公开(公告)号：IN3104CHN2014A

公开(公告)日：2015-07-03

申请号：IN3104CHN2014

申请日：2014-04-24

Applicant: QUALCOMM INC

Inventor： ZHANG XIAOPENG ,  QIU GANG ,  WANG JILAI ,  ZHOU XIAOMING ,  SHEN LIANG ,  ALEKSIC MILIVOJE ,  VADDADI SUNDEEP ,  ZHOU SHAOJIE

IPC: G06T7/00

Abstract: Systems and methods for multispectral imaging are disclosed. The multispectral imaging system can include a near infrared (NIR) imaging sensor and a visible imaging sensor. The disclosed systems and methods can be implemented to improve alignment between the NIR and visible images. Once the NIR and visible images are aligned various types of multispectral processing techniques can be performed on the aligned images.

公开(公告)号：IN1386MUN2014A

公开(公告)日：2015-04-03

申请号：IN1386MUN2014

申请日：2014-07-09

Applicant: QUALCOMM INC

Inventor： VADDADI SUNDEEP ,  CHIMALAMARRI KRISHNAKANTH S ,  HONG JOHN H ,  LEE CHONG U

IPC: G01B11/25

Abstract: In some embodiments first information indicative of an image of a scene is accessed. One or more reference features are detected the reference features being associated with a reference object in the image. A transformation between an image space and a real world space is determined based on the first information. Second information indicative of input from a user is accessed the second information identifying an image space distance in the image space corresponding to a real world distance of interest in the real world space. The real world distance of interest is then estimated based on the second information and the determined transformation.

公开(公告)号：IN117CHN2013A

公开(公告)日：2015-10-02

申请号：IN117CHN2013

申请日：2013-01-04

Applicant: QUALCOMM INC

Inventor： BAHETI PAWAN KUMAR ,  VADDADI SUNDEEP ,  SWAMINATHAN ASHWIN ,  REZNIK YURIY ,  HAMSICI ONUR C ,  CHARI MURALI RAMASWAMY ,  HONG JOHN H ,  LEE CHONG UK

IPC: G06K9/46 ,  G06K9/68

Abstract: In one example an apparatus includes a processor configured to extract a first set of one or more keypoints from a first set of blurred images of a first octave of a received image calculate a first set of one or more descriptors for the first set of keypoints receive a confidence value for a result produced by querying a feature descriptor database with the first set of descriptors wherein the result comprises information describing an identity of an object in the received image and extract a second set of one or more keypoints from a second set of blurred images of a second octave of the received image when the confidence value does not exceed a confidence threshold. In this manner the processor may perform incremental feature descriptor extraction which may improve computational efficiency of object recognition in digital images.

公开(公告)号：WO2012016168A2

公开(公告)日：2012-02-02

申请号：PCT/US2011045942

申请日：2011-07-29

Applicant: QUALCOMM INC ,  BAHETI PAWAN KUMAR ,  VADDADI SUNDEEP ,  SWAMINATHAN ASHWIN ,  REZNIK YURIY ,  HAMSICI ONUR C ,  CHARI MURALI RAMASWAMY ,  HONG JOHN H ,  LEE CHONG UK

Inventor： BAHETI PAWAN KUMAR ,  VADDADI SUNDEEP ,  SWAMINATHAN ASHWIN ,  REZNIK YURIY ,  HAMSICI ONUR C ,  CHARI MURALI RAMASWAMY ,  HONG JOHN H ,  LEE CHONG UK

IPC: G06K9/46

CPC classification number: G06K9/6857 ,  G06K9/4671

Abstract: In one example, an apparatus includes a processor configured to extract a first set of one or more keypoints from a first set of blurred images of a first octave of a received image, calculate a first set of one or more descriptors for the first set of keypoints, receive a confidence value for a result produced by querying a feature descriptor database with the first set of descriptors, wherein the result comprises information describing an identity of an object in the received image, and extract a second set of one or more keypoints from a second set of blurred images of a second octave of the received image when the confidence value does not exceed a confidence threshold. In this manner, the processor may perform incremental feature descriptor extraction, which may improve computational efficiency of object recognition in digital images.

Abstract translation: 在一个示例中，一种装置包括处理器，该处理器被配置为从接收到的图像的第一八度的第一组模糊图像中提取第一组一个或多个关键点，计算第一组的一个或多个描述符 接收关于通过用第一组描述符查询特征描述符数据库产生的结果的置信度值，其中所述结果包括描述所接收的图像中的对象的身份的信息，并且从第二组关键点提取一个或多个关键点 当置信度值不超过置信度阈值时，所接收图像的第二八度音阶的第二组模糊图像。 以这种方式，处理器可以执行递增特征描述符提取，这可以提高数字图像中对象识别的计算效率。

