公开(公告)号：JP2001195592A

公开(公告)日：2001-07-19

申请号：JP2000350169

申请日：2000-11-16

Applicant: ST MICROELECTRONICS SRL

Inventor： DE PONTI MAURO ,  SCHETTINI RAIMONDO ,  BRAMBILLA CARLA ,  VALSASNA ANNA ,  CIOCCA GIANLUIGI

IPC: G06T7/00

Abstract: PROBLEM TO BE SOLVED: To provide a digital image sorting method by contents, by which the complexity of calculation required for analyzing the image of a large number of pixels is avoided. SOLUTION: The set of N pieces of low level features for describing the semantic contents of the image composed of a quantity provided from the image is defined while using a known logical and mathematical expression (step 10), and the image of a sorting target is indexed (20) for extracting a feature vector formed from the values of these N low level features. Next, a feature space (vector space) specified by N selected features is divided into the finite number of areas (30). Each of these divided areas is the locus of a point in the relevant feature space, where the value of at least one component of the feature vector satisfies a prescribed relation with each of thresholds. A sorting area including the feature vector is identified (40) from the area dividing such a feature space and an image class made correspond thereto is identified (50).

公开(公告)号：IT1311443B1

公开(公告)日：2002-03-12

申请号：ITTO990996

申请日：1999-11-16

Applicant: ST MICROELECTRONICS SRL

Inventor： DE PONTI MAURO ,  SCHETTINI RAIMONDO ,  BRAMBILLA CARLA ,  VALSASNA ANNA ,  CIOCCA GIANLUIGI

IPC: G06F17/30 ,  G06K9/00

Abstract: The classification method involves the following steps: defining a set of low-level features describing the semantic content of the image, said features being quantities obtainable from the image by means of logico-mathematical expressions that are known beforehand, and the choice of said features depending upon the image classes used for the classification; indexing an image to be classified, with the purpose of extracting therefrom a feature vector, the components of which consist of the values assumed, in the image, by said low-level features; splitting the feature space defined by the low-level features into a plurality of classification regions, to each one of said regions there being associated a respective image class, and each classification region being the locus of the points of the feature space defined by a finite set of conditions laid on at least one component of the feature vector; associating the feature vector to the feature space; identifying, among the classification regions, a specific classification region containing the feature vector extracted from the image to be classified; and identifying the image class associated to the specific classification region identified.

公开(公告)号：ITVI20120291A1

公开(公告)日：2014-04-30

申请号：ITVI20120291

申请日：2012-10-29

Applicant: ST MICROELECTRONICS SRL

Inventor： BIANCO SIMONE ,  BRUNA ARCANGELORANIERI ,  NACCARDI FILIPPO ,  SCHETTINI RAIMONDO

公开(公告)号：ITTO990996A1

公开(公告)日：2001-05-16

申请号：ITTO990996

申请日：1999-11-16

Applicant: ST MICROELECTRONICS SRL

Inventor： DE PONTI MAURO ,  SCHETTINI RAIMONDO ,  BRAMBILLA CARLA ,  VALSASNA ANNA ,  CIOCCA GIANLUIGI

IPC: G06F17/30 ,  G06K9/00

Abstract: The classification method involves the following steps: defining a set of low-level features describing the semantic content of the image, said features being quantities obtainable from the image by means of logico-mathematical expressions that are known beforehand, and the choice of said features depending upon the image classes used for the classification; indexing an image to be classified, with the purpose of extracting therefrom a feature vector, the components of which consist of the values assumed, in the image, by said low-level features; splitting the feature space defined by the low-level features into a plurality of classification regions, to each one of said regions there being associated a respective image class, and each classification region being the locus of the points of the feature space defined by a finite set of conditions laid on at least one component of the feature vector; associating the feature vector to the feature space; identifying, among the classification regions, a specific classification region containing the feature vector extracted from the image to be classified; and identifying the image class associated to the specific classification region identified.

公开(公告)号：ITTO990996D0

公开(公告)日：1999-11-16

申请号：ITTO990996

申请日：1999-11-16

Applicant: ST MICROELECTRONICS SRL

Inventor： DE PONTI MAURO ,  SCHETTINI RAIMONDO ,  BRAMBILLA CARLA ,  VALSASNA ANNA ,  CIOCCA GIANLUIGI

IPC: G06F17/30 ,  G06K9/00

Abstract: The classification method involves the following steps: defining a set of low-level features describing the semantic content of the image, said features being quantities obtainable from the image by means of logico-mathematical expressions that are known beforehand, and the choice of said features depending upon the image classes used for the classification; indexing an image to be classified, with the purpose of extracting therefrom a feature vector, the components of which consist of the values assumed, in the image, by said low-level features; splitting the feature space defined by the low-level features into a plurality of classification regions, to each one of said regions there being associated a respective image class, and each classification region being the locus of the points of the feature space defined by a finite set of conditions laid on at least one component of the feature vector; associating the feature vector to the feature space; identifying, among the classification regions, a specific classification region containing the feature vector extracted from the image to be classified; and identifying the image class associated to the specific classification region identified.