公开(公告)号：US08818103B2

公开(公告)日：2014-08-26

申请号：US13254234

申请日：2010-03-03

申请人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura ,  Yukihito Furuhashi ,  Taiji Mine

发明人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura ,  Yukihito Furuhashi ,  Taiji Mine

IPC分类号： G06K9/46 ,  G06K9/62 ,  G06F7/00 ,  G06F12/00 ,  G06F17/30

CPC分类号： G06K9/6211 ,  G06F17/30259

摘要： An image retrieval method comprising: a step of extracting at least one query feature vector from a query image on which a subject of the image retrieval is captured, the query feature vector representing a local feature of the query image; a step of accessing an image data base in which a plurality of reference images are stored previously, each reference image being stored in conjunction with learning images generated therefrom and reference feature vectors representing local features of the reference image and the learning images; a comparing step of comparing the query feature vector with the reference feature vectors stored in conjunction with each reference image using an approximate nearest neighbor search to find a reference feature vector approximately nearest to the query feature vector; and a selecting step of selecting a reference image with which the found reference feature vector is stored in conjunction from the reference images as a retrieval result wherein: the learning image is generated by adding a defocus and/or a motion-blur effect likely to occur on capturing the subject to each reference image, the reference feature vectors are extracted from each reference image and the learning image corresponding to the reference image respectively using the scale-space approach, the query feature vector is extracted from the query image using the scale-space approach, and each of the above steps is executed by a computer.

摘要翻译： 一种图像检索方法，包括：从其中捕获图像检索对象的查询图像中提取至少一个查询特征向量的步骤，所述查询特征向量表示查询图像的本地特征; 存储先前存储多个参考图像的图像数据库的步骤，每个参考图像与从其生成的学习图像一起存储，参考特征向量表示参考图像和学习图像的局部特征; 比较步骤，使用近似最近邻搜索将查询特征向量与结合每个参考图像存储的参考特征向量进行比较，以找到近似于查询特征向量的参考特征向量; 以及选择步骤，从参考图像中选择结合存储所找到的参考特征向量的参考图像作为检索结果，其中：通过添加散焦和/或可能发生的运动模糊效果来生成学习图像 在对每个参考图像拍摄对象时，分别使用缩放空间方法从每个参考图像和对应于参考图像的学习图像中提取参考特征向量，使用缩放空间方法从查询图像中提取查​​询特征向量， 空间方法，并且上述步骤中的每一个由计算机执行。

公开(公告)号：US20120051628A1

公开(公告)日：2012-03-01

申请号：US13254234

申请日：2010-03-03

申请人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura ,  Yukihito Furuhashi ,  Taiji Mine

发明人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura ,  Yukihito Furuhashi ,  Taiji Mine

IPC分类号： G06K9/62

CPC分类号： G06K9/6211 ,  G06F17/30259

摘要： An image retrieval method comprising: a step of extracting at least one query feature vector from a query image on which a subject of the image retrieval is captured, the query feature vector representing a local feature of the query image; a step of accessing an image data base in which a plurality of reference images are stored previously, each reference image being stored in conjunction with learning images generated therefrom and reference feature vectors representing local features of the reference image and the learning images; a comparing step of comparing the query feature vector with the reference feature vectors stored in conjunction with each reference image using an approximate nearest neighbor search to find a reference feature vector approximately nearest to the query feature vector; and a selecting step of selecting a reference image with which the found reference feature vector is stored in conjunction from the reference images as a retrieval result wherein: the learning image is generated by adding a defocus and/or a motion-blur effect likely to occur on capturing the subject to each reference image, the reference feature vectors are extracted from each reference image and the learning image corresponding to the reference image respectively using the scale-space approach, the query feature vector is extracted from the query image using the scale-space approach, and each of the above steps is executed by a computer.

