-
公开(公告)号：KR1020140082126A
公开(公告)日：2014-07-02
申请号：KR1020120151616
申请日：2012-12-23
Applicant: 전남대학교산학협력단
IPC: G06T7/00
CPC classification number: G06K9/6218 , G06K9/6212 , G06T7/90
Abstract: The present invention relates to a system and a method for image classification using a color histogram. The method for image classification of the present invention comprises the steps of normalizing an image to a certain size; quantizing RGB values of the normalized image to certain levels; extracting a color histogram based on the quantized RGB values and corresponding frequencies as axes by measuring the frequencies of the quantized RGB values; and determining similarity by comparing the color histogram of the image with a color histogram of another image, thereby classifying multiple images by similar images using dominant colors of the images.
Abstract translation: 本发明涉及一种使用颜色直方图进行图像分类的系统和方法。 本发明的图像分类方法包括将图像归一化为一定尺寸的步骤; 将归一化图像的RGB值量化到一定水平; 通过测量量化的RGB值的频率，基于量化的RGB值和相应的频率作为轴来提取颜色直方图; 以及通过将图像的颜色直方图与另一图像的颜色直方图进行比较来确定相似性，从而通过使用图像的主要颜色的相似图像对多个图像进行分类。
-
公开(公告)号：KR101215569B1
公开(公告)日：2012-12-26
申请号：KR1020110146589
申请日：2011-12-29
Applicant: 전남대학교산학협력단
CPC classification number: G06K9/50 , G06K9/6202 , G06T7/12
Abstract: PURPOSE: A label area division method is provided to enable a user to receive or purchase a product including a label by recognizing a label area from an input image. CONSTITUTION: An importance area creation unit creates a predetermined area of an input image as an importance area(S1000). A vertical edge detection unit detects two vertical edges from the importance area and creates an area which is covered with the vertical edges as a label candidate area(S2000). A label area encoding unit creates encoded label area code by calculating the similarity of a direction and a size between pixels in a boundary of the label candidate area(S4000). A label area separation unit separates an area matched with the label area code from the input image(S5000). [Reference numerals] (S1000) Creating an importance area; (S2000) Creating vertical edges and a label candidate area which is covered by the vertical edges; (S3000) Detecting horizontal edges which cross with vertical edges; (S4000) Creating a label area code connected to vertical edges and horizontal edges; (S5000) Dividing a label area matching with a label code area in an inputted image
Abstract translation: 目的：提供一种标签区域分割方法，以使用户能够通过从输入图像中识别标签区域来接收或购买包括标签的产品。 构成：重要区域创建单元创建输入图像的预定区域作为重要区域（S1000）。 垂直边缘检测单元从重要性区域检测两个垂直边缘，并创建被垂直边缘覆盖的区域作为标签候选区域（S2000）。 标签区域编码单元通过计算标签候选区域的边界中的方向和像素之间的大小的相似度来创建编码标签区域代码（S4000）。 标签区域分离单元将与标签区域代码匹配的区域与输入图像分离（S5000）。 （S1000）创建重要区域; （S2000）创建垂直边缘和由垂直边缘覆盖的标签候选区域; （S3000）检测与垂直边缘交叉的水平边缘; （S4000）创建连接到垂直边缘和水平边缘的标签区域代码; （S5000）在输入图像中划分与标签代码区域匹配的标签区域
-
-
公开(公告)号：KR101811581B1
公开(公告)日：2017-12-26
申请号：KR1020160152159
申请日：2016-11-15
Applicant: 주식회사 셀바스에이아이 , 전남대학교산학협력단
CPC classification number: G06K9/00449 , G06K9/2054 , G06K9/344
Abstract: 본개시에일 실시예에따른문서이미지에서하나이상의셀을포함하는표를인식하는장치가개시된다. 상기장치는상기문서이미지에서표 영역이미지를인식하는이미지인식모듈, 상기인식된표 영역이미지에서라인성분을추출한라인이미지를생성하고, 상기표 영역이미지에서텍스트성분을추출한텍스트이미지를생성하고, 그리고상기텍스트이미지에기초하여텍스트블록이미지를생성하는전처리모듈, 상기라인이미지에기초하여라인기반셀 구조를생성하고, 상기텍스트블록이미지에기초하여텍스트블록기반셀 구조를생성하고, 그리고상기라인기반셀 구조와상기텍스트블록기반셀 구조를비교하여, 결과셀 구조를생성하는셀 구조분석모듈및 상기라인이미지, 상기텍스트이미지및 상기결과셀 구조중 적어도하나에기초하여결과표를생성하는결과표 생성모듈을포함할수 있다.
