首页> 外文会议>International Conference on Circuits, Power and Computing Technologies >Detect and separate localization text in various complicated-colour image
【24h】

Detect and separate localization text in various complicated-colour image

机译:在各种复杂彩色图像中检测和分开本地化文本

获取原文

摘要

Text detection and localization in Multi-color images is important for content-based image analysis. This conflict is occurring due to the various backgrounds, the inhomogeneous illumination, and the variations of text font, size and line orientation. In this the image is converted into grayscale image then median filter is used to discard the unwanted noise in the gray scale image. Edge detection is mentioned by using canny edge detection and sobel edge detector. Canny edge detection is used to measure edges of the horizontal and vertical axis present around the text region in gray scale image. Sobel edge detector is used to measure edges of the overall boundaries of the horizontal and vertical axis present in the grayscale image. A mask is a dark and light image of the same dimensions as the original image (or the region of interest we are working in). Group of the pixels in the mask can have therefore a value of 0 (black) or 1 (white). When executing operations on the image the mask is used to restrict the result to the pixels that are 1 (selected, active, and white) in the mask. Morphological operations are defined by moving a structuring element over a binary image to be modified in such a way that it is centered over an image pixel at some point. The process of removing certain details in an image which is smaller than certain preference shape is called Morphological image processing and the preference shape is called structuring element. The goal of the connected component analysis is to detect the large sized connected foreground region or object. This is one of the important operations in motion detection. The pixels that are collectively connected can be clustered into changing or moving objects by analyzing their connectivity. The recognition of the text region is referred to the optical Character Recognition which compares the text with databases which consists of different types of character and extracts the text and mentioned in the text notepad docum- nt.
机译:多色图像中的文本检测和定位对于基于内容的图像分析非常重要。由于各种背景,不均匀的照明和文本字体,尺寸和线路方向的变化,因此发生这种冲突。在此,图像被转换成灰度图像,然后中值滤波器用于丢弃灰度图像中的不需要的噪声。通过使用Canny Edge检测和Sobel Edge检测器提及边缘检测。罐内边缘检测用于测量灰度图像周围的水平和垂直轴的边缘。 Sobel边缘检测器用于测量灰度图像中存在的水平和垂直轴的总边界的边缘。掩模是与原始图像(或我们在工作区域)相同的尺寸的暗和光图像。因此,掩模中的像素的组可以具有0(黑色)或1(白色)的值。当在图像上执行操作时,掩码用于将结果限制到掩码中的1(选择,有效和白色)的像素。通过在待修改的二进制图像上移动结构化元件来定义形态学操作,以使其在某个点在图像像素上以图像像素为中心。去除小于某些偏好形状的图像中的某些细节的过程称为形态学图像处理,并且偏好形状称为结构元件。连接分量分析的目标是检测大型连接的前景区域或物体。这是运动检测中的重要操作之一。通过分析它们的连接,可以将统称连接的像素聚集成改变或移动物体。识别文本区域被称为光学字符识别,它将文本与数据库进行比较,该数据库由不同类型的字符组成,并在文本记事本Docum-NT中提及文本并提及。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号