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. 发明授权

US10296813B2 Training method and apparatus for neural network for image recognition 有权
公开(公告)号：US10296813B2
公开(公告)日：2019-05-21
申请号：US15254249
申请日：2016-09-01
申请人： FUJITSU LIMITED
发明人： Li Chen , Song Wang , Wei Fan , Jun Sun , Naoi Satoshi
IPC分类号： G06K9/62 , G06T5/20 , G06N3/08
摘要： A training method and a training apparatus for a neutral network for image recognition are provided. The method includes: representing a sample image as a point set in a high-dimensional space, a size of the high-dimensional space being a size of space domain of the sample image multiplied by a size of intensity domain of the sample image; generating a first random perturbation matrix having a same size as the high-dimensional space; smoothing the first random perturbation matrix; perturbing the point set in the high-dimensional space using the smoothed first random perturbation matrix to obtain a perturbed point set; and training the neutral network using the perturbed point set as a new sample. With the training method and the training apparatus, classification performance of a conventional convolutional neural network is improved, thereby generating more training samples, reducing influence of overfitting, and enhancing generalization performance of the convolutional neural network.

2. 发明授权

US11144817B2 Device and method for determining convolutional neural network model for database 有权
公开(公告)号：US11144817B2
公开(公告)日：2021-10-12
申请号：US15799533
申请日：2017-10-31
申请人： FUJITSU LIMITED
发明人： Li Sun , Song Wang , Wei Fan , Jun Sun
IPC分类号： G06N3/04 , G06N3/08
摘要： A device for determining a CNN model for a database according to the present disclosure includes: a selecting unit configured to select at least two CNN models from multiple CNN models whose classification capacity is known; a fitting unit configured to fit, based on the classification capacity and first parameters of the at least two CNN models, a curve taking classification capacity and a first parameter as variables; a predicting unit configured to predict, based on the curve, a first parameter of other CNN models; and a determining unit configured to determine a CNN model applicable to the database from the multiple CNN models. With the device and the method according to the present disclosure, there is no need to train all the CNN models, thereby greatly reducing the amount of computation and simplifying the process of designing the CNN model.

3. 发明授权

US09881228B2 Method and apparatus for removing a mark in a document image 有权
公开(公告)号：US09881228B2
公开(公告)日：2018-01-30
申请号：US14965548
申请日：2015-12-10
申请人： FUJITSU LIMITED
发明人： Song Wang , Wei Fan , Jun Sun
IPC分类号： G06K9/46 , G06T5/00
CPC分类号： G06K9/4604 , G06K9/38 , G06K9/4638 , G06K9/6218 , G06T5/002 , G06T2207/30176
摘要： A method for removing a mark in a document image includes: extracting connected components from a binary image corresponding to the document image; clustering the connected components based on grayscale features of the connected components to obtain one clustering center; searching, within numerical ranges of a clustering radius R and a grayscale threshold T, for a combination (R, T) which causes an evaluation value based on the grayscale features of the connected components to be higher than a first evaluation threshold; and removing the mark in the document image based on the grayscale threshold in the combination. The method and an apparatus according to the invention can remove a mark in a document image effectively and accurately.

4. 发明授权

US09204011B1 Apparatus and method for extracting a background luminance map of an image, de-shading apparatus and method 有权
标题翻译：用于提取图像的背景亮度图的设备和方法，去遮蔽装置和方法
公开(公告)号：US09204011B1
公开(公告)日：2015-12-01
申请号：US14717273
申请日：2015-05-20
申请人： FUJITSU LIMITED
发明人： Yun Zheng , Wei Fan , Jun Sun
IPC分类号： H04N1/407 , H04N1/409 , G06T5/00 , H04N1/401
CPC分类号： H04N1/401 , G06K9/38 , G06T5/00 , G06T5/008 , H04N1/407 , H04N1/4072 , H04N1/409
摘要： An apparatus and a method for extracting a background luminance map of an image, and a de-shading apparatus and method. The apparatus includes: a luminance extracting unit, configured to extract luminance values everywhere in the image to obtain a luminance map; a separating unit, configured to separate background from foreground of the image based on the luminance map, to obtain an initial background luminance map; a top and bottom luminance obtaining unit, configured to extract top and bottom luminance of the initial background luminance map, and in the case of a part of the top and/or bottom luminance being missing, to supplement the missing part utilizing existing data of the top and/or bottom luminance to obtain complete top and bottom luminance; and an interpolation unit, configured to perform interpolation on the whole image based on the complete top and bottom luminance, to obtain the background luminance map of the image.
摘要翻译：一种用于提取图像的背景亮度图的装置和方法，以及去遮蔽装置和方法。该装置包括：亮度提取单元，被配置为提取图像中的任何地方的亮度值以获得亮度图; 分离单元，被配置为基于亮度图从背景与图像的前景分离，以获得初始背景亮度图; 顶部和底部亮度获取单元，被配置为提取初始背景亮度图的顶部和底部亮度，并且在顶部和/或底部亮度的一部分丢失的情况下，使用现有的数据来补充缺失部分顶部和/或底部亮度以获得完整的顶部和底部亮度; 以及内插单元，被配置为基于完整的顶部和底部亮度对整个图像执行插值，以获得图像的背景亮度图。

