METHOD OF MULTI-LEVEL FACIAL IMAGE RECOGNITION AND SYSTEM USING THE SAME |
USA
|
Abstract
A quadrature mirror filter is applied to decompose an image into at least two sub-images each having a different resolution. These decomposed sub-images pass through self-organizing map neural networks for performing a non-supervisory classification learning. In a testing stage, the recognition process is performed from sub-images having a lower resolution. If the image can not be identified in this low resolution, the possible candidates are further recognized in a higher level of resolution.
|
對影像切割的方法 |
USA
|
Abstract
The method proposed by the present invention divide an image into a plurality of cluster regions, each has its own principle color. The method mentioned above includes the following steps. At first, acquire a three dimensional histogram of the image by counting the number of pixels of the image based on three dimensional color coordinate. Then smooth the three dimensional histogram by a numerical method to obtain a function. Next, define a plurality of cluster regions on the image. Subsequently, assign the color of each pixel of the image as a principle color of a first cluster region among the plurality of cluster regions. The principle color is equal to average of all pixels within the cluster region, the average is equal to A/B, wherein A is summation of color of pixels within the first cluster region, and B is equal to the number of pixels within the first cluster region. Each of the plurality of cluster regions has a corresponding principle color. Finally, joint an unclassified pixel to a cluster region of the plurality of cluster regions according to minimum distance from the unclassified pixel to the plurality of cluster regions. The unclassified pixel is outside the plurality of cluster regions before this joint step, after all pixels of the image have being jointed to one of the plurality of cluster regions, the image is divided into the plurality of cluster regions.
|
RAPID CHECKING METHOD FOR DETERMINING WHETHER AN OBJECT IS LOCATED WITHIN A FIELD OF VISION |
USA
|
Abstract
A rapid checking method for determining whether an object is located within a field of vision is provided. A computer serves to process the states of a 3D image so as to determine whether the object is located within the field of vision. An efficient calculation process is used. In the invention, the 3D coordinates of eight vertexes of an object and the boundary box formed by equations of six planes are employed in operation. By determining the relationship of the vertexes and the planes, it is determined whether an object is within the field of vision.
|
DIGITAL IMAGE PROCESSING DEVICE AND DIGITAL CAMERA USING THIS DEVICE |
USA
|
Abstract
A digital image processing device, more particularly, a digital camera, which includes: a static digital image extracting device used to extract the static digital image and output; a compressing device used to compress the static digital image using wavelet transform to produce a compressed image data file; a memory card used to store the compressed image data file; a central processing unit used to calculate and compare the storage space left in the memory card, storage space used in the memory card and the size of the compressed image data file, when the storage space left in the memory card is less than the size of the compressed image data file, part of the image data file stored in the memory is eliminated to store the compressed image data file, otherwise the compressed image data file is directly stored in the memory card; a decompressing device used to decompress the compressed image data file to produce an image data file; and an image display device used to display the image data file.
|
AUTOMATIC IMAGE REPLACEMENT AND REBUILDING SYSTEM AND METHOD THEREOF |
USA
|
Abstract
An automatic image replacement and rebuilding system and method thereof which can replace and rebuild the image automatically. The system includes an input module; a processing module; a storage module; and an output module. The input module inputs external image data into the system. The processing module is coupled to the input module to receive the image data, then analyzes the shading value and the direction of the sample object in the image data, and outputs an access signal and an output signal. The storage module coupled to the processing module to receive the access signal, and is stored or read out on a replaced image, and the output module coupled to the processing module to receive the output signal, and outputs the image data. In addition, the automatic image replacement and rebuilding method, which is suitable for a system with a processing module and a storage module, by analyzing the shading value and the direction of the image of the sample object having a white background and dark grid characteristics, so that the image can be replaced and rebuilt.
|
METHOD OF RAPIDLY DETERMINING THE TRANSMISSION TIME AND RANGE OF A POSITION MESSAGE UNDER INTERNET VIRTUAL REALITY ENVIRONMENT |
USA
|
Abstract
A method of rapidly determining the transmission time and range of a position message under an Internet virtual reality environment is employs a timer scheme for updating the virtual reality environment by using a timer to count a time value. A virtual scene is divided into plural blocks. A block where a user is in and blocks neighboring to the block are defined as a low interactive area of interest, which is divided into plural sub-blocks. A sub-block where the user is in and sub-blocks neighboring to the sub-block are defined as a high interactive area of interest. A message is transmitted to update the virtual reality environment based on different settings of the low interactive area of interest and the high interactive area of interest when the timer is up and the virtual reality environment reaches a predetermined inconsistency, wherein the time value for the high interactive area is shorter than that for the low interactive area.
|
METHOD FOR TRANSFORMING VIDEO DATA BY WAVELET TRANSFORM SIGNAL PROCESSING |
USA
|
Abstract
A method to transform video data having an interest region by wavelet transform signal processing. First, the video data is transformed to high frequency image data and low frequency image data. Then, first image data recording first information responding to the interest region of the video data and second image data recording second information responding to the interest region of the video data from the high frequency image data are received. The first image data and the low frequency image data are combined to form interest region data. The interest region data and the second image data are combined by bit plane coding to generate display data. Finally, a display device displays the display data.
|
APPARATUS AND METHOD FOR CONNECTING AVIONICS CONTROL PANELS AND TRAINING/SIMULATION SYSTEMS |
USA
|
Abstract
A device to connect a control panel of an avionics equipment with a training and simulation apparatus to process and display signals that correspond to output signals of the control panel. The device includes means coupled to a keyboard of the control panel for decoding signals from the control panel keyboard and means coupled to the means for decoding the keyboard signals to convert the keyboard signals into computer keyboard signals. The device additionally includes adaptor means to provide the computer keyboard signals to the computer.
|
METHOD FOR THE TEST OF A TESTING PLATFORM |
USA
|
Abstract
A method for the test of a testing platform comprises a series of steps performed by a computer. Wherein the computer software structure includes a tested object menu, a plurality of test programs, a test instruction subprogram, a speech instruction subprogram, a driver program database, and a manager for receiving a test results. The method according to the present invention has several stages so that another tested object is further tested, the current test program can be used. Moreover, a speech prompt function is further provided, which may match new computer software construction.
|
DRIVING DEVICE FOR FLUORESCENT TUBE |
USA
|
Abstract
A driving device for a fluorescent tube has a high frequency oscillator which outputs a high frequency AC signal. A pulse width modulator is connected to the high frequency oscillator for outputting a PWM harmonic frequency signal. A first power switch is connected to the pulse width modulator for being turned off during a positive half-cycle of the PWM harmonic frequency signal and being turned on during a negative half-cycle of the PWM harmonic frequency signal. A second power switch is connected to the pulse width modulator for being turned on during the positive half-cycle of the PWM harmonic frequency signal and being turned off during the negative half-cycle of the PWM harmonic frequency signal. A piezoelectric transformer includes a primary winding having two input terminals connected to the first power switch and the second power switch, respectively, and a center terminal connected to the output terminal of the pulse width modulator.
|