42 results about "Tongue region" patented technology

Systems and methods selectively stabilize tissue in the oral cavity. The systems and methods implant a first implant element in a bone region within an oral cavity and implant a second implant element in a tongue region within the oral cavity. The systems and methods resist posterior movement of the tongue within the oral cavity by coupling the first implant element to the second implant element within the oral cavity. By coupling, the systems and methods apply force, e.g., tension, between the first and second implant elements.

A HGA comprises a slider and a suspension with a flexure having a tongue region for supporting the slider. A read / write transducer and a piezoelectric element are formed oppositely. The connecting points of the curve beams and the inner tongue of the tongue region are in mirror positions to a center point of the inner tongue, and the connecting points of each curve beam are located at opposite sides of a center axis of the flexure. The slider has multiple electrical pads electrically connecting with the read / write transducer. The inner tongue has multiple electrical pads. The flexure has multiple inner leads electrically connected with the electrical pads of the inner tongue formed on the curve beams. The structure of the HGA prevents the read / write transducer from damaged and cause the manufacture of the HGA simpler. The present invention also discloses a suspension and a disk drive unit.

The invention discloses an automatic detection system and method for a tongue image in the traditional Chinese medicine technology, wherein the system and method are applied to a computer. The method comprises: an image acquisition device is controlled to shoot a clear tongue image; the tongue image is segmented to obtain a tongue image; a color feature and a texture feature of a coating on the tongue are extracted from a tongue region and feature values of tongue coating thickness, tongue coating dryness, and tongue coating greasyness are analyzed based on the color feature and texture feature of the coating; the length and the width of the tongue print are detected and are quantified into a feature value for describing a tongue print situation; the tongue surface size is detected and is quantified into a feature value for describing the thin degree of the tongue body; and the features of the tongue coating thickness, the tongue coating dryness, the tongue coating greasyness, the tongue print situation, and the tongue thin degree are combined into a multi-dimensional tongue image feature vector. The image acquisition device can be controlled automatically to shoot the clear tongue image and various attributes of the tongue body can be analyzed and identified automatically, so that the reference basis can be provided for the tongue diagnosis in the traditional Chinese medicine technology.

Systems and methods for accurate position adjustment of thin-film PZT micro-actuators are provided. In certain example embodiments, a head gimbal assembly is provided. A suspension includes a suspension flexure located at one end of the suspension, and the suspension flexure includes a tongue region. At least one micro-actuator is mounted to a micro-actuator mounting region of the tongue region of the suspension flexure. Any gap between the slider and the at least one micro-actuator may be reduced and / or eliminated. A slider is substantially fully mounted on an insulating layer, and the insulating layer separates the slider from the at least one micro-actuator. The suspension flexure may be located substantially fully underneath the slider, and the tongue region of the suspension flexure may be substantially equally weighted around a dimple formed on the suspension. The rigidity of the suspension flexure may be increased by one or more supports formed thereon.

The invention discloses a physique analysis method based on face and tongue image acquisition, which comprises the following steps: a doctor initially inputs personal information of a patient user andestablishes a personal electronic medical record for the patient user to archive; Under the standard illumination environment, the images of the patient's face and tongue were collected, and then theimages were preprocessed. The facial contour region is segmented to extract the color and luster information of the five internal organs feature regions, and then the tongue region of the front tongue image is segmented to extract the tongue quality and tongue coating information of the tongue body. Based on the YCbCr color space, the color and luster of the five internal organs characteristic regions were analyzed, and then the tongue quality and tongue coating information were analyzed, and the results of physique test were obtained synthetically. Generate and archive physique analysis reports. The method can improve the efficiency of physician's constitution diagnosis.

A suspension for a HGA includes a flexure having a tongue region and an out trigger. The out trigger has a pair of rigid beams and a pair of spring beams. The pair of spring beams enables the flexure to possess favorable flexibility and assists a slider to have enough displacement. The pair of rigid beams provides the flexure enough stiffness which not only successfully prevents the HGA with the out trigger from suspension deformation and dimple separation, but also assists to improve shock performance of the HGA. Furthermore, such configuration of the suspension provides the slider a good static performance and thus improves dynamic performance of the HGA. The present invention also discloses a HGA with the suspension, a manufacturing method of the HGA, and a drive unit with such HGA.

The invention discloses a tongue area recognition method and device and a computer storage medium, and relates to the field of artificial intelligence. The tongue area recognition method comprises thefollowing steps: acquiring a tongue image; performing image recognition on the tongue image through a target detection algorithm to obtain target tongue information; inputting the target tongue information into a trained Unet segmentation network for region segmentation to obtain tongue root information, tongue middle information, tongue tip information and tongue edge information. The Unet segmentation network adopts full-scale jump connection, and each layer of the Unet segmentation network is provided with a 1 * 1 convolution kernel; inputting the tongue root information, the tongue middleinformation, the tongue tip information, the tongue edge information and the target tongue information into a trained deep learning network for classification processing to obtain a tongue attribute,a tongue root attribute, a tongue middle attribute, a tongue tip attribute and a tongue edge attribute; in this way, the accuracy of tongue area recognition can be improved to assist in tongue diagnosis judgment.

