专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20120294544A1 IMAGE COMPRESSION METHOD 审中-公开
标题翻译：图像压缩方法
公开(公告)号：US20120294544A1
公开(公告)日：2012-11-22
申请号：US13369051
申请日：2012-02-08
申请人： Heng-Yao Lin , Pai-Chu Hsieh , Te-Wei Lee
发明人： Heng-Yao Lin , Pai-Chu Hsieh , Te-Wei Lee
IPC分类号： G06K9/36
CPC分类号： H04N19/91
摘要： An image compression method is provided. Residuals of a lot of pixel data are calculated. Respective compression costs of the residuals of the pixel data are estimated according to a lot of category residual relationships. The compression costs are compared and one of the category residual relationships is selected. The residuals of the pixel data are compressed according to the selected category residual relationship.
摘要翻译：提供了图像压缩方法。计算大量像素数据的残差。根据许多类别残差关系来估计像素数据的残差的各自的压缩成本。比较压缩成本，选择类别残差关系之一。根据所选择的类别残差关系来压缩像素数据的残差。

2. 发明授权

US08824814B2 Pixel data compression and decompression method 有权
标题翻译：像素数据压缩和解压缩方法
公开(公告)号：US08824814B2
公开(公告)日：2014-09-02
申请号：US13424051
申请日：2012-03-19
申请人： Te-Wei Lee , Pai-Chu Hsieh , Heng-Yao Lin
发明人： Te-Wei Lee , Pai-Chu Hsieh , Heng-Yao Lin
IPC分类号： G06K9/36 , H04N7/26 , G06K9/46
CPC分类号： H04N19/00533 , H04N19/423 , H04N19/44
摘要： A pixel data decompression method for decompressing a frame is provided. The method includes: loading memory storage addresses corresponding to to-be-decompressed blocks; reading and decompressing a pixel line of a current block according to a memory storage address of the current block; updating the memory storage address of the current block according to a decompression result; reading a memory storage address of a next block if decompression of a pixel line of the frame is not completed; and repeating the above steps until decompression of the frame is completed.
摘要翻译：提供了用于解压缩帧的像素数据解压缩方法。该方法包括：加载对应于待解压缩块的存储器存储地址; 根据当前块的存储器存储地址读取和解压缩当前块的像素线; 根据解压缩结果更新当前块的存储器存储地址; 如果帧的像素线的解压缩未完成，则读取下一个块的存储器存储地址; 并重复上述步骤直到帧的解压缩完成。

3. 发明申请

US20070014350A1 Method and apparatus for low-memory high-performance discrete cosine transform coefficient prediction 有权
标题翻译：用于低内存高性能离散余弦变换系数预测的方法和装置
公开(公告)号：US20070014350A1
公开(公告)日：2007-01-18
申请号：US11314788
申请日：2005-12-19
申请人： Chien-Yu Lin , Tzu-Hsin Wang , Pai-Chu Hsieh
发明人： Chien-Yu Lin , Tzu-Hsin Wang , Pai-Chu Hsieh
IPC分类号： H04N11/04
CPC分类号： H04N19/18 , H04N19/105 , H04N19/13 , H04N19/146 , H04N19/176 , H04N19/593
摘要： A method and an apparatus for discrete cosine transform coefficient prediction are disclosed. The method comprises the steps as follows: providing a discrete cosine transformed block; performing an AC prediction mode variable length coding and a non-AC prediction mode variable length coding simultaneously; providing a first memory block and a second memory block for storing the bit strings of two coding; and using another coding as long as one of first and second memory blocks reaches its capacity limit.
摘要翻译：公开了一种用于离散余弦变换系数预测的方法和装置。该方法包括以下步骤：提供离散余弦变换块; 同时执行AC预测模式可变长度编码和非AC预测模式可变长度编码; 提供第一存储器块和用于存储两个编码的位串的第二存储器块; 并且只要第一和第二存储器块之一达到其容量限制，则使用另一编码。

4. 发明申请

US20060256121A1 Method and apparatus for storing image data 审中-公开
标题翻译：用于存储图像数据的方法和装置
公开(公告)号：US20060256121A1
公开(公告)日：2006-11-16
申请号：US11197819
申请日：2005-08-05
申请人： Pai-Chu Hsieh , Tzu-Hsin Wang , Chien-Yu Lin
发明人： Pai-Chu Hsieh , Tzu-Hsin Wang , Chien-Yu Lin
IPC分类号： G06F12/02
CPC分类号： G06T1/60
摘要： A method and an apparatus for saving image data are disclosed. This method adopts a pixel processor to separate the pixel values of an image data into a plurality of pixel groups according to the highest sub-sample rate and the data width of the memory. Then, those pixel groups are stored in different memory blocks of the memory after being recombined such that sub-sample images can be efficiently obtained under a condition that only full-resolution image is stored, thus improving disadvantage of storing low-resolution image separately in prior art. Besides, the data of those pixel groups are arranged alternately such that every memory block can be opened evenly and alternately when storing/reading sub-sample and original image, thus enhancing the efficiency of memory use.
摘要翻译：公开了一种用于保存图像数据的方法和装置。该方法采用像素处理器，根据最高子采样率和存储器的数据宽度将图像数据的像素值分离成多个像素组。然后，这些像素组被重新组合后存储在存储器的不同存储块中，使得在仅存储全分辨率图像的条件下可以有效地获得子样本图像，从而改善了将低分辨率图像分开存储的缺点现有技术此外，这些像素组的数据被交替布置，使得当存储/读取子样本和原始图像时，每个存储块可以均匀和交替地打开，从而提高存储器使用的效率。