公开(公告)号：WO2011069021A3

公开(公告)日：2011-08-18

申请号：PCT/US2010058805

申请日：2010-12-02

Applicant: QUALCOMM INC ,  VADDADI SUNDEEP ,  HONG JOHN H ,  HAMSICI ONUR C ,  REZNIK YURIY ,  LEE CHONG U

Inventor： VADDADI SUNDEEP ,  HONG JOHN H ,  HAMSICI ONUR C ,  REZNIK YURIY ,  LEE CHONG U

IPC: G06K9/64

CPC classification number: G06K9/6211

Abstract: A method for feature matching in image recognition is provided. First, image scaling may be based on a feature distribution across scale spaces for an image to estimate image size/resolution, where peak(s) in the keypoint distribution at different scales is used to track a dominant image scale and roughly track object sizes. Second, instead of using all detected features in an image for feature matching, keypoints may be pruned based on cluster density and/or the scale level in which the keypoints are detected. Keypoints falling within high-density clusters may be preferred over features falling within lower density clusters for purposes of feature matching. Third, inlier-to-outlier keypoint ratios are increased by spatially constraining keypoints into clusters in order to reduce or avoid geometric consistency checking for the image.

Abstract translation: 提供了一种图像识别中的特征匹配方法。 首先，图像缩放可以基于用于图像的尺度空间上的特征分布来估计图像尺寸/分辨率，其中使用不同尺度的关键点分布中的峰值来跟踪主要图像尺度并粗略地跟踪对象尺寸。 第二，不是使用图像中的所有检测到的特征来进行特征匹配，而是可以基于簇密度和/或检测关键点的比例级别来修剪关键点。 落入高密度簇内的关键点可能优于落入低密度簇内的特征，用于特征匹配。 第三，通过空间约束关键点进入群集来增加从早到晚的关键点比例，以便减少或避免图像的几何一致性检查。

公开(公告)号：WO2013022656A3

公开(公告)日：2014-03-13

申请号：PCT/US2012049055

申请日：2012-07-31

Applicant: QUALCOMM INC ,  REZNIK YURIY ,  HAMSICI ONUR C ,  VADDADI SUNDEEP ,  HONG JOHN H ,  LEE CHONG U

Inventor： REZNIK YURIY ,  HAMSICI ONUR C ,  VADDADI SUNDEEP ,  HONG JOHN H ,  LEE CHONG U

IPC: G06K9/46 ,  G06F17/18 ,  G06T9/00

CPC classification number: G06T9/00 ,  G06F17/18 ,  G06F17/30247 ,  G06K9/4671 ,  G06K2009/226 ,  G06T9/005

Abstract: Methods and devices for coding of feature locations are disclosed. In one embodiment, a method of coding feature location information of an image includes generating a hexagonal grid, where the hexagonal grid includes a plurality of hexagonal cells, quantizing feature locations of an image using the hexagonal grid, generating a histogram to record occurrences of feature locations in each hexagonal cell, and encoding the histogram in accordance with the occurrences of feature locations in each hexagonal cell. The method of encoding the histogram includes applying context information of neighboring hexagonal cells to encode information of a subsequent hexagonal cell to be encoded in the histogram, where the context information includes context information from first order neighbors and context information from second order neighbors of the subsequent hexagonal cell to be encoded.

Abstract translation: 公开了用于编码特征位置的方法和装置。 在一个实施例中，一种编码图像的特征位置信息的方法包括生成六边形网格，其中六边形网格包括多个六边形单元格，使用六边形网格量化图像的特征位置，生成直方图以记录特征的出现 每个六边形单元格中的位置，并根据每个六边形单元格中特征位置的出现来对直方图进行编码。 对直方图进行编码的方法包括应用相邻六边形单元格的上下文信息来编码在直方图中要编码的随后的六边形单元格的信息，其中上下文信息包括来自一阶邻居的上下文信息和来自第二阶邻居的上下文信息 六角形单元格进行编码。

公开(公告)号：WO2011069023A3

公开(公告)日：2011-07-28

申请号：PCT/US2010058807

申请日：2010-12-02

Applicant: QUALCOMM INC ,  HAMSICI ONUR C ,  REZNIK YURIY ,  HONG JOHN H ,  VADDADI SUNDEEP ,  LEE CHONG U

Inventor： HAMSICI ONUR C ,  REZNIK YURIY ,  HONG JOHN H ,  VADDADI SUNDEEP ,  LEE CHONG U

IPC: G06K9/46

CPC classification number: G06K9/4671

Abstract: A method for generating a feature descriptor is provided. A set of pre-generated sparse projection vectors is obtained. A scale space for an image is also obtained, where the scale space having a plurality scale levels. A descriptor for a keypoint in the scale space is then generated based on a combination of the sparse projection vectors and sparsely sampled pixel information for a plurality of pixels across the plurality of scale levels.

Abstract translation: 提供了一种用于生成特征描述符的方法。 获得一组预先生成的稀疏投影矢量。 还获得图像的比例空间，其中比例空间具有多个比例级别。 然后基于多个比例级别上的多个像素的稀疏投影矢量和稀疏采样像素信息的组合生成比例空间中的关键点的描述符。