Image analysis – Histogram processing – With pattern recognition or classification
Reexamination Certificate
2000-12-28
2003-04-29
Tran, Phuoc (Department: 2621)
Image analysis
Histogram processing
With pattern recognition or classification
C382S165000, C382S305000
Reexamination Certificate
active
06556710
ABSTRACT:
TECHNICAL FIELD
This invention relates to image searching techniques.
BACKGROUND
The process of finding and retrieving images stored on electronic media (e.g., a computer, the Internet) has become increasingly difficult for a variety of reasons. For instance, with the explosive growth of the Internet, the number of searchable images available on the Internet has dramatically increased. With the increased number of images, the abilities of conventional systems, methods, and computer programs to perform searching and retrieval functions in an efficient, useful, and timely manner have been challenged.
The ability of conventional systems, methods, and computer programs to efficiently find and retrieve desired images in a database has been hampered by poor organization and inconsistent formatting of the images being searched and/or retrieved. Similar problems also may be experienced by other electronic applications involving a large quantity of images that may be searched for and retrieved. These problems may be compounded when the desired search result includes multiple formats (e.g., images and text).
SUMMARY
In one general aspect, a search for an image includes receiving a set of features corresponding to image characteristics and comparing the set of features to features indexed for multiple stored images. A search result is produced based on the comparison.
Implementations may include one or more of the following features. For example, text that corresponds to image characteristics and/or posterized image characteristics may be received, and features from the text corresponding to image characteristics and/or posterized image characteristics may be identified. Additionally or alternatively, at least a portion of an image may be received, and features of the received image portion corresponding to image characteristics and/or posterized image characteristics may be identified. Based on the identified features, a joint histogram or a posterized joint histogram may be computed.
The received set of features may be compared to features that have been indexed in an inverted index for multiple stored images. The inverted index typically includes a relationship between features and images including those features. Additionally or alternatively, the received set of features may be compared to a list of features, where the list of features includes a relationship between features and images including those features.
The received set of features also may be compared to features that have been indexed in an inverted index for multiple stored posterized images. The inverted index may include a relationship between features and posterized images including those features. Additionally or alternatively, the received set of features may be compared to a list of features, where the list of features includes a relationship between features and posterized images including those features.
The search result may include retrieving a set of images for display in a ranked manner based on similarity metrics, or the like. The set of images may be ranked in order from an image that is most relevant to the set of features to an image that is least relevant to the set of features.
Additionally or alternatively, the received set of features may correspond to posterized image characteristics. These features are generally compared to indexed features that correspond to posterized image characteristics for multiple stored images. Similarly, the received set of features may correspond to image characteristics, where at least one of the received set of features are posterized to create posterized image characteristics. Posterizing may include computing a first histogram vector based on the received set of features that correspond to image characteristics, where the first histogram vector includes one or more vector elements, each representing information for a different received feature. Multiple subsets within at least one of the vector elements included in the first histogram vector are then identified, and a second histogram vector including a vector element for each of the identified subsets is identified. Data within the vector elements of the second histogram vector represent the posterized image characteristics.
In another general aspect, retrieving images and producing a ranked result includes receiving a feature corresponding to an image characteristic and/or a posterized image characteristic, and thereafter searching an index of features corresponding to indexed images to identify selected indexed images that include the received feature. Selected indexed images then may be retrieved depending on whether the selected indexed images satisfy at least one similarity metric. A ranked list of images is ultimately produced based on images satisfying the metric(s).
Implementations may include one or more of the following features. For example, multiple features corresponding to image characteristics and/or posterized image characteristics may be received, and thereafter used to search the index of features to determine if the indexed images include at least one of the received features.
A joint histogram and/or a posterized joint histogram may be computed based on the received features. The index may include joint histogram features, or the index may include posterized joint histogram features. The index of joint histogram features and/or posterized joint histogram features may be searched to determine if the indexed images include at least one of the computed joint histogram features and/or the posterized joint histogram features.
These general and specific aspects may be implemented using a system, a method, or a computer program, or any combination of systems, methods, and computer programs.
Other features and advantages will be apparent from the description and drawings, and from the claims.
REFERENCES:
patent: 028739 (1860-06-01), Lee et al.
patent: 5049986 (1991-09-01), Aono et al.
patent: 5226175 (1993-07-01), Deutsch et al.
patent: 5404435 (1995-04-01), Rosenbaum
patent: 5559899 (1996-09-01), Schmidt
patent: 5579471 (1996-11-01), Barber et al.
patent: 5623681 (1997-04-01), Rivette et al.
patent: 5857033 (1999-01-01), Kim
patent: 5933194 (1999-08-01), Kim et al.
patent: 5949556 (1999-09-01), Tamai
patent: 6181817 (2001-01-01), Zabih et al.
patent: 6243713 (2001-06-01), Nelson et al.
patent: 6295533 (2001-09-01), Cohen
International Search Report (International Application No. PCT/US01/47646).
Pass et al., “Comparing Images Using Joint Histograms,” Multimedia Systems 7(3): 234-240 (1999).
Pass Gregory S.
Wood Frank
America Online Inc.
Fish & Richardson P.C.
Sherali Ishrat
Tran Phuoc
LandOfFree
Image searching techniques does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image searching techniques, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image searching techniques will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3003439