186results about How to "Reduce misidentification" patented technology

A method for speech recognition includes providing a source of geographical information within a vehicle. The geographical information pertains to a current location of the vehicle, a planned travel route of the vehicle, a map displayed within the vehicle, and / or a gesture marked by a user on a map. Words spoken within the vehicle are recognized by use of a speech recognition module. The recognizing is dependent upon the geographical information.

A feature extraction method includes: the step of grouping a cluster of features, in which an internal of the respective features is less than or equal to a predetermined grouping interval, to form a feature group, for a plurality of features of which feature information including at least information of a position and a feature type is included in a predetermined feature information storage unit; the step of excluding the feature, not suitable for use in an image recognition process of the feature with respect to image information, from the cluster of the features within the feature group; and the step of extracting a part or all of one or more of the features within the feature group remaining as a result of the exclusion step as a target feature suitable for the use in the image recognition process.

In conventional systems using an onboard camera disposed rearward of a vehicle for recognizing an object surrounding the vehicle, the object is recognized by the camera disposed rearward of the vehicle. In the image recognized by the camera, a road surface marking taken by the camera appears at a lower end of a screen of the image, which makes it difficult to predict a specific position in the screen from which the road surface marking appears. Further, an angle of depression of the camera is large, and it is a short period of time to acquire the object. Therefore, it is difficult to improve a recognition rate and to reduce false recognition. Results of recognition (type, position, angle, recognition time) made by a camera disposed forward of the vehicle, are used to predict a specific timing and a specific position of a field of view of a camera disposed rearward of the vehicle, at which the object appears. Parameters of recognition logic of the rearwardly disposed camera and processing timing are then optimally adjusted. Further, luminance information of the image from the forwardly disposed camera is used to predict possible changes to be made in luminance of the field of view of the rearwardly disposed camera. Gain and exposure time of the rearwardly disposed camera are then adjusted.

When a transfer signal according to ultrasonic echoes is wirelessly transmitted from an ultrasonic probe to an ultrasonic diagnostic apparatus main body, the main body and the probe are reliably connected without false recognition. An ultrasonic diagnostic apparatus includes an ultrasonic probe and an ultrasonic diagnostic apparatus main body, and the ultrasonic probe includes a probe ID transport unit having a transport distance shorter than that of a first wireless communication unit for transporting a probe ID for identification of itself in contact or noncontact to an outside, the ultrasonic diagnostic apparatus main body includes a probe ID acquiring unit for acquiring the probe ID transported from the probe ID transport unit, and a second wireless communication unit receives the transfer signal from the ultrasonic probe having the probe ID acquired by the probe ID acquiring unit.

A system and method of identifying and communicating with an interventional medical device is provided. The system includes a novel catheter handle operably coupled to the shaft of a catheter at the distal end and an electrophysiology recording system at the proximal end. The catheter handle includes a visual identification system visible through a lens on the catheter handles and a microcontroller with a memory structure, the memory structure including catheter identification information. The visual identification system includes a light emitting diode, which is visible from the outside of the catheter handle through the lens. When a user of the EP recording system selects a desired color for the coupled catheter, the desired color is communicated to the microcontroller in the catheter handle and the microcontroller causes the visual identification to display the requested color so that the desired color is visible by the user.

The invention discloses an electronic smoking device with a fingerprint identification function and usage. The electronic smoking device comprises a shell, an atomizer and a power supply unit, a smokable material source is loaded in the shell, the atomizer is used for producing the smokable material source into smoke inhaled by a person, a microcontroller and a fingerprint identifier are arranged in the shell, and the microcontroller is connected with a starting switch for starting the atomizer; the fingerprint identifier is provided with a fingerprint acquisition surface, and the fingerprint acquisition surface is located on the outer surface of the shell and can touch the fingerprint of a finger of the user; the fingerprint identifier is used for identifying whether the acquired fingerprint information of the user is matched with prestored fingerprint data and outputting a feedback signal to the microcontroller; only when the fingerprint information of the user is matched with the prestored fingerprint data, can the microcontroller activate the starting switch, and the user obtains the permission for operation.

The invention relates to a voice recognition method and method based on sound source localization and an intelligent household appliance. The voice recognition method comprises the following steps of: when the intelligent household appliance enters a voice recognition state, acquiring a voice signal, and controlling a timer to start timing; if the voice signal is acquired when the time recorded by the timer is less than or equal to a preset time, performing location according to the voice signal, and determining position information a sound source corresponding to the voice signal; and determining whether to control the intelligent household appliance to perform recognition of the voice signal or not according to the position information and prestored voice recognition area information, thereby removing interferential voice signals, effectively improving recognition rate of the voice recognition system, and promoting interactive experience of a user.

