会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 2. 发明授权
    • Method for regenerating the background of digital images of a video stream
    • 用于再现视频流的数字图像的背景的方法
    • US08948516B2
    • 2015-02-03
    • US13518369
    • 2010-12-17
    • Estelle Lesellier
    • Estelle Lesellier
    • G06K9/00G06T7/20H04N19/23
    • G06T7/2053G06T7/194G06T7/223G06T7/254G06T2207/10016G06T2207/20021G06T2207/30232H04N19/23
    • The invention relates to a method for regenerating the background of digital images of a video stream comprising steps consisting in: —setting an initial background image, —cutting the unit images of the video stream into blocks b (i, j, t) and of the background image into corresponding blocks Bo (i, j, t). The method is essentially characterized in that it further includes steps consisting in: —selecting one block Bo of the background image and/or b of the frame image, and •calculating the space correlation thereof, with: •at least one block Bo of the background image at a time (t), and/or at another time (t−a), and/or •at least one block b of the frame image at a time (t), and/or at another time (t−a), and/or—updating the background image according to the calculation of the space correlation.
    • 本发明涉及一种用于再现视频流的数字图像的背景的方法,包括以下步骤: - 设置初始背景图像,将视频流的单位图像切割成块b(i,j,t)和 将背景图像转换成相应的块Bo(i,j,t)。 该方法的特征在于,其还包括以下步骤: - 选择背景图像的一个块Bo和帧图像的b,以及•计算其空间相关性,其中:•至少一个块Bo 和/或在时间(t)和/或在另一时间(t-a)的时间(t-a),和/或在帧时间(t)和/ a),和/或根据空间相关性的计算更新背景图像。
    • 3. 发明申请
    • Method for Regenerating the Background of Digital Images of a Video Stream
    • 再生视频流数字图像背景的方法
    • US20120275706A1
    • 2012-11-01
    • US13518369
    • 2010-12-17
    • Estelle Lesellier
    • Estelle Lesellier
    • G06K9/46
    • G06T7/2053G06T7/194G06T7/223G06T7/254G06T2207/10016G06T2207/20021G06T2207/30232H04N19/23
    • The invention relates to a method for regenerating the background of digital images of a video stream comprising steps consisting in: -setting an initial background image, -cutting the unit images of the video stream into blocks b (i, j, t) and of the background image into corresponding blocks Bo (i, j, t). The method is essentially characterized in that it further includes steps consisting in: -selecting one block Bo of the background image and/or b of the frame image, and •calculating the space correlation thereof, with: •at least one block Bo of the background image at a time (t), and/or at another time (t−a), and/or .at least one block b of the frame image at a time (t), and/or at another time (t−a), and/or -updating the background image according to the calculation of the space correlation.
    • 本发明涉及一种用于再现视频流的数字图像的背景的方法,包括以下步骤: - 设置初始背景图像,将视频流的单位图像切割成块b(i,j,t)和 将背景图像转换成相应的块Bo(i,j,t)。 该方法的特征在于,其还包括以下步骤: - 选择背景图像的一个块Bo和帧图像的b,以及•计算其空间相关性,其中:•至少一个块Bo 和/或在时间(t)和/或另一时间(t-a)的时间(t-a),和/或至少一个帧图像在时间(t) a),和/或根据空间相关性的计算更新背景图像。
    • 4. 发明申请
    • Dynamic detection of blocking artefacts
    • 阻塞伪影的动态检测
    • US20050271287A1
    • 2005-12-08
    • US10522467
    • 2003-07-09
    • Estelle LesellierCarolina Miro SorollaVincent Ruol
    • Estelle LesellierCarolina Miro SorollaVincent Ruol
    • H03M7/30H04N7/26H04N7/30G06K9/36G06K9/46
    • H04N19/527H04N19/86H04N19/865
    • The present invention relates to a method of processing a sequence of digital images, intended to detect, in a dynamical manner, a grid comprising blocking artefacts. Said method comprises the steps of, a) detecting (100) a current spatial grid (SG(t)) within a portion of the image constituted by a current field (FLD(t)), b) determining (200) a current reference grid (RG(t)) from a current spatial grid (SG(t)) and a preceding reference grid (RG(t-1)) supplied by a memory MEM (150), the current reference grid (RG(t)) being subsequently stored temporarily in the memory MEM (150), and c) correcting COR (300) blocking artefacts which are present in the current field (FLD(t)) from the preceding reference grid (RG(t-1)) so as to supply a processed field (PPP(t)).
