公开(公告)号：US07027637B2

公开(公告)日：2006-04-11

申请号：US10080805

申请日：2002-02-21

Applicant: Tong Fang ,  Ming Fang

Inventor： Tong Fang ,  Ming Fang

IPC: G06K9/00

CPC classification number: G01R1/0483 ,  G01R31/2889 ,  G06T7/0004 ,  G06T7/73 ,  G06T7/77 ,  G06T2207/30141 ,  G06T2207/30148

Abstract: A method for determining a number of balls in a projection space comprises determining a projection of a portion of a ball grid array, determining at least one local maximum of the projection space for a given threshold, and determining at least a distance between adjacent maximum. The method further comprises determining an inter-peak histogram of the distances, determining an inter-ball distance for each pair of adjacent balls that has the maximum value of the inter-peak distance histogram corresponding to the pair of adjacent balls, and determining a position of a first ball and a position of a last ball. The method comprises verifying the position of the first ball and the position of the last ball based on a general inter-ball distance, and determining the number of balls.

公开(公告)号：US06996295B2

公开(公告)日：2006-02-07

申请号：US10125200

申请日：2002-04-18

Applicant: Jenn-Kwei Tyan ,  Ming Fang

Inventor： Jenn-Kwei Tyan ,  Ming Fang

IPC: G06K9/36 ,  G06F7/00

CPC classification number: G06K9/00476 ,  G06F17/30253 ,  G06K9/2054 ,  G06K2209/01 ,  Y10S707/99931 ,  Y10S707/99933

Abstract: A system and method for electronic archival of paper-based technical drawings includes a system having a processor, a form learning unit in signal communication with the processor for learning the form of a model drawing image, a form localization unit in signal communication with the processor for localizing the form of an input drawing image, an optical character recognition unit in signal communication with the processor for optically recognizing identification text of the input drawing image, and a result verification unit in signal communication with the processor for verifying the results of the recognized identification text of the input drawing image; and further includes a method for providing a model drawing image, learning the form of the provided model drawing image, receiving an input drawing image, localizing the form of the input drawing image, optically recognizing identification text of the input drawing image, verifying the results of the recognized identification text of the input drawing image, and storing the input drawing image and the corresponding verified identification text into a drawing database.

公开(公告)号：US06744933B2

公开(公告)日：2004-06-01

申请号：US09764586

申请日：2001-01-18

Applicant: Shang-Hong Lai ,  Ming Fang

Inventor： Shang-Hong Lai ,  Ming Fang

IPC: G06K932

CPC classification number: G06T7/001 ,  G06T7/37 ,  G06T2207/30108

Abstract: A method for matching images includes the step of providing a template image and an input image. A Laplacian-of-Gaussian filtered log (LOG-log) image function is computed with respect to the template image and the input image to obtain a Laplacian-of-Gaussian filtered template image and a Laplacian-of-Gaussian filtered input image, respectively. An energy function is minimized to determine estimated geometric transformation parameters and estimated photometric parameters for the input image with respect to the template image. The energy function is formed by weighting non-linear least squared differences of data constraints corresponding to locations of both the Laplacian-of-Gaussian filtered template image and the Laplacian-of-Gaussian filtered input image. The estimated geometric transformation parameters and the estimated photometric parameters are output for further processing. The method allows for image matching under non-uniform illumination variations.

Abstract translation: 用于匹配图像的方法包括提供模板图像和输入图像的步骤。 计算相对于模板图像和输入图像的拉普拉斯高斯滤波日志（LOG-log）图像函数，以分别获得拉普拉斯高斯滤波模板图像和拉普拉斯高斯滤波输入图像 。 能量函数被最小化以确定关于模板图像的输入图像的估计的几何变换参数和估计的测光参数。 能量函数是通过加权对应于高斯拉普拉斯高斯滤波模板图像和拉普拉斯高斯滤波输入图像的位置的数据约束的非线性最小平方差来形成的。 输出估计的几何变换参数和估计的光度参数用于进一步处理。 该方法允许在不均匀照明变化下的图像匹配。

