公开(公告)号：US20200160717A1

公开(公告)日：2020-05-21

申请号：US16278213

申请日：2019-02-18

Applicant: VIA Technologies, Inc.

Inventor： Juan HE

IPC: G08G1/16 ,  G06K9/00

Abstract: A method for vehicle blind zone detection, applied to an electronic device coupled to one or more cameras arranged on a vehicle, the method comprising: setting a capture zone in a current frame image captured by the camera and detecting an object entering the capture zone in the current frame image, wherein the object meeting a capture criterion and a location information of the object meeting the capture criterion are added into a tracking list; performing tracking operations on an existing object, which has been detected and thus added to the tracking list, in one or more previous frame images preceding the current frame image captured by the camera, to obtain a new location information of the existing object in the current frame image, and determining whether to have the existing object remained in the tracking list in accordance with the new location information of the existing object and the detection scope; and making a warning determination in accordance with the location information in the current frame image for all the objects remained in the tracking list.

公开(公告)号：US20230186506A1

公开(公告)日：2023-06-15

申请号：US17699467

申请日：2022-03-21

Applicant: VIA Technologies, Inc.

Inventor： Chaochin CHANG ,  JinBo YIN ,  Juan HE

IPC: G06T7/70 ,  G06V10/25 ,  G06V10/82 ,  G06T7/20 ,  G06T3/40

CPC classification number: G06T7/70 ,  G06V10/25 ,  G06V10/82 ,  G06T7/20 ,  G06T3/40

Abstract: There is provided an object detection device and an object detection method. The processor of the object detection device defines respective overall image areas of a plurality of first sensed images from a plurality of original sensed images as first regions of interest; the processor defines respective partial image areas of a plurality of second sensed images from the plurality of original sensed images as second regions of interest, and crops out a plurality of third sensed images; the processor inputs the plurality of first sensed images and the plurality of third sensed images to a deep neural network learning model, so that the deep neural network learning model outputs image information of a target object image in the plurality of first sensed images and the plurality of third sensed images, respectively. By this, a function of detecting an object in front with high reliability is provided by means of image detection.