:2019-02-26
Applicant: SZ DJI TECHNOLOGY CO., LTD.
Inventor: Shaojie SHEN , Xianliang WU , Liansheng CHEN
CPC classification number: G06T7/74 , A63H27/12 , A63H30/04 , B64C39/024 , B64C2201/123 , B64C2201/127 , B64D47/08 , G01V8/10 , G06K9/0063 , G06K9/4661 , G06K9/6202 , G06T7/73 , G06T2207/10021 , G06T2207/10032 , G06T2207/30204 , G06T2207/30232 , G06T2207/30261
Abstract: A method for supporting image processing for a movable object includes acquiring one or more images captured by an imaging device borne by the movable object. The imaging device is at least partially blocked by an obstructive object attached to the movable object. The method further includes applying a template to the one or more images to obtain one or more projected locations of the obstructive object within the one or more images and detecting at least portion of the obstructive object at the one or more projected locations within the one or more images.
-
-