Industrial Technology Research Institute

Industrial Technical Research Institute

Guo-Zua Wu

Zu-Wen Chao

Yi-Jung Wang

Meng-Han Tsai

Chih-Chi Chang

Der-Ray Huang

Oscal Tzyh-Chiang Chen

Chien-Ju Li

Wei-Cheng Chen

Jau-Jiu Ju

Wen-Hsing Hsu

Wei-Zheng Lu

Kun-Ta Wu

Jian-Ru Li

Chien-Hsin Huang

Kuo-Tung Tiao

Rung-Ywan Tsai

Yuan-Chin Lee

Tai-Ting Huang

Nai-Heng Tseng

Ji-Da Chen

Chang-Lin Hu

Hsin-Han Shen

Shi-Shen Wang

Wei-Cheng Lu

Sung-Bo Chen

Yuan-Mao Hung

Yi-Chun Wang

Shu-Ching Chen

Kuei-Hung Cheng

Chwan-Chan Chen

Shin-Hui Huang

I-Yen Lin

Ken-Tsu Hu

Oscal Tzyh Chiang Chen

Chwan-Chang Chen

Kun-Da Wu

Chia-Yang Tsai

S Sam Tsai

Jheng-You Lin

Hsueh-Maing Hang

Ti-Hao Chiang

Yin-Tsung Hwang

An ultrasonic imaging device includes an ultrasonic generating unit and an ultrasonic imaging processing unit. The ultrasonic generating unit repeatedly turns on N of M ultrasonic array elements of the ultrasonic probe multiple times as a group of array elements for linear scanning, and each scan is to emit an ultrasonic signal by each group of array elements and receive an echo signal of the ultrasonic signal. The ultrasonic imaging processing unit extracts a central echo signal from the echo signal in each scan to form a channel signal, and according to the arrival time of the echo signals, each central echo signal in the channel signal is delayed and summed to generate a modified channel signal. The modified channel signal is subjected to image synthesis process to obtain an ultrasonic image.

Ultrasonic imaging device and imaging method thereof

A method for detecting vascular obstruction and a system using the same are provided. The method includes steps of: detecting a blood vessel through a probe to generate a reference signal before the blood vessel is obstructed, wherein the probe is configured to transmit or receive ultrasonic waves; detecting the blood vessel through the probe to generate a detection signal; performing Fourier transformation on the reference signal to generate a reference power spectrum, and performing Fourier transformation on the detection signal to generate a detection power spectrum; transforming the reference power spectrum into a reference time-frequency spectrogram, and transforming the detection power spectrum into a detection time-frequency spectrogram; judging a similarity between the reference time-frequency spectrogram and the detection time-frequency spectrogram, and judging whether the blood vessel is obstructed or not according to the similarity.

Method for detecting vascular obstruction and system using the same

An ultrasound imaging system includes a beam receiving circuit and a back-end circuit. The beam receiving circuit receives a plurality of digitized echo signals. The back-end circuit is coupled to the beam receiving circuit for outputting a plurality of compressed delay timing parameters corresponding to a plurality of channels to the beam receiving circuit. The beam receiving circuit decompresses the compressed delay timing parameters into a plurality of delay timing parameters, and processes the digitized echo signals into an ultrasound beamforming value according to the delay timing parameters corresponding to the channels. The back-end circuit synthesizes an ultrasound image according to the ultrasound beamforming value outputted from the beam receiving circuit.

Ultrasound imaging system and method thereof, and beamforming circuit

A method of image reconstruction is provided. When an object is not placed in a measurement space, first intensity images of an electromagnetic wave passing through the measurement space and corresponding to photon energy levels are measured. When the object is placed in the measurement space, second intensity images of the electromagnetic wave passing through the object and corresponding to photon energy levels are measured. In a database, data including an attenuation coefficient of each substance respectively having components irradiated by the electromagnetic wave corresponding to each photon energy level and a thickness of each substance in a transmission direction of the electromagnetic wave corresponding to each photon energy level are provided. Attenuation images of the object respectively corresponding to the components are calculated according to the data and the first and second intensity images. An image reconstruction system and a method and system of image construction are also provided.

Method and system of image reconstruction and method and system of image construction

The disclosure provides an ultrasound image registration apparatus and a method thereof suitable for registering two ultrasound images partially overlapped with each other. The apparatus comprises: a first-stage image-developing processing module, a second-stage image-developing processing module and a registration module. The first-stage image-developing processing module performs beam-forming processing on two ultrasound images so as to generate two raw images. The second-stage image-developing processing module connects the first-stage module for performing envelope detection processing and compression processing on the raw images so as to generate two developed ultrasound images. The registration module connects the two image-developing processing modules for respectively obtaining coordinate information of at least one feature point of the raw images as the initial values of an image registration procedure, and for performing speckles-reducing processing on the developed ultrasound images and using the speckles-reduced ultrasound images to perform the image registration procedure.

Ultrasound image registration apparatus and method thereof

An image processing method and a system for processing the same are provided. The image processing method includes the following steps. A first image having several first areas and a second image having several second areas are provided. Each first area has a first feature point having the largest or the smallest grey value in the first area. Each second area has a second feature point having the largest or the smallest grey value in the second area. A first relationship between the first feature points and a second relationship between the second feature points are created. The first and the second feature points are paired by a microprocessor according to the first and the second relationship.

