公开(公告)号：US6091460A

公开(公告)日：2000-07-18

申请号：US948589

申请日：1997-10-10

申请人： Yoshiko Hatano ,  Tadashi Kasezawa ,  Koji Okazaki ,  Takashi Shinohara ,  Hiromu Hasegawa ,  Hidetoshi Mishima

发明人： Yoshiko Hatano ,  Tadashi Kasezawa ,  Koji Okazaki ,  Takashi Shinohara ,  Hiromu Hasegawa ,  Hidetoshi Mishima

IPC分类号： G06T9/00 ,  H04N5/14 ,  H04N7/26 ,  H04N7/46

CPC分类号： H04N19/105 ,  H04N19/107 ,  H04N19/112 ,  H04N19/137 ,  H04N19/142 ,  H04N19/179 ,  H04N19/51 ,  H04N19/57 ,  H04N19/577 ,  H04N19/87 ,  H04N5/145

摘要： In an image signal encoding method for encoding an image signal using motion compensation, a motion vector is found by means of a block matching method, a first distortion SEmc of motion compensated prediction associated with the motion vector, and a second distortion SEnomc of prediction without motion compensation are detected, the motion vector is used for for inter-picture prediction encoding when SEnomc>SEmc+K, with K being a constant greater than 0, and a vector having a value zero is used, in place of the motion vector, for inter-picture prediction encoding when SEnomc.ltoreq.SEmc+K. In another aspect, a speed of motion in a sequence of pictures is detected, and for a part of the sequence of pictures detected to contain a quick motion, the prediction encoding is performed using one-way prediction encoding, and for a part of the sequence of pictures without a quick motion, prediction encoding is performed using bi-directional prediction encoding. In another aspect, the image signal is subsampled for each field, and a motion vector is determined using the field-subsampled image signal.

摘要翻译： 在用于使用运动补偿对图像信号进行编码的图像信号编码方法中，通过块匹配方法，与运动矢量相关联的运动补偿预测的第一失真SEmc以及无运动补偿预测的第二失真SEnomc，找到运动矢量 检测到运动补偿时，运动矢量用于SEnomc> SEmc + K时的画面间预测编码，其中K为大于0的常数，使用值为零的矢量代替运动矢量， 用于当SEnomc

公开(公告)号：US09661333B2

公开(公告)日：2017-05-23

申请号：US13522122

申请日：2010-11-29

申请人： Toshimitsu Tatsuka ,  Hiromu Hasegawa ,  Takeaki Komuro ,  Masato Yamada

发明人： Toshimitsu Tatsuka ,  Hiromu Hasegawa ,  Takeaki Komuro ,  Masato Yamada

IPC分类号： H04N11/04 ,  H04N7/12 ,  H04N19/174 ,  H04N19/147 ,  H04N19/196 ,  H04N19/149 ,  H04N19/61 ,  H04N19/124

CPC分类号： H04N19/174 ,  H04N19/124 ,  H04N19/147 ,  H04N19/149 ,  H04N19/196 ,  H04N19/61

摘要： An image processor includes an encoder that performs encoding including quantization on an image signal and a controller that controls a quantization parameter for quantization. The controller determines a quantization parameter of a currently target macroblock as an increase or decrease from a reference value, and determines the increase or decrease based on a difference between a target amount of code for a predetermined number of macroblocks fewer than a total number of macroblocks within one frame and a generated amount of code of the predetermined number of macroblocks processed immediately before. The controller can further determine the increase or decrease, based on pixel information of the currently target macroblock such as an activity evaluation value.

公开(公告)号：US09443318B2

公开(公告)日：2016-09-13

申请号：US14240912

申请日：2012-07-20

申请人： Hiromu Hasegawa

发明人： Hiromu Hasegawa

IPC分类号： G06K9/36 ,  G06T7/00 ,  G06K9/46

CPC分类号： G06K9/4633 ,  G06K9/4652 ,  G06T7/13 ,  G06T7/168 ,  G06T2207/10024 ,  G06T2207/20061

摘要： An apparatus and method to detect a line segment or arc using Hough transform. A Hough transform unit performs contour extraction on brightness image data to generate contour image data, with pixels having a pixel value of 0 to 255, performs the Hough transform on points in the contour image data, and counts additional values represented by pixel values of points in the contour image data in a Hough table. The Hough transform unit performs contour extraction on first to third component data to generate first to third contour data with pixels having a pixel value of 0 to 255, performs the Hough transform on points in the first to third contour data, and counts additional values represented by pixel values of points in the first to third contour data in the Hough table. The detection unit comprehensively evaluates the counts to detect a line segment or arc.

