公开(公告)号：US20120133789A1

公开(公告)日：2012-05-31

申请号：US13012802

申请日：2011-01-25

申请人： Yi-Ta Wu ,  Hung-I Pai ,  Kung-Ming Lan

发明人： Yi-Ta Wu ,  Hung-I Pai ,  Kung-Ming Lan

IPC分类号： H04N5/225 ,  G06K9/46

CPC分类号： G06K9/4652 ,  G06K9/6223 ,  G06K9/6276

摘要： A data clustering method, a data clustering device using the same, and an image processing apparatus and a data processing apparatus equipped with the data clustering device are provided. The method includes sorting a plurality of data point to be clustered and generating a processing sequence based on the sorting, wherein each of the data point has at least one feature value. The method also includes using a non-iterative mechanism to cluster the data points into a plurality of data clusters according to the processing sequence. The method further includes optimizing the generated data clusters. Accordingly, the data clustering method can fast cluster the data points.

摘要翻译： 提供数据聚类方法，使用该数据聚类装置的数据聚类装置，以及配备有数据聚类装置的图像处理装置和数据处理装置。 该方法包括对要聚类的多个数据点进行排序，并基于排序生成处理序列，其中每个数据点具有至少一个特征值。 该方法还包括使用非迭代机制根据处理顺序将数据点聚类成多个数据簇。 该方法还包括优化所生成的数据集群。 因此，数据聚类方法可以快速聚集数据点。

公开(公告)号：US20130136350A1

公开(公告)日：2013-05-30

申请号：US13612845

申请日：2012-09-13

申请人： Hung-I Pai ,  Yi-Ta Wu ,  Kung-Ming Lan

发明人： Hung-I Pai ,  Yi-Ta Wu ,  Kung-Ming Lan

IPC分类号： G06K9/68

CPC分类号： G06T7/254 ,  G06T7/11 ,  G06T7/194 ,  G06T2207/10024 ,  G06T2207/30196 ,  G06T2207/30232 ,  G06T2207/30236

摘要： An apparatus and a method for fast foreground detection are provided. A foreground characteristic value calculation module calculates a foreground characteristic value by using an image intensity of each pixel in a target image and a background intensity of a corresponding pixel in a background model. A first filter determines a first threshold and a second threshold according to at least one scenic factor for capturing the target image and filters out non-foreground pixels having their foreground characteristic value between the first threshold and the second threshold from pixels in the target image. A second filter determines an image difference condition and a chroma similarity condition according to the scenic factor and filters out non-foreground pixels having their foreground characteristic value satisfying the image difference condition and the chroma similarity condition from pixels left by the first filter. Pixels left by the second filter are served as foreground pixels.

摘要翻译： 提供了一种用于快速前景检测的装置和方法。 前景特征值计算模块通过使用目标图像中的每个像素的图像强度和背景模型中相应像素的背景强度来计算前景特征值。 第一滤波器根据用于捕获目标图像的至少一个景观因素确定第一阈值和第二阈值，并从目标图像中的像素滤除具有其第一阈值和第二阈值之间的其前景特征值的非前景像素。 第二滤波器根据风景因素确定图像差异条件和色度相似度条件，并且从第一滤波器剩余的像素滤除具有满足图像差异条件和色度相似度条件的前景特征值的非前景像素。 第二个滤镜留下的像素作为前景像素。

公开(公告)号：US09058540B2

公开(公告)日：2015-06-16

申请号：US13012802

申请日：2011-01-25

申请人： Yi-Ta Wu ,  Hung-I Pai ,  Kung-Ming Lan

发明人： Yi-Ta Wu ,  Hung-I Pai ,  Kung-Ming Lan

IPC分类号： H04N5/225 ,  G06K9/46 ,  G06K9/62

CPC分类号： G06K9/4652 ,  G06K9/6223 ,  G06K9/6276

摘要： A data clustering method, a data clustering device using the same, and an image processing apparatus and a data processing apparatus equipped with the data clustering device are provided. The method includes sorting a plurality of data point to be clustered and generating a processing sequence based on the sorting, wherein each of the data point has at least one feature value. The method also includes using a non-iterative mechanism to cluster the data points into a plurality of data clusters according to the processing sequence. The method further includes optimizing the generated data clusters. Accordingly, the data clustering method can fast cluster the data points.

