G - Physics – 06 – F
Patent
G - Physics
06
F
G06F 17/30 (2006.01) G06T 7/00 (2006.01)
Patent
CA 2626323
With the advent of the digital camera and memory media that can hold thousands of images, the consumer is faced with the task of searching through many thousands of images to select images that are appropriate for his or her needs. The selection of photo images based on both content and quality can be made considerably easier for the consumer through the implementation of an automated tagging algorithm. The characteristics of an image can be detected and classified in an automated process. An image is processed and features of the image are extracted. These features and the values associated with them are examined by classification algorithms ("classifiers") that achieve the detection of specific image characteristics. Image characteristics include but are not limited to the presence of objects, lighting environments and image quality. These characteristics can be combined with metadata available from EXIF files or the like produced by the camera used to capture the image. Automated learning capability is included in the automated tagging algorithm. The algorithm can learn based on frequent user behavior by developing a user profile, including but not limited to information regarding frequently used camera models, camera settings, lighting environments and subject matter. The user profile permits the algorithm to make tagging decisions with greater accuracy. Each tag is described as a combination of image characteristics, metadata and user profile data (where a combination may be based on a minimum of one image characteristic or one item of metadata or one item of user profile data). The characteristics, metadata and user profile data associated with an image are compared to a database of tags in which each tag corresponds to a specific combination of characteristics, metadata and user profile data. An image is automatically assigned all of the tags described by the characteristics, metadata and user profile data associated with the image.
Guzzwell John A.
Hale Rodney D.
Lefeuvre Edythe P.
Guzzwell John A.
Hale Rodney D.
Lefeuvre Edythe P.
Na
LandOfFree
Automated tagging of photo images does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automated tagging of photo images, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automated tagging of photo images will most certainly appreciate the feedback.
Profile ID: LFCA-PAI-O-1926653