63results about How to "Reduce false recognition rate" patented technology

The invention provides an access control system based on face recognition and a recognition method, and the system comprises a multi-modal image collection subsystem which integrates a visible light camera, a depth camera and an infrared thermal imaging camera and is used for synchronously collecting a visible light image, a depth image and an infrared thermal imaging image of a face; the image preprocessing unit is used for carrying out corresponding preprocessing on the image; the feature extraction network module is used for extracting image features; the multi-modal feature fusion and alignment module is used for calculating a mutual information matrix by adopting a feature fusion strategy based on mutual information maximization and adaptive weighting, determining an adaptive weight and performing weighted summation, and performing feature alignment by applying a method based on canonical correlation analysis; and the face recognition classification and verification module verifies the image by adopting a multi-level fusion strategy. According to the method, the false recognition rate and the rejection rate are effectively reduced, the strong anti-cheating capability is achieved, and cheating attacks of various fake face means can be accurately recognized and resisted.

The invention relates to the technical field of voice recognition, in particular to a mixed recognition processing method and device, equipment and a medium, and the method comprises the steps: recognizing command word data based on a command word recognition model, selecting command word data with an error rate lower than a preset range, generating a corresponding phoneme embedding representation, and forming a phoneme embedding dictionary; determining a mixed recognition command word corresponding to the current command word, and generating a mixed recognition list; identifying an audio embedding sequence corresponding to the current command word; obtaining a phoneme embedding sequence of each mixed recognition command word in the mixed recognition list based on a phoneme embedding dictionary; and calculating the similarity between an audio embedding sequence and each phoneme embedding sequence to obtain an identification result. The method and the device can improve recognition accuracy and reduce mixed recognition of command words.

The invention discloses a method and device for recognizing and analyzing PDF texts and pictures based on OCR and a storage medium, and particularly relates to the field of text and picture processing.The method for recognizing and analyzing PDF texts and pictures based on OCR comprises the following steps that S100, PDF documents are loaded, page analysis is conducted, page elements are extracted, and page layout information is processed; and step S200, for the extracted picture or the PDF page generated by scanning, performing high-precision OCR recognition by using a deep learning model, and converting characters in the image into editable texts. According to the method, efficient and accurate analysis of complex texts and picture contents in the PDF document is realized, the technical scheme not only improves the analysis precision and efficiency and enhances the flexibility and expandability of the system, but also promotes the development of digital and intelligent application of the document, and meanwhile, by saving resources and time cost, the method and the system have good application prospects. And remarkable economic benefits are brought to enterprises and individual users.

The invention belongs to the technical field of natural language processing, and discloses a natural language processing platform and method based on deep learning, and the platform comprises a data collection module which is used for collecting font data, text data, image data and environment data; the data processing module is used for preprocessing the collected font data, text data, image data and environment data and obtaining a comprehensive feature data set; the character recognition module is used for predicting character recognition results through a character recognition model and judging whether the character recognition results are consistent or not; the semantic recognition module is used for generating corresponding character paraphrases through a semantic recognition model; collecting character recognition results and corresponding character paraphrases through the software terminal; the voice output module is used for converting the character recognition result and the paraphrases of the corresponding characters into natural voices and outputting the natural voices by adopting a TTS technology; through the implementation of the technology, students can be helped to better learn and understand.

The invention relates to the technical field of electrical fire identification, in particular to an electrical fire intelligent identification method and system based on a visual image, and the system comprises a visual image fire identification platform, a fire information base, an infrared image analysis unit, a region identification unit, a feature matching unit, a dynamic management and control unit and an alarm response unit. According to the method, regional fire monitoring feedback analysis is performed on the infrared feature image preliminarily, so that the suspicious risk condition of each monitoring point in the target electrical equipment can be known, and data support is provided for subsequent analysis; whether the target electrical equipment has a fire or not is analyzed from the two aspects of infrared feature image processing and monitoring video image feature extraction, and through the mode of combining the two fire analysis methods, the fire recognition error rate is reduced, the target electrical equipment with a disaster is analyzed deeply, and the fire recognition efficiency is improved. Reasonable management and control can be carried out according to the dynamic change trend of the fire, and the fire management efficiency can be improved.

