公开(公告)号：US08325998B2

公开(公告)日：2012-12-04

申请号：US12328891

申请日：2008-12-05

Applicant: Tai-Chang Yang ,  Hong-Long Chou ,  Yin-Pin Chang

Inventor： Tai-Chang Yang ,  Hong-Long Chou ,  Yin-Pin Chang

IPC: G06K9/00

CPC classification number: G06K9/00228

Abstract: A multidirectional face detection method is for detecting a face in a picture under detection at different positions. The face detection method includes the steps. A selecting window sets to sequentially select different sub-image patterns from the picture under detection. A facial feature weight calculates and it is calculated according to a feature value of the pixels in a sub-image pattern selected by the selecting window, thereby determining if the sub-image pattern has any features similar to the face. A facial edge weight calculates for made on the picture under detection according to a boundary value of the pixels in the sub-image pattern selected by the selecting window, so as to determine if the part of area of the picture under detection has any facial-boundaries. Profile detection is performed to respectively mark the facial-boundaries in the sub-image patterns with arc segments respectively for the sub-image patterns having the facial-boundaries.

Abstract translation: 多方向面部检测方法用于在不同位置检测图像中的脸部。 面部检测方法包括以下步骤。 选择窗口设置为从检测到的图片中顺序地选择不同的子图像图案。 面部特征权重计算，并且根据由选择窗口选择的子图像图案中的像素的特征值来计算，从而确定子图像图案是否具有与面部相似的任何特征。 面部边缘权重根据由选择窗口选择的子图像图案中的像素的边界值，在检测到的图像上进行计算，以确定被检测图像的部分区域是否具有任何面部边缘权重， 边界 执行轮廓检测以分别用于具有弧段的子图像图案中的面部边界，用于具有面部边界的子图像图案。

公开(公告)号：US20120327232A1

公开(公告)日：2012-12-27

申请号：US13207857

申请日：2011-08-11

Applicant: Tai-Chang Yang ,  Yin-Pin Chang ,  Wen-Yan Chang ,  Hong-Long Chou

Inventor： Tai-Chang Yang ,  Yin-Pin Chang ,  Wen-Yan Chang ,  Hong-Long Chou

IPC: H04N7/18 ,  G06K9/00

CPC classification number: G06F3/011 ,  B60N2/0248 ,  B60N2002/0268 ,  B60N2002/0272 ,  B60R16/037 ,  B60R16/0373 ,  G06F3/017 ,  G06K9/00845

Abstract: An automobile equipment control system and a control method thereof. The system comprises an input module, a storage module, an image capturing module, a first recognizing module, a second recognizing module and a processing module. The input module is used to input a plurality of vehicle equipment setting values. The storage module is used to save a plurality of facial characteristic values and a plurality of setting values of vehicle equipment. The image capturing module is used to capture the image of the user. The first recognizing module analyzes the facial image in the image and extracts a facial characteristic point from the facial image. The second recognizing module analyzes the user's hand gesture, the user's head gestures or what the user speaks so as to decide which setting value of vehicle equipment is adopted to control the vehicle equipment.

Abstract translation: 一种汽车设备控制系统及其控制方法。 该系统包括输入模块，存储模块，图像捕获模块，第一识别模块，第二识别模块和处理模块。 输入模块用于输入多个车辆设备设定值。 存储模块用于保存车辆设备的多个面部特征值和多个设定值。 图像捕获模块用于捕获用户的图像。 第一识别模块分析图像中的面部图像并从面部图像提取面部特征点。 第二识别模块分析用户的手势，用户的头部手势或用户说话的方式，以决定采用车辆设备的设置值来控制车辆设备。

公开(公告)号：US08194979B2

公开(公告)日：2012-06-05

申请号：US12344326

申请日：2008-12-26

Applicant: Yin-Pin Chang ,  Hong-Long Chou ,  Yu-Chen Huang

Inventor： Yin-Pin Chang ,  Hong-Long Chou ,  Yu-Chen Huang

IPC: G06K9/00 ,  G06K9/40

CPC classification number: H04N9/646 ,  H04N1/60

Abstract: A method of correcting a false-color pixel in a digital image is used to correct a false-color pixel generated after being processed by a color correction matrix (CCM). The method includes the following steps. Obtaining a raw image, for being processed by the CCM operation to generate a processed image; comparing pixels of the raw image with that of the processed image according to a color difference ratio, to find out a false-color pixel in the pixels of the processed image; then, performing a color correction on the false-color pixel at the corresponding position of the raw image and the processed image. The false-color pixel is corrected according to the pixel color at the corresponding position of the raw image.

