339 results about "Adaptive weighting" patented technology

An information retrieval system for automatically retrieving information related to the context of an active task being manipulated by a user. The system observes the operation of the active task and user interactions, and utilizes predetermined criteria to generate context representation of the active task that are relevant to the context of the active task. The information retrieval system then processes the context representation to generate queries or search terms for conducting an information search. The information retrieval system determines the relevance of a word to the context by utilizing an adaptive weighting system. The information retrieval system assigns varying weights to different attributes of a word and calculates an accumulated weight of the word by accumulating all weights assigned to the word. The attributes may include word size, style, location of the word, etc. The system then ranks the importance of words based on their respective accumulated weight, and chooses words that rank within a predetermined number from the top to form search terms to conduct an information search using various data sources.

A method schedules packets in a router of a packet-switched network. The network has a set of service classes including a premium service, an assured service, and a best-effort service. The router includes one queue for each service class. Each queue stores packets to be transmitted according to the associated service class. The method measures an exponential weighted moving average queue length of the queue associated with the premium service each time a packet is stored in that queue. Bandwidth is increased if the average increases over a minimum threshold. The bandwidth remains below an upper limit once the average reaches a maximum threshold.

An adaptive weighted round robin scheduling apparatus and method schedules variable-length frame transmissions from a plurality of output queues having different transmission priorities by first allocating, for each queue, a number of bandwidth segments for a bandwidth cycle and a number of transmission opportunities for a round robin cycle, and then processing the queues consecutively in a round-robin fashion, beginning with a highest priority queue, until none of the queues has any bandwidth remaining. More specifically, during each iteration of a round robin cycle, a queue is permitted to transmit a frame if the queue has at least one remaining transmission opportunity, the queue has a frame ready for transmission, and the queue has at least one remaining bandwidth segment, and furthermore the number of transmission opportunities for the queue is decremented by at least one. Upon transmitting a frame, the number of bandwidth segments for the queue is decreased by the number of bandwidth segments in the frame. If a queue has no frame ready for transmission, then the queue may be either penalized, in which case the number of bandwidth segments for the queue is reduced, or forced to forfeit its bandwidth segments, in which case any remaining bandwidth segments are reallocated to other queues and the number of bandwidth segments and the number of transmission opportunities for the queue are set to zero.

The invention discloses an automatic image annotation method based on semi-supervised learning. The method comprises the steps that a data set is divided into a training data set, an unlabeled data set and a test set; the SIFT feature and the HOG feature of a training sample are extracted to train an LDA_SVM classifier; color and texture features are extracted to train a neural network; unlabeleddata are used to enable two classifiers to label and predict the same unlabeled sample simultaneously; according to the contribution of the classifiers to the unlabeled sample classification accuracy,the classification results of two classifiers are weighted and fused by an adaptive weighted fusion policy, so as to acquire the final prediction label probability vector of the sample; and finally two classifiers are updated by the sample with high confidence and the predictive label thereof until the preset maximum number of iterations is reached. According to the invention, the method can makefull use of the unlabeled sample to excavate the inherent law of the image feature; the number of annotation samples required for the classifier training is effectively reduced; the annotation effectis great.

There is provided a method of selecting a pitch lag value from a plurality of pitch lag candidates for coding a speech signal. The method comprises identifying the plurality of pitch lag candidates from a frame of the speech signal using correlation; classifying the speech signal to obtain a voice classification; determining whether one or more of the plurality of pitch lag candidates are in a temporal neighborhood of one or more previous pitch lag values; favoring the one or more of the plurality of pitch lag candidates determined to be in the temporal neighborhood of the one or more previous pitch lag values, by adaptive weighting, over other ones of the plurality of pitch lag candidates; and selecting the pitch lag value based on the voice classification and the one or more of the plurality of pitch lag candidates favored by the adaptive weighting.

The invention discloses an image segmentation and adaptive weighting-based stereo matching method. The method comprises the steps of parallax initialization and parallax optimization. The parallax initialization comprises the steps of adopting corrected left and right images as a reference image and a target image respectively, and segmenting the image; based on the constructed combination segmentation information and an adaptive-weighting cost function, calculating matching costs E (p, pd, d) between a current to-be-matched pixel pint p in the left image and all candidate matching points pd in the right image, and selecting a candidate matching point of a minimum matching cost as an optimal matching point for the pixel pint p; repeating the above steps till all pixel points in the left image are traversed in the raster scan order. In this way, an initial parallax image is obtained. The parallax optimization comprises the steps of fitting the parallax plane of the obtained initial parallax image, suppressing the abnormity of the initial parallax image, and recovering the edge of the initial parallax image. The method is advantaged in that unreliable points in the initial parallax image obtained based on the calculation of the optimal matching point are re-corrected, while abnormal small areas are merged to adjacent normal areas. Meanwhile, edge pixels are recovered, and the parallax error is eliminated. The matching accuracy is improved.