The voice recognition processing apparatus includes a voice acquirer, a first voice recognizer, a storage device, and a recognition result determiner. The voice acquirer acquires a user's voice, and outputs voice information. The first voice recognizer converts the voice information into first information. The storage device previously stores a dictionary in which an exclusion vocabulary is registered. The recognition result determiner compares the first information with the exclusion vocabulary to determine whether the first information includes a word that agrees with a word included in the exclusion vocabulary. The recognition result determiner determines that the first information is information to be rejected, when the first information includes the word that agrees with a word included in the exclusion vocabulary, and determines that the first information is information to be executed, when the first information does not include the word that agrees with a word included in the exclusion vocabulary.

An elevator landing call register comprising a landing call registering button (11), a speech input device (12), a speech recognizing device (13) for receiving a speech signal of a user and recognizing the destination floor, and a call control unit (15) for receiving the outputs from the registering button, the speech input device, and the speech recognizing device and reporting to an elevator control device (2). When the call registering button (11) is operated in predetermined procedures, the call control unit (15) enables the speech recognition function of the speech recognizing device and reports the destination floor recognized by the speech recognizing device to the elevator control device.

A game system includes a game apparatus and a controller. The controller is furnished with an acceleration sensor for detecting accelerations in at least two axis directions. Game processing corresponding to the kind of an acceleration input by means of the controller is executed. For determining the kind, reference timing when acceleration in a first-axis direction is below a threshold value to take on a minimum value is detected. Then, it is determined whether or not an angle between acceleration change vectors before and after the reference timing is equal to or more than a predetermined angle. When the angle is not equal to or more than the predetermined angle, it is determined that the acceleration input is an acceleration input in any one of the two-axis directions, and when the angle is equal to or more than the predetermined angle, it is determined that the acceleration input is an acceleration input in a direction including the two-axis directions as components.

The invention provides an automatic e-commerce website duplicate picture detection method based on an SIFT algorithm. The method comprises the following steps: 1) carrying out screenshot pretreatment on pictures to be processed in a website, and keeping 15%-85% of upper and lower intervals of the pictures; 2) scale space construction: through scale transform of the pictures, obtaining a scale space representation sequence of the images under multiple scales; 3) key point positioning: at each candidate position, determining location and scale through a fitting fine model; 4) key point direction determination: based on image local gradient direction, allocating one or more direction to each key point position; 5) key point description: in neighborhood surrounding each key point, measuring image local gradient in the selected scale; 6) key matching: carrying out pair-wise comparison through a describer in the two images and finding out a plurality of pairs of matched feature points; and 7) similarity calculation: through a customized picture similarity calculating formula, judging whether the pictures are duplicate pictures.

In conventional systems using an onboard camera disposed rearward of a vehicle for recognizing an object surrounding the vehicle, the object is recognized by the camera disposed rearward of the vehicle. In the image recognized by the camera, a road surface marking taken by the camera appears at a lower end of a screen of the image, which makes it difficult to predict a specific position in the screen from which the road surface marking appears. Further, an angle of depression of the camera is large, and it is a short period of time to acquire the object. Therefore, it is difficult to improve a recognition rate and to reduce false recognition. Results of recognition (type, position, angle, recognition time) made by a camera disposed forward of the vehicle, are used to predict a specific timing and a specific position of a field of view of a camera disposed rearward of the vehicle, at which the object appears. Parameters of recognition logic of the rearwardly disposed camera and processing timing are then optimally adjusted. Further, luminance information of the image from the forwardly disposed camera is used to predict possible changes to be made in luminance of the field of view of the rearwardly disposed camera. Gain and exposure time of the rearwardly disposed camera are then adjusted.

A display device includes a laminated wiring formed of a low-resistance conductive film, and a low-reflection film mainly containing Al and functioning as an antireflective film which are sequentially arranged on a transparent substrate, a wiring terminal part provided at an end part of the laminated wiring and has the same laminated structure as that of the laminated wiring, and an insulating film for covering the laminated wiring and the wiring terminal part, in which the insulating film side serves as a display surface side, the wiring terminal part has a first opening part penetrating the insulating film and the low-reflection film and reaching the low-resistance conductive film, and an outer peripheral portion of the first opening part has a laminated structure of the low-resistance conductive film, the low-reflection film, and the insulating film, in at least one part.

Disclosed are a method and a radar apparatus for transmitting a transmission signal at a controlled timing in order to avoid signal interference, thereby exactly detecting a target object without misrecognition.