摘要翻译： 一种图像检索方法，包括：从其中捕获图像检索对象的查询图像中提取至少一个查询特征向量的步骤，所述查询特征向量表示查询图像的本地特征; 存储先前存储多个参考图像的图像数据库的步骤，每个参考图像与从其生成的学习图像一起存储，参考特征向量表示参考图像和学习图像的局部特征; 比较步骤，使用近似最近邻搜索将查询特征向量与结合每个参考图像存储的参考特征向量进行比较，以找到近似于查询特征向量的参考特征向量; 以及选择步骤，从参考图像中选择结合存储所找到的参考特征向量的参考图像作为检索结果，其中：通过添加散焦和/或可能发生的运动模糊效果来生成学习图像 在对每个参考图像拍摄对象时，分别使用缩放空间方法从每个参考图像和对应于参考图像的学习图像中提取参考特征向量，使用缩放空间方法从查询图像中提取查​​询特征向量， 空间方法，并且上述步骤中的每一个由计算机执行。

公开(公告)号：US08340451B2

公开(公告)日：2012-12-25

申请号：US12989906

申请日：2009-04-27

申请人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura

发明人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura

IPC分类号： G06K9/00 ,  H04N11/02 ,  G10L19/12

CPC分类号： G06F17/30247 ,  G06K9/4671

摘要： Provided is a method for constructing an image database for object recognition, which includes a feature extraction step of extracting local descriptors from object images which are to be stored in an image database, a scalar quantization step of quantizing a numeric value indicating each dimension of each of the local descriptors into a predetermined number of bit digits, and a storing step of organizing each of the local descriptors after the quantization to be able to be searched for in the closest vicinity, giving to the local descriptor an identifier of the image from which the local descriptor has been extracted, and storing the local descriptor to which the identifiers are given in the image database. The storing step comprises extracting the local descriptors from the object images when a search query is given, scalar-quantizing each dimension, determining a local descriptor in the closest vicinity of each of the local descriptors from the image database, and storing each local descriptors so as to be able to identify one image by majority vote processing from the images including any determined local descriptor. The scalar quantization step comprises quantizing each dimension of each of the local descriptors into 8 bits or less. Also provided are a processing program for the method and a processing device for performing the processing.

摘要翻译： 提供一种用于构建用于对象识别的图像数据库的方法，其包括从要存储在图像数据库中的对象图像中提取局部描述符的特征提取步骤，量化表示每个维度的每个维度的数值的数值的标量量化步骤 将本地描述符转换成预定数量的位数，以及存储步骤，在量化之后组织每个本地描述符以便能够在最近的附近被搜索，向本地描述符给出图像的标识符， 已经提取了本地描述符，并且存储在图像数据库中给出标识符的本地描述符。 存储步骤包括当提供搜索查询时从对象图像中提取局部描述符，对每个维进行标量量化，从图像数据库中确定每个局部描述符最接近的局部描述符，并且存储每个局部描述符 以便能够从包括任何确定的本地描述符的图像通过多数投票处理识别一个图像。 标量量化步骤包括将每个局部描述符的每个维度量化为8位或更少。 还提供了用于该方法的处理程序和用于执行处理的处理装置。

公开(公告)号：US08199973B2

公开(公告)日：2012-06-12

申请号：US12439493

申请日：2007-08-01

申请人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura

发明人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura

IPC分类号： G06K9/00 ,  H04N7/16

CPC分类号： G06K9/6276 ,  G06K9/4671 ,  G06K9/6211 ,  G06K9/6231

摘要： For object recognition based on nearest neighbor search of local descriptors such as SIFT, it is important to keep the nearest neighbor search efficient to deal with a huge number of descriptors. The present invention provides methods of efficient recognition. In one embodiment, the method is based on the observation that the level of accuracy of nearest neighbor search for correct recognition depends on images to be recognized. The method is characterized by the mechanism that multiple recognizers with approximate nearest neighbor search are cascaded in the order of the level of approximation so as to improve the efficiency by adaptively controlling the level to be applied depending on images. In another embodiment the method is characterized by excluding local descriptors with low discriminability when a plenty of local descriptors are present in the vicinity and a plenty of distance calculation are required.

摘要翻译： 对于基于SIFT等局部描述符的最近邻搜索的对象识别，保持最近邻搜索有效处理大量描述符很重要。 本发明提供了有效识别的方法。 在一个实施例中，该方法基于如下观察：最近邻搜索正确识别的精度水平取决于要识别的图像。 该方法的特征在于具有近似最近邻搜索的多个识别器以近似级别的级联级联，以通过根据图像自适应地控制要应用的级别来提高效率。 在另一个实施例中，该方法的特征在于当附近存在大量局部描述符并且需要大量的距离计算时，排除具有低识别能力的局部描述符。