    • 本发明涉及一种处理数字图像序列的方法,该方法旨在以动态方式检测包含阻挡伪影的网格。 所述方法包括以下步骤:a)在由当前场(FLD(t))构成的图像的一部分内检测(100)当前空间网格(SG(t)),b)确定(200)当前参考 当前参考网格(RG(t))由存储器MEM(150)提供的来自当前空间网格(SG(t))和前一个参考网格(RG(t-1))的网格(RG(t) 随后临时存储在存储器MEM(150)中,以及c)从先前的参考网格(RG(t-1))校正COR(300)阻塞存在于当前场(FLD(t))中的伪像的COR(300),以便 以提供处理的字段(PPP(t))。
    • 5. 发明申请
    • Digital Image Processing Method and Device for Lightening Said Image
    • 用于减轻所述图像的数字图像处理方法和装置
    • US20120148155A1
    • 2012-06-14
    • US13379867
    • 2010-06-18
    • Estelle Lesellier
    • Estelle Lesellier
    • G06K9/00
    • G06T5/40G06T5/009G06T2207/10016G06T2207/20012H04N1/4074
    • A method of processing a digital image, said image comprising a plurality of pixels, pixels, the method comprising a computation step (S1) wherein a histogram of the distribution of the number of pixels of the image as a function of their luminance is computed, a step (S2) for lightening the image based on said histogram comprising a subdivision (S20) of the pixels of the image into a first set of pixels having luminance values between a low threshold and a high threshold and into a second set of pixels having luminance values greater than said high threshold, a first luminance processing operation (S21) on the pixels of the first set of pixels and a second luminance processing operation (S22) on the pixels of the second set of pixels, the two luminance processing operations (S21, S22) being different, the first processing operation (S21) comprising an increase in the luminance of the pixels of the image.
    • 一种处理数字图像的方法,所述图像包括多个像素,所述图像包括计算步骤(S1),其中计算作为其亮度的函数的图像的像素数目的分布的直方图, 基于包括图像的像素的细分(S20)的图像的步骤(S2),其包括具有低阈值和高阈值之间的亮度值的第一组像素,并且具有第二组像素,具有 亮度值大于所述高阈值,第一组像素的像素上的第一亮度处理操作(S21)和第二组像素的像素的第二亮度处理操作(S22),两个亮度处理操作 S21,S22)不同,第一处理操作(S21)包括图像的像素的亮度的增加。
    • 6. 发明授权
    • Method and device for post-processing digital images
    • 用于后处理数字图像的方法和装置
    • US07373010B2
    • 2008-05-13
    • US10474112
    • 2002-04-08
    • Estelle LesellierJorge E. CaviedesCarolina Miro Sorolla
    • Estelle LesellierJorge E. CaviedesCarolina Miro Sorolla
    • G06K9/40
    • G06T5/10G06T5/002G06T2207/20052
    • A method of post-processing digital images frequency transforming a set of the pixels to a set of transformed coefficients; extracting original low-frequency coefficients and original high-frequency coefficients from the set of transformed coefficients; correcting the original low-frequency coefficients and the original high-frequency coefficients for delivering a set of corrected transformed coefficients including corrected low-frequency coefficients and corrected high-frequency coefficients; and combining the set of corrected transformed coefficients and the original high-frequency coefficients for delivering a set of combined transformed coefficients. The set of combined transformed coefficients are inverse frequency transformed for delivering a processed set of pixels that are ready for display on a screen.
