公开(公告)号：KR1020140030735A

公开(公告)日：2014-03-12

申请号：KR1020120097259

申请日：2012-09-03

Applicant: 삼성전자주식회사

Inventor： 안원석 ,  한승훈

IPC: H04N13/00

CPC classification number: G06T15/10 ,  H04N13/261

Abstract: A device and a method for a display are disclosed. The display device according to the present invention comprises: a receiving unit for receiving an image; a grouping unit which analyzes the received image and groups a plurality of frames that constitute the received image based on the analysis result; a depth assigning unit which determines at least two key frames among the plurality of frames constituting at least one group, and assigns depth for each object included in the determined key frame; and an image generating unit which generates a 3D image that corresponds to the frames except the key frame based on the depth value assigned to the key frame. Thus, the display device can assign the depth value with high quality to the object included in the frames which constitute the received image. [Reference numerals] (110) Receiving unit; (120) Grouping unit; (130) Depth assigning unit; (140) Image generating unit; (150) Image analyzing unit; (160) Scaling unit; (170) Control unit; (180) Storage unit

Abstract translation: 公开了一种用于显示器的装置和方法。 根据本发明的显示装置包括：接收单元，用于接收图像; 分组单元，其基于分析结果分析接收的图像并对构成所接收的图像的多个帧进行分组; 深度分配单元，其确定构成至少一个组的所述多个帧中的至少两个关键帧，并且分配所确定的关键帧中包括的每个对象的深度; 以及图像生成单元，其基于分配给所述关键帧的深度值，生成与所述关键帧以外的帧对应的3D图像。 因此，显示装置可以将高质量的深度值分配给构成接收图像的帧中所包括的对象。 （附图标记）（110）接收单元; （120）分组单位; （130）深度分配单位; （140）图像生成单元; （150）图像分析单元; （160）缩放单元; （170）控制单元; （180）存储单元

公开(公告)号：KR1020130027727A

公开(公告)日：2013-03-18

申请号：KR1020110091122

申请日：2011-09-08

Applicant: 삼성전자주식회사

Inventor： 권오재 ,  한승훈

IPC: H04N13/00

CPC classification number: G06T7/55 ,  G06T2207/10016 ,  G06T2207/10028 ,  H04N13/264

Abstract: PURPOSE: A depth information generation device, a depth information generation method, and a storage medium thereof are provided to economically generate depth information by reducing depth information generation time. CONSTITUTION: A first receiving unit(110) receives a plurality of frames which constructs a two-dimensional image from a source providing device. A first object tracking unit(121) of a first depth information generation unit(120) tracks a second object selected from a second frame based on a first object. A first object location change size calculation unit(123) calculates the size of position change by comparing the position of the second object with the position of the first object. A first depth value calculation unit(125) generates the second depth information of the second object from the first depth information of the first object based on a comparison result. [Reference numerals] (100) Depth information generation device; (110) First receiving unit; (120) First depth information generation unit; (121) First object tracking unit; (123) First object location change size calculation unit; (125) First depth value calculation unit

Abstract translation: 目的：提供深度信息生成装置，深度信息生成方法及其存储介质，以通过减少深度信息生成时间来经济地生成深度信息。 构成：第一接收单元（110）从源提供装置接收构成二维图像的多个帧。 第一深度信息生成单元（120）的第一对象跟踪单元（121）基于第一对象跟踪从第二帧选择的第二对象。 第一对象位置变更尺寸计算部（123）通过将第二对象的位置与第一对象的位置进行比较来计算位置变更的大小。 第一深度值计算单元（125）基于比较结果从第一对象的第一深度信息生成第二对象的第二深度信息。 （附图标记）（100）深度信息生成装置; （110）第一接收单元; （120）第一深度信息生成单元; （121）第一物体跟踪单元; （123）第一对象位置变化大小计算单元; （125）第一深度值计算单元

公开(公告)号：KR1020130002049A

公开(公告)日：2013-01-07

申请号：KR1020110063045

申请日：2011-06-28

Applicant: 삼성전자주식회사

Inventor： 한승훈 ,  맹완주 ,  이상윤 ,  이홍석

IPC: G02F1/33 ,  G02F1/11

CPC classification number: G02F1/33 ,  G01N29/2418 ,  G01N29/34 ,  G02F1/0072 ,  G02F2202/32 ,  G02F2203/10 ,  G03H1/2294 ,  G03H2001/0224 ,  G03H2225/21 ,  G01N29/348 ,  G02F1/113

Abstract: PURPOSE: An acoustic-optic device with a multilayered nanostructure, an optical scanner using same and a display device using same are provided to increase the diffraction angle of output light by using acoustic-optic medium with a multilayered nanostructure. CONSTITUTION: An acoustic-optic device(100) includes acoustic-optic medium(10) with a multilayered nanostructure and a sound wave generator(16). The sound wave generator supplies sound waves to the acoustic-optic medium. The acoustic-optic medium includes at least one layer(12,13,15). In a second layer, two or more materials(13a,13b) having different dielectric constants are alternately arranged. A first layer is formed with a first material(12a) on the lower surface of the second layer. A third layer is formed with a fourth material(15a) on the upper surface of the second layer.

