217results about How to "Improve classification" patented technology

A magnetic anomaly sensing system and method uses at least four triaxial magnetometer (TM) sensors with each of the TM sensors having X,Y,Z magnetic sensing axes. The TM sensors are arranged in a three-dimensional array with respective ones of the X,Y,Z magnetic sensing axes being mutually parallel to one another. The three-dimensional array defines a geometry that forms at least one single-axis gradiometer along each of the X,Y,Z magnetic sensing axes. Information sensed by the TM sensors is to generate scalar magnitudes of a magnetic anomaly field measured at each of the TM sensors, comparisons of the scalar magnitudes to at least one threshold value, distance to a source of the magnetic anomaly field using the scalar magnitudes when the threshold value(s) is exceeded, and a magnetic dipole moment of the source using the distance.

A radar microchip for short range detection with phased array radar uses phase shifters along with an antenna array to steer the transmitted and received radar beams along orthogonal axes to achieve a 3D scan. Individual phase shifters connected to antenna cells that transmit and receive the radar beams steer the radar along two orthogonal axes by controlling the phase of the radar. The radar then detects where the two beams overlap. The antenna cells are further aligned along these orthogonal axes. An isolation barrier between the phase shifters of the transmitted and received signals reduces cross coupling on the radar microchip.

The embodiment of the invention provides a garbage classification method, device and system. The problem that the garbage classification effect needs to be improved in the prior art is solved. The garbage classification method includes the steps that identity information of a user is verified and the user passing the verification is registered as a legal user; a garbage bag acquisition request sent by the user and the identity information of the user sending the garbage bag acquisition request are received; whether the user sending the garbage bag acquisition request is a legal user or not is judged, if yes, a garbage bag issuing permission control instruction is sent to a garbage bag issuing machine, and the garbage bag issuing machine issues garbage bags with class labels to the legal user; and control instruction execution feedback sent by the garbage bag issuing machine is received and a unique identification code is correlated with the identity information of the user sending the garbage bag acquisition request and then stored. By the utilization of the garbage classification method, device and system, user garbage classification convenience can be remarkably improved, implementation is convenient, and application and popularization are easy.

The invention discloses a multi-scale fast human face detection method for convolutional neural network feature fusion. The method comprises the following steps: step 1, based on a model structure ofan SSD fast target detection method, a feature extraction method in the SSD is transformed, a feature fusion method is added, and a modified detection model is obtained; step 2, human face detection training is carried out on the detection model transformed in the step 1, and a trained deep neural network model is obtained; step 3, the to-be-detected picture is calculated by using a deep neural network model trained in the step 2, to obtain a model output result. According to the invention, the human face in the image can be quickly recognized and the human face can be accurately positioned, so that the human face is separated from the complex background, and a foundation is provided for identity verification and tracking of people in an image.

A method of and a system for sharp object detection using computed tomography images are provided. The method comprises identifying voxels corresponding to individual objects; performing eigen-analysis and generating eigen-projection of an identified object; computing an axial concavity ratio of the identified object; computing a pointness measurement of the identified object; computing a flat area of the identified object; calculating a sharpness score of the identified object; and declaring the identified object as a threat if the sharpness score is greater than a pre-defined threshold.

The invention discloses a disk fault detection method using multi-model prediction, which extracts multiple characteristics of disk SMART indexes through a sequential data processing means, and establishes a classification model to predict disk state. Step 1, data input: acquiring a data set composed of monitoring data of a plurality of disks in a period of time; step 2, SMART screening: adoptingmutation point detection mode to select SMART index; step 3, feature engineering: using SMART index as the input of the user-defined feature extraction module to extract the features of the SMART index, then extracting the corresponding parameter configuration, and transmitting the parameter configuration to the feature extraction module as a parameter, so as to extract the feature sets of the training set and the test set; step 4, data set balance: desampling the negative sample which occupies a large amount by adopting dimension reduction clustering; step 5, algorithm selection and modeling:on the basis of the step 4, training the classification model and testing whether the current disk belongs to the normal state or the fault state that needs to be replaced.