A method for speech recognition includes providing a source of geographical information within a vehicle. The geographical information pertains to a current location of the vehicle, a planned travel route of the vehicle, a map displayed within the vehicle, and / or a gesture marked by a user on a map. Words spoken within the vehicle are recognized by use of a speech recognition module. The recognizing is dependent upon the geographical information.

The invention relates to an augmented reality-based Chinese Pinyin combination recognition and reading learning system. The augmented reality-based Chinese Pinyin combination recognition and reading learning system includes recognition and reading Pinyin cards with pinyin pictures, a Pinyin picture image recognition module, a Pinyin combination processing module, a Pinyin recognition and reading module and a multimedia module; the Pinyin picture image recognition module calls the Pinyin combination processing module to process obtained Pinyin data when encountering a plurality of Pinyin cards; the output end of the Pinyin picture image recognition module is connected with the input end of the Pinyin recognition and reading module, so that multimedia unit marks formed by initial consonant combinations and final vowel combinations of the Pinyin can be obtained from the recognition and reading Pinyin cards with the Pinyin pictures and are recognized; the output end of the Pinyin recognition and reading module is connected with the input end of the multimedia module, so that specific pronunciations of Pinyin combinations and multimedia images related to the pronunciations can be outputted; the recognition and reading Pinyin cards with the Pinyin pictures are cards printed with Pinyin initial consonant images or final vowel images; and each card is printed with a Pinyin picture with unique identification features.

The invention discloses a commodity identification method and device, and belongs to the technical field of the Internet. In the invention, each image in the plurality of images is identified throughthe commodity identification model. Since the commodity recognition model is obtained by training a plurality of sample images containing hands and commodities, the hands and the commodities containedin the images can be detected according to the commodity recognition model. On the basis, a plurality of target images containing the hand and the commodity held by the hand can be determined according to the detection result of the image obtained by the commodity identification model. Therefore, the commodity identification result is determined according to the position information of the hand and the commodity held by the hand in each target image and the acquisition time of each target image; and equivalently, other interference information contained in the image is filtered, and only thecategory of the commodity held by the hand is concerned, so that error identification of other irrelevant commodities is reduced, and the commodity identification accuracy is improved.

A medical apparatus including a main unit including an X-ray CT imaging unit configured to image a patient and a display unit provided on the X-ray CT imaging unit and configured to display a name of the patient to be imaged, and a console unit remote from the main unit and including an input unit configured to input condition information of the main unit and individual information including the name of the patient. The console also includes a control unit configured to extract at least one item including the name of the patient in the individual information from the inputted condition information and the inputted individual information. The display unit is configured to display the extracted item including the name of the patient on a screen.

The embodiment of the application discloses a training method and device for an image recognition model, which obtains a plurality of application images of an application and determines the color distribution characteristics corresponding to each application image respectively, and performs the image clustering according to the color distribution characteristics of the plurality of application images to obtain a plurality of image sets. In the same set of clustered images, the color distribution features of the applied images are similar to each other, which generally accords with the first threshold, so that the applied images in the same set of images are likely to belong to the same set of application scenes. Based on the image set obtained by the clustering mentioned above, the image recognition model is trained according to a class of image sets, so that the image recognition model can be used to recognize the application images in the application scenes corresponding to this class of image sets with high accuracy, thus reducing the possibility of false recognition results.

The invention provides a target identification method based on a deep learning network and a hot-line robot. Weights and offsets on each layer in a convolution neural network are initialized; two-dimensional image data is input for convolution to obtain C1-layer data; the maximum pooling treatment is performed for the C1-layer data to reduce the magnitude of the data to obtain S2-layer data; the convolution is performed for the S2-layer data to obtain C3-layer data; the maximum pooling treatment is performed for the C3-layer data to further reduce the magnitude of the data to obtain S4-layer data; the convolution is performed for the S4-layer data to obtain C5-layer data; the C5-layer data serves as input to totally connect with a H6 layer, and an output layer is totally connected with the H6 layer to output learning results of target characteristics; and the weights and the offsets are finely adjusted by applying a back propagation algorithm to finish the network learning. The method is high in target identification rate and excellent in robustness under complicated background.

A pressurized flow of breathable gas is delivered to the airway of a subject in accordance with a therapy regimen. One or more fluid parameters of the pressurized flow of breathable gas are adjusted based on the therapy regimen. The therapy regimen dictates that such adjustments be made based on the respiratory state of the subject. Transitions in respiratory state are identified without relying on measurement or estimation of flow at or near the airway of subject. Transitions in respiratory state are identified based on changes in the first time derivative of flow at or near the airway of the subject. In one embodiment, an effort parameter is determined that approximates the second time derivative of flow. Based on comparisons of the effort parameter to a dynamic threshold, transitions in respiratory state are identified.