Abstract translation: 使用校正数字图像中的假色像素的方法来校正由颜色校正矩阵（CCM）处理之后生成的假色像素。 该方法包括以下步骤。 获取原始图像，以便由CCM操作处理以产生经处理的图像; 根据色差比对原始图像的像素与处理图像的像素进行比较，以找出处理图像的像素中的假色像素; 然后对原始图像和处理图像的对应位置处的伪色像素执行颜色校正。 根据原始图像的对应位置处的像素颜色来校正假色像素。

公开(公告)号：US07840088B2

公开(公告)日：2010-11-23

申请号：US11732249

申请日：2007-04-03

Applicant: Yu-Chen Huang ,  Yin-Pin Chang

Inventor： Yu-Chen Huang ,  Yin-Pin Chang

IPC: G06K9/00 ,  G06K9/40

CPC classification number: G06T5/002 ,  G06T2207/20012 ,  G06T2207/20028

Abstract: A method of reducing image noise is provided. In the method, during the process of reducing image noise by a bilateral filter, the image noise filtering intensity of a domain filter in the bilateral filter is adjusted according to whether a regional area of the image has a frame boundary, and the image noise filtering intensity of a range filter in the bilateral filter is adjusted through the intensity of a regional area, so as to improve the image noise reduction effect of the bilateral filter.

Abstract translation: 提供了降低图像噪声的方法。 在该方法中，在通过双边滤波器降低图像噪声的过程中，根据图像的区域是否具有帧边界来调整双边滤波器中的域滤波器的图像噪声滤波强度，并且图像噪声滤波 通过区域的强度调整双边滤波器中的范围滤波器的强度，从而提高双边滤波器的图像降噪效果。

公开(公告)号：US08532337B2

公开(公告)日：2013-09-10

申请号：US13077013

申请日：2011-03-31

Applicant: Yin-Pin Chang ,  Wen-Yan Chang ,  Hong-Long Chou

Inventor： Yin-Pin Chang ,  Wen-Yan Chang ,  Hong-Long Chou

IPC: G06K9/00

CPC classification number: G06T7/248 ,  G06T2207/10024

Abstract: An object tracking method includes steps of obtaining multiple first classifications of pixels within a first focus frame in a first frame picture, wherein the first focus frame includes an object to be tracked and has a first rectangular frame in a second frame picture; performing a positioning process to obtain a second rectangular frame; and obtaining color features of pixels around the second rectangular frame sequentially and establishing multiple second classifications according to the color feature. The established second classifications are compared with the first classifications sequentially to obtain an approximation value, compared with a predetermined threshold. The second rectangular frame is progressively adjusted, so as to establish a second focus frame. By analyzing color features of the pixels of the object and with a classification manner, the efficacy of detecting a shape and size of the object so as to update information of the focus frame is achieved.

Abstract translation: 对象跟踪方法包括以下步骤：在第一帧画面中获得第一焦点帧内的像素的多个第一分类，其中第一焦点帧包括要跟踪的对象，并且在第二帧画面中具有第一矩形帧; 执行定位处理以获得第二矩形框; 并且顺序地获得第二矩形框周围的像素的颜色特征，并根据颜色特征建立多个第二分类。 将所建立的第二分类顺序与第一分类进行比较，以获得与预定阈值相比的近似值。 逐渐调整第二矩形框架，以便建立第二焦点框架。 通过分析对象的像素的颜色特征和分类方式，实现检测对象的形状和大小以更新聚焦框的信息的功效。