公开(公告)号：US06546137B1

公开(公告)日：2003-04-08

申请号：US09236725

申请日：1999-01-25

Applicant: Shang-Hong Lai ,  Ming Fang

Inventor： Shang-Hong Lai ,  Ming Fang

IPC: G06K964

CPC classification number: G06T7/0004 ,  G06K9/6203 ,  G06T7/74

Abstract: A fast localization with advanced search hierarchy system for fast and accurate object localization in a large search space is based on an assumption that surrounding regions of a pattern within a search range are always fixed. The FLASH system comprises a hierarchical nearest-neighbor search system and an optical-flow based energy minimization system. The hierarchical nearest-neighbor search system produces rough estimates of the transformation parameters for the optical-flow based energy minimization system which provides very accurate estimation results and associated confidence measures.

Abstract translation: 利用高级搜索层次系统的快速定位，可以在大型搜索空间中快速准确地对对象定位，这是基于搜索范围内的图案的周围区域始终是固定的假设。 FLASH系统包括分层最近邻搜索系统和基于光流的能量最小化系统。 分层最近邻搜索系统产生基于光流的能量最小化系统的变换参数的粗略估计，其提供非常精确的估计结果和相关置信度度量。

公开(公告)号：US06445812B1

公开(公告)日：2002-09-03

申请号：US09235997

申请日：1999-01-22

Applicant: Shang-Hong Lai ,  Ming Fang

Inventor： Shang-Hong Lai ,  Ming Fang

IPC: G06K900

CPC classification number: G06T7/001 ,  G06K9/2036 ,  G06T5/008 ,  G06T2207/10152 ,  G06T2207/20064 ,  G06T2207/30164

Abstract: An illumination compensation system for correcting smooth intensity variations due to illumination changes is based on an assumption that an underlining image reflectance function is approximately a piecewise constant and that an image irradiance function is spatially smooth. The system first takes the logarithm of an image brightness function. Gradient constraints are then computed using a finite difference. Reliable gradient constraints are selected based on a local uniformity test. A process is subsequently applied to estimate the logarithmic irradiance function. A logarithmic irradiance function is subtracted from the logarithmic image brightness function and an exponential operation of the above subtracted image function is taken and an illumination compensated image is outputted from the system.

Abstract translation: 用于校正由于照明变化引起的平滑强度变化的照明补偿系统基于下划线图像反射函数近似分段常数并且图像辐照度函数在空间上平滑的假设。 系统首先采用图像亮度函数的对数。 然后使用有限差分计算梯度约束。 基于局部均匀性测试选择可靠的梯度约束。 随后应用过程来估计对数辐照度函数。 从对数图像亮度函数中减去对数辐照函数，并采用上述减法图像函数的指数运算，并从系统输出照明补偿图像。

公开(公告)号：USD441132S1

公开(公告)日：2001-04-24

申请号：US29128891

申请日：2000-08-31

Applicant: Zheng Xiong Xie ,  Ming Fang

Designer： Zheng Xiong Xie ,  Ming Fang

公开(公告)号：USD439698S1

公开(公告)日：2001-03-27

申请号：US29128836

申请日：2000-08-30

Applicant: Ming Fang

Designer： Ming Fang

公开(公告)号：US06208138B1

公开(公告)日：2001-03-27

申请号：US09095792

申请日：1998-06-11

Applicant: Shang-Hong Lai ,  Ming Fang

Inventor： Shang-Hong Lai ,  Ming Fang

IPC: G01V300

CPC classification number: G01R33/3875

Abstract: An intensity inhomogeneity correction system includes a log operator which performs a log operation on the intensity values of the original MR image. A regular grid generator partitions the image domain and then a compute and select establishes reliable orientation constraints at grid locations. A bias field surface reconstructor reconstructs a bias field surface from selected orientation constraints with a thin-plate spline by using a preconditioned conjugate gradient. An intensity inhomogeneities remover subtracts an estimated bias function from the original log image and an exponential operation on the bias-field corrected log image provides a corrected image.