The invention discloses a shot-based television advertisement recognition method. The method comprises the following steps: firstly, decomposing continuous video data into shots; then extracting the features of each shot by a convolution neural network; performing dimension reduction processing on feature vectors; classifying the shots into advertisements and non-advertisements by a trained classification model; and after advertisement shots are classified, recognizing the content of the advertisement shots by using image and audio features in the shots, that is, recognizing the shots are advertisements of which company or product. The aim of the invention is to monitor commercial advertisements broadcast in television broadcasting; the advertisements can be recognized in various programssuch as movies, TV shows, variety shows, news and sports, and among different programs; the calculation cost can be lowered; and the efficiency is increased.

The embodiment of the invention discloses a protocol identification method, a protocol identification device and a protocol identification system. The method comprises the following steps of: extracting an identification tuple of a message, wherein the identification tuple at least comprises a network address; searching an algorithm application list corresponding to the identification tuple in an identification table; and performing content identification on the message by using an algorithm in the algorithm application list to acquire a protocol type of the message. According to the implementation mode, the identification tuple which at least includes an Internet protocol (IP) address and is extracted from the message is adopted for searching the corresponding algorithm, and the searched algorithm is used for content identification of the message, so that the identification precision is improved and the error identification of the message is reduced.

The invention discloses a speech recognition control device, which is suitable for being connected with more than one intelligent furniture. The speech recognition control device comprises a speech information acquisition module, a speech recognition module and a first information receiving / sending module. The speech information acquisition module and the first information receiving / sending module are connected with the speech recognition module; the speech information acquisition module is used for collecting speech information, and sending the collected speech information to the speech recognition module; the speech recognition module is used for carrying out speech recognition on the received speech information, and generating an instruction according to recognition result; and the first information receiving / sending module is used for sending the instruction to the intelligent furniture corresponding to the instruction. The speech recognition control device can enable a user to have a better user experience. The invention also discloses an intelligent furniture system.

The invention discloses a visual detecting and counting method for the vehicle flow rate under a complex road condition, belongs to the technical field of machine vision, and solves the problems that a visual detection method for counting the vehicle flow rate by counting the area of moving objects is extremely low in the counting accuracy and greatly influenced by the image quality under complex roads in which vehicles, pedestrians and non-motor vehicles are mixed. The method comprises that initial information is set; video images are read in a cyclic way, and an MSRCR algorithm is used to realize illumination equalization and color balance at the same time; an AdaptiveMedian algorithm is used for background modeling, and a movement foreground is extracted; opening operation is carried out on the foreground, and bonded moving objects are separated while noises are removed; a contour is searched, the area is calculated, and the contour is filled if the area is lower than a threshold; vehicles in the foreground are identified by utilizing Haar characteristics and an AdaBoost classifer; and vehicles in each frame of foreground are identified, and positions of the vehicles are recorded. Thus, the method is used to extract moving objects under the complex road condition accurately, identify the vehicles specifically, and counting the vehicles in the real-time collection video.

The invention discloses a method and a device of combined verification of an image capable of visual recognition and a two-dimensional code. The method includes: before code scanning, carrying out, bya user, checking and comparison of whether a first image is consistent with a real scene; after confirming, starting a camera function of mobile intelligent equipment; carrying out scanning and recognition on a first two-dimensional code through software APP; carrying out autofocus photographing on the first image at the same time, wherein the first image is at least bound to one first two-dimensional code; sending the first two-dimensional code of code scanning and recognition and the first image of photographing to a website server; receiving the first two-dimensional code and the first image by the website server, and carrying out comparison and verification with a stored second two-dimensional code and second image of the website server; and checking a binding relationship between thefirst two-dimensional code and the first image. The method can enable security, of using static two-dimensional codes for payment, to be greatly improved.

The invention discloses a license plate character recognition method based on data enhancement and data generation. According to the method, a simulated license plate picture can be randomly generated; after the simulated license plate picture is obtained by the image generator of a cyclic adversarial network, and the simulated license plate picture is mixed with a simulated license plate pictureand an enhanced data picture to generate training data, so that optimization training can be trained on a license plate recognition network; and finally, the trained recognition network is subjected to rapid license plate recognition. According to the method of the invention, a license plate can be correctly and rapidly identified under the influence of low light, low resolution, motion blur and other severe conditions. Meanwhile, a large amount of data needed by neural network training is obtained in a data enhancement and data generation mode, and a large amount of manually-labeled data is not needed.