The invention relates to the technical field of identity verification, in particular to an identity verification method and system based on multi-modal data fusion, and the method comprises the following steps: based on user login information, comparing the login information with pre-stored information, recognizing a user identity, and extracting pre-stored face data and iris data of a user to obtain a user identity information extraction result. According to the method, the accuracy of identity verification is improved by combining face and iris recognition technologies, optimizing image quality and comparing pre-stored data, the possibility of wrong recognition and identity fraud can be remarkably reduced, and flexible response can be performed according to different security requirements by dynamically adjusting a matching degree verification threshold value; according to the method, the adaptability and the protection degree of the verification process are enhanced, the false identification rate and the false rejection rate can be effectively reduced, meanwhile, a stricter security verification process is provided, higher-level personal identity protection is realized, identity embezzlement and other related fraudulent behaviors are effectively prevented, and the trust of consumers to digital services is enhanced.

The invention relates to the technical field of gate channel target image recognition, and particularly discloses a gate channel target recognition detection method and system, and the method comprises the steps: collecting image data, carrying out the image preprocessing steps, such as illumination compensation and noise removal, extracting the brightness features and texture features in an image, and calculating the visibility coefficient of a target region. According to the method, high identifiability of the target is kept under various illumination conditions, the final identification accuracy of the target in different environments is evaluated by comprehensively analyzing the visibility coefficient and the identification confidence of the target, so that the image preprocessing and target identification algorithms are dynamically adjusted, the target can be rapidly and accurately identified, and the identification accuracy of the target is improved. The target identification efficiency in the gate channel is improved, the high efficiency and robustness of the system under the complex illumination condition are ensured, and the method has practical value and application prospect.

The invention relates to the field of campus energy-saving management, and discloses a smart campus energy-saving management method and system based on the Internet of Things, and the method comprises the following steps: transmitting collected power utilization data to a cloud server for storage; ultrasonic signals of the campus building are obtained and transmitted to the cloud server; defects existing in the campus building are recognized, and defect repairing is carried out; formulating energy-saving strategies of classes and public places; if the daily power consumption of the class or the public place exceeds the daily energy-saving power consumption target power, feeding back to a manager; the system comprises a power utilization data acquisition module, an ultrasonic data acquisition module, a model construction module, a power utilization equipment adjustment module and an energy-saving feedback module. The defects of the building structure can be found and repaired in time, potential safety hazards can be prevented, the thermal insulation performance of the campus building can be guaranteed by repairing the campus building, the place where water possibly leaks can be found in time, and energy consumption of the campus building is reduced.

The invention discloses a multi-mode defect automatic classification method, device and equipment and a storage medium. The method comprises the following steps: acquiring a microscope image, an X-ray image and defect feature information data of a to-be-detected product; first classification processing is preferentially carried out based on the microscope image, when the first classification processing result meets a first preset reliability condition, the first classification processing result is taken as a final classification result, and when the first classification processing result does not meet the first preset reliability condition, the final classification result is taken as a final classification result. Performing feature fusion based on the X-ray image and the defect feature information data, and executing second classification processing; when the second classification processing result meets a second preset reliability condition, the second classification processing result serves as a final classification result, and when the second classification processing result does not meet the second preset reliability condition, defect type judgment is conducted based on key parameters in the defect feature information data, and the final classification result is output; the classification accuracy and efficiency can be improved, and the error recognition rate is reduced.