摘要翻译： 使用霍夫变换检测线段或弧的装置和方法。 霍夫变换单元对亮度图像数据执行轮廓提取，以生成具有像素值为0至255的像素的轮廓图像数据，对轮廓图像数据中的点执行霍夫变换，并对由点的像素值表示的附加值进行计数 在Hough表中的轮廓图像数据中。 霍夫变换单元在第一至第三分量数据上执行轮廓提取，以生成具有像素值为0至255的像素的第一至第三轮廓数据，对第一至第三轮廓数据中的点执行霍夫变换，并计数表示的附加值 在霍夫表中第一到第三轮廓数据中的点的逐像素值。 检测单元综合评估检测线段或弧线的计数。

公开(公告)号：US20140219504A1

公开(公告)日：2014-08-07

申请号：US14343182

申请日：2012-07-20

申请人： Yuki Haraguchi ,  Hiromu Hasegawa

发明人： Yuki Haraguchi ,  Hiromu Hasegawa

IPC分类号： G06K9/00 ,  G06T7/00

CPC分类号： G06K9/00369 ,  G06K9/00805 ,  G06T11/60 ,  G06T2210/22

摘要： It's an object of the invention to provide an object detection device capable of detecting an object for detection in an input image with high precision. In an object detection device 1, a detection window setting unit 11 receives a photographic image 21 photographed by a camera. The detection window setting unit 11 sets a detection window area in the photographic image 21 and generates a normal window image 22 by cutting out the detection window area from the photographic image 21. An image processing unit 12 performs image processing such as enlargement and reduction, etc. on the photographic image 21. Each modified window images 23 is cut out from each of the enlarged photographic image 21 and the reduced photographic image 21. A degree calculation unit 13 calculates matching rates indicating a possibility the object for detection in the window image for each window images on the basis of feature data 51 indicating a feature of the object for detection. A determination unit 14 determines whether or not the detection object is present in the detection window area on the basis of the matching rates of each window image.

摘要翻译： 本发明的目的是提供一种能够高精度地检测输入图像中的检测对象的物体检测装置。 在物体检测装置1中，检测窗口设定部11接收由照相机拍摄的摄影图像21。 检测窗口设定单元11在摄影图像21中设置检测窗口区域，并通过从摄影图像21切出检测窗口区域来生成正常窗口图像22.图像处理单元12进行诸如放大和缩小的图像处理， 每个经修改的窗口图像23从放大的照片图像21和缩小的摄影图像21中的每一个切出。度数计算单元13计算指示在窗口图像中检测对象的可能性的匹配率 基于指示用于检测的对象的特征的特征数据51对于每个窗口图像。 确定单元14基于每个窗口图像的匹配率来确定检测对象是否存在于检测窗口区域中。

公开(公告)号：US20120321207A1

公开(公告)日：2012-12-20

申请号：US13582056

申请日：2011-02-24

申请人： Hiromu Hasegawa ,  Toshimitsu Tatsuka

发明人： Hiromu Hasegawa ,  Toshimitsu Tatsuka

IPC分类号： G06K9/36

CPC分类号： H04N19/0003 ,  H04N19/107 ,  H04N19/176 ,  H04N19/184 ,  H04N19/196

摘要： It is an object of the present invention to provide an image coding technique for suppressing degradation in image quality, in which the time and space where intra macroblocks appear are dispersed. A numerical value (Ftk) is generated from the lower-order six bits of the frame number (Ft) of a coding object frame. A numerical value (Fs) is generated by shifting the numerical value (Ftk) leftward by two bits. An exclusive OR of the numerical value (Ftk) and the numerical value (Fs) is calculated, to thereby generate a numerical value (A). A numerical value (Ytk) is generated from the lower-order six bits of the Y coordinate (Yt) of a coding object macroblock. The upper-order bits of the numerical value (Ytk) and the lower-order bits thereof are inverted, to thereby generate a numerical value (Yr). Further, an exclusive OR of the numerical value (Yr) and the numerical value (A) is calculated, to thereby generate a numerical value (B). A numerical value (Xtk) is generated from the lower-order six bits of the X coordinate (Xt) of the coding object macroblock. When the numerical value (Xtk) and the numerical value (B) are identical to each other, the coding object macroblock is intra-coded.