A suspension for a HGA includes a suspension flexure and an out trigger. The suspension flexure has a tongue region having a first flexure part in a tip portion thereof for mounting a trailing edge of a slider and a PZT mounting region for mounting PZT elements. One end of the out trigger is connected to the suspension flexure, and the other end of the out trigger has a trigger support portion which extends above the tongue region so as to support the leading edge of the slider. The structure of the out trigger is capable of stiffening the tongue region of the suspension flexure and, in turns, avoiding the deformation problem and the dimple separation problem during the manufacturing process or in vibration or shock events. The invention also discloses a HGA with the suspension, manufacturing methods of the HGA and a disk drive unit having such HGA.

The invention discloses a tongue coating tongue quality analysis method and system, computer equipment and a storage medium. The method comprises the following steps: acquiring an image of a tongue region from a face image comprising the tongue region according to a tongue positioning model; Separating the tongue part from the image of the tongue part area by using an elliptical template to form atongue part image; Dividing the tongue image into a tongue image and a tongue coating image according to a coating separation algorithm; And matching the tongue image with a pre-stored tongue color classification model, and matching the tongue coating image with the tongue coating color classification model to obtain tongue classification and tongue coating classification. By adopting the method,the speed and the accuracy of separating the tongue coating from the tongue in the image can be improved.

The invention relates to a pump body of a centrifugal pump, which comprises a pressure tapping hole on the pump body and a baffle tongue inside the pump body; the position of the pressure tapping hole is designed in a mode that an auxiliary circle in the central section of the pump body taking the tongue tip of the baffle tongue as a center of the circle is tangential with the inner wall of the pump body; and the extension line of the connecting line between the tongue tip and the tangent point is used as the centerline of the pressure tapping hole. The pump body has the advantages that: firstly, the pump body can judge the working condition of the operation of the centrifugal pump through monitoring pressure pulsation amplitude value; and under the same working condition, the change of the pressure pulsation amplitude value also reflects the change of the position of a blade relative to the baffle tongue; and secondly, the condition of pressure pulsation in a baffle tongue region reflects the stability of the operation of the centrifugal pump in corresponding degree; therefore, the pump body can effectively avoid the operation of the centrifugal pump under the working condition of large pressure pulsation and prolong the service life of the centrifugal pump.

The invention relates to the field of image detection, capable of extracting a tongue region from a face image, carrying out recognition on the extracted tongue region based on a trained convolutionalneural network model, and outputting a corresponding physique type, wherein the convolutional neural network model needs less data to be processed, and is less interfered by background data, and theidentification speed is higher, and the precision is higher. The invention particularly discloses computer equipment, a tongue photo physique recognition device and a storage medium. The computer equipment is implemented by the following steps: acquiring data of a face image; acquiring data of upper lip key points and chin key points in the face image based on a face key point detector according to the data of the face image; according to the data of the upper lip key points and the chin key points, extracting data of a tongue body region from the data of the face image; and based on the trained convolutional neural network model, identifying the physique type corresponding to the tongue body region according to the data of the tongue body region so as to output the data of the physique type.

The invention provides a tongue image segmentation method based on color decomposition and the threshold technology. The method comprises steps that S1, a tone component of a tongue image in the HSI color space is transformed, and tone components of a tongue and an upper lip after transformation and the neighbor tissue are made to have tone difference; S2, image threshold segmentation of the transformed tone component is carried out to obtain the binary segmentation result, after morphological operation of the binary segmentation result is carried out, the initial tongue region containing thetrue tongue and the upper lip is extracted; and S3, the initial tongue region is corrected, two image threshold segmentation methods are selected, one of segmentation results of the two image threshold segmentation methods is utilized to separate the true tongue region from the upper lip region, and the final segmentation result of the tongue image is obtained. The method is advantaged in that themethod is simple and effective, and image segmentation performance is substantially improved.

An intelligent tongue segmentation method based on depth learning relates to image processing. According to the method, tongue image acquisition standards are formulated, and tongue image data are collected in standard environment, wherein the acquisition standards include equipment aspect, light aspect, position aspect and patient requirement aspect, etc.; the tongue image data is segmented artificially, and the data set is divided into a training set and a test set to verify the effectiveness of the algorithm; a tongue Body Segmentation Network TS-NET encoder acquires the low-level featuresof tongue image and locates the pixels, the Tongue Body Segmentation Network TS-NET decoder fuses the high-level and low-level features and classifies the pixels to obtain the complete tongue region.

The invention discloses a method and system for determining image sharpness based on a deep learning algorithm, and the method comprises the steps: obtaining an original face image data set comprisinga tongue region, and carrying out the marking of sharpness; performing data enhancement processing on the labeled original face image data set to obtain a face image data set subjected to data enhancement; building a deep network on the basis of a framework of a residual error network; establishing an association relationship between an optimizer and the residual network to form an image definition judgment model, and training and testing the image definition judgment model by using the face image data set subjected to data enhancement to determine a trained image definition judgment model; and preprocessing a to-be-detected face image containing the tongue region, and analyzing the preprocessed face image by using the trained image definition judgment model to determine the image definition of the to-be-detected face image containing the tongue region.

