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
    • 4. 发明公开
    • 영상 디인터레이싱 방법 및 그 장치
    • 视频互连方法及其设备
    • KR1020100106695A
    • 2010-10-04
    • KR1020090024781
    • 2009-03-24
    • 한양대학교 산학협력단중앙대학교 산학협력단
    • 강근화김주혁정제창
    • H04N7/01G06T5/00
    • H04N7/0142H04N7/0135H04N7/0145
    • PURPOSE: A video Interlacing method and an apparatus thereof are provided to obtain more excellent performance by adaptively adopting an interpolation method depending on a circumference after judging an area to be interpolated through the detection of a texture. CONSTITUTION: If pixels are inputted, a device judges whether or not an edge or texture exists in a current pixel region(S10,S11). If the current region is judged as a texture region, the device calculates a current pixel value through a pixel value inclination filter(S12,S13). If the current region is judged as a flat region, the device calculates the current pixel value through a median filter(S14,S15). The device interpolates the current pixels based on a calculated pixel value(S16).
    • 目的:提供一种视频交错方法及其装置,以通过在通过纹理检测判断要内插的区域之后,根据周围自适应地采用插值方法来获得更优异的性能。 构成:如果输入像素,则设备判断当前像素区域中是否存在边缘或纹理(S10,S11)。 如果当前区域被判断为纹理区域,则该装置通过像素值倾斜滤波器计算当前像素值(S12,S13)。 如果当前区域被判断为平坦区域,则装置通过中值滤波器计算当前像素值(S14,S15)。 该装置基于计算出的像素值内插当前像素(S16)。
    • 9. 发明公开
    • 계수의 생성 장치 및 생성 방법, 클래스 구성의 생성 장치및 생성 방법, 정보 신호 처리 장치, 및 각 방법을실행하기 위한 프로그램
    • 用于生产参数的装置和方法,装置和用于产生类结构的方法,信息信号处理装置和用于实施有效方法的程序,以有效地产生每个类别的参数数据
    • KR1020040103428A
    • 2004-12-08
    • KR1020040038395
    • 2004-05-28
    • 소니 주식회사
    • 곤도데쯔지로미야이다께시기꾸찌다이스께지까오까시즈오나까니시다까시나까무라요시아끼하가쯔기히꼬
    • H04N7/01
    • H04N7/0135H04N7/0125H04N7/0145
    • PURPOSE: An apparatus and a method for producing a parameter, an apparatus and a method for producing a class structure, an information signal processing apparatus and programs for implementing the respective methods are provided to effectively produce parameter data of each class in an arbitrary class structure by only one time learning. CONSTITUTION: A class structure producing member(401) selects one from (n-i) characteristics, where i(i = an integer) denotes previously selected characters and produces (n-i) class structures including (i+1) characteristics, respectively. A class structure selection member(402) selects an optimum class structure from the (n-i) class structures by using an arbitrary evaluation value. A selection finish characteristics number discrimination member(403) discriminates whether the number of the selected characteristics is r. When the number of the selected characteristics is r, an object class structure output member(404) outputs information on the selected characteristics of r to an output terminal(405).
    • 目的:提供用于产生类结构的装置和方法,用于产生类结构的装置和方法,用于实现各种方法的信息信号处理装置和程序,以有效地产生任意类结构中的每个类的参数数据 只有一次学习。 构成:类结构生成构件(401)从(n-i)个特征中选择一个,其中i(i =整数)表示先前选择的字符,并分别产生包括(i + 1)特性的(n-i)类结构。 类结构选择构件(402)通过使用任意评估值从(n-i)类结构中选择最优类结构。 选择完成特征数辨别构件(403)判断所选择的特性的数量是否为r。 当所选特征的数量为r时,对象类结构输出构件(404)将关于所选择的特性的信息输出到输出端子(405)。
    • 10. 发明公开
    • 화상 정보 변환 장치
    • 图像信息变换装置
    • KR1020000035112A
    • 2000-06-26
    • KR1019990047452
    • 1999-10-29
    • 소니 주식회사
    • 곤도데추지로호시노다카야나카야히데오이노우에사토시치카오카시즈오
    • H04N7/01
    • H04N9/641H04N7/0145
    • PURPOSE: An image information transforming apparatus is provided to prevent a drop of a screen quality and to reduce a circuit scale. CONSTITUTION: An image information transforming apparatus comprises two area extracting sections (101,105), a pattern detector(102), a class code determining section(103), a coefficient memory(104) and a prediction calculation section(106). An input NTSC complex signal is applied to the area extracting sections(101,105). The area extracting section(101) extracts a class tab(a pixel for a class categorization existing a predetermined area) from the received complex signal, and transfers the extracted class tab to the pattern detector(102). The pattern detector(102) executes a class categorization to the class tab on the basis of a form of the class tap or pixel data, and transfers the result of the class categorization to the class code determining section(103). The class code determining section(103) determines a class code on the basis of the result from the detector(102), and transfers the determined class code to the coefficient memory(104). The coefficient memory(104) transfers previously stored predictive coefficient to the prediction calculation section(106) on the basis of the class code. The area extracting section(105) extracts a prediction tab from the received complex signal, and the prediction calculation section(106) generates a component color image signal(Y) based on the prediction tab from the area extracting section(105) and the prediction coefficient from the memory(104).
    • 目的:提供一种图像信息变换装置,以防止屏幕质量下降并减小电路规模。 构成:图像信息变换装置包括两个区域提取部分(101,105),模式检测器(102),类别代码确定部分(103),系数存储器(104)和预测计算部分(106)。 输入NTSC复合信号被施加到区域提取部分(101,105)。 区域提取部(101)从接收到的复信号中提取类别标签(用于类别分类存在预定区域的像素),并将提取的类标签传送到图案检测器(102)。 模式检测器(102)根据类别抽头或像素数据的形式对类标签执行类分类,并将类分类的结果传送给类代码确定部(103)。 等级代码确定部分(103)根据检测器(102)的结果确定类代码,并将确定的类代码传送给系数存储器(104)。 系数存储器(104)基于类代码将预先存储的预测系数传送到预测计算部(106)。 所述区域提取部(105)从所接收的复信号中提取预测标签,所述预测计算部(106)基于来自所述区域提取部(105)的预测标签,生成分量彩色图像信号(Y) 系数(104)。