公开(公告)号：US08085983B2

公开(公告)日：2011-12-27

申请号：US12344796

申请日：2008-12-29

Applicant: Yin-Pin Chang ,  Tai-Chang Yang ,  Hong-Long Chou

Inventor： Yin-Pin Chang ,  Tai-Chang Yang ,  Hong-Long Chou

IPC: G06K9/00

CPC classification number: G06K9/3233 ,  G06K9/00221

Abstract: A method of adjusting selected window size of an image object is applicable for tracking a target object in a video. The video includes a plurality of frames, and the target object has a display range changing with the playback of each frame. According to a variation trend of the display range of the target object, whether a variation times corresponding to the variation trend reaches a threshold value or not is recorded, and then the selected window size is reset, such that the target object is enclosed with a selected window having a size closer to the target object.

Abstract translation: 调整图像对象的选定窗口大小的方法适用于跟踪视频中的目标对象。 视频包括多个帧，并且目标对象具有随着每个帧的重放而改变的显示范围。 根据目标对象的显示范围的变化趋势，记录与变化趋势对应的变化时间是否达到阈值，然后重新设定所选择的窗口尺寸，使得目标对象被包围在 所选择的窗口具有更接近目标对象的尺寸。

公开(公告)号：US20100074479A1

公开(公告)日：2010-03-25

申请号：US12350378

申请日：2009-01-08

Applicant: Hong-Long CHOU ,  Tai-Chang YANG ,  Yin-Pin CHANG

Inventor： Hong-Long CHOU ,  Tai-Chang YANG ,  Yin-Pin CHANG

IPC: G06K9/00

CPC classification number: G06K9/6282 ,  G06K9/00221 ,  G06K9/6807

Abstract: A hierarchical face recognition training method and a hierarchical face recognition method thereof for performing a face feature recognition on an image under detection. The method includes a training process and a recognition process. The recognition method includes the steps. A plurality of training samples is obtained. The training samples are subdivided into a plurality of sub-image categories according to a plurality of angle intervals, and the training of a plurality of face features performs on a corresponding sub-image detector of each of the sub-image categories. The training measures performed repeatedly to generate sub-image categories at a sub-level of the sub-image categories. The training method includes the steps. An image under detection is loaded. A similarity of each of sub-image detectors compares according to the image under detection, and the sub-image detector having the highest similarity is selected. The face recognition measures performed repeatedly on the selected sub-image detector.

Abstract translation: 一种用于对检测到的图像执行面部特征识别的分层面部识别训练方法和分层面部识别方法。 该方法包括训练过程和识别过程。 识别方法包括步骤。 获得多个训练样本。 训练样本根据多个角度间隔被细分为多个子图像类别，并且多个脸部特征的训练在每个子图像类别的相应子图像检测器上执行。 重复执行的训练措施，以在子图像类别的子级别生成子图像类别。 训练方法包括步骤。 检测到的图像被加载。 每个子图像检测器的相似度根据检测到的图像进行比较，并且选择具有最高相似度的子图像检测器。 在所选择的子图像检测器上重复执行面部识别措施。

公开(公告)号：US20090169103A1

公开(公告)日：2009-07-02

申请号：US12344326

申请日：2008-12-26

Applicant: Yin-Pin CHANG ,  Hong-Long Chou ,  Yu-Chen Huang

Inventor： Yin-Pin CHANG ,  Hong-Long Chou ,  Yu-Chen Huang

IPC: G06K9/40 ,  G06K9/00

CPC classification number: H04N9/646 ,  H04N1/60

Abstract: A method of correcting a false-color pixel in a digital image is used to correct a false-color pixel generated after being processed by a color correction matrix (CCM). The method includes the following steps. Obtaining a raw image, for being processed by the CCM operation to generate a processed image; comparing pixels of the raw image with that of the processed image according to a color difference ratio, to find out a false-color pixel in the pixels of the processed image; then, performing a color correction on the false-color pixel at the corresponding position of the raw image and the processed image. The false-color pixel is corrected according to the pixel color at the corresponding position of the raw image.