Abstract translation: 本文公开了根据一个实施例的用于识别包括文档图像中的一个或多个单元格的表格的设备。 该设备包括：图像识别模块，用于识别文档图像中的表格区域图像;线条图像从识别出的表格区域图像中提取线条分量;文本图像，从表格区域图像中提取文本分量; 预处理模块，用于基于文本图像生成文本块图像，基于线图像的基于线的单元结构，基于文本块图像的基于文本块的单元结构， 通过基于小区结构比较的结构和文本块，单元结构，分析模块的单元结构，和的基础上的结果，至少在所述线图像和文本图像，并且生成的所得到的单元结构可以包括用于产生一个结果列表中的结果表生成模块的一个 有。
-
Patent Agency Ranking
公开(公告)号：KR101635738B1
公开(公告)日：2016-07-20
申请号：KR1020140181620
申请日：2014-12-16
Applicant: 전남대학교산학협력단 , 주식회사 셀바스에이아이
IPC: G06T7/00
Abstract: 본발명의일 양상에따른문서구조를분석하기위한방법이개시된다. 상기방법은, 입력된이미지에사전결정된이진화(binarization) 기법을적용함으로써이진화된이미지를생성하는단계; 상기생성된이진화된이미지에연결요소알고리즘(connected component algorithm)을적용함으로써연결요소들을생성하는단계; 상기생성된연결요소들의기하학적성질(geometrical property)에기초하여상기연결요소들각각의크기값을특성화하는단계; 상기연결요소들의특성화된크기값에적어도부분적으로기초하여퍼지에너지매트릭스(FEM:Fuzzy Energy Matrix)를생성하는단계; 및상기생성된퍼지에너지매트릭스를기초로하여상기입력된이미지에서의텍스트부분과비-텍스트부분을구별하여인식하는단계를포함할수 있다.
-
公开(公告)号：KR101451097B1
公开(公告)日：2014-10-15
申请号：KR1020120151616
申请日：2012-12-23
Applicant: 전남대학교산학협력단
IPC: G06T7/00
Abstract: 본 발명은 색상 히스토그램을 이용한 이미지 분류 시스템 및 방법에 관한 것으로, 본 발명의 이미지 분류 방법은, 이미지를 일정 크기로 정규화하고, 정규화한 이미지의 RGB 값을 일정 단계로 양자화한 후, 양자화한 RGB 값의 빈도수를 측정하여, 양자화한 RGB 값 및 대응하는 빈도수를 축으로 하는 색상 히스토그램을 추출하며, 이미지의 색상 히스토그램과 다른 이미지의 색상 히스토그램을 비교하여 유사도를 판단하는 과정을 포함한다. 이를 통해, 이미지의 지배적인 색상을 이용하여 복수의 이미지를 유사한 이미지끼리 분류할 수 있다.
-
公开(公告)号：KR1020140047331A
公开(公告)日：2014-04-22
申请号：KR1020120113457
申请日：2012-10-12
Applicant: 전남대학교산학협력단
IPC: G06T7/00
CPC classification number: G06K9/6223 , G06K9/00718 , G06K9/3233
Abstract: The present invention relates to an automatic object segmentation method for automatically segmenting an object from a background using a block clustering algorithm. The automatic object segmentation method is capable of improving user′s convenience and efficiency by automation to supplement the weakness of providing information on the object by a user in the existing GrabCut implementation. The automatic object segmentation method of the present invention comprises an object interest region estimation step of estimating an interest region of an object for image information including segmentation information of a background and the object. In the object interest region estimation step, cluster dispersion information of an image is analyzed to distinguish the cluster according to the size of a dispersion region, and the image is segmented into the predetermined number of blocks and is determined as an object cluster and a background cluster depending on the object area and the background area within an individual block, and a color mean value of the block.
