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
    • 1. 发明申请
    • NOISE SUPPRESSION METHOD FOR DIGITAL IMAGE
    • 用于数字图像的噪声抑制方法
    • US20110069886A1
    • 2011-03-24
    • US12649126
    • 2009-12-29
    • Chung Fang ChienChe Hung Lin
    • Chung Fang ChienChe Hung Lin
    • G06K9/40G06K9/46
    • G06T5/002G06T2207/20012
    • A noise suppression method for a digital image is described, which utilizes an adjacent area of a target pixel in the digital image to conduct a denoising process on the target pixel. The noise suppression method includes the following steps. A feature-weighted processing procedure is performed on each original pixel in the digital image to convert the original pixel into a feature-weighted pixel. According to a position of the target pixel in the digital image, a reference pixel is selected from a corresponding position in the feature-weighted image. Similarities of the reference pixel relative to the other feature-weighted pixels are respectively computed. Through the computed similarities, a feature-weighted pixel for compensation is selected from the feature-weighted pixels other than the reference pixel. Finally, noise suppression is performed on the reference pixel by using the feature-weighted pixel for compensation.
    • 描述了一种用于数字图像的噪声抑制方法,其利用数字图像中的目标像素的相邻区域对目标像素进行去噪处理。 噪声抑制方法包括以下步骤。 对数字图像中的每个原始像素执行特征加权处理过程,以将原始像素转换成特征加权像素。 根据数字图像中的目标像素的位置,从特征加权图像中的相应位置中选择参考像素。 分别计算参考像素相对于其他特征加权像素的相似度。 通过计算的相似度,从除参考像素之外的特征加权像素中选择用于补偿的特征加权像素。 最后,通过使用用于补偿的特征加权像素对参考像素执行噪声抑制。
    • 2. 发明申请
    • COMPENSATION METHOD FOR REMOVING IMAGE NOISE
    • 用于去除图像噪声的补偿方法
    • US20100246949A1
    • 2010-09-30
    • US12482706
    • 2009-06-11
    • Chung-Fang ChienChe Hung Lin
    • Chung-Fang ChienChe Hung Lin
    • G06K9/00
    • G06T3/4015
    • A compensation method for removing an image noise is applicable to compensate for noises generated by Bayer pattern color interpolation. The compensation method includes the following steps selecting a partial region in the Bayer pattern to set a selection region; selecting a plurality of color units from the selection region one by one, in which each of the color units at least includes a green pixel, a red pixel, and a blue pixel; sequentially performing a color weight adjustment on the green pixel, the red pixel, and the blue pixel; and compensating for the pixel in the selection region according to each color weight, so as to generate a recovered pixel.
    • 用于去除图像噪声的补偿方法适用于补偿由拜耳图案颜色插值产生的噪声。 补偿方法包括以下步骤:选择拜耳图案中的部分区域以设置选择区域; 从选择区域逐个选择多个颜色单元,其中每个颜色单元至少包括绿色像素,红色像素和蓝色像素; 对所述绿色像素,所述红色像素和所述蓝色像素依次进行色彩权重调整; 并且根据每个颜色权重补偿选择区域中的像素,以便产生恢复的像素。
    • 3. 发明申请
    • SHARPNESS PROCESSING METHOD AND SYSTEM FOR DIGITAL IMAGE
    • 数字图像的锐化处理方法和系统
    • US20100321524A1
    • 2010-12-23
    • US12709758
    • 2010-02-22
    • Che Hung LinChung Ta Wu
    • Che Hung LinChung Ta Wu
    • H04N9/73H04N5/208H04N9/64
    • H04N5/208H04N5/23229H04N5/3572
    • A sharpness processing method and system for a digital image are applied to an image capturing device for capturing a raw image. The method includes the steps. An image processing procedure is performed on the raw image to generate a registered image. A first sharpness enhancement table and a second sharpness enhancement table respectively sharpness attenuation at an outer-edge region and a central region of the registered image are loaded, and a range between the outer-edge region and the central region is defined as a plurality of critical regions. A sharpness enhancement weight table is loaded, and weight values of each critical region respectively corresponding to the first sharpness enhancement table and the second sharpness enhancement table are searched, for obtaining a sharpness enhanced value of each critical region. An image sharpness processing procedure is performed on the critical region, for correct the registered image into a digital image.
    • 用于数字图像的锐度处理方法和系统被应用于用于捕获原始图像的图像捕获装置。 该方法包括步骤。 对原始图像执行图像处理过程以生成登记图像。 加载分别在登记图像的外边缘区域和中心区域的锐度衰减的第一锐度增强表和第二锐度增强表,并且外边缘区域和中心区域之间的范围被定义为多个 关键地区。 加载锐度增强权重表,并且搜索分别对应于第一锐度增强表和第二锐度增强表的每个临界区的权重值,以获得每个临界区的锐度增强值。 在临界区域执行图像锐度处理过程,以将登记的图像校正为数字图像。