A glass surface stain image recognition method applied to a window-cleaning robot relates to the technical field of image recognition, and comprises the following steps: acquiring image data, environmental perception data and tactile data of a target glass area according to a sensor terminal of the window-cleaning robot; constructing a shooting parameter lookup table, and updating shooting parameters of the sensor terminal in real time according to the environment sensing data in the current acquisition period and the shooting parameter lookup table; constructing a stain feature recognition model, and performing pre-estimated stain type labeling on the image data of the current acquisition period according to the stain feature recognition model; and performing correctness detection on the estimated stain type marked in the target glass area according to the tactile data of the current acquisition period, if a wrong stain type exists in the target glass area, obtaining the recognition correct rate of the stain feature recognition model, and performing local feature enhanced recognition operation or retraining operation according to the recognition correct rate. And the stain identification capability of the window cleaning robot is obviously improved.

The invention provides a biological multi-modal feature recognition method and device based on dynamic fuzzy reasoning. The method comprises the steps that an obtained face image, a fingerprint image and an iris image of a target object are preprocessed; performing feature extraction on the preprocessed face image, the fingerprint image and the iris image, and determining a three-dimensional Euclidean distance of the face, the fingerprint and the iris between the target object and an object to be verified in a database according to a feature extraction result; performing fuzzy reasoning based on the three-dimensional Euclidean distance of the face, the fingerprint and the iris to obtain a recognition score; comparing the identification score with a preset decision threshold value, when the identification score is greater than or equal to the decision threshold value, determining that the target object is matched with the to-be-verified object, otherwise, determining that the target object is not matched with the to-be-verified object; wherein the decision threshold value is obtained by training through a reward / punishment mechanism based on preset target recognition precision. The method can effectively cope with various environment changes, improves the recognition accuracy, and is suitable for application scenes needing extremely high safety.

The invention discloses a small sample face recognition method based on prototype calibration and adaptive fine tuning, which comprises the following steps: 1) pre-training an encoder on a large-scale face database, capturing rich semantic information and class distribution, constructing an evaluation set, and ensuring that training identity information is not overlapped with the constructed evaluation set; 2) initializing the weight of the classifier through a weight imprinting method, ensuring smooth transition, and inducing a discrimination structure of the encoder; 3) calculating the cosine similarity between the fine-tuning prototype and the pre-training prototype, fusing the pre-training information through weighted average, and calibrating the fine-tuning prototype with deviation; 4) introducing an adaptive interval loss function, adjusting the penalty intensity of the sample, and performing selective fine adjustment on the BatchNorm layer and the classifier weight; 5) using the domain perception similarity NAC to calibrate unknown category rejection, enhancing the similarity relation with the domain sample, and obtaining the final matching score. The method can improve the recognition precision and rejection capability of the model in a few-sample learning scene.

The present application relates to the technical field of sample pretreatment automation, and discloses a system of a full-automatic sampling tube cap screwing and pipetting workstation and a running method thereof.The system comprises a motion point definition module, a cap screwing adaptation identification module, a label scanning module, a pipetting planning and execution module, a cap screwing torque calling module and a real-time torque acquisition module.The system realizes continuous scanning and reliable identification of the cap profile in the motion state through a dynamic coordinate system and key motion points, dynamically calls matched torque parameters to execute cap screwing according to the identification result, and simultaneously acquires real-time torque data.The system synchronously completes sample label information analysis, pipetting path planning and pressure monitoring.The scheme realizes adaptive perception and closed-loop cap screwing control of the cap state, and improves the robustness of cap identification, the consistency of cap screwing action and the reliability of the overall process.

The embodiment of the application relates to the computer technical field, and provides an audio recognition method, related device and storage medium, the audio recognition method comprises the following steps: obtaining audio to be processed; performing frequency domain conversion on the audio to be processed to obtain target frequency domain acoustic characteristics of the audio to be processed, and performing feature extraction on the audio to be processed to obtain target audio attribute characteristics of the audio to be processed; performing feature fusion on the target frequency domain acoustic characteristics and the target audio attribute characteristics to obtain target audio characteristics; performing recognition on the target audio characteristics to obtain a recognition result, and the recognition result is used for indicating whether the audio to be processed is fake. The embodiment of the application can fuse the target frequency domain acoustic characteristics and the target audio attribute characteristics to obtain target audio characteristics containing richer and more comprehensive audio representation, so that accurate recognition can be carried out based on the target audio characteristics, and the accuracy of fake trace detection on the audio is improved.