Abstract translation: 目的：提供具有多层纳米结构的声光装置，使用其的光学扫描器和使用其的显示装置，以通过使用具有多层纳米结构的声光介质来增加输出光的衍射角。 构成：声光装置（100）包括具有多层纳米结构的声光介质（10）和声波发生器（16）。 声波发生器将声波提供给声光介质。 声光介质包括至少一层（12,13,15）。 在第二层中，交替布置具有不同介电常数的两种或更多种材料（13a，13b）。 第一层在第二层的下表面上形成有第一材料（12a）。 第三层在第二层的上表面上形成有第四材料（15a）。

公开(公告)号：KR1020120055102A

公开(公告)日：2012-05-31

申请号：KR1020100116622

申请日：2010-11-23

Applicant: 삼성전자주식회사

Inventor： 김신행 ,  한승훈 ,  손영욱

IPC: H04N13/00

CPC classification number: G06T5/50 ,  G06T5/006 ,  H04N13/139 ,  H04N2013/0081 ,  H04N2013/0085

Abstract: PURPOSE: An image processing apparatus and an image processing method are provided to reduce unnatural motion deviation between image frames. CONSTITUTION: An image receiving unit(100) receives a plurality of image frames with a first image and a second image corresponding to both eyes of a user. An image processing unit(200) detects a pixel of a second image corresponding to one or more pixels of the first image. The image processing unit compensates the first image based on pixel information of the detected second image.

Abstract translation: 目的：提供一种图像处理装置和图像处理方法，以减少图像帧之间的不自然的运动偏差。 构成：图像接收单元（100）接收具有对应于用户双眼的第一图像和第二图像的多个图像帧。 图像处理单元（200）检测与第一图像的一个或多个像素对应的第二图像的像素。 图像处理单元基于检测到的第二图像的像素信息补偿第一图像。

公开(公告)号：KR1020100069507A

公开(公告)日：2010-06-24

申请号：KR1020080128203

申请日：2008-12-16

Applicant: 삼성전자주식회사

Inventor： 이미연 ,  한승훈

IPC: H04N5/21

CPC classification number: G06T5/002 ,  G06T2207/10016 ,  G06T2207/20182 ,  H04N5/21

Abstract: PURPOSE: An image noise elimination method and an apparatus are provided so that the uncorrelated processing is proceed about the flicker noise component of the video signal. The video noise is eliminated. CONSTITUTION: A signal separator part(110) divides the video signal into the signal factor and noise component. The uncorrelated signal processing unit(120) changes the noise component into the temporally and spatially incoherent noise component. The adder(130) adds the signal factor and incoherent noise component as described above. The filter section is proceed the motion-compensated low pass filtering about the video signal outputted in above mentioned addition part.

Abstract translation: 目的：提供一种图像噪声消除方法和装置，使得关于视频信号的闪烁噪声分量进行不相关的处理。 视频噪声消除。 构成：信号分离器部分（110）将视频信号分成信号因子和噪声分量。 不相关信​​号处理单元（120）将噪声分量改变为时间和空间上的非相干噪声分量。 加法器（130）如上所述相加信号因子和非相干噪声分量。 滤波器部分进行关于上述加法部分输出的视频信号的运动补偿低通滤波。

公开(公告)号：KR1020080070216A

公开(公告)日：2008-07-30

申请号：KR1020070007945

申请日：2007-01-25

Applicant: 삼성전자주식회사

Inventor： 김대현 ,  한승훈 ,  양승준 ,  이안나

IPC: H04N19/51 ,  H04N19/137 ,  H04N19/176

CPC classification number: H04N19/52 ,  H04N19/61

Abstract: A method and an apparatus for estimating a motion vector using a motion vector of an adjacent block are provided to reduce a block artifact effectively by smoothly connecting blocks. A block dividing unit(210) divides a current block into a plurality of sub-blocks. A motion vector estimating unit(220) estimates motion vectors of the divided sub-blocks by using motion a plurality of vectors of blocks adjacent to the divided sub-blocks and the weight values set for the pixels of the adjacent blocks. A repeating unit(230) detects the number of pixels included in the current block, and repeatedly performs the operation of the block dividing unit and the motion vector estimating unit based on the number of detected pixels until a motion vector of the pixels of the current block are estimated.