The invention relates to an intelligent classification dustbin. The intelligent classification dustbin is characterized by comprising a base, a rotating driving mechanism, a center shaft, a rotary table, a classification bin and a bin lid unit, the base and the bottom of the center shaft are fixedly installed, the rotating driving mechanism comprises a driving plate and a driving motor, the driving plate is rotatably installed at the bottom of the center shaft, the rotary table which is located above the driving plate is rotatably installed on the center shaft, the rotary table is connected with the driving plate through a connecting column, the classification bin is installed on the rotary table and comprises multiple bin bodies, the bin lid unit is installed on the upper portion of the center shaft, the bin lid unit comprises a support, a bin lid covering the classification bin and an intelligent starting control panel located on the bin lid, the support is horizontally fixed on the top of the center shaft, the bin lid is installed on the support, and the bin lid is provided with an inlet opening capable of being opened and closed automatically. According to the intelligent classification dustbin, the structural design is scientific and reasonable, garbage can be subjected to detailed classification and recycled, maximized recycling of the garbage can be achieved, bacterial infection can be avoided, and odor dispersion is prevented.

a method for improving stacking schema for classification tasks, according to which predictive models are built, based on stacked-generalization meta-classifiers. Classifications are combined to build a new scheme from at least two layers and multiclass classification problems are converted into binary classification problems. One-against-all class binarization and regression learners are used for each class model and ensemble classifiers are improved using stacking. Accuracy differences, accuracy ratio, and runtime classification in multiclass datasets are also improved and the class of a value is then predicted.

The invention discloses an intelligent garbage classification monitoring device, which comprises a garbage temporary storage box, a door, a card reader, a card, a photographing assembly and a server, wherein the door is locked through a locking mechanism; the card reader is arranged on the inner side wall of the outer box body, and the chip of the card reader is in communication connection with the locking mechanism of the door; the photographing assembly comprises a first camera arranged around the garbage temporary storage box and a second camera arranged on the inner side wall at the top part of the outer box body; and the server is in communication connection with a weighing device, the card reader and the photographing assembly through a network module. Through real-time monitoring and acquiring picture information of people throwing garbage everyday and carrying out comprehensive analysis on the data, a reward is given to a resident with correct garbage classification and throwing, a resident with wrong classification is punished, the resident is promoted to automatically classify the garbage, and a virtuous cycle is realized. Besides, the invention also provides a system for using the intelligent garbage classification monitoring device for garbage classification management, visual management is carried out on garbage classification, and the working efficiency is improved.

The invention discloses a short text topic modeling method based on word semantic similarity. The method comprises: according to word semantic similarity provided by external, establishing a similar word set of short text centralized words; determining the number of topics used in modeling; randomly distributing the topic of each short text; through a Gibbs sampling process, iteratively determining the topic of each short text and the distribution of the words in the topic; according to a final distribution result of the above variable, feeding back the word distribution under each topic and the topic associated to each short text. The method preferably solves problems of sparse information contents of short texts and unclear semantic expression. According to the model result, short texts can be preferably expressed as topic vectors, and the topic vectors are used as final feature vectors of a short essay. The topic vector-based expression has good semantic interpretability, and can be used as algorithm basis of various applications. The method can be widely applied in various short text data, and has wide actual meaning and commercial values.

The invention provides an information processing device. The information processing device comprises a storage unit and a data processing unit. The storage unit is used for receiving and storing dataand commands. The data processing unit is connected with the storage unit, and is used for receiving the data and the commands both sent by the storage unit, extracting and calculating key features included in the data and generating multi-dimensional vectors according to calculation results. The invention further provides an information processing method. The information processing device and theinformation processing method have the advantages that human faces are recognized accurately and rapidly; offline neural network operation is supported, and human face recognition and corresponding control works can be realized when user terminals/front ends are offline without assisted calculation of cloud servers; high portability is achieved, the information processing device and the information processing method can be applied to various application scenes and equipment, and design cost is saved greatly.

A system and method for the use of sensors and processors of existing, distributed systems, operating individually or in cooperation with other systems, networks or cloud-based services to enhance the detection and classification of sound events in an environment (e.g., a home), while having low computational complexity. The system and method provides functions where the most relevant features that help in discriminating sounds are extracted from an audio signal and then classified depending on whether the extracted features correspond to a sound event that should result in a communication to a user. Threshold values and other variables can be determined by training on audio signals of known sounds in defined environments, and implemented to distinguish human and pet sounds from other sounds, and compensate for variations in the magnitude of the audio signal, different sizes and reverberation characteristics of the environment, and variations in microphone responses.

A method for categorising items being data entities stored in a computer system, the method comprising performing categorisation in such a manner that an item and a category are linked if a determined quantification of a relation between said item and said category fulfils a predefined criterion, the said method utilising a list of categories on which the categorisation is to be based, for each category comprised in the list of categorises at least one categorisation function for determining quantification for at least one relation between the category and an item, such as a number, a colour, and/or a text; the quantification of the relation(s) being determined by executing the categorisation function(s), for each item to be categorised, item data to be used for executing the categorisation function(s), the said method comprising, selecting a first set of categorisation functions and a first set of item data, (A) executing the categorisation function(s) comprised in the first set of categorisation functions on item data comprised in the first set of item data thereby determining a first set of quantification of relation(s), and (B) determining whether one or more of the quantification of relations determined fulfil(s) a predefined linking criterion and in case the linking criterion is observed then linking the item and category in question, and optionally selecting a new first set of categorisation functions and a new first set of item data and repeating step (A) and (B) for these new sets.

