Abstract: A face annotation method and a face annotation system are provided. The face annotation method is adapted for a current owner to annotate contacts in online social networks. The face annotation method comprising: providing a pyramid database access control module which consists of a plurality of pyramid database units and performs a first batch of access control procedure and a non-first batch of access control procedure, wherein the pyramid database unit is constructed according to social relationship information; providing a multiple-kernel learning face recognition module implemented through the use of a MKL classifier unit which uses a MKL algorithm to achieve a face identification; and if the MKL-FR model is not able to identify query faces, providing a multiple-kernel learning face recognition fusion module to perform a collaborative face recognition strategy by utilizing a user who features a highest priority rule within a collaborative face recognition framework.

Abstract: A method and image processing apparatus for image visibility restoration are provided. The method includes the following steps: receiving an input hazy image including input pixels; obtaining edge information of each of the input pixels according to a median filtering operation and a dark channel; determining a transmission map according to each of the input pixels and atmospheric light associated with the input hazy image in each color channel; obtaining a refined transmission map according to the edge information and the transmission map; adjusting the refined transmission map by performing a gamma correction operation thereon to obtain an enhanced transmission map; determining a color difference value corresponding to each of the color channels; recovering scene radiance for each of the input pixels in each of the color channels according to the corresponding color difference value, the enhanced transmission map, and the atmospheric light to produce and output a de-hazed image.

Abstract: The present disclosure proposes a method of moving object detection in variable bit-rate video steams based on probabilistic neural networks, and the method features a background generation module and a moving object detection module. The background generation module produces a model of background images which express properties of variable bit-rate video streams. The moving object detection module distinguishes a moving object in both low and high bit-rate video steams in an efficient manner. The detection result is generated by calculating the output value of the probabilistic neural networks.

Abstract: A method for moving object detection based on a Principal Component Analysis-based Radial Basis Function network (PCA-based RBF network) includes the following steps. A sequence of incoming frames of a fixed location delivered over a network are received. A plurality of Eigen-patterns are generated from the sequence of incoming frames based on a Principal Component Analysis (PCA) model. A background model is constructed from the sequence of incoming frames based on a Radial Basis Function (RBF) network model. A current incoming frame is received and divided into a plurality of current incoming blocks. Each of the current incoming blocks is classified as either a background block or a moving object block according to the Eigen-patterns. Whether a current incoming pixel of the moving object blocks among the current incoming blocks is a moving object pixel or a background pixel is determined according to the background model.

Abstract: A method for moving object detection based on a Fisher's Linear Discriminant-based Radial Basis Function Network (FLD-based RBF network) includes the following steps. A sequence of incoming frames of a fixed location delivered over a network are received. A plurality of discriminant patterns are generated from the sequence of incoming frames based on a Fisher's Linear Discriminant (FLD) model. A background model is constructed from the sequence of incoming frames based on a Radial Basis Function (RBF) network model. A current incoming frame is received and divided into a plurality of current incoming blocks. Each of the current incoming blocks is classified as either a background block or a moving object block according to the discriminant patterns. Whether a current incoming pixel of the moving object blocks among the current incoming blocks is a moving object pixel or a background pixel is determined according to the background model.

Abstract: A method for moving object detection based on a Cerebellar Model Articulation Controller (CMAC) network includes the following steps. A time series of incoming frames of a fixed location delivered over a network is received. A CMAC network is constructed from the time series of incoming frames, where the CMAC network includes an input space, an association memory space, a weight memory space, and an output space. A current frame is received and divided into a plurality of current blocks. Each of the current blocks is classified as either a background block or a moving object block according to the CMAC network. Whether a target pixel of the moving object blocks is a moving object pixel or a background pixel is determined according to an output of the CMAC network in the output space.

Abstract: The invention discloses a method for solving carpool matching problem and a carpool server using the same. The method includes the following steps: generating a carpool population according to carpool requests received from passengers and drivers; performing a routing procedure to each of the segments, such that a segment fitness value of each of the segments is maximum; performing a calculating procedure to calculate a fitness value of each of the carpool matching results; performing a sorting procedure to sort the carpool matching results as first results and second results; performing a modifying procedure to modify each of the renewed second results by swapping two slots of each of the renewed second results; and performing an updating procedure to update the carpool population according to the first results and the modified second results.

Abstract: The invention discloses a carpool service providing method and a carpool server using the same. The method includes the following steps: generating a carpool population matrix according to a plurality of carpool requests received from a plurality of passengers and drivers; generating a first carpool matching result and a second carpool matching result according to the carpool population matrix; performing a routing procedure to each of the first and second segments; respectively comparing the segment fitness value of one of the second segments with the segment fitness value of the corresponding first segment and updating the second carpool matching result by replacing the one of the second segments with the corresponding first segment if the segment fitness value of the one of the second segments is worse than the segment fitness value of the corresponding first segment; updating the carpool population matrix according to the updated second carpool matching result.

Abstract: The invention discloses an image processing method and an imager processing apparatus using the same. The method includes the following steps: receiving an training image; finding a minimum difference among the differences; determining whether the minimum difference is larger than a first threshold; if no, generating a first output value according to the first pixel, the background candidates and a plurality of weightings corresponding to the background candidates; updating a first background candidate corresponding to the minimum difference; updating a first weighting related to the first background candidate; if yes, adding the first pixel as a new background candidate to the background candidates and adding a new weighting corresponding to the new background candidate to the weightings; and detecting whether a moving object existing in an incoming image according to the background candidates and the weightings.