The invention discloses a self-adaptive weighted stereo matching algorithm, comprising the steps of: obtaining a binocular stereo image; selecting a pixel supporting window, calculating the real supporting window of each pixel, wherein the supporting windows formed from the supported pixels have different sizes and shapes; calculating the supporting weight in the supporting window; calculating collected energy cost functions; calculating the parallax value of the pixels; and obtaining an original left-to-right parallax map. The invention also discloses a stereo display and collecting device and a system. The self-adaptive weighted stereo matching algorithm, the stereo display and collecting device and the system can be applied to the technical field of stereo display to improve the effect on stereo matching, as well as obtaining necessary three-dimensional information through less data volume.

The invention provides a self-adaptive real-time denoising scheme for a 3D digital video image, which is used for filtering and denoising a video image in a space domain two-dimensional mode and a time domain one-dimensional mode. The method comprises the following steps of: detecting the motion intensity of the video image, estimating a noise standard difference of the image and setting two different threshold value limits according to an estimation result of the noise standard difference; and adjusting parameters in time domain filtering in a self-adaptive mode according to a relation between the motion intensity and the limits and weighting a time domain filtering result and a space domain filtering result according to the relation in a self-adaptive mode to obtain a denoised output result. In the scheme, the computational load and memory space are small, the 3D digital video image can be denoised effectively in real time, borders and details of the image are protected, and the watching comfort of the 3D digital video image is enhanced greatly.

The invention relates to a multi-characteristic layered fusion correlation filtering robustness tracking method. In order to improve the robustness of target tracking, aiming at the multi-characteristic fusion problem in correlation filtering tracking, a multi-characteristic layered fusion strategy is provided. HOG characteristics, CN characteristics and color histogram characteristics are extracted from a target area and a surrounding background area respectively. A self-adaption weighting fusion strategy is adopted for integration of characteristic response diagrams of the HOG characteristics and the CN characteristics. A fusion result of a layer and the obtained characteristic response diagrams based on the color histogram characteristics are subjected to second-layer integration, a fixed parameter fusion strategy is adopted for fusion of the characteristic response diagrams. According to the method, under the premise that the tracking accuracy is guaranteed, the tracking robustnessis superior to other algorithms. When multiple characteristics are adopted for a correlation filtering tracking algorithm, the layered fusion strategy has certain reference.

A system, apparatus and method for a multi-stage Parallel Residue Compensation (PRC) receiver for enhanced suppression of the Multiple Access Interference (MAI) in Code Division Multiple Access (CDMA) systems. The accuracy of the interference estimation is improved with a set of weights computed from an adaptive Normalized Least Mean Square (NLMS) algorithm. In order to reduce complexity, the commonality of the multi-code processing is extracted and used to derive a structure of PRC to avoid direct interference cancellation. The derived PRC structure reduces the interference cancellation architecture from a complexity that is proportional to the square of the number of users to a complexity that is linear with respect to the number of users. The complexity is further reduced by replacing dedicated multiplier circuits with simple combinational logic.

The invention discloses a morphology and PCA (principal component analysis) based contourlet fusion method for infrared and visible light images. The method includes: firstly, subjecting two images to morphology-hat transformation respectively and subjecting the two obtained images to contourlet decomposition to form a high-frequency image and a low-frequency image respectively; then, fusing the low-frequency image by a PCA method; fusing the high-frequency image by means of average gradient adaptive weighted processing; reconstructing a fused image by means of contourlet reverse transformation. The morphology and PCA based contourlet fusion method for the infrared and visible light images is used for image fusion of the infrared and visible light images, so that the fused image is high in contrast ratio, rich in information and clear in texture detail.

The invention discloses a device and a method for fusing sensor information. The sensor-information fusion method comprises the steps of adopting a self-adaptive weighted-data fusion algorithm to allocate corresponding weight numbers for induction information from a plurality of sensors, adopting a Kalman filtering algorithm to perform optimal estimation to the induction information of the corresponding weight numbers so as to acquire the locally fused induction information, adopting a D-S theory estimation algorithm to perform interval estimation to the uncertain induction information in the locally fused induction information, adopting a multi-Bayesian estimation algorithm to perform individual association probability distribution for the induction information after interval estimation so as to synthesize a joint posteriori probability distribution function, and utilizing the joint distribution function to output a final fusion value of the induction information. The invention has the advantages of improving the reliability, confidence and efficiency of sensor information fusion.