Abstract translation: 提供了使用相邻块的运动矢量来估计运动矢量的方法和装置，以通过平滑地连接块来有效地减少块伪影。 块分割单元（210）将当前块划分成多个子块。 运动矢量估计单元（220）通过使用运动与分割子块相邻的多个块的矢量和为相邻块的像素设置的权重值来估计划分的子块的运动矢量。 重复单元（230）检测包括在当前块中的像素的数量，并且基于检测到的像素的数量重复地执行块分割单元和运动矢量估计单元的操作，直到当前像素的像素的运动矢量 块估计。

公开(公告)号：KR1020080067172A

公开(公告)日：2008-07-18

申请号：KR1020070004402

申请日：2007-01-15

Applicant: 삼성전자주식회사

Inventor： 김대현 ,  한승훈 ,  양승준 ,  이미연

IPC: H04N19/43

CPC classification number: G06T3/4007 ,  G06T5/002 ,  G06T5/20 ,  G06T2207/10016 ,  G06T2207/20192

Abstract: An MC method and apparatus are provided to recognize the characteristic of artifacts which appears in MC, configure an MC filter bank using filters corresponding to the characteristic, mix interpolation images outputted from each filter according to the characteristic of an image, thereby suppressing the generation of artifacts and holding the resolution of the image. An MC(Motion Compensation) filter bank(20) performs MC using plural MC filters from an input image(22) and a predicted motion vector(24). An MC weight generator(40) calculates weight(w1-wN) to mix each MC result from the input image, the predicted motion vector, and other image input information(26). A mixer(30) calculates a final MC image(Fout) from the calculated weight and MC images(f1-fN).

Abstract translation: 提供了一种MC方法和装置来识别出现在MC中的人为因素的特征，使用与特征相对应的滤波器来配置MC滤波器组，根据图像的特性混合从每个滤波器输​​出的插值图像，从而抑制 文物和保持图像的分辨率。 MC（运动补偿）滤波器组（20）使用来自输入图像（22）和预测运动矢量（24）的多个MC滤波器来执行MC。 MC权重发生器（40）计算从输入图像，预测运动矢量和其他图像输入信息（26）中混合每个MC结果的权重（w1-wN）。 混合器（30）根据计算的权重和MC图像（f1-fN）计算最终MC图像（Fout）。

公开(公告)号：KR100670003B1

公开(公告)日：2007-01-19

申请号：KR1020040113795

申请日：2004-12-28

Applicant: 삼성전자주식회사

Inventor： 한승훈

IPC: G06T7/00 ,  H04N5/21

CPC classification number: G06K9/4647 ,  G06T7/11 ,  G06T2207/20008

Abstract: 적응형 문턱치를 이용한 영상의 평탄 영역 검출장치 및 그 방법이 개시된다. 본 발명에 따른 적응형 문턱치를 이용한 영상의 평탄 영역 검출장치는 입력영상 전체 영역의 표준편차를 산출하는 전역 표준편차 산출부, 입력영상을 소정 개수의 영역으로 분할한 후, 분할된 각 영역별로 국부영역 표준편차를 산출하는 국부영역 표준편차 산출부 및 입력영상의 복잡도에 기초하여 산출된 적응형 문턱치를 이용하여 영상의 평탄 영역과 특징 영역을 분리하는 평탄영역 결정부를 포함한다. 본 발명에 의하면 영상의 복잡도에 따라 적응적으로 산출되는 문턱치를 이용하여 영상의 평탄 영역과 특징 영역을 효과적으로 분리하여 영상의 평탄 영역을 정확히 검출할 수 있는 장점이 있다. 따라서, 평탄 영역을 이용하는 영상처리분야에 다양하게 응용될 수 있다.
평탄 영역, 적응형 문턱치, 전역 표준편차, 복잡도 상수

Abstract translation: 使用自适应阈值的图像的均匀区域检测器及其方法。 均匀区域检测器包括全局区域标准偏差计算部分，用于计算输入图像的整个区域的全局区域标准偏差，局部区域标准偏差计算部分，以将输入图像划分为预定数量的局部区域，并计算 每个局部区域的局部区域标准偏差，以及均匀区域确定部分，用于使用基于输入图像的熵计算的自适应阈值将均匀区域与输入图像中的特征区域分离。 因此，通过使用基于输入图像的熵自适应计算的阈值，可以通过有效地将均匀区域与输入图像的特征区域分离来精确地检测均匀区域，使得均匀区域可以应用于 图像处理。