摘要翻译： 提供数据聚类方法，使用该数据聚类装置的数据聚类装置，以及配备有数据聚类装置的图像处理装置和数据处理装置。 该方法包括对要聚类的多个数据点进行排序，并基于排序生成处理序列，其中每个数据点具有至少一个特征值。 该方法还包括使用非迭代机制根据处理顺序将数据点聚类成多个数据簇。 该方法还包括优化所生成的数据集群。 因此，数据聚类方法可以快速聚集数据点。

公开(公告)号：US08712149B2

公开(公告)日：2014-04-29

申请号：US13612845

申请日：2012-09-13

申请人： Hung-I Pai ,  Yi-Ta Wu ,  Kung-Ming Lan

发明人： Hung-I Pai ,  Yi-Ta Wu ,  Kung-Ming Lan

IPC分类号： G06K9/00

CPC分类号： G06T7/254 ,  G06T7/11 ,  G06T7/194 ,  G06T2207/10024 ,  G06T2207/30196 ,  G06T2207/30232 ,  G06T2207/30236

摘要： An apparatus and a method for fast foreground detection are provided. A foreground characteristic value calculation module calculates a foreground characteristic value by using an image intensity of each pixel in a target image and a background intensity of a corresponding pixel in a background model. A first filter determines a first threshold and a second threshold according to at least one scenic factor for capturing the target image and filters out non-foreground pixels having their foreground characteristic value between the first threshold and the second threshold from pixels in the target image. A second filter determines an image difference condition and a chroma similarity condition according to the scenic factor and filters out non-foreground pixels having their foreground characteristic value satisfying the image difference condition and the chroma similarity condition from pixels left by the first filter. Pixels left by the second filter are served as foreground pixels.

摘要翻译： 提供了一种用于快速前景检测的装置和方法。 前景特征值计算模块通过使用目标图像中的每个像素的图像强度和背景模型中相应像素的背景强度来计算前景特征值。 第一滤波器根据用于捕获目标图像的至少一个景观因素确定第一阈值和第二阈值，并从目标图像中的像素滤除具有其第一阈值和第二阈值之间的其前景特征值的非前景像素。 第二滤波器根据风景因素确定图像差异条件和色度相似度条件，并且从第一滤波器剩余的像素滤除具有满足图像差异条件和色度相似度条件的前景特征值的非前景像素。 第二个滤镜留下的像素作为前景像素。

公开(公告)号：US20120275524A1

公开(公告)日：2012-11-01

申请号：US13096691

申请日：2011-04-28

申请人： Cheng-Chang LIEN ,  En-Jung Fam ,  Yue-Min Jiang ,  Hung-I Pai ,  Kung-Ming Lan

发明人： Cheng-Chang LIEN ,  En-Jung Fam ,  Yue-Min Jiang ,  Hung-I Pai ,  Kung-Ming Lan

IPC分类号： H04N7/26

CPC分类号： H04N19/48 ,  G06T7/246 ,  G06T2207/10016 ,  G06T2207/20021 ,  H04N19/139 ,  H04N19/146 ,  H04N19/17 ,  H04N19/176 ,  H04N19/18

摘要： Methods and systems are disclosed for processing compressed video images. A processor detects a candidate object region from the compressed video images. The candidate object region includes a moving object and a shadow associated with the moving object. For each data block in the candidate object region, the processor calculates an amount of encoding data used to encode temporal changes in the respective data block. The processor then identifies the shadow in the candidate object region composed of data blocks each having the amount of encoding data below a threshold value.