摘要翻译： 本发明的一个目的是提供一种图像编码技术，用于抑制图像质量的劣化，其中内部宏块出现的时间和空间被分散。 从编码对象帧的帧号（Ft）的低位6位生成数值（Ftk）。 通过将数值（Ftk）向左移位两位来产生数值（Fs）。 计算数值（Ftk）和数值（Fs）的异或，从而生成数值（A）。 从编码对象宏块的Y坐标（Yt）的低位6位生成数值（Ytk）。 数值（Ytk）及其低位的高位被反转，从而生成数值（Yr）。 此外，计算数值（Yr）和数值（A）的异或，从而生成数值（B）。 从编码对象宏块的X坐标（Xt）的低6位生成数值（Xtk）。 当数值（Xtk）和数值（B）彼此相同时，编码对象宏块被帧内编码。

公开(公告)号：US20110134997A1

公开(公告)日：2011-06-09

申请号：US13057111

申请日：2009-03-06

申请人： Nobumasa Narimatsu ,  Hiromu Hasegawa

发明人： Nobumasa Narimatsu ,  Hiromu Hasegawa

IPC分类号： H04N7/12

CPC分类号： H04N19/124 ,  H04N19/149 ,  H04N19/162 ,  H04N19/172 ,  H04N19/177 ,  H04N19/40 ,  H04N19/61

摘要： A technique for suppressing a significant variation of a quantization step value and enabling a stable rate control to be performed. A function is used for calculating a quantization step conversion factor from a bit rate ratio is a straight line with an inclination of −1, intersecting a function at a reference point. The function is a monotone decreasing exponential function. A reference bit rate ratio (R0) is expressed as R0=T/S by using a total bit rate (S) of a first stream and a total target bit rate (T) of a second stream. The function appropriately represents a relation between the bit rate ratio and the quantization step conversion factor in coding conversion but has a large rate of variation in an area where the bit rate ratio is about 0.5. The function has a small rate of variation and can suppress a significant variation of the quantization step conversion value.

摘要翻译： 一种用于抑制量化步长值的显着变化并实现稳定速率控制的技术。 使用函数来计算比特率的量化步长转换因子是与参考点的函数相交的具有-1的倾斜度的直线。 该函数是单调递减指数函数。 通过使用第一流的总比特率（S）和第二流的总目标比特率（T）将参考比特率比（R0）表示为R0 = T / S。 该功能适当地表示编码转换中的比特率比与量化级转换因子之间的关系，但是在比特率为约0.5的区域中具有较大的变化率。 该功能具有较小的变化率，并且可以抑制量化步长转换值的显着变化。

公开(公告)号：US20090225825A1

公开(公告)日：2009-09-10

申请号：US12367658

申请日：2009-02-09

申请人： Makoto SAITO ,  Hiromu Hasegawa

发明人： Makoto SAITO ,  Hiromu Hasegawa

IPC分类号： H04N7/26

CPC分类号： H04N19/132 ,  H04N19/15 ,  H04N19/162 ,  H04N19/177 ,  H04N19/196 ,  H04N19/40

摘要： A reference differential value calculation part calculates a reference differential value by subtracting an accumulated value of the target amounts of codes from an accumulated value of the amounts of generated codes from the first period to the (n−1) period. A comparison part compares a cumulative differential value obtained by subtracting the cumulative amount of generated codes from the cumulative target amount of codes of the first to m-th picture frames in the n period with the reference differential value. If the cumulative differential value is larger than the reference differential value, stuffing bits are added to the m-th picture frame to be processed by the amount of codes obtained by subtracting the reference differential value from the cumulative differential value.

摘要翻译： 参考差分值计算部分从第一周期到第（n-1）个周期的生成代码量的累计值中减去目标代码量的累加值来计算参考差分值。 比较部分将通过从n个周期中的第一至第m个图像帧的累积目标量的累加量减去生成的代码获得的累积差分值与参考差分值进行比较。 如果累积微分值大于参考差分值，则将填充比特添加到通过从累积微分值减去参考差分值而获得的代码量来处理的第m个图像帧。

公开(公告)号：US09426474B2

公开(公告)日：2016-08-23

申请号：US12995826

申请日：2009-02-26

申请人： Hiromu Hasegawa

发明人： Hiromu Hasegawa

IPC分类号： H04N7/26 ,  H04N19/124 ,  H04N19/172 ,  H04N19/61 ,  H04N19/146 ,  H04N19/154 ,  H04N19/40

CPC分类号： H04N19/124 ,  H04N19/146 ,  H04N19/154 ,  H04N19/172 ,  H04N19/40 ,  H04N19/61

摘要： A transcoder that controls the amount of generated codes of an output stream toward a target bit rate without degradation of image quality. The transcoder decodes a first stream and encodes the decoded image again to thereby output a second stream. The transcoder calculates a distortion evaluation value from the image obtained by decoding the first stream and an image obtained by decoding the second stream. Assuming that a ratio between the distortion evaluation value and a total target distortion evaluation value is determined as a target distortion ratio, a target setting bit rate of a second stream in the period can be obtained by multiplying a total target bit rate of the second stream by the target distortion ratio. Alternatively, the target setting bit rate can be obtained by adjusting the target distortion ratio with an appropriate function and adding the target distortion ratio to the total target bit rate of the second stream.