5. 发明授权

US07787537B2 Method and apparatus for low-memory high-performance discrete cosine transform coefficient prediction 有权
标题翻译：用于低内存高性能离散余弦变换系数预测的方法和装置
公开(公告)号：US07787537B2
公开(公告)日：2010-08-31
申请号：US11314788
申请日：2005-12-19
申请人： Chien-Yu Lin , Tzu-Hsin Wang , Pai-Chu Hsieh
发明人： Chien-Yu Lin , Tzu-Hsin Wang , Pai-Chu Hsieh
IPC分类号： H04B1/66 , G06F7/00
CPC分类号： H04N19/18 , H04N19/105 , H04N19/13 , H04N19/146 , H04N19/176 , H04N19/593
摘要： A method and an apparatus for discrete cosine transform coefficient prediction are disclosed. The method comprises the steps as follows: providing a discrete cosine transformed block; performing an AC prediction mode variable length coding and a non-AC prediction mode variable length coding simultaneously; providing a first memory block and a second memory block for storing the bit strings of two coding; and using another coding as long as one of first and second memory blocks reaches its capacity limit.
摘要翻译：公开了一种用于离散余弦变换系数预测的方法和装置。该方法包括以下步骤：提供离散余弦变换块; 同时执行AC预测模式可变长度编码和非AC预测模式可变长度编码; 提供第一存储器块和用于存储两个编码的位串的第二存储器块; 并且只要第一和第二存储器块之一达到其容量限制，则使用另一编码。

6. 发明授权

US07826533B2 Method for object edge detection in macroblock and method for deciding quantization scaling factor 有权
标题翻译：宏块中物体边缘检测方法及决定量化比例因子的方法
公开(公告)号：US07826533B2
公开(公告)日：2010-11-02
申请号：US11188040
申请日：2005-07-22
申请人： Pai-Chu Hsieh , Tzu-Hsin Wang , Chien-Yu Lin
发明人： Pai-Chu Hsieh , Tzu-Hsin Wang , Chien-Yu Lin
IPC分类号： H04N7/18
CPC分类号： H04N19/14 , H04N19/126 , H04N19/176 , H04N19/60
摘要： A method for object edge detection in a macroblock and a method for deciding quantization scaling factor are disclosed. This method calculates and compares the pixel means and means of absolutely difference of a plurality of pixel blocks in a macroblock to achieve the purpose of detecting if there is an object edge or flat area in a macroblock. In the meantime, the image structure of macroblock is analyzed and classified and its related messages are used for bit rate control, such that the visual effect of the compressed image can be appropriately enhanced.
摘要翻译：公开了一种用于宏块中的对象边缘检测的方法以及用于决定量化缩放因子的方法。该方法计算并比较宏块中的多个像素块的像素平均值和绝对差异的平均值，以实现宏块中是否存在物体边缘或平坦区域的目的。同时，对宏块的图像结构进行分析分类，并将其相关消息用于比特率控制，从而可以适当增强压缩图像的视觉效果。

7. 发明授权

US07558429B2 Image processing method and method for detecting differences between different image macro-blocks 有权
标题翻译：用于检测不同图像宏块之间的差异的图像处理方法和方法
公开(公告)号：US07558429B2
公开(公告)日：2009-07-07
申请号：US11161758
申请日：2005-08-16
申请人： Chien-Yu Lin , Tzu-Hsin Wang , Pai-Chu Hsieh
发明人： Chien-Yu Lin , Tzu-Hsin Wang , Pai-Chu Hsieh
IPC分类号： G06K9/36 , G06K9/46
CPC分类号： H04N19/137 , H04N19/132 , H04N19/159 , H04N19/172 , H04N19/176 , H04N19/61
摘要： An image processing method and a method for detecting differences between different image macro-blocks are provided. The image processing method includes steps of: providing one of a plurality of macro-blocks of a processed image; performing a method for detecting the differences between the macro-blocks of different images, and there should be at least two images before the processed image in displaying time sequence for obtaining a macro-block difference table; and in the process of image encoding, if the processed image is a P-picture, determining whether the macro-block is a skip mode or not, by reference to the macro-block table in an auxiliary manner.
摘要翻译：提供了一种用于检测不同图像宏块之间的差异的图像处理方法和方法。该图像处理方法包括以下步骤：提供处理图像的多个宏块中的一个; 执行用于检测不同图像的宏块之间的差异的方法，并且在显示用于获得宏块差异表的时间序列中的处理图像之前应该存在至少两个图像; 并且在图像编码的过程中，如果处理的图像是P图像，则以辅助方式参考宏块表来确定宏块是否是跳过模式。