摘要翻译： 公开了用于处理压缩视频图像的方法和系统。 处理器从压缩视频图像中检测候选对象区域。 候选对象区域包括移动对象和与移动对象相关联的阴影。 对于候选对象区域中的每个数据块，处理器计算用于对相应数据块中的时间变化进行编码的编码数据量。 然后，处理器识别由每个具有低于阈值的编码数据量的数据块组成的候选对象区域中的阴影。

公开(公告)号：US09001206B2

公开(公告)日：2015-04-07

申请号：US13214415

申请日：2011-08-22

申请人： Shen-Zheng Wang ,  San-Lung Zhao ,  Hung-I Pai ,  Kung-Ming Lan ,  En-Jung Farn

发明人： Shen-Zheng Wang ,  San-Lung Zhao ,  Hung-I Pai ,  Kung-Ming Lan ,  En-Jung Farn

IPC分类号： G08B13/196 ,  G08B29/04

CPC分类号： G08B13/196 ,  G08B29/046

摘要： A cascadable camera tampering detection transceiver module has a processing unit and a storing unit, an information controlling module and an analyzing module. The storing unit stores a transceiving module. The detection module analyzes input video, detects camera tampering events, synthesizes the input video with the image of camera tampering result, and outputs the synthesized video. When the input video is an output from the detection module, the detection module separates the camera tampering result from the input video, and the result can be used to simplify or enhance the subsequent video analysis. Performing the existing analysis repeatedly may be avoided, and the user may re-define the detection conditions in this manner. When the camera tampering result is transmitted in the video channel, the detection module transmits the camera tampering result, and hence the detection module may be used in combination with surveillance devices having image output or input interfaces.

摘要翻译： 可级联摄像机篡改检测收发模块具有处理单元和存储单元，信息控制模块和分析模块。 存储单元存储收发模块。 检测模块分析输入视频，检测相机篡改事件，使用相机篡改结果的图像合成输入视频，并输出合成视频。 当输入视频是检测模块的输出时，检测模块将相机篡改结果与输入视频分离，结果可用于简化或增强后续视频分析。 可以避免重复执行现有分析，并且用户可以以这种方式重新定义检测条件。 当在视频通道中发送相机篡改结果时，检测模块发送相机篡改结果，因此检测模块可以与具有图像输出或输入接口的监视设备结合使用。

公开(公告)号：US08494218B2

公开(公告)日：2013-07-23

申请号：US12703352

申请日：2010-02-10

申请人： Yi-Yuan Chen ,  Kung-Ming Lan ,  Hung-I Pai ,  Jen-Hui Chuang ,  Chii-Yah Yuan

发明人： Yi-Yuan Chen ,  Kung-Ming Lan ,  Hung-I Pai ,  Jen-Hui Chuang ,  Chii-Yah Yuan

IPC分类号： G06K9/00 ,  H04N7/00

CPC分类号： G06K9/2027

摘要： A light information receiving method, a method and a unit for the recognition of light-emitting objects are provided. The light information receiving method includes the following steps. A light-emitting object array is captured to obtain a plurality of images, wherein the light-emitting object array includes at least one light-emitting object. A temporal filtering process is performed to the images to recognize a light-emitting object. A light-emitting status of the light-emitting object array is recognized according to the light-emitting object location. A decoding process is performed according to the light-emitting status to output an item of information.

摘要翻译： 提供了一种用于识别发光物体的光信息接收方法，方法和单元。 光信息接收方法包括以下步骤。 捕获发光物体阵列以获得多个图像，其中发光物体阵列包括至少一个发光物体。 对图像执行时间滤波处理以识别发光物体。 根据发光对象位置识别发光对象阵列的发光状态。 根据发光状态进行解码处理，输出信息。

公开(公告)号：US20120154581A1

公开(公告)日：2012-06-21

申请号：US13214415

申请日：2011-08-22

申请人： Shen-Zheng Wang ,  San-Lung Zhao ,  Hung-I Pai ,  Kung-Ming Lan ,  En-Jung Farn

发明人： Shen-Zheng Wang ,  San-Lung Zhao ,  Hung-I Pai ,  Kung-Ming Lan ,  En-Jung Farn

IPC分类号： H04N7/18

CPC分类号： G08B13/196 ,  G08B29/046

摘要： A cascadable camera tampering detection transceiver module has a processing unit and a storing unit, an information controlling module and an analyzing module. The storing unit stores a transceiving module. The detection module analyzes input video, detects camera tampering events, synthesizes the input video with the image of camera tampering result, and outputs the synthesized video. When the input video is an output from the detection module, the detection module separates the camera tampering result from the input video, and the result can be used to simplify or enhance the subsequent video analysis. Performing the existing analysis repeatedly may be avoided, and the user may re-define the detection conditions in this manner. When the camera tampering result is transmitted in the video channel, the detection module transmits the camera tampering result, and hence the detection module may be used in combination with surveillance devices having image output or input interfaces.