The present application relates to the technical field of face recognition, and particularly relates to a recognition method and system based on a dynamic face database, a computing device and a storage medium, comprising a face tracking server configured to acquire face information obtained by a channel camera and a regional camera, and to merge the face information obtained by the channel camera and the regional camera; a central face server configured to store face information left by a user when registering remotely, and to respectively identify face information collected by the face tracking server from the channel camera and the regional camera, and to send the identification results to a channel face server and a regional face server; the present application reduces the number of face databases by using face tracking and face recognition technology, forms a face server with multiple face databases, improves the recognition efficiency by accurately simplifying the face databases, and reduces the false recognition rate of face recognition.

The application discloses a portrait inspection and identification method based on gait cycle decomposition and multi-phase force analysis, and comprises the following steps: extracting the skeleton key points of a target object in a target object image set in a comparison video and the confidence score of each skeleton key point; extracting the features of the target object in the target object image set according to the skeleton key points of the target object, and obtaining the target object features, wherein the target object features comprise basic features, gait cycle features, gait stage features and gait parameter features; integrating the basic features, the gait cycle features and the gait stage features to obtain comprehensive features; respectively calculating the similarity of the target object features and the comprehensive features in the comparison video to obtain similarity calculation results and overall similarity calculation results; and obtaining the identity identification result of the target object according to the similarity calculation results and the overall similarity calculation results.

The invention discloses an API path aggregation identification method and device and a storage medium, and relates to the technical field of electronic digital data processing, and the method comprises the steps: carrying out the preliminary aggregation identification of API path data based on the path structure features of the API path data; if the preliminary aggregation identification is not successful, performing deep aggregation identification on the API path data based on word segmentation features and / or statistical features of the API path data; and generating an aggregation identification result of the API path data. According to the method and the device, the API path aggregation identification accuracy is improved.

The invention provides a self-service vending method, a vending device, a vending room and a storage medium, the method is applied to the vending room, the vending room is provided with a delivery channel, and the method comprises the following steps: obtaining activity information of a user entering the vending room; information of taking the commodity by the user is acquired, and first prompt information is pushed, so that the user places the commodity on the inner side of the delivery channel; identifying commodity information on the delivery channel; generating first order information; pushing the first order information and confirming the first order information by a user; acquiring payment information and opening a door; by setting the delivery channel, the user puts the commodities into the delivery channel after selecting the commodities and identifies and confirms the commodities, so that the intelligence of commodity settlement is improved, the commodity identification error rate caused by misoperation of the user is reduced, and the consumption experience of the user is improved.

This application provides a training method for a fingerprint recognition network, a fingerprint recognition method, an apparatus, and a medium. The training method for the fingerprint recognition network includes: determining fingerprint feature matching maps based on a first initial fingerprint image and a second initial fingerprint image; forming a training sample set from multiple fingerprint feature matching maps; and training an original neural network based on the training sample set to obtain a fingerprint recognition network. The technical solution provided by this application solves the problem of difficulty in distinguishing whether two different fingerprint images belong to the same finger, thus reducing the false acceptance and false recognition rates of fingerprint recognition.

The invention relates to an unmanned aerial vehicle image high consequence area building identification method, system and device based on priori knowledge and a medium, and the method is characterized in that the method comprises the steps: obtaining a to-be-identified remote sensing image collected by an unmanned aerial vehicle at a set height and route parameters; inputting the obtained to-be-recognized remote sensing image into a pre-trained multi-scale space attention aggregation model to obtain a building probability graph of a high-consequence area building recognition result; wherein the multi-scale space attention aggregation model is constructed based on a U-Net model and is provided with a multi-scale space attention aggregation module and a self-adaptive loss function, and the multi-scale space attention aggregation module is used for carrying out multi-scale space attention aggregation processing on an input feature map; according to the method, the building in the remote sensing image acquired by the unmanned aerial vehicle can be accurately identified, and the method can be widely applied to the technical field of computer vision and remote sensing image processing.

