专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

BR112016013533B8 composição, método para acentuar o perfil de fragrância de uma composição, método para produzir um produto para o consumidor, produto ou artigo de perfumaria para consumidor e método para modificar ou melhorar as propriedades de odor de uma superfície corporal 未知
公开(公告)号：BR112016013533B8
公开(公告)日：2021-04-13
申请号：BR112016013533
申请日：2014-12-11
申请人： PROCTER & GAMBLE
发明人： CHRISTELLE MARIE SANDRINE BONNET , DAVID THOMAS STANTON , FABIENNE PASTOR , JONATHAN RICHARD STONEHOUSE , JOSE MARIA VELAZQUEZ MENDOZA , LYNETTE ANNE MAKINS HOLLAND , ORESTE TODINI , WILLIAM EOGHAN STAITE
IPC分类号： A61Q15/00 , A61K8/34 , A61K8/37 , A61K8/60 , A61Q19/00 , A61Q19/10
摘要： composições de fragrância. a presente invenção refere-se a uma composição de fragrância tendo intensidade e/ou longevidade melhoradas do perfil de fragrância, compreendendo menos que cerca de 30% em peso de materiais de fragrâncias de baixa volatilidade (isto é, notas de base) tendo uma pressão de vapor

2. 发明专利

GB2591178B Machine learning based imaging method of determining authenticity of a consumer good 未知
公开(公告)号：GB2591178B
公开(公告)日：2022-07-27
申请号：GB202018278
申请日：2020-11-20
申请人： PROCTER & GAMBLE
发明人： JONATHAN RICHARD STONEHOUSE , BOGUSLAW OBARA
IPC分类号： G06V20/80 , G06K7/14 , G06K19/06
摘要： An economical and accurate machine learning based imaging method of classifying a consumer good as authentic is provided. The machine learning based imaging method leverages machine learning and the use of steganographic features on the authentic consumer good.

3. 发明专利

BR112016013533B1 composição, método para acentuar o perfil de fragrância de uma composição, método para produzir um produto para o consumidor, produto ou artigo de perfumaria para consumidor e método para modificar ou melhorar as propriedades de odor de uma superfície corporal 未知
公开(公告)号：BR112016013533B1
公开(公告)日：2020-12-15
申请号：BR112016013533
申请日：2014-12-11
申请人： PROCTER & GAMBLE
发明人： CHRISTELLE MARIE SANDRINE BONNET , DAVID THOMAS STANTON , FABIENNE PASTOR , JONATHAN RICHARD STONEHOUSE , JOSE MARIA VELAZQUEZ MENDOZA , LYNETTE ANNE MAKINS HOLLAND , ORESTE TODINI , WILLIAM EOGHAN STAITE
IPC分类号： A61Q15/00 , A61K8/34 , A61K8/37 , A61K8/60 , A61Q19/00 , A61Q19/10
摘要： composições de fragrância. a presente invenção refere-se a uma composição de fragrância tendo intensidade e/ou longevidade melhoradas do perfil de fragrância, compreendendo menos que cerca de 30% em peso de materiais de fragrâncias de baixa volatilidade (isto é, notas de base) tendo uma pressão de vapor

4. 发明专利

MXPA04010081A SILICONAS. 未知
公开(公告)号：MXPA04010081A
公开(公告)日：2004-12-13
申请号：MXPA04010081
申请日：2003-04-22
申请人： PROCTER & GAMBLE
发明人： JONATHAN RICHARD STONEHOUSE
IPC分类号： A61K8/89 , A61K8/893 , A61K8/898 , A61Q5/00 , A61Q5/08 , A61Q5/10 , A61Q5/12 , A61K8/00 , C08G77/382 , C08G77/46 , D06M15/647 , D06M101/12 , A61K7/11 , A61K7/13 , A61K7/135 , C08G77/14 , C08G77/26 , C08G77/38
摘要： En un primer aspecto, la presente invencion se relaciona con siliconas organomodificadas que tienen sustituyentes que comprenden oxigeno, de manera que el contenido de oxigeno de los sustituyentes sea de 1% a 10% del peso de la silicona con grupos funcionales agregados y el contenido de la cadena principal de silicona es de 67 a 95% en peso. En un aspecto adicional, la invencion se relaciona con composiciones topicas que comprenden estas siliconas organomodificadas que se depositan en una variedad de fibras, en especial el cabello, con diferentes niveles de danos. En un aspecto adicional, la invencion se relaciona con un estuche para el tratamiento del cabello que comprende las composiciones topicas.

5. 发明专利

GB2607160A Artificial intelligence based steganographic systems and methods for analyzing pixel data of a product to detect product counterfeiting 未知
公开(公告)号：GB2607160A
公开(公告)日：2022-11-30
申请号：GB202203498
申请日：2022-03-14
申请人： PROCTER & GAMBLE
发明人： JONATHAN RICHARD STONEHOUSE
IPC分类号： G06V10/764 , G06K7/14 , G06V10/82 , G06V20/52 , G06V20/80
摘要： A trained artificial intelligence, such as a neural network (NN) (Fig.4A) that detects the presence of a steganographic feature 500d1af in an image 500d1a of a product 500d1ap, thus classifying the product as authentic or counterfeit goods. The NN is implemented through a computer application. NN training data comprises images with and without the steganographic features. Authentic steganographic features may be hidden in: barcodes; QR codes; data matrix codes; scannable codes; alphanumeric characters; printed symbols; textual characters; fonts; graphical alterations; or packaging alterations (Figs.5A-5G). Steganographic feature authenticity may depend on alignment of the steganographic feature to a two-dimensional data matrix. The training images may be synthetic; created by augmenting, altering, deleting, or annotating steganographic features of a known image. The training images may be an original image that has been cropped and have a reduced pixel count. Training images may comprise multiple angles or perspectives of the product. Training images may comprise a third subset of images depicting real counterfeit features. The steganographic app may display, on a screen, an indication of whether the product is authentic or counterfeit 210. The steganographic app may select a set of authentic steganographic features for printing (130,Fig.1) on a different version of the product.

6. 发明专利

GB2591178A Machine learning based imaging method of determining authenticity of a consumer good 未知
公开(公告)号：GB2591178A
公开(公告)日：2021-07-21
申请号：GB202018278
申请日：2020-11-20
申请人： PROCTER & GAMBLE
发明人： JONATHAN RICHARD STONEHOUSE , BOGUSLAW OBARA
IPC分类号： G06K9/00 , G06K7/14 , G06K19/06
摘要： Method of determining counterfeit consumer products using a machine learning classification model, comprising: inputting an image of the consumer good into the model 11 and outputting a classification from the model indicating a likelihood that the consumer good is authentic or non-authentic 76. The model is trained with images from a plurality of different camera types thereby avoiding camera bias in the trained model. Images of authentic products comprise a visible steganographic feature (Fig. 2) and/or a manufacturing line variable printing (batch) code (Fig. 3). The steganographic features may comprise differing font styles and/or a change in location of a character or punctuation (25-37, Fig. 2A-H). The Manufacturing Line Variable Printing Code may comprise: alphanumeric characters, non-alphanumeric characters; pattern boxes; and/or dotted columns (41, 43, 45, Fig. 3 respectively). A product specification for an authentic item may comprise: production codes; batch codes; brand name; product line label; artwork, ingredient list; and/or usage instructions. The training dataset may be spatially manipulated, by geometric or colour distortion, before training the machine learning classifier (Fig. 7). The image of the consumer product to be authenticated may be spatially manipulated before classification.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式