5. 发明授权

US10891476B2 Method, system, and neural network for identifying direction of a document 有权
公开(公告)号：US10891476B2
公开(公告)日：2021-01-12
申请号：US15876334
申请日：2018-01-22
申请人： Fujitsu Limited
发明人： Ruochen Wang , Wei Fan , Jun Sun
IPC分类号： G06K9/00 , G06N3/08 , G06K9/62 , G06K9/32 , G06N3/04
摘要： A method, system, and neural network for identifying direction of a document where the method comprises: extracting a text line in the document; calculating a first normal direction result indicative of the text line probably being in a normal direction and a first upside-down direction result indicative of the text line probably being in a direction upside-down with respect to the normal direction; calculating a second normal direction result indicative of the text line after being rotated by 180 degrees probably being in the normal direction and a second upside-down direction result indicative of the text line after being rotated by 180 degrees probably being in the direction upside-down with respect to the normal direction; and determining the direction of the document according to the first normal direction result, the first upside-down direction result, the second normal direction result and the second upside-down direction result.

6. 发明授权

US10614560B2 Apparatus and method for image processing 有权
公开(公告)号：US10614560B2
公开(公告)日：2020-04-07
申请号：US15838591
申请日：2017-12-12
申请人： Fujitsu Limited
发明人： Wei Liu , Wei Fan , Jun Sun
IPC分类号： G06T5/50 , G06T7/11 , G06T3/40 , G06T1/00 , G06T5/00 , G06T7/33
CPC分类号： G06T5/50 , G06T1/0007 , G06T3/4007 , G06T3/4038 , G06T5/002 , G06T7/11 , G06T7/33 , G06T2200/32 , G06T2207/20221 , G06T2210/21
摘要： An image processing apparatus and an image processing method where the apparatus includes: a seam line acquisitor configured to acquire a seam line of a first image in images to be stitched; a segmenter configured to segment, in an extension direction of the seam line of the first image, a center region including the seam line in the first image into blocks, where an intersection point of a block boundary with the seam line of the first image is located in a strong gradient region; a matcher configured to perform keypoint matching based on the intersection point to find a coordinate point corresponding to the intersection point in a position region corresponding to the seam line in a second image in the images to be stitched; and a fuser configured to fuse the first image with the second image based on the intersection point and the coordinate point.

7. 发明申请

US20180260943A1 APPARATUS AND METHOD FOR IMAGE PROCESSING 审中-公开
公开(公告)号：US20180260943A1
公开(公告)日：2018-09-13
申请号：US15838591
申请日：2017-12-12
申请人： Fujitsu Limited
发明人： Wei LIU , Wei Fan , Jun Sun
IPC分类号： G06T5/50 , G06T7/11 , G06T3/40 , G06T5/00 , G06T1/00
CPC分类号： G06T5/50 , G06T1/0007 , G06T3/4007 , G06T3/4038 , G06T5/002 , G06T7/11 , G06T7/33 , G06T2200/32 , G06T2207/20221 , G06T2210/21
摘要： An image processing apparatus and an image processing method where the apparatus includes: a seam line acquisitor configured to acquire a seam line of a first image in images to be stitched; a segmenter configured to segment, in an extension direction of the seam line of the first image, a center region including the seam line in the first image into blocks, where an intersection point of a block boundary with the seam line of the first image is located in a strong gradient region; a matcher configured to perform keypoint matching based on the intersection point to find a coordinate point corresponding to the intersection point in a position region corresponding to the seam line in a second image in the images to be stitched; and a fuser configured to fuse the first image with the second image based on the intersection point and the coordinate point.