公开(公告)号：US20100027895A1

公开(公告)日：2010-02-04

申请号：US12439493

申请日：2007-08-01

申请人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura

发明人： Kazuto Noguchi ,  Koichi Kise ,  Masakazu Iwamura

IPC分类号： G06K9/62

CPC分类号： G06K9/6276 ,  G06K9/4671 ,  G06K9/6211 ,  G06K9/6231

摘要： The invention provides a method for actualizing a high-speed object recognition process in an object recognition which extracts feature vectors from an input image where an object is captured, the feature vectors representing the object with the large number of the feature vectors; and retrieves, among an image database, one or more images whose features are matched. The invention also provides a method for saving a memory capacity required for the image database.Suggested are methods each of the methods being capable of reducing a processing time required to perform a task which recognizes the object described by a plurality of the feature vectors with use of an approximate nearest neighbor search technique. One of the suggested methods is actualized by excluding such feature vectors that a large number of feature vectors are present in the vicinity and a large number of distance calculations, therefore, are required. Another suggested method is actualized by only performs voting using a hash table without performing any distance calculation. Moreover, another suggested method is actualized by connecting recognizers in cascaded manner each of which uses the nearest neighbor search technique, varying a degree of approximation applied to the object recognition in accordance with the input image, thereby significantly increasing efficiency.

摘要翻译： 本发明提供了一种实现物体识别中的高速物体识别过程的方法，该对象识别从对象被捕获的输入图像中提取特征向量，表示具有大量特征向量的对象的特征向量; 并且在图像数据库中检索其特征匹配的一个或多个图像。 本发明还提供了一种用于保存图像数据库所需的存储容量的方法。 建议的方法是每种方法能够减少执行使用近似最近邻搜索技术来识别由多个特征向量描述的对象的任务所需的处理时间。 所提出的方法之一是通过排除在附近存在大量特征向量并且需要大量距离计算的特征向量来实现的。 另一个建议的方法是通过使用哈希表执行投票而不执行任何距离计算来实现的。 此外，通过以级联方式连接识别器来实现另一种建议的方法，每个方法使用最近邻搜索技术，根据输入图像改变应用于对象识别的近似程度，从而显着提高效率。

公开(公告)号：US20140086492A1

公开(公告)日：2014-03-27

申请号：US14119775

申请日：2012-05-15

申请人： Masakazu Iwamura ,  Koichi Kise

发明人： Masakazu Iwamura ,  Koichi Kise

IPC分类号： G06K9/62 ,  G06F17/30

CPC分类号： G06K9/62 ,  G06F16/50 ,  G06F16/56

摘要： The present application relates to a method whereby a plurality of characteristic vectors which are extracted from image data are logged in a database together with the image data for approximate nearest neighbor searching, and has as an objective reducing computation time and memory use. L groups of K hash tables are generated, and each characteristic vector is respectively logged with each hash table. With one group as a copy destination, another group as a copy source, and each respective division by combination of logging bin of the K hash tables of each group as a bucket: 1) a given characteristic vector is focused on; 2) another characteristic vector which is logged in the same bucket in the copy source as the characteristic vector is identified; 3) a characteristic vector is selected in which a number of groups in which the other characteristic vector is logged in the same bucket as the characteristic vector which is focused on is greater than or equal to a prescribed threshold; and 4) when the characteristic vector which is selected in 3) is not logged in each bin of the copy destination in which the characteristic vector being focused on is logged, the characteristic vector is logged in each bin. After focusing on a prescribed number of characteristic vectors and executing 1)-4) foregoing for each characteristic vector, the copy source hash tables are deleted.

摘要翻译： 本申请涉及一种方法，其中从图像数据中提取的多个特征向量与用于近似最近邻搜索的图像数据一起记录在数据库中，并且具有减少计算时间和存储器使用的目的。 生成L组K个哈希表，并且每个特征向量分别与每个散列表一起记录。 将一组作为复制目的地，另一组作为复制源，并将每个组的K个哈希表的日志文件夹的组合分别作为一个桶：1）给定的特征向量集中在; 2）识别特征向量时在复制源中记录在同一桶中的另一个特征向量; 3）选择特征向量，其中与聚焦的特征向量相同的桶中记录另一特征向量的组数大于或等于规定阈值; 和4）当记录在3）中选择的特征向量未记录在其中聚焦的特征向量的复制目的地的每个仓时，在每个仓中记录特征向量。 在专注于规定数量的特征向量并且对每个特征向量执行1）-4）之后，删除复制源哈希表。