Abstract translation: 使用校正数字图像中的假色像素的方法来校正由颜色校正矩阵（CCM）处理之后生成的假色像素。 该方法包括以下步骤。 获取原始图像，以便由CCM操作处理以产生经处理的图像; 根据色差比对原始图像的像素与处理图像的像素进行比较，以找出处理图像的像素中的假色像素; 然后对原始图像和处理图像的对应位置处的伪色像素执行颜色校正。 根据原始图像的对应位置处的像素颜色来校正假色像素。

公开(公告)号：US20090169066A1

公开(公告)日：2009-07-02

申请号：US12328891

申请日：2008-12-05

Applicant: Tai-Chang YANG ,  Hong-Long CHOU ,  Yin-Pin CHANG

Inventor： Tai-Chang YANG ,  Hong-Long CHOU ,  Yin-Pin CHANG

IPC: G06K9/00

CPC classification number: G06K9/00228

Abstract: A multidirectional face detection method is used to detect a face in a picture under detection at different positions. The face detection method includes the steps. A selecting window sets to sequentially select different sub-image patterns from the picture under detection. A facial weight calculates and it is calculated according to a feature value of the pixels in a sub-image pattern selected by the selecting window, thereby determining if the sub-image pattern has any features similar to the face. A facial edge weight calculates for made on the picture under detection according to a boundary value of the pixels in the sub-image pattern selected by the selecting window, so as to determine if the part of area of the picture under detection has any facial-boundaries. A profile detection performs to respectively mark the facial-boundaries in the sub-image patterns with a arc segments respectively for the sub-image patterns having the facial-boundaries.

Abstract translation: 多方向面部检测方法用于检测在不同位置检测到的图像中的脸部。 面部检测方法包括以下步骤。 选择窗口设置为从检测到的图片中顺序地选择不同的子图像图案。 面部重量计算，并且根据由选择窗口选择的子图像图案中的像素的特征值来计算，从而确定子图像图案是否具有与面部相似的任何特征。 面部边缘权重根据由选择窗口选择的子图像图案中的像素的边界值，在检测到的图像上进行计算，以确定被检测图像的部分区域是否具有任何面部边缘权重， 边界 轮廓检测执行分别用具有面部边界的子图像图案的弧段分别标记子图像图案中的面部边界。

公开(公告)号：US20090167900A1

公开(公告)日：2009-07-02

申请号：US12344783

申请日：2008-12-29

Applicant: Yin-Pin CHANG ,  Tai-Chang YANG ,  Hong-Long CHOU

Inventor： Yin-Pin CHANG ,  Tai-Chang YANG ,  Hong-Long CHOU

IPC: H04N5/262

CPC classification number: H04N5/2628 ,  G06T3/403

Abstract: An image processing method for adjusting a scale of a face is implemented by a microprocessor of a digital camera. The microprocessor performs a face detection to find the position of a face in the image, and sets a zoom block according to an entire column/row of the position of the face, so as to zoom in/out the scale of the face. When the scale of the face is adjusted, an entire column of pixels are added/deleted with constrain proportions from two sides of the zoom block, for repairing the pixels altered during the zoom in/out of the face so as to remain the whole image undistorted. This method finds the position of the face accurately, and adjusts the pixels at the position of the face, so as to effectively adjust the scale of the face, and avoid disproportion between the face and the body in the image.

Abstract translation: 用于调节脸部的刻度的图像处理方法由数字照相机的微处理器实现。 微处理器执行面部检测以找到图像中的脸部的位置，并且根据脸部位置的整个列/列来设置缩放块，以便放大/缩小脸部的比例。 当调整脸部的比例时，以缩放块两侧的约束比例添加/删除整列像素，以修复在放大/缩小脸部期间改变的像素，以保持整个图像 无失真 该方法精确地找到脸部的位置，并且调整脸部位置处的像素，以有效地调节面部的比例，并避免图像中脸部和身体之间的不成比例。