公开(公告)号：US20180048829A1

公开(公告)日：2018-02-15

申请号：US15642303

申请日：2017-07-05

Applicant: QUALCOMM Incorporated

Inventor： Victor CHAN

IPC: H04N5/243 ,  H04N5/235 ,  G06K9/62 ,  G06K9/46

CPC classification number: H04N5/243 ,  G06K9/00624 ,  G06K9/4604 ,  G06K9/4642 ,  G06K9/4647 ,  G06K9/6202 ,  G06K9/6228 ,  G06K9/6267 ,  G06K2009/4666 ,  H04N5/2353

Abstract: Methods, apparatuses, computer-readable medium, and systems are disclosed for performing automatic exposure control (AEC). In one embodiment, a first digital image is captured while applying a first set of values to one or more exposure control parameters. At least one computer vision (CV) operation is performed using image data from the first digital image, thereby generating a first set of CV features from a set of possible CV features. A mask is obtained comprising a value for each feature of the set of possible CV features. Using the mask, a first measure of abundance is obtained of relevant CV features among the first set of CV features extracted from the first digital image. Based on the first measure of abundance of relevant CV features, an updated set of values is generated for applying to the one or more exposure control parameters for capturing a subsequent digital image of the scene.

公开(公告)号：US20170083788A1

公开(公告)日：2017-03-23

申请号：US14863364

申请日：2015-09-23

Applicant: QUALCOMM Incorporated

Inventor： Victor CHAN ,  Alok GOVIL ,  Gerhard REITMAYR

IPC: G06K9/62 ,  G06K9/46

CPC classification number: G06K9/6201 ,  G06K9/46 ,  G06K9/6267 ,  G06K2009/4666

Abstract: Methods, systems, computer-readable media, and apparatuses for incremental object detection using a staged process and a band-pass feature extractor are presented. At each stage of the staged process, a different band of features from a plurality of bands of features in image data can be extracted using a dual-threshold local binary pattern operator, and compared with features of a target object within the band for a partial decision. The staged process exits if a rejection decision is made at any stage of the staged process. If no rejection decision is made in each stage of the staged process, the target object is detected. Features extracted at each stage may be from a different image for some applications.

公开(公告)号：US20210125315A1

公开(公告)日：2021-04-29

申请号：US16664641

申请日：2019-10-25

Applicant: QUALCOMM Incorporated

Inventor： Victor CHAN

IPC: G06T5/00 ,  G06T7/00 ,  G06T5/50

Abstract: Due to complexity constraints, most computer vision (CV) features are not suitable for certain applications such as low-power always-on applications. To address such issues, a contrast-adaptive normalized pixel difference (CA-NPD) is proposed. Unlike conventional NPD techniques, the CA-NPD smooths a metric surface, optimizes contrast discriminability, and enables context sensitive normalization.