公开(公告)号：US20120230592A1

公开(公告)日：2012-09-13

申请号：US13148850

申请日：2010-02-09

申请人： Masakazu Iwamura ,  Koichi Kise

发明人： Masakazu Iwamura ,  Koichi Kise

IPC分类号： G06K9/46 ,  G06K9/68

CPC分类号： G06K9/3283 ,  G06F17/30256 ,  G06K9/723 ,  G06K2209/01 ,  G06T7/66

摘要： A pattern recognition apparatus including: an extracting section for extracting, from a query image that is composed of at least one piece of pattern component and previously undergoes a geometric transformation, the pattern component; a feature acquiring section for acquiring a geometric invariant feature of the pattern component as a query feature, the query feature being represented by at least three feature points including first, second and third feature points, each feature point locating on the pattern component and being retrieved from the pattern component based on a predetermined rule; a comparing section for comparing the query feature with a plurality of reference features, each reference feature representing different reference patterns prepared as candidates for pattern recognition; and a pattern determination section for determining, as a recognition result, a specific reference pattern out of the candidates based on a similarity of features therebetween and, wherein: each reference feature is represented using feature points retrieved from each reference pattern based on the same rule as that of the query feature, and based on the predetermined rule, a position of the first feature point is specified out of points which locate on the pattern component and are invariant to the geometric transformation, a position of the second feature point is specified using a characteristic regarding a shape of the pattern component, the characteristic being invariant to the geometric transformation, and a position of the third feature point is specified from a predetermined value being invariant to the geometric transformation and from the specified positions of the first and second feature points.

摘要翻译： 一种图案识别装置，包括：提取部分，用于从由至少一个图案分量组成并且先前经历几何变换的查询图像中提取图案分量; 特征获取部分，用于获取所述图案部件的几何不变特征作为查询特征，所述查询特征由至少三个特征点表示，所述特征点包括第一，第二和第三特征点，每个特征点定位在所述图案部件上并被检索 基于预定规则从图案分量; 比较部分，用于将查询特征与多个参考特征进行比较，每个参考特征表示准备为模式识别候选的不同参考图; 以及模式确定部分，用于基于它们之间的特征的相似度来确定候选者中的特定参考模式作为识别结果，并且其中：基于相同规则从每个参考模式检索的特征点来表示每个参考特征 作为查询特征，并且基于预定规则，从位于图案分量上的点指定第一特征点的位置，并且对于几何变换是不变的，第二特征点的位置使用 关于图形分量的形状，对几何变换不变的特性以及第三特征点的位置的特征是从对于几何变换不变的预定值和从第一和第二特征的指定位置 积分

公开(公告)号：US08422793B2

公开(公告)日：2013-04-16

申请号：US13148850

申请日：2010-02-09

申请人： Masakazu Iwamura ,  Koichi Kise

发明人： Masakazu Iwamura ,  Koichi Kise

IPC分类号： G06K9/50

CPC分类号： G06K9/3283 ,  G06F17/30256 ,  G06K9/723 ,  G06K2209/01 ,  G06T7/66

摘要： A pattern recognition apparatus including: an extracting section for extracting, from a query image that is composed of at least one piece of pattern component and previously undergoes a geometric transformation, the pattern component; a feature acquiring section for acquiring a geometric invariant feature of the pattern component as a query feature, the query feature being represented by at least three feature points including first, second and third feature points, each feature point locating on the pattern component and being retrieved from the pattern component based on a predetermined rule; a comparing section for comparing the query feature with a plurality of reference features, each reference feature representing different reference patterns prepared as candidates for pattern recognition; and a pattern determination section for determining, as a recognition result, a specific reference pattern out of the candidates based on a similarity of features therebetween and, wherein: each reference feature is represented using feature points retrieved from each reference pattern based on the same rule as that of the query feature, and based on the predetermined rule, a position of the first feature point is specified out of points which locate on the pattern component and are invariant to the geometric transformation, a position of the second feature point is specified using a characteristic regarding a shape of the pattern component, the characteristic being invariant to the geometric transformation, and a position of the third feature point is specified from a predetermined value being invariant to the geometric transformation and from the specified positions of the first and second feature points.