摘要翻译： 代码转换器，其将输出流的生成代码的量控制到目标比特率，而不降低图像质量。 代码转换器对第一流进行解码并再次对解码图像进行编码，从而输出第二流。 代码转换器从通过解码第一流获得的图像和通过解码第二流获得的图像来计算失真评估值。 假设失真评估值和总目标失真评估值之间的比率被确定为目标失真比率，则可以通过将第二流的总目标比特率相乘来获得该周期中的第二流的目标设置比特率 通过目标失真比。 或者，可以通过利用适当的功能调整目标失真比并将目标失真比与第二流的总目标比特率相加来获得目标设定比特率。

公开(公告)号：US09071837B2

公开(公告)日：2015-06-30

申请号：US13401198

申请日：2012-02-21

申请人： Hiromu Hasegawa ,  Nobuyuki Takasu ,  Makoto Saito

发明人： Hiromu Hasegawa ,  Nobuyuki Takasu ,  Makoto Saito

IPC分类号： H04N19/114 ,  H04N19/169 ,  H04N19/15 ,  H04N19/115 ,  H04N19/61 ,  H04N19/126 ,  H04N19/157 ,  H04N19/162 ,  H04N19/177 ,  H04N19/40

CPC分类号： H04N19/146 ,  H04N19/114 ,  H04N19/115 ,  H04N19/124 ,  H04N19/126 ,  H04N19/15 ,  H04N19/157 ,  H04N19/162 ,  H04N19/169 ,  H04N19/177 ,  H04N19/40 ,  H04N19/61

摘要： If the number of frames in a GOP of an input stream is not less than 15, the GOP is determined as a control unit time. If the number of frames in the GOP is less than 15, the following GOP is connected thereto until the number of frames becomes not less than 15 and the connected GOPs are determined as a control unit time. After correcting the control unit time, the average input bit rate Sn in each control unit time is obtained, and by using information on the input stream including the average input bit rate Sn, a quantization step value of an output stream is calculated.

摘要翻译： 如果输入流的GOP中的帧数不小于15，则将GOP确定为控制单元时间。 如果GOP中的帧数小于15，则连接到下一个GOP，直到帧的数量变得不小于15，并且所连接的GOP被确定为控制单元时间。 在校正控制单元时间之后，获得每个控制单元时间内的平均输入比特率Sn，并且通过使用包括平均输入比特率Sn的输入流的信息，计算输出流的量化步长值。

公开(公告)号：US20140193096A1

公开(公告)日：2014-07-10

申请号：US14240912

申请日：2012-07-20

申请人： Hiromu Hasegawa

发明人： Hiromu Hasegawa

IPC分类号： G06T7/00

CPC分类号： G06K9/4633 ,  G06K9/4652 ,  G06T7/13 ,  G06T7/168 ,  G06T2207/10024 ,  G06T2207/20061

摘要： An apparatus and method to detect a line segment or arc using Hough transform. A Hough transform unit performs contour extraction on brightness image data to generate contour image data, with pixels having a pixel value of 0 to 255, performs the Hough transform on points in the contour image data, and counts additional values represented by pixel values of points in the contour image data in a Hough table. The Hough transform unit performs contour extraction on first to third component data to generate first to third contour data with pixels having a pixel value of 0 to 255, performs the Hough transform on points in the first to third contour data, and counts additional values represented by pixel values of points in the first to third contour data in the Hough table. The detection unit comprehensively evaluates the counts to detect a line segment or arc.

摘要翻译： 使用霍夫变换检测线段或弧的装置和方法。 霍夫变换单元对亮度图像数据执行轮廓提取，以生成具有像素值为0至255的像素的轮廓图像数据，对轮廓图像数据中的点执行霍夫变换，并对由点的像素值表示的附加值进行计数 在Hough表中的轮廓图像数据中。 霍夫变换单元在第一至第三分量数据上执行轮廓提取，以生成具有像素值为0至255的像素的第一至第三轮廓数据，对第一至第三轮廓数据中的点执行霍夫变换，并计数表示的附加值 在霍夫表中第一到第三轮廓数据中的点的逐像素值。 检测单元综合评估检测线段或弧线的计数。