The application relates to a multi-protocol adaptive control method, system, device and medium of an LED dimming power supply, which comprises the following steps: receiving a dimming signal output by an external dimming device, and preprocessing the dimming signal to obtain a target input signal; extracting protocol identification features from the target input signal, and constructing a protocol feature vector; matching the protocol feature vector with a preset protocol template library to generate a candidate protocol set and corresponding initial confidence; dynamically correcting the initial confidence by combining historical recognition results, signal quality indexes and output response consistency to determine a target protocol type; calling corresponding protocol analysis rules and control parameter templates according to the target protocol type to generate target dimming control parameters; and adjusting LED drive output based on the target dimming control parameters. The application has the effect of improving the adaptive identification capability of the LED dimming power supply for various dimming protocols.

The present application belongs to the technical field of photon counting imaging, and particularly relates to a pulse recognition method based on three-channel cross verification. The method comprises the following steps: S1: obtaining time sequence synchronized signal sequences of three channels output by a charge sensitive amplifier, and pre-processing the signal sequences of the channels to obtain respective deconvolution signals of the three channels; S2: designing a threshold based on the respective deconvolution signals of the three channels, and performing three-channel cross recognition on each pulse event by using the threshold to obtain a pile-up discrimination group and an amplitude extraction group; and S3: recording a pulse event with an amplitude to be extracted into a non-pile-up amplitude extraction group by using the relationship among a time interval of each pulse event, a pulse shaping width and an amplitude extraction width, and obtaining three-channel pulse amplitudes of each pulse event contained in the non-pile-up amplitude extraction group. The present application saves the time cost of debugging, and reduces the instability of results caused by human factors.

The invention discloses a Hash and feature selection-based software and hardware equipment fingerprint fine-grained generation method, and belongs to the technical field of power network software and hardware equipment fingerprint biomarker generation. Based on the information entropy and the self-correlation coefficient, the features used for generating the device fingerprints are screened out from the numerous features, representative and stable features are extracted from the device fingerprints, and selected feature parameters are converted into hash codes used for the device fingerprint generation and matching step; different devices are identified and distinguished hierarchically and accurately through a fine-grained device fingerprint generation step based on feature selection, a hash code and a Bayesian network.

The application discloses a carotid artery mis-segmentation processing method and device for robot autonomous scanning, and the method comprises the following steps: calculating an average height; controlling the ultrasonic probe to start moving from the starting point of longitudinal scanning, and acquiring a current ultrasonic image in real time; calculating the height mean value and the height standard deviation of the longitudinal profile of the blood vessel; judging whether the longitudinal profile of the blood vessel is a jugular vein; if yes, calculating the moving distance along the X-axis direction of the tool coordinate system at the next moment and the exertion strength at the next moment; otherwise, taking the preset step length and the preset strength as the moving distance and the exertion strength at the next moment respectively, determining the moving direction at the next moment based on the trend of the longitudinal profile of the blood vessel; controlling the ultrasonic probe to move to the position at the next moment, and re-executing the above process until the ultrasonic probe moves to the end point of the longitudinal scanning. The method effectively reduces the adverse effect of carotid artery mis-segmentation on scanning, and ensures that the carotid artery can be continuously and stably tracked.

This application relates to the field of image processing technology and provides a method and electronic device for air gesture recognition. The air gesture recognition method includes: acquiring information of a hand image captured by a front-facing camera at a first moment; acquiring the screen content display direction of the electronic device at the first moment; determining a target preprocessing operation corresponding to the screen content display direction; performing the target preprocessing operation on the hand in the hand image to obtain an image to be recognized; and recognizing air gestures based on multiple frames of images to be recognized. This enables the finger orientation recognized by the electronic device during air gesture recognition to be consistent with the finger orientation of the user in the actual physical space, thereby ensuring compatibility with air gesture recognition under various screen content display directions and reducing the probability of misrecognition of air gestures.