Abstract: The present disclosure proposes a method of moving object detection in variable bit-rate video steams based on probabilistic neural networks, and the method features a background generation module and a moving object detection module. The background generation module produces a model of background images which express properties of variable bit-rate video streams. The moving object detection module distinguishes a moving object in both low and high bit-rate video steams in an efficient manner. The detection result is generated by calculating the output value of the probabilistic neural networks.

Abstract: A certification method comprises steps of: providing a reliable time clock on a first electronic device; when data of the digital file are generated on the first electronic device, reading a reliable time count from the reliable time clock and adding the reliable time count into the digital file; generating a first abstract code from the digital file; generating a signature of the digital file by encrypting the first abstract code; and, sending the digital file and the signature to a second electronic device. In addition, electronic devices corresponding to the certification method are also disclosed herein.

Abstract: A high-performance block-matching VLSI architecture with low memory bandwidth for power-efficient multimedia devices is disclosed. The architecture uses several current blocks with the same spatial address in different current frames to search the best matched blocks in the search window of the reference frame based on the best matching algorithm (BMA) to implement the process of motion estimation in video coding. The scheme of the architecture using several current blocks for one search window greatly increases data reuse, accelerates the process of motion estimation, and reduces the data bandwidth and the power consumption.

Abstract: A video editing method for a video editing device is provided. The method includes: obtaining an editing segment of a video, wherein the editing segment follows a first segment of the video and precedes a second segment of the video, the first segment includes a first picture, the second segment includes a second picture, and a short-term reference index of the second picture points to the first picture; setting a header of a third picture in the first segment, such that the first picture is moved from a short-term reference picture list to a long-term reference picture list when the header of the third picture is decoded; changing the short-term reference index of the second picture into a long-term reference index, wherein the long-term reference index points to the first picture in the long-term reference picture list; and encoding the video. Accordingly, the speed of editing the video is increased.

Abstract: A video decoding device is provided. The device includes an error checking unit and a decoding unit. The error checking unit checks the error state of a video data, so as to produce an error information. Then, the decoding unit selectively performs an error concealment process and a decoding process for the video data according to the error information. The video decoding device of the present invention can perform error checking process, error concealment process, and video decoding process for the video data.

Abstract: A high safety vehicular transportation system is used in an individual carpool car with at least one user and comprises a user module, a rescue unit and a server center. The user module stores and provides carpool data, stores and provides states of individual carpool car, submits a carpool request and comprises a save our soul (SOS) system. The SOS system transmits a distress signal. The rescue unit provides emergency rescue assistance if any user is in a dangerous situation. The server center receives, stores and provides carpool data, has an instantaneous tracking capability, provides carpool service to at least one user, calls the rescue unit if the server center receives the distress signal, receives the carpool request and generates at least one carpool mission to fulfill the carpool request. An operational method for a high safety vehicular transportation system is also disclosed.

Abstract: A method for performing temporal error concealment is provided. The method includes the following steps: detecting a damaged macroblock having four 8×8 damaged blocks; obtaining motion vectors of a plurality of 4×4 neighboring blocks surrounding the damaged macroblock; and for each of the 8×8 damaged blocks, determining a predicted motion vector according to motion vectors of six of the plurality of 4×4 neighboring blocks which are closest to the 8×8 damaged block.

Abstract: A motion detection method for complex scenes has steps of receiving an image frame including a plurality of pixels, each of the pixel including a first pixel information; performing a multi-background generation module based on the plurality of pixels; generating a plurality of background pixels based on the multi-background generation module; performing a moving object detection module; and deriving the background pixel based on the moving object detection module.

Abstract: The invention discloses an image processing apparatus. The image processing apparatus includes an image statistic computation circuitry, a reconfigurable circuitry and a luminance transformation circuitry. The image statistic computation circuitry computes a probability density function corresponding to an inputted image; generates a first luminance histogram by subsampling a luminance histogram related to the probability density function in a first period. The reconfigurable circuitry computes a weighting distribution function according to the first luminance histogram in a second period after the first period; computes a smoothed cumulative density function according to the weighting distribution function in a third period after the second period; computes a gamma transform function in a fourth period after the third period.

Abstract: A method for assisting in video compression using rotation operation and an image capturing device thereof are provided. In the method, a rotation operation of the image capturing device is detected, and a rotation angle formed by the rotation operation is calculated. The image capturing device is rotated by the rotation angle so as to capture a current frame. In addition, whether a reference frame having a rotation angle information exists in a reference frame list stored in the image capturing device is determined. If not, a rotational motion vector is calculated based on the rotation angle. The current frame is divided into a plurality of current blocks. Among each of the current blocks, a matching block corresponding to the current block is searched in a reference frame of the reference frame list by using the rotational motion vector, and accordingly, the frame data of the current frame is compressed.

Abstract: A method for accessing a file in a network storage device is disclosed. The method includes receiving a file transmitting request from a first electronic device via the network for accessing the file; transmitting an authorization request corresponding to the file transmitting request to a second electronic device via the network; receiving a reply corresponding to the authorization request from the second electronic device; and transmitting the file to the first electronic device via the network when the reply notices that the file is accessible.