公开(公告)号：KR1020060075204A

公开(公告)日：2006-07-04

申请号：KR1020040113795

申请日：2004-12-28

Applicant: 삼성전자주식회사

Inventor： 한승훈

IPC: G06T7/00 ,  H04N5/21

CPC classification number: G06K9/4647 ,  G06T7/11 ,  G06T2207/20008

Abstract: 적응형 문턱치를 이용한 영상의 평탄 영역 검출장치 및 그 방법이 개시된다. 본 발명에 따른 적응형 문턱치를 이용한 영상의 평탄 영역 검출장치는 입력영상 전체 영역의 표준편차를 산출하는 전역 표준편차 산출부, 입력영상을 소정 개수의 영역으로 분할한 후, 분할된 각 영역별로 국부영역 표준편차를 산출하는 국부영역 표준편차 산출부 및 입력영상의 복잡도에 기초하여 산출된 적응형 문턱치를 이용하여 영상의 평탄 영역과 특징 영역을 분리하는 평탄영역 결정부를 포함한다. 본 발명에 의하면 영상의 복잡도에 따라 적응적으로 산출되는 문턱치를 이용하여 영상의 평탄 영역과 특징 영역을 효과적으로 분리하여 영상의 평탄 영역을 정확히 검출할 수 있는 장점이 있다. 따라서, 평탄 영역을 이용하는 영상처리분야에 다양하게 응용될 수 있다.
평탄 영역, 적응형 문턱치, 전역 표준편차, 복잡도 상수

Abstract translation: 使用自适应阈值的图像的均匀区域检测器及其方法。 均匀区域检测器包括全局区域标准偏差计算部分，用于计算输入图像的整个区域的全局区域标准偏差，局部区域标准偏差计算部分，以将输入图像划分为预定数量的局部区域，并计算 每个局部区域的局部区域标准偏差，以及均匀区域确定部分，用于使用基于输入图像的熵计算的自适应阈值将均匀区域与输入图像中的特征区域分离。 因此，通过使用基于输入图像的熵自适应计算的阈值，可以通过有效地将均匀区域与输入图像的特征区域分离来精确地检测均匀区域，使得均匀区域可以应用于 图像处理。

公开(公告)号：KR1020020003608A

公开(公告)日：2002-01-15

申请号：KR1020000034259

申请日：2000-06-21

Applicant: 삼성전자주식회사

Inventor： 이동훈 ,  한승훈 ,  신헌

IPC: H01L21/8242

Abstract: PURPOSE: A method of fabricating a merged DRAM with logic for improving a refresh characteristic is provided to improve a refresh characteristic of a DRAM without degrading a transistor characteristic of a logic circuit. CONSTITUTION: An interlayer dielectric(110) is formed on a semiconductor substrate(100). A photo-resist layer is formed on the interlayer dielectric(110). A photo-resist pattern having an opening is formed on the interlayer dielectric(110) by using a photo-lithography method. A contact hole(115) is formed by etching an exposed portion of the interlayer dielectric(110). A drain region or a source region(105) is exposed by the contact hole(115). The photo-resist pattern is removed by performing an ashing process and a strip process. A nitride layer(130') is formed on a surface of the whole structure. A nitride spacer(130') is formed on a sidewall of the contact hole(115) by removing a part of the nitride layer(130'). A refresh characteristic of a DRAM is improved by performing the ashing process.

Abstract translation: 目的：提供一种制造具有用于提高刷新特性的逻辑的合并DRAM的方法，以改善DRAM的刷新特性，而不降低逻辑电路的晶体管特性。 构成：在半导体衬底（100）上形成层间电介质（110）。 在层间电介质（110）上形成光致抗蚀剂层。 通过使用光刻法在层间电介质（110）上形成具有开口的光刻胶图案。 通过蚀刻层间电介质（110）的暴露部分形成接触孔（115）。 漏极区域（105）由接触孔（115）露出。 通过进行灰化处理和剥离处理来去除光刻胶图案。 在整个结构的表面上形成氮化物层（130'）。 通过去除氮化物层（130'）的一部分，在接触孔（115）的侧壁上形成氮化物间隔物（130'）。 通过执行灰化处理来提高DRAM的刷新特性。