A lightweight rule induction method is described that generates compact Disjunctive Normal Form (DNF) rules. Each class may have an equal number of unweighted rules. A new example is classified by applying all rules and assigning the example to the class with the most satisfied rules. The induction method attempts to minimize the training error with no pruning. An overall design is specified by setting limits on the size and number of rules. During training, cases are adaptively weighted using a simple cumulative error method. The induction method is nearly linear in time relative to an increase in the number of induced rules or the number of cases. Experimental results on large benchmark datasets demonstrate that predictive performance can rival the best reported results in the literature.

The invention relates to a multi-layer convolution feature self-adaptive fusion moving target tracking method, and belongs to the field of computer vision. The method comprises the following steps: firstly, initializing a target area in a first frame of image, and utilizing a trained deep network framework VGG-19 to extract first and fifth layers of convolution features of the target image block,and obtaining two templates through learning and training of a related filter; Secondly, extracting features of a detection sample from the prediction position and the scale size of the next frame andthe previous frame of target, and carrying out convolution on the features of the detection sample and the two templates of the previous frame to obtain a response graph of the two-layer features; calculating the weight of the obtained response graph according to an APCE measurement method, and adaptively weighting and fusing the response graph to determine the final position of the target; And after the position is determined, estimating the target optimal scale by extracting the directional gradient histogram features of the multiple scales of the target. According to the method, the targetis positioned more accurately, and the tracking precision is improved.

The invention discloses a traditional Chinese medicinal material microscopic image noise filtering system and method adopting a pulse coupling neural network. A pulse coupling neural network model suitable for processing tissue image information is adopted for detecting traditional Chinese medicinal material microscopic images. When small-density pulse noise pollution happens to the traditional Chinese medicinal material microscopic images, adaptive weighted filtering processing is carried out; when large-density pulse noise pollution happens to the traditional Chinese medicinal material microscopic images, the introduction dual-structure element mathematical morphology with edge detailed information kept is adopted for secondary filtering. Earlier foundations are laid for further improving quality testing, recognition and identification of traditional Chinese medicinal materials; higher noisy point detection performance is achieved, the noise fallout ratio and omission ratio are low, and the noise detection precision is high; detection time is short, and automatism is high; noise is removed, meanwhile, noise interference can be effectively filtered out, and image edge details and other information can be protected well.

The present invention provides an automatic color balancing method for digital images by essentially performing adaptive weighting of surface reflectance and illuminant spectra components of the image. The adaptive weighting mechanism is derived from fuzzy logic based inference methods, taking advantage of its ability to perform inferences from data by providing a computational framework for knowledge that is in linguistic form. It also makes it amenable for implementation on hardware because of the commercially available fuzzy logic chips that already exist in real-world systems such as camcorders for image stabilization, washing machines, etc. The present invention enables much more efficient and robust color segmentation that forms core components for several computer vision algorithms.

Techniques and systems are provided for processing 360-degree video data. For example, a picture of the 360-degree video data can be obtained. The picture can include samples projected from a three-dimensional format to a two-dimensional format. A weight value can be determined for at least one sample of the picture. The weight value can be determined based at least on a position of the at least one sample in the picture. At least one adaptive parameter can be determined for the at least one sample using the determined weight value. The at least one adaptive parameter can include one or more of an adaptive weighted distortion, an adaptive weighted quantization parameter value, or an adaptive weighted lambda value. A cost associated with coding the at least one sample using one or more coding modes can be determined using the at least one adaptive parameter of the at least one sample.

The invention discloses a quick binocular stereo matching method based on superpixel segmentation. The method comprises the following steps that: (1) adopting a SLIC (Simple Linear Iterative Clustering) superpixel segmentation method to carry out regional division on an original reference diagram and a target diagram; (2) on the basis of a local matching algorithm of adaptive weighting, calcuating an original parallax spatial diagram; (3) carrying out regional parallax plane fitting based on a confidence point; (4) applying a clustering algorithm to combine adjacent regional parallax planes; and (5) on the basis of a superpixel region, constructing an energy cost function to carry out stereo matching. A stereo matching unit is the superpixel region combined on the basis of the edge information of an image, so that the algorithm is guaranteed to be suitable for a large no-texture region, a depth boundary can be accurately positioned, a matching cost better than the matching cost of a traditional stereo matching method is obtained and is combined with a parallax postprocessing method to effectively obtain a high-accuracy parallax diagram, and the method exhibits good instantaneity.