Method for processing image and system processing the same

An optical equipment for inspecting and addressing a specimen is disclosed. The optical equipment comprises an optical device and a processing module. The optical device comprises a light source, a sample inspecting device and an address detecting device. The sample inspecting device comprises a first objective lens and a first detector. A beam is focused on a sample placed in an inspected site of a specimen by the first objective lens. The address detecting device comprises a second objective lens and a second detector. A beam is focused on the address coding site by the second objective lens. The processing module controls the beam to be focused on the sampling points of the inspected site to generate first optical signals, and simultaneously controls the beam of the light source to be focused on the corresponding address codes of the address coding site to generate second optical signals.

Optical equipment and registration method

A coding mode selection method for selecting a coding mode for motion vector estimation is disclosed. The method includes the following steps. First, a first coding mode is selected to perform motion vector estimation for a macroblock having M sub-blocks. M motion vectors corresponding to the M sub-blocks are separately calculated. It is determined whether correlations for the M motion vectors match to a predetermined condition. If not, the first coding mode is selected and the M motion vectors are utilized to perform a motion vector estimation; and if so, a second coding mode is selected and the N motion vectors corresponding to the second coding mode are utilized to perform the motion vector estimation, wherein the N motion vectors are generated by using the M motion vectors and 1&#x2266;N&#x3c;M.

Motion vector coding mode selection method and coding mode selection apparatus and machine readable medium using the same

A non-volatile data storage system including a first non-volatile storage medium, a second non-volatile storage medium, and a microprocessor is provided. The first non-volatile storage medium includes a popular data address recording area for recording logic addresses of popular data in the first non-volatile storage medium. The microprocessor is coupled to the first non-volatile storage medium and the second non-volatile storage medium. When the non-volatile data storage system boots up, the microprocessor copies the popular data from the first non-volatile storage medium to the second non-volatile storage medium according to the popular data address recording area. The popular data is accessed in the second non-volatile storage medium instead of the first non-volatile storage medium.

Non-volatile data storage system and method thereof

An integrated transformation apparatus is provided. The apparatus includes a first multiplexer, a second multiplexer, and a transformation unit. The first multiplexer retrieves point data from columns or rows of a multi-dimensional matrix and input data. The second multiplexer retrieves transformation coefficients corresponding to the point data. The transformation unit transforms data blocks of the multi-dimensional matrix to a plurality of sub data blocks according to the input data, the point data, and the transformation coefficients.

Integrated conversion method and apparatus

The present invention provides a medium for storing a subtitle data structure of an audio/video program and a method for displaying the same. The subtitle data structure of the audio/video program includes a plurality of character packages and a background image package, wherein each character package further has a header that includes the start and the end of a presenting time stamp (PTS), X-Y coordinates of the character, X length, Y length and a color (or colors) of the character, as well as pixel data for displaying the character. Compared with prior art, using the subtitle data structure of the present invention can considerably save the required medium capacity for storing the subtitle data.

Medium for storing subtitle data structure of an audio/ video program and method for displaying the same

A method for processing video image is provided to perform a motion vector filtering process on the image after being processed by a motion estimation process. The motion vector filtering process adjusts the motion vectors derived for each block in a video sequence, so that the motion vector of each block and the actual moving direction of the corresponding object in the image tend to conformity. Then, the blocks with similar motion vectors are merged and blocks with radical motion vectors are corrected, thereby reducing distortion of the output image and the number of bits used to represent the motion information, which in turns increases the efficiency of compression process.

Method for processing motion information

A watermark coding method and its recording medium utilize image interleaving and wavelet-based transformation to enhance the robustness of the image. The watermark series to be inserted is converted, encrypted, and modulated to increase the difficulty in decoding and restoring. Finally, the watermark is distributed and hidden in various frequency bands of the image in a frequency hopping way, generating a robust watermarked image. When used in a recording medium, side information produced during coding is recorded in the lead-in area of the recording medium to reduce the chances of illegal copying.

Watermark coding method and recording medium thereof

A method for encoding images according to object shapes is provided. The method enables reduced usage of storage space when an image encoding system encodes a received image frame. The method uses a bottom-up and sequential approach for encoding coefficients. The coefficients in each subband are numbered using a tree structure, wherein level N subband is designated as a root of the whole tree, level (N-1) subbands are designated as roots of subtrees, and level (N=1) subbands are designated as terminal nodes of subtrees. The terminal nodes for one of the subtrees are numbered first, and then this subtree is numbered in the bottom-up sequence. After that, the terminal nodes for another subtree are numbered, and this numbering sequence continues to number all the other subtrees until the root of the tree has been numbered, so as to allow a bottom-up encoding process for the coefficient to be performed.

Method for encoding images according to object shapes

A method for repairing video/audio files of multimedia video/audio disks is provided. The disclosed method may repair the fields with errors in the video/audio files after recording by utilizing the rewritable characteristic of rewritable optical disks. The errors caused by the video/audio instantaneous editing and recording platform may be reduced according to the disclosed method such that the repaired films may play fast forward/backward smoothly, perform Time Search function accurately, and display presentation time correctly, and are compatible to players of different specifications.

Method for repairing video/audio files of multimedia video/audio disks

Growing community of inventors

Taichung, Taiwan

Guo-Zua Wu