Abstract translation: 强度不均匀性校正系统包括对原始MR图像的强度值执行对数运算的对数运算符。 常规网格生成器分割图像域，然后计算和选择在网格位置建立可靠的方向约束。 偏置场表面重建器通过使用预条件共轭梯度，用薄板样条从选定的取向约束重构偏置场表面。 强度不均匀性去除器从原始对数图像中减去估计偏差函数，并且偏置场校正日志图像上的指数运算提供校正图像。

公开(公告)号：US06175643B1

公开(公告)日：2001-01-16

申请号：US08993230

申请日：1997-12-18

Applicant: Shang-Hong Lai ,  Ming Fang

Inventor： Shang-Hong Lai ,  Ming Fang

IPC: G06K900

CPC classification number: G06T3/4046 ,  G06T3/4053

Abstract: An adaptive hierarchical neural network based system with online adaptation capabilities has been developed to automatically adjust the display window width and center for MR images. Our windowing system possesses the online training capabilities that make the adaptation of the optimal display parameters to personal preference as well as different viewing conditions possible. The online adaptation capabilities are primarily due to the use of the hierarchical neural networks and the development of a new width/center mapping system. The large training image set is hierarchically organized for efficient user interaction and effective re-mapping of the width/center settings in the training data set. The width/center values are modified in the training data through a width/center mapping function, which is estimated from the new width/center values of some representative images adjusted by the user. The width/center mapping process consists of a global spline mapping for the entire training images as well as a first-order polynomial sequence mapping for the image sequences selected in the user's new adjustment procedure.

Abstract translation: 已经开发了具有在线适应能力的基于自适应层次神经网络的系统，以自动调整MR图像的显示窗口宽度和中心。 我们的窗口系统具有在线培训功能，可以使最佳显示参数适应个人偏好以及可能的不同观看条件。 在线适应能力主要是由于使用分层神经网络和开发新的宽度/中心映射系统。 大型训练图像集被分层组织，用于有效的用户交互，并有效地重新映射训练数据集中的宽度/中心设置。 通过宽度/中心映射函数在训练数据中修改宽度/中心值，该宽度/中心映射函数根据由用户调整的一些代表性图像的新的宽度/中心值估计。 宽度/中心映射过程由用于整个训练图像的全局样条映射以及在用户的新调整过程中选择的图像序列的一阶多项式序列映射组成。

公开(公告)号：US5963676A

公开(公告)日：1999-10-05

申请号：US796331

申请日：1997-02-07

Applicant: Zhenyu Wu ,  Ming Fang

Inventor： Zhenyu Wu ,  Ming Fang

IPC: G06T5/00 ,  G06K9/40

CPC classification number: G06T5/009 ,  G06T5/002 ,  G06T2207/10116 ,  G06T2207/20016 ,  G06T2207/30101

Abstract: A multi-scale approach to enhancement is taken where enhancement masks are generated at different scales and then combined together using a pyramid scheme. Each mask is computed from applying directional sensitive Laplacian kernels, which is responsible for extracting images of low contrast, followed by an adaptive non-linear mapping. This non-linearity is crucial in virtually eliminating edge overshoots. Furthermore, the input image may also be preprocessed by an edge-preserving smoothing filter to reduce the noise effect.

Abstract translation: 在增强掩模以不同尺度生成然后使用金字塔方案组合在一起的情况下，采用多尺度增强方法。 每个掩码都是通过应用方向敏感的拉普拉斯算子来计算的，它们负责提取低对比度的图像，然后是自适应非线性映射。 这种非线性在几乎消除边缘过冲中至关重要。 此外，输入图像也可以由边缘保持平滑滤波器预处理，以减少噪声效应。