A discriminating method of parallel series patter and its application in the machinery smell, the characteristics is that regarding the big scale smell sorts and intension confirm problem as assortment problem, and then regarding as approximation problem, giving settling respectively for the mode which is made up of classifier layer and function approximation mode layer. It includes the four implementation steps: (1) n pieces one output neural net of classifier layer takes charge of ensuring the types of n kinds of smell, one to one correspondence. The training subset of a network is made up of all representative smell samples and the part apposition smell samples. (2) n pieces one output neural net of function approximation mode layer takes charge of ensuring the intensity of n kinds of smell, one to one correspondence. The training subset of a network is made up of all representative smell samples. (3) The suppositional balance of imbalance training subset. (4) n+1 pieces one output neural net puts up decision. By using the invention, the machinery olfactometer can solve the simultaneous estimation problem of thousands upon thousands smell types and intensity.

An apparatus for classifying videos according to an embodiment of the present invention comprises: an audio extractor for extracting an audio signal upon receiving video information; an audio signal classifier for outputting primary classification information from the audio signal; and a video classifier for performing secondary classification on video data of the video information using the primary classification information.

Methods and devices are provided for use in a streaming process (130) of a Virtual Reality [VR] video to a VR rendering device (120). The streaming process has a configurable setup, each configured setup providing respective different image data of a scene. The VR rendering device is arranged to render a current view of the scene based on a current head position. The streaming process is executed according to a first configured setup providing first image data needed to render the current view. EEG data is determined by measuring brain signals (115) and a head movement is predicted based on the EEG data. Target image data is determined as needed for rendering a future view based on the predicted head movement. A new configured setup for providing the target image data is determined. By adapting the streaming process based on the prediction the target image data can be provided in time and efficiently using available bandwidth.

The invention belongs to the technical field of biomedicines, and in particular relates to application of seaweed polysaccharides. The seaweed polysaccharides can be taken as preparations for preparing anti-viral or immune medicines. Cell models and in-vivo animal experiments show that the seaweed polysaccharides can be used for significantly enhancing the animal immunity, and can also be used for promoting the generation of cell factors, the typing of T lymphocytes and the proliferation of mouse spleen cells so as to ensure that cellular immune reaction can be activated. The functional seaweed polysaccharides disclosed by the invention are natural polysaccharides which are extracted from seaweeds such as large seaweeds and kelps, gulfweeds, grateloupia sparsa, eucheuma muricatum, ulva pertusa kjellm, enteromorpha, gracilaria, ascophyllum nodosum and scytosiphon lomentaria, and can also be low-molecular-weight seaweed polysaccharides or seaweed oligosaccharides which are prepared by degrading the seaweed polysaccharides by virtue of different preparation methods. According to the application disclosed by the invention, a polysaccharide extract of single seaweed or a polysaccharide extract mixture of a variety of seaweeds can be taken as a novel anti-viral and immune enhancing agent to be applied to feeds of livestock, poultry, fish, shrimps and shellfish, and has wide application prospects.

A system for controlling a menu based augmented reality (AR) Graphical User Interface (GUI) according to predefined head movement positions, comprising a head mounted AR display and one or more hardware processors adapted to execute a code, the code comprising code instructions to present one or more selection menus of a GUI displayed by the head mounted AR display, the selection menu(s) comprising one or more control display objects, code instructions to detect one or more predefined discrete head movement positions of the head mounted display by analyzing sensory data received from one or more orientation sensors monitoring orientation of the head mounted display, each of the predefined discrete head movement positions maps one of a plurality of navigation actions and code instructions to apply a respective navigation action mapped by the detected predefined discrete head movement position(s) on a currently pointed control object of the control display objects.

A time sequence mode representation-based weighted directed complicated network construction method comprises the steps of adopting a zero-mean normalization method to normalize an original time sequence; dividing a new time sequence into n sections in an equal probability manner, using the characters in a set character string to represent the sections, and representing the new time sequence into a character string sequence; moving a sliding window of which the length is 1 from left to right from the first character of the character string sequence, every time the sliding window moves one step, dividing the character string sequence into ((n-1)+1) fragments of which the lengths are all 1, and regarding each fragment as a mode; taking the different modes as the nodes of a complicated network, determining the connection edge weights and directions between the nodes of the complicated network according to the conversion frequency and the conversion directions between the nodes, and mapping the character string sequence into the weighted directed complicated network; and calculating the network topology statistical characteristics of the weighted directed complicated network. The method of the present invention enables the classification or identification precision of the time sequence signals to be improved remarkably.