8. 发明授权

US09697444B2 Convolutional-neural-network-based classifier and classifying method and training methods for the same 有权
公开(公告)号：US09697444B2
公开(公告)日：2017-07-04
申请号：US14448049
申请日：2014-07-31
申请人： FUJITSU LIMITED
发明人： Chunpeng Wu , Wei Fan , Yuan He , Jun Sun
IPC分类号： G06K9/46 , G06K9/62 , G06N3/04
CPC分类号： G06K9/6272 , G06K9/4628 , G06K9/6267 , G06N3/0454 , G06N3/0481
摘要： The present invention relates to a convolutional-neural-network-based classifier, a classifying method by using a convolutional-neural-network-based classifier and a method for training the convolutional-neural-network-based classifier. The convolutional-neural-network-based classifier comprises: a plurality of feature map layers, at least one feature map in at least one of the plurality of feature map layers being divided into a plurality of regions; and a plurality of convolutional templates corresponding to the plurality of regions respectively, each of the convolutional templates being used for obtaining a response value of a neuron in the corresponding region.

9. 发明授权

US09524559B2 Image processing device and method 有权
标题翻译：图像处理装置及方法
公开(公告)号：US09524559B2
公开(公告)日：2016-12-20
申请号：US14946147
申请日：2015-11-19
申请人： FUJITSU LIMITED
发明人： Shufu Xie , Wei Liu , Wei Fan , Jun Sun
IPC分类号： G06K9/34 , G06T7/00 , G06T11/60 , G06T5/00
CPC分类号： G06T7/0081 , G06K9/4647 , G06K2209/01 , G06T5/005 , G06T7/11 , G06T7/194 , G06T11/60 , G06T2207/10024 , G06T2207/30176
摘要： An image processing device includes: an image acquiring unit for acquiring an image containing pixels; a picture searching unit for searching for a picture region in the image; a background acquiring unit for performing a background connected region acquiring operation on a non-picture region of the image in which the picture region is not included and the picture region respectively, to obtain a non-picture background connected region and a picture background connected region respectively; a foreground acquiring unit for removing the non-picture background connected region and the picture background connected region from the non-picture region and the picture region respectively, to obtain a non-picture foreground region of the non-picture region and a picture foreground region of the picture region respectively; and a merging unit for merging the non-picture foreground region of the non-picture region and the picture foreground region of the picture region to obtain an ultimate image foreground region.
摘要翻译：一种图像处理装置，包括：图像获取单元，用于获取包含像素的图像; 图像搜索单元，用于搜索图像中的图像区域; 背景获取单元，用于对不包括图像区域的图像的非图像区域和图像区域分别执行背景连接区域获取操作，以获得非图像背景连接区域和图像背景连接区域分别; 从非图像区域和图像区域分别去除非图像背景连接区域和图像背景连接区域的前景获取单元，以获得非图像区域的非图像前景区域和图像前景区域的图片区域; 以及用于合并非图像区域的非图像前景区域和图像区域的图像前景区域的合并单元，以获得最终图像前景区域。

10. 发明授权

US11049017B2 Device and method for determining convolutional neural network model 有权
公开(公告)号：US11049017B2
公开(公告)日：2021-06-29
申请号：US15723457
申请日：2017-10-03
申请人： FUJITSU LIMITED
发明人： Li Sun , Song Wang , Wei Fan , Jun Sun
IPC分类号： G06N3/08 , G06N3/10 , G06N3/04
摘要： Provided are device and method for determining a Convolutional Neural Network (CNN) model. The device for determining the CNN model includes: a first determination unit configured to determine complexity of a database including multiple samples; a second determination unit configured to determine a classification capability of a CNN model applicable to the database based on the complexity of the database; a third determination unit configured to acquire classification capability of each candidate CNN model; and a matching unit configured to determine the CNN model applicable to the database based on the classification capability of each candidate CNN model. With the device and method for determining the CNN module, a design process of CNN model can be simplified.

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