公开(公告)号：US10902539B2

公开(公告)日：2021-01-26

申请号：US15446811

申请日：2017-03-01

Applicant: Digimarc Corporation

Inventor： Tony F. Rodriguez ,  Osama M. Alattar ,  Hugh L. Brunk ,  Joel R. Meyer ,  William Y. Conwell ,  Ajith Mulki Kamath

IPC: G06K9/62 ,  G06T1/00 ,  G06K9/00 ,  G06F21/16 ,  G06K9/32

Abstract: A sequence of images depicting an object is captured, e.g., by a camera at a point-of-sale terminal in a retail store. The object is identified, such as by a barcode or watermark that is detected from one or more of the images. Once the object's identity is known, such information is used in training a classifier (e.g., a machine learning system) to recognize the object from others of the captured images, including images that may be degraded by blur, inferior lighting, etc. In another arrangement, such degraded images are processed to identify feature points useful in fingerprint-based identification of the object. Feature points extracted from such degraded imagery aid in fingerprint-based recognition of objects under real life circumstances, as contrasted with feature points extracted from pristine imagery (e.g., digital files containing label artwork for such objects). A great variety of other features and arrangements—some involving designing classifiers so as to combat classifier copying—are also detailed.

公开(公告)号：US20170243317A1

公开(公告)日：2017-08-24

申请号：US15446811

申请日：2017-03-01

Applicant: Digimarc Corporation

Inventor： Tony F. Rodriguez ,  Osama M. Alattar ,  Hugh L. Brunk ,  Joel R. Meyer ,  William Y. Conwell ,  Ajith Mulki Kamath

IPC: G06T1/00 ,  G06K9/62

CPC classification number: G06T1/0021 ,  G06F21/16 ,  G06K9/00671 ,  G06K9/3216 ,  G06K9/6256 ,  G06K9/6259 ,  G06K9/627

Abstract: A sequence of images depicting an object is captured, e.g., by a camera at a point-of-sale terminal in a retail store. The object is identified, such as by a barcode or watermark that is detected from one or more of the images. Once the object's identity is known, such information is used in training a classifier (e.g., a machine learning system) to recognize the object from others of the captured images, including images that may be degraded by blur, inferior lighting, etc. In another arrangement, such degraded images are processed to identify feature points useful in fingerprint-based identification of the object. Feature points extracted from such degraded imagery aid in fingerprint-based recognition of objects under real life circumstances, as contrasted with feature points extracted from pristine imagery (e.g., digital files containing label artwork for such objects). A great variety of other features and arrangements—some involving designing classifiers so as to combat classifier copying—are also detailed.

公开(公告)号：US20150055855A1

公开(公告)日：2015-02-26

申请号：US14449821

申请日：2014-08-01

Applicant: Digimarc Corporation

Inventor： Tony F. Rodriguez ,  Osama M. Alattar ,  Hugh L. Brunk ,  Joel R. Meyer ,  William Y. Conwell ,  Ajith Mulki Kamath

IPC: G06K9/62

CPC classification number: G06T1/0021 ,  G06F21/16 ,  G06K9/00671 ,  G06K9/3216 ,  G06K9/6256 ,  G06K9/6259 ,  G06K9/627

Abstract: A sequence of images depicting an object is captured, e.g., by a camera at a point-of-sale terminal in a retail store. The object is identified, such as by a barcode or watermark that is detected from one or more of the images. Once the object's identity is known, such information is used in training a classifier (e.g., a machine learning system) to recognize the object from others of the captured images, including images that may be degraded by blur, inferior lighting, etc. In another arrangement, such degraded images are processed to identify feature points useful in fingerprint-based identification of the object. Feature points extracted from such degraded imagery aid in fingerprint-based recognition of objects under real life circumstances, as contrasted with feature points extracted from pristine imagery (e.g., digital files containing label artwork for such objects). A great variety of other features and arrangements—some involving designing classifiers so as to combat classifier copying—are also detailed.

Abstract translation: 描绘对象的图像序列例如通过零售商店中的销售点终端处的相机被捕获。 识别对象，例如通过从一个或多个图像检测到的条形码或水印。 一旦对象的身份被知道，这样的信息被用于训练分类器（例如，机器学习系统）以从其他捕获的图像识别对象，包括可能由于模糊，劣质照明等而降级的图像。在另一个 处理这种退化的图像以识别在对象的基于指纹的识别中有用的特征点。 从这种退化的图像提取的特征点有助于在现实生活环境下的对象的基于指纹的识别，与从原始图像提取的特征点（例如，包含用于这些对象的标签图案的数字文件）相反。 其他各种功能和布置也有所不同，其中一些涉及设计分类器，以防止分类器复制。