The invention provides a method for enhancing control recognition and reducing control misrecognition of Bluetooth equipment, which comprises the following steps of: S1, setting a g-sensor and a touch device in the Bluetooth equipment in a set manner, and performing synchronous detection on the operation of a user; s2, during knocking, x / y / z three-axis data collected by a g-sensor is read, calculation is carried out by using an algorithm, and whether the knocking is effective knocking is judged; meanwhile, detecting whether the knocking is effective knocking or not through a touch device; and S3, combining detection results of the g-sensor and the touch, if the g-sensor and the touch are identified as effective knocks, regarding the knocks as effective knocks and performing corresponding functions, and if one of the g-sensor and the touch is invalid knocks, regarding the knocks as invalid knocks and not performing corresponding functions. According to the invention, during user control, the recognition accuracy and the control experience of control can be effectively improved, and the error recognition rate of user control caused by error recognition during use is reduced.

The invention provides a protocol identification method and system of photovoltaic equipment, computer equipment and a storage medium. According to the method, firstly, the address where the photovoltaic equipment possibly exists is predicted to improve the query efficiency and save the query time, and the target Modbus protocol is selected from the pre-constructed Modbus protocol library to poll the predicted address to determine whether the photovoltaic equipment exists in the predicted address and whether the existing photovoltaic equipment is the target photovoltaic equipment; and under the condition that the photovoltaic equipment exists in the predicted address, the target Modbus protocol is sent to the photovoltaic equipment again, so that the type of the photovoltaic equipment is further verified, the error recognition rate is reduced, the recognition accuracy of the protocol matching process of the photovoltaic equipment is improved, and the efficiency of the whole protocol automatic recognition process is ensured.

The present application relates to the field of image processing and target image recognition technology of civil optical equipment such as intelligent traffic, city supervision and commercial monitoring, and particularly relates to a multi-target characteristic recognition method based on photoelectric image, which aims to improve the adaptability and accuracy of target recognition in target optical imaging, and the method obtains target motion image through photoelectric equipment, divides the target into relative static, straight line motion and turning state according to a preset threshold, extracts target motion trajectory, judges target state, respectively carries out straight line and arc line fitting on the target trajectory of straight line motion and turning state, analyzes the overall motion state of multiple targets in the imaging image based on the motion state trajectory, and adopts corresponding individual motion trajectory statistical analysis method for different states, the multi-state adaptive analysis capability of the method significantly improves the recognition accuracy, and the recognition accuracy of the target with frequently changing state in a complex scene is improved by more than 30%.

This invention discloses a signal recognition method, apparatus, computer device, and storage medium. The method includes: acquiring a signal indicating ventricular fibrillation in a human body; performing time-frequency analysis on the signal to obtain the frequency content of different frequency bands corresponding to the i-th time point within a preset Hertz range; where i ranges from 1 to N, and N is the total number of time points; and identifying the type of the signal based on the frequency content of different frequency bands corresponding to each time point, wherein the type includes non-shockable ventricular tachycardia signals and ventricular fibrillation signals. In this technical solution, by utilizing the different frequency content of non-shockable ventricular tachycardia signals and ventricular fibrillation signals, the two signals can be effectively distinguished, thereby reducing the probability of misidentification of non-shockable ventricular tachycardia signals.

The invention discloses an electric power Internet of Things terminal equipment identification method and system based on network traffic, and belongs to the technical field of electric power Internet of Things terminal equipment identification. Traffic data acquisition is carried out on accessed terminal equipment, feature extraction is carried out on the acquired traffic data, and model training is carried out according to known equipment types; according to the method, automatic identification of the newly accessed terminal equipment is realized without modifying the configuration of the field terminal equipment or installing an agent program, the field deployment difficulty is greatly reduced, the vector is constructed through the multi-dimensional flow characteristics, the terminal equipment is identified from a multi-dimensional angle, the probability of misidentification is reduced, the identification accuracy is ensured, and the whole-process automatic identification is realized, so that the efficiency is improved. Manual participation is not needed, and the recognition efficiency is remarkably improved.