8. 发明申请

US20060257038A1 Method for object edge detection in macroblock and method for deciding quantization scaling factor 有权
标题翻译：宏块中物体边缘检测方法及决定量化比例因子的方法
公开(公告)号：US20060257038A1
公开(公告)日：2006-11-16
申请号：US11188040
申请日：2005-07-22
申请人： Pai-Chu Hsieh , Tzu-Hsin Wang , Chien-Yu Lin
发明人： Pai-Chu Hsieh , Tzu-Hsin Wang , Chien-Yu Lin
IPC分类号： G06K9/36 , G06K9/00 , G06K9/48
CPC分类号： H04N19/14 , H04N19/126 , H04N19/176 , H04N19/60
摘要： A method for object edge detection in a macroblock and a method for deciding quantization scaling factor are disclosed. This method calculates and compares the pixel means and means of absolutely difference of a plurality of pixel blocks in a macroblock to achieve the purpose of detecting if there is an object edge or flat area in a macroblock. In the meantime, the image structure of macroblock is analyzed and classified and its related messages are used for bit rate control, such that the visual effect of the compressed image can be appropriately enhanced.
摘要翻译：公开了一种用于宏块中的对象边缘检测的方法以及用于决定量化缩放因子的方法。该方法计算并比较宏块中的多个像素块的像素平均值和绝对差异的平均值，以实现宏块中是否存在物体边缘或平坦区域的目的。同时，对宏块的图像结构进行分析分类，并将其相关消息用于比特率控制，从而可以适当增强压缩图像的视觉效果。

9. 发明授权

US07778329B2 Mode decision system and method for motion vector fine search 有权
标题翻译：运动矢量精细搜索的模式决策系统和方法
公开(公告)号：US07778329B2
公开(公告)日：2010-08-17
申请号：US11206121
申请日：2005-08-18
申请人： Pai-Chu Hsieh , Chien-Yu Lin
发明人： Pai-Chu Hsieh , Chien-Yu Lin
IPC分类号： H04N7/12
CPC分类号： H04N5/145 , H04N19/43 , H04N19/53
摘要： A mode decision system and method for motion vector fine search. Each macroblock in a frame has a pre-determined clock number. The system includes a first subtractor, an operation level generator and a determinator. The first subtractor has a first input terminal to receive a clock number and a second input terminal to receive the pre-determined clock number in order to subtract the clock number from the pre-determined clock number and accordingly generate a clock difference. The operation level generator is connected to the first subtractor in order to generate an operation level based on the clock difference. The determinator selects one or more modes in accordance with the operation level and accordingly performs the fine search on the macroblock.
摘要翻译：一种用于运动矢量精细搜索的模式决策系统和方法。帧中的每个宏块具有预定的时钟数。该系统包括第一减法器，操作电平发生器和确定器。第一减法器具有接收时钟号的第一输入端和接收预定时钟号的第二输入端，以便从预定的时钟数中减去时钟数，从而产生时钟差。操作电平发生器连接到第一减法器，以便基于时钟差产生操作电平。确定器根据操作级别选择一个或多个模式，并相应地对宏块执行精细搜索。

10. 发明申请

US20060245657A1 IMAGE PROCESSING METHOD AND METHOD FOR DETECTING DIFFERENCES BETWEEN DIFFERENT IMAGE MACRO-BLOCKS 有权
标题翻译：用于检测不同图像宏块之间的差异的图像处理方法和方法
公开(公告)号：US20060245657A1
公开(公告)日：2006-11-02
申请号：US11161758
申请日：2005-08-16
申请人： Chien-Yu Lin , Tzu-Hsin Wang , Pai-Chu Hsieh
发明人： Chien-Yu Lin , Tzu-Hsin Wang , Pai-Chu Hsieh
IPC分类号： G06K9/36
CPC分类号： H04N19/137 , H04N19/132 , H04N19/159 , H04N19/172 , H04N19/176 , H04N19/61
摘要： An image processing method and a method for detecting differences between different image macro-blocks are provided. The image processing method includes steps of: providing one of a plurality of macro-blocks of a processed image; performing a method for detecting the differences between the macro-blocks of different images, and there should be at least two images before the processed image in displaying time sequence for obtaining a macro-block difference table; and in the process of image encoding, if the processed image is a P-picture, determining whether the macro-block is a skip mode or not, by reference to the macro-block table in an auxiliary manner.
摘要翻译：提供了一种用于检测不同图像宏块之间的差异的图像处理方法和方法。该图像处理方法包括以下步骤：提供处理图像的多个宏块中的一个; 执行用于检测不同图像的宏块之间的差异的方法，并且在显示用于获得宏块差异表的时间序列中的处理图像之前应该存在至少两个图像; 并且在图像编码的过程中，如果处理的图像是P图像，则以辅助方式参考宏块表来确定宏块是否是跳过模式。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式