The invention discloses a method for fusing sensor grid data based on grid clustering, which comprises the following steps: allocating tasks to all sensor nodes of each wireless sensor network, and then running an adaptive genetic algorithm based on the grid clustering in each wireless sensor network; according to perceptive data of all the sensor nodes, finding out an optimal sensor node and then finding out an optimal fusing routing with the minimum communication cost; primarily fusing data by a mobile agent along the optimal fusing routing and then sending the primarily fused result to nodes of convergent points of the network; and adopting an adaptive weighted fusion algorithm by a grid layer to carry out deep fusion on the data so as to obtain the final fused result which is providedfor a user for making decisions. The method can be applied to a sensor grid environment and is used to process the data sufficiently so as to obtain the good fused result. The grid is used for deeplyfusing the data, thus reducing energy consumption of the wireless sensor networks and prolonging the service life of sensor grids.

The present invention relates to adaptive weighting feedback correcting and filtering method for combined SINS / GPS navigation system. The method includes the first utilizing observed position and speed quantities the GPS provides in Kalman filtering to estimate the SINS position error, speed error, attitude error and device error; and the subsequent calculating the measurability of the system state quantities in the system measurability analysis process and adaptive weighting feedback correction on the SINS with the normalized system measurability as the weighting feedback factor. The present invention has the advantages of high precision, less divergence and high navigation precision on airplane, missile, ship and ground vehicle.

The disclosed removing interlace method using motion detection and adaptive weighting filter comprises: filtering the last field, current field and follow-up field of the image signal; storing in SDRAM as pre-set format; based on pixel waiting for interpolation in current field, calculating the motion level of last and follow-up field pixel, then calculating the motion information weighting value for mixing; mixing two interpolation by motion information weighting, and outputting the mixed value. This invention can hold image space feature and vertical details without precise motion information estimation, reduces design complexity, and improves design reliability.

A radar system includes a transmit antenna array having subarrays disposed at predetermined positions. An orthogonal waveform signal is directed to a corresponding one of the subarrays. On receive, an adaptive processor derives a plurality of adaptive weight factors from a plurality of receive signals and applies them to the receive signals to obtain a jammer cancelled signal. That signal is separated into its orthogonal waveform components by passing it through a bank of correlators. The correlator system provides a plurality of unique receive signals substantially corresponding one-to-one to the unique transmit subarrays. The receive beamformer derives an angular estimate of at least one target relative to boresight from the unique receive signals.

There are provided a method and apparatus for adaptive weight selection for motion compensated prediction. The apparatus includes an encoder (100) for encoding a picture by deriving a set of weighting parameters, selecting at least one weighting parameter in the set based upon a selection criteria, and applying the selected at least one weighting parameter to a reference picture used to encode the picture.

The invention discloses a video image character recognition method based on submesh characteristic adaptive weighting. Aiming at the technical problems in the conventional video image character recognition, the invention provides an image character binaryzation method based on edge characteristics. After a binary image of a character is acquired, the image character is subjected to regional mesh division by an elastic mesh method; and before the elastic mesh characteristic of the image character is calculated, the relevance of stroke structures of adjacent meshes is extracted. By the method, the stability and the robustness of extracting the characteristic of the character are improved, so that the recognition rate of video image characters is increased.

The invention relates to a Backlash self-adaptive filter and a method for modeling and compensating the hysteresis thereof, belonging to the technical field of modeling and control of hysteresis non-linear systems. The filter comprises a summator module, an error calculating module, a plurality of self-adaptive weighting modules and a plurality of Backlash operator modules with the same width. The whole designing scheme avoids using a hysteresis item and is only related with current inputs, meanwhile, the hysteresis non-linearity can be reflected, and the modeling effect is more precise; and the self-adaptive inverse control of the filter can effectively compensate the hysteresis non-linearity of a piezoelectric ceramic actuator. The self-adaptive filter and the modeling and self-adaptive inverse control method based on the filter can be applied to systems having the characteristics of hysteresis non-linearity, such as intelligent materials of magnetostrictive materials, excitation motors, piezoelectric chips, piezoelectric ceramics, and the like.

The present invention discloses a method of adaptive weighting space-time transmit diversity, comprising: symbols to be transmitted at a transmitting end being output after being space-time encoded according to certain regulation, and the current transmission powers of the two antennae being real-timely adjusted according to the given transmission power weight values; after the transmission powers are determined, the space-time encoded output signals being sent out at two independent antennas by transmitting antenna array in terms of the current transmission power; the fading characters of two wireless channels being estimated by the receiver at a receiving end according to the present received signals and then fed back to the transmitting end; the transmitting end receiving and obtaining the fading amplitude characters of two wireless channels through feedback channel, calculating new adaptive weight values of transmission powers, and adjusting transmission powers according to these weight values. The present invention also discloses a system for implementing the above method of adaptive weighting Space-time Transmit Diversity. In term of the method and structure, the transmission powers can be dynamically adjusted in adaptive weighting manner, in order to boost system performance gain.