摘要翻译： 一种图案识别装置，包括：提取部分，用于从由至少一个图案分量组成并且先前经历几何变换的查询图像中提取图案分量; 特征获取部分，用于获取所述图案部件的几何不变特征作为查询特征，所述查询特征由至少三个特征点表示，所述特征点包括第一，第二和第三特征点，每个特征点定位在所述图案部件上并被检索 基于预定规则从图案分量; 比较部分，用于将查询特征与多个参考特征进行比较，每个参考特征表示准备为模式识别候选的不同参考图; 以及模式确定部分，用于基于它们之间的特征的相似度来确定候选者中的特定参考模式作为识别结果，并且其中：基于相同规则从每个参考模式检索的特征点来表示每个参考特征 作为查询特征，并且基于预定规则，从位于图案分量上的点指定第一特征点的位置，并且对于几何变换是不变的，第二特征点的位置使用 关于图形分量的形状，对几何变换不变的特性以及第三特征点的位置的特征是从对于几何变换不变的预定值和从第一和第二特征的指定位置 积分

公开(公告)号：US08036497B2

公开(公告)日：2011-10-11

申请号：US11885567

申请日：2006-02-15

申请人： Koichi Kise ,  Tomohiro Nakai ,  Masakazu Iwamura

发明人： Koichi Kise ,  Tomohiro Nakai ,  Masakazu Iwamura

IPC分类号： G06K9/60 ,  G06F17/30

CPC分类号： G06F17/30247 ,  G06K9/00442 ,  G06K9/00483

摘要： A document/image retrieval method for retrieving a document/image corresponding to a captured digital image from a database by comparing features calculated based on feature points of the captured digital image with features preliminarily calculated based on feature points of each of documents and/or images stored in the database, the method comprising: extracting the feature points from the captured digital image; defining a local set of feature points for each of the extracted feature points; selecting feature points from the defined local set to define a feature point subset of the local set; determining invariant values as values characterizing the defined subset for combinations of the feature points in the subset, the invariant values being invariant to a geometric transformation; calculating a feature by combining the determined invariant values; and performing a voting process on the documents and/or images in the database based on the preliminarily calculated features of the documents and/or images; whereby the document/image corresponding to the captured digital image is retrieved from the database.

摘要翻译： 一种文档/图像检索方法，用于通过将基于所捕获的数字图像的特征点计算的特征与基于每个文档和/或图像的特征点预先计算的特征进行比较，从数据库中检索与捕获的数字图像相对应的文档/图像 存储在数据库中，该方法包括：从捕获的数字图像中提取特征点; 为每个提取的特征点定义一组局部特征点; 从定义的本地集合中选择特征点以定义本地集合的特征点子集; 将不变值确定为表征所述子集中特征点的组合的所定义子集的值，所述不变值对于几何变换是不变的; 通过组合确定的不变值来计算特征; 并且基于所述文档和/或图像的预先计算的特征，对所述数据库中的文档和/或图像执行投票处理; 从而从数据库中检索对应于所捕获的数字图像的文档/图像。

公开(公告)号：US20080177764A1

公开(公告)日：2008-07-24

申请号：US11885567

申请日：2006-02-15

申请人： Koichi Kise ,  Tomohiro Nakai ,  Masakazu Iwamura

发明人： Koichi Kise ,  Tomohiro Nakai ,  Masakazu Iwamura

IPC分类号： G06F17/30

CPC分类号： G06F17/30247 ,  G06K9/00442 ,  G06K9/00483

摘要： A document/image retrieval method for retrieving a document/image corresponding to a captured digital image from a database by comparing features calculated based on feature points of the captured digital image with features preliminarily calculated based on feature points of each of documents and/or images stored in the database, the method comprising: extracting the feature points from the captured digital image; defining a local set of feature points for each of the extracted feature points; selecting feature points from the defined local set to define a feature point subset of the local set; determining invariant values as values characterizing the defined subset for combinations of the feature points in the subset, the invariant values being invariant to a geometric transformation; calculating a feature by combining the determined invariant values; and performing a voting process on the documents and/or images in the database based on the preliminarily calculated features of the documents and/or images; whereby the document/image corresponding to the captured digital image is retrieved from the database.

摘要翻译： 一种文档/图像检索方法，用于通过将基于所捕获的数字图像的特征点计算的特征与基于每个文档和/或图像的特征点预先计算的特征进行比较，从数据库中检索与捕获的数字图像相对应的文档/图像 存储在数据库中，该方法包括：从捕获的数字图像中提取特征点; 为每个提取的特征点定义一组局部特征点; 从定义的本地集合中选择特征点以定义本地集合的特征点子集; 将不变值确定为表征所述子集中特征点的组合的所定义子集的值，所述不变值对于几何变换是不变的; 通过组合确定的不变值来计算特征; 并且基于所述文档和/或图像的预先计算的特征，对所述数据库中的文档和/或图像执行投票处理; 从而从数据库中检索对应于所捕获的数字图像的文档/图像。