Abstract translation: 本发明涉及一种使用块聚类算法从背景中自动分割对象的自动对象分割方法。 自动对象分割方法能够通过自动化来提高用户的便利性和效率，以补充用户在现有的GrabCut实现中提供关于对象的信息的弱点。 本发明的自动对象分割方法包括：对象感兴趣区域估计步骤，用于估计包含背景和对象的分割信息的图像信息的对象的兴趣区域。 在对象感兴趣区域估计步骤中，分析图像的群集色散信息以根据色散区域的大小来区分簇，并且将图像分割成预定数量的块并且被确定为对象簇和背景 取决于单个块内的对象区域和背景区域以及块的颜色平均值。
-
公开(公告)号：KR1020130078130A
公开(公告)日：2013-07-10
申请号：KR1020110146905
申请日：2011-12-30
Applicant: 전남대학교산학협력단
IPC: G06T7/00
CPC classification number: G06K9/00577 , G06T7/10
Abstract: PURPOSE: An object dividing method using a gaussian mixture model and an RGB clustering is provided to automatically divide an object domain of an object material on an input image without a direct handling by a user. CONSTITUTION: A first sub image generator leads a computer equipment to calculate normal distributions of pixels of each input image, and to produce a first sub image generator (S1200). A second sub image generator leads the computer equipment to calculate color mean values of each candidate area, to calculate Euclidean distance between the each pixel and the each color mean value, and to produce a second sub image (S1300). An object divider leads the computer equipment to produce a result image comprising the object division area comprising the pixels which are positioned on an overlapped spot among each pixel of the first and second object candidate area (S1400). [Reference numerals] (S1100) Output image format is converted to RGB color format; (S1200) First sub image generator leads a computer equipment to calculate normal distributions of pixels of each input image, and to produce a first sub image generator; (S1300) Second sub image generator leads the computer equipment to calculate color mean values of each candidate area, to calculate Euclidean distance between the each pixel and the each color mean value, and to produce a second sub image; (S1400) Object divider leads the computer equipment to produce a result image comprising the object division area comprising the pixels which are positioned on an overlapped spot among each pixel of the first and second object candidate area; (S1500) Result images are converted into a divided binary area, a largest binary area is set as an object area, other areas are set as noises and removed; (S1600) Pixels matched with each pixel coordiate of an object area are created as an object block
Abstract translation: 目的：提供使用高斯混合模型和RGB聚类的对象分割方法，用于在用户直接处理的情况下，自动划分输入图像上的对象材料的对象域。 构成：第一子图像生成器引导计算机设备计算每个输入图像的像素的正态分布，并产生第一子图像生成器（S1200）。 第二子图像生成器引导计算机设备计算每个候选区域的颜色平均值，以计算每个像素与每个颜色平均值之间的欧几里德距离，并产生第二子图像（S1300）。 对象分割器引导计算机设备产生包括包括位于第一和第二对象候选区域的每个像素之间的重叠点上的像素的对象分割区域的结果图像（S1400）。 （附图标记）（S1100）输出图像格式被转换为RGB颜色格式; （S1200）第一子图像生成器引导计算机设备来计算每个输入图像的像素的正态分布，并产生第一子图像生成器; （S1300）第二子图像生成器引导计算机设备计算每个候选区域的颜色平均值，计算每个像素之间的欧几里德距离和每个颜色平均值，并产生第二子图像; （S1400）对象分割器引导计算机设备产生包括对象分割区域的结果图像，该对象分割区域包括位于第一和第二对象候选区域的每个像素之间的重叠点上的像素; （S1500）将结果图像转换为分割二进制区域，将最大二进制区域设置为对象区域，其他区域设置为噪声并移除; （S1600）创建与对象区域的每个像素协调匹配的像素作为对象块
-
公开(公告)号：KR1020120041468A
公开(公告)日：2012-05-02
申请号：KR1020100102937
申请日：2010-10-21
Applicant: 전남대학교산학협력단
Abstract: PURPOSE: A system and a method for detecting interstitial lung diseases at a CT image are provided to automatically detect and quantify the initial interstitial lung disease using the CT image. CONSTITUTION: A lung dividing unit(10) divides and extracts a lung area from a CT image. A texture feature point extracting unit(20) extracts one or more texture features in an interest region. A classifying unit(30) classifies corresponding pixel into positivity or negativity with regard to interstitial lung diseases. A detection volume calculating unit(40) calculates the volume of all areas which are classified into positivity. A classification score calculating unit(50) calculates a volume ratio of a positive area to the whole area of a lung.
Abstract translation: 目的：提供一种用于CT图像检测间质性肺疾病的系统和方法，以使用CT图像自动检测和量化初始间质性肺病。 构成：肺分裂单元（10）从CT图像中分离并提取肺部区域。 纹理特征点提取单元（20）提取兴趣区域中的一个或多个纹理特征。 分类单元（30）将相应像素分类为间质性肺疾病的阳性或阴性。 检测量计算单元（40）计算分类为阳性的所有区域的体积。 分类分数计算单元（50）计算正面积与肺的整个面积的体积比。
-
-
-
-
-
-
-
-
Search Results
29
records
(took 0.028 seconds)