摘要翻译： 可级联摄像机篡改检测收发模块具有处理单元和存储单元，信息控制模块和分析模块。 存储单元存储收发模块。 检测模块分析输入视频，检测相机篡改事件，使用相机篡改结果的图像合成输入视频，并输出合成视频。 当输入视频是检测模块的输出时，检测模块将相机篡改结果与输入视频分离，结果可用于简化或增强后续视频分析。 可以避免重复执行现有分析，并且用户可以以这种方式重新定义检测条件。 当在视频通道中发送相机篡改结果时，检测模块发送相机篡改结果，因此检测模块可以与具有图像输出或输入接口的监视设备结合使用。

公开(公告)号：US20110044500A1

公开(公告)日：2011-02-24

申请号：US12703352

申请日：2010-02-10

申请人： Yi-Yuan Chen ,  Kung-Ming Lan ,  Hung-I Pai ,  Jen-Hui Chuang ,  Chii-Yah Yuan

发明人： Yi-Yuan Chen ,  Kung-Ming Lan ,  Hung-I Pai ,  Jen-Hui Chuang ,  Chii-Yah Yuan

IPC分类号： G06K9/00

CPC分类号： G06K9/2027

摘要： A light information receiving method, a method and a unit for the recognition of light-emitting objects are provided. The light information receiving method includes the following steps. A light-emitting object array is captured to obtain a plurality of images, wherein the light-emitting object array includes at least one light-emitting object. A temporal filtering process is performed to the images to recognize a light-emitting object. A light-emitting status of the light-emitting object array is recognized according to the light-emitting object location. A decoding process is performed according to the light-emitting status to output an item of information.

摘要翻译： 提供了一种用于识别发光物体的光信息接收方法，方法和单元。 光信息接收方法包括以下步骤。 捕获发光物体阵列以获得多个图像，其中发光物体阵列包括至少一个发光物体。 对图像执行时间滤波处理以识别发光物体。 根据发光对象位置识别发光对象阵列的发光状态。 根据发光状态进行解码处理，输出信息。

公开(公告)号：US20110128388A1

公开(公告)日：2011-06-02

申请号：US12754617

申请日：2010-04-06

申请人： Hung-I Pai ,  Shang-Chih Hung ,  Chii-Yah Yuan ,  Yi-Yuan Chen ,  Kung-Ming Lan

发明人： Hung-I Pai ,  Shang-Chih Hung ,  Chii-Yah Yuan ,  Yi-Yuan Chen ,  Kung-Ming Lan

IPC分类号： H04N17/00

CPC分类号： G01B11/03 ,  G01B21/042 ,  G06T7/80 ,  G06T2207/30204 ,  G06T2207/30232

摘要： A camera calibration system including a coordinate data generation device and a coordinate data recognition device is provided. The coordinate data generation device generates a plurality of map coordinate data corresponding to a plurality of real positions in a real scene. The coordinate data recognition device receives an image plane of the real scene from a camera to be calibrated and receives the map coordinate data from the coordinate data generation device. Besides, the coordinate data recognition device recognizes image positions corresponding to the real positions in the image plane and calculates image coordinate data corresponding to the image positions. Moreover, the coordinate data recognition device calculates a coordinate transform matrix corresponding to the camera according to the image coordinate data and the map coordinate data. Thereby, the camera calibration system can finish the calibration of the camera quickly.

摘要翻译： 提供了包括坐标数据生成装置和坐标数据识别装置的相机校准系统。 坐标数据生成装置生成与真实场景中的多个实际位置对应的多个地图坐标数据。 坐标数据识别装置从要校准的摄像机接收真实场景的图像平面，并从坐标数据生成装置接收地图坐标数据。 此外，坐标数据识别装置识别与图像平面中的实际位置相对应的图像位置，并且计算与图像位置相对应的图像坐标数据。 此外，坐标数据识别装置根据图像坐标数据和地图坐标数据来计算与照相机对应的坐标变换矩阵。 因此，相机校准系统可以快速完成相机的校准。