    • 一种后处理数字图像的方法,其将一组像素频率变换成一组经变换的系数; 从变换系数集中提取原始低频系数和原始高频系数; 校正原始低频系数和原始高频系数,用于传送包括校正的低频系数和校正的高频系数的一组校正后的变换系数; 以及组合所述经校正的经变换的系数组和所述原始高频系数,用于递送一组组合的变换系数。 组合变换系数的集合被逆频变换,用于递送准备好在屏幕上显示的经处理的像素组。
    • 9. 发明申请
    • Method of detecting blocking artefacts
    • 检测阻塞伪影的方法
    • US20060078155A1
    • 2006-04-13
    • US10518253
    • 2003-06-16
    • Estelle Lesellier
    • Estelle Lesellier
    • H04N7/16G06K9/00
    • H04N19/86H04N5/21H04N19/865
    • The invention relates to a method of detecting blocking artefacts within a sequence of digital images. The method comprises a step of high-pass filtering (110) a portion of a digital image, intended to supply at least one card of discontinuity pixels. It also comprises a step of detecting (120) blocking artefacts from the at least one card of discontinuity pixels. Finally, it comprises a step of searching (130), HPF within the portion of the digital image, a set of grid rows, a grid row having a density of discontinuity pixels which is substantially higher than that of its neighboring rows. Such a method of detecting blocking artefacts is particularly efficient and allows, for example, a better correction of blocking artefacts in the grid rows.
    • 本发明涉及一种检测数字图像序列内的阻塞伪影的方法。 该方法包括高通滤波(110)数字图像的一部分,旨在提供至少一个不连续像素的卡的步骤。 它还包括检测(120)阻挡来自不连续像素的至少一个卡的伪像的步骤。 最后,它包括在数字图像的部分内搜索(130)HPF,一组网格行,具有基本上高于其相邻行的不连续像素密度的网格行的步骤。 这种检测阻塞伪影的方法是特别有效的,并且允许例如更好地校正网格行中的阻塞伪影。
    • 10. 发明授权
    • Estimation of picture motion blurriness
    • 图像运动模糊度的估计
    • US09288393B2
    • 2016-03-15
    • US14113447
    • 2012-04-30
    • Antoine ChoulyEstelle Lesellier
    • Antoine ChoulyEstelle Lesellier
    • H04N5/232G06T5/00G06T7/20
    • H04N5/23254G06T5/003G06T7/20G06T2207/10016G06T2207/20201G06T2207/30241
    • The invention relates to a method for estimating a blurriness value based on a series of at least two frames acquired within a period of time greater than or equal to a given exposure time, the frames being grouped in at least one couple of frames, characterized in that it comprises determining, for each couple of frames of the series, first and second values corresponding to a displacement according to a first direction and a second direction, between the first frame and the second frame of a couple of frames, determining a first blur measure for one direction among the first and second directions by calculating a metric based on the displacement values for each couple of frames of at least a subset of the series, determining a second blur measure for the other direction among the first and second directions by calculating a metric applied to the displacement values for each couple of frames of the series of frames and determining a blurriness value based on the first and second blur measures.
    • 本发明涉及一种用于基于在大于或等于给定曝光时间的时间段内获取的至少两帧的一系列来估计模糊度的方法,所述帧被分组为至少一对帧,其特征在于 其包括为所述系列的每一对帧确定对应于两帧中的第一帧和第二帧之间的根据第一方向和第二方向的位移的第一和第二值,确定第一模糊 通过基于该系列的至少一个子集的每一对的帧的位移值计算度量来测量第一和第二方向之间的一个方向,通过计算第一和第二方向之间的另一个方向来确定第二模糊度量 该度量应用于该系列帧的每对帧的位移值,并且基于第一和第二模糊确定模糊度值 措施。