A HGA comprises a slider and a suspension with a flexure having a tongue region for supporting the slider. A read / write transducer and a piezoelectric element are formed oppositely. The connecting points of the curve beams and the inner tongue of the tongue region are in mirror positions to a center point of the inner tongue, and the connecting points of each curve beam are located at opposite sides of a center axis of the flexure. The slider has multiple electrical pads electrically connecting with the read / write transducer. The inner tongue has multiple electrical pads. The flexure has multiple inner leads electrically connected with the electrical pads of the inner tongue formed on the curve beams. The structure of the HGA prevents the read / write transducer from damaged and cause the manufacture of the HGA simpler. The present invention also discloses a suspension and a disk drive unit.

Systems and methods for accurate position adjustment of thin-film PZT micro-actuators are provided. In certain example embodiments, a head gimbal assembly is provided. A suspension includes a suspension flexure located at one end of the suspension, and the suspension flexure includes a tongue region. At least one micro-actuator is mounted to a micro-actuator mounting region of the tongue region of the suspension flexure. Any gap between the slider and the at least one micro-actuator may be reduced and / or eliminated. A slider is substantially fully mounted on an insulating layer, and the insulating layer separates the slider from the at least one micro-actuator. The suspension flexure may be located substantially fully underneath the slider, and the tongue region of the suspension flexure may be substantially equally weighted around a dimple formed on the suspension. The rigidity of the suspension flexure may be increased by one or more supports formed thereon.

The invention provides a tongue photo processing method and device, and belongs to the technical field of computers. In the method, a tongue picture input by a user is obtained, then position information of a tongue in the tongue picture is determined according to the tongue picture, an area occupied by the tongue in the tongue picture is extracted based on the position information of the tongue in the tongue picture, a target tongue picture is obtained, and finally, tongue picture recognition is performed on the target tongue picture according to a pre-trained tongue picture feature recognition model, thus obtaining tongue picture characteristics in the target tongue picture. Therefore, the position information of the tongue in the tongue picture is determined firstly, and the target tongue picture is extracted, so that interference on the recognition process due to existence of images outside the tongue region can be avoided, waste of processing resources due to existence of the images outside the tongue region can also be avoided, and the processing speed of tongue picture feature recognition can be improved. Moreover, the tongue picture features are recognized through the tongue picture feature recognition model, so that different input tongue pictures can be processed according to the same recognition standard, and the credibility of a judgment result can be improved.

The embodiment of the invention provides a tongue image analysis method and device, and the method comprises the steps: obtaining a patient image collected when each patient stretches out of a tongue,and extracting a corresponding tongue region image from each patient image according to a pre-trained tongue segmentation model. And for each extracted tongue region image, determining a foreground pixel region and a background pixel region in the tongue region image, and extracting a corresponding tongue image from the tongue region image according to the foreground pixel region and the background pixel region, so that the tongue image of the patient can be accurately extracted. Then, each tongue image is marked according to the patient category corresponding to each tongue image; a tongue image sample set is generated, a deep learning algorithm is adopted to classify and identify the tongue image of the patient, whether the patient suffers from the fatty liver or not is rapidly monitored, the patient can monitor the fatty liver at any time without going to a hospital for inquiry treatment, and the patient can use the device conveniently.

The invention discloses a mobile terminal tongue picture acquisition method, device and apparatus, and the method comprises the steps: carrying out tongue picture shooting through a rear camera of a mobile terminal in response to a target object, and collecting initial tongue picture images; detecting the initial tongue picture images based on an image recognition model to obtain a detection result, the image recognition model being a model trained based on a tongue picture training sample, and the image recognition model being capable of recognizing a tongue region of the tongue picture images; determining whether the detection result meets a target acquisition condition, and if so, storing the tongue picture images meeting the target acquisition condition; if not, generating adjusting prompt information, so that the tongue picture images meeting the target acquisition condition are obtained based on the adjusting prompt information; and determining a target tongue picture image in the tongue picture images meeting the target acquisition condition. According to the invention, the convenience and high efficiency of tongue picture acquisition by the mobile terminal are realized, and the accuracy of image acquisition is improved.

The present invention provides a decorative material for a floor having a feeling of uniform smoothness, flaw resistance and water resistance(moisture absorption resistance) and moreover having excellent design, and a method for manufacturing the same. In particular, the invention provides a decorative material using a wooden plate as a base material, which is characterized in that (1) the wooden plate has at least one of a tongue and a small hole formed by tongue and groove working, at a side face, (2) a synthetic resin layer with a thickness of 100 mum or more is laminated on the upper face of the wooden plate, and (3) a decorative sheet is laminated on the faces shown in (i)-(iii); (i) the upper face of the synthetic resin layer, (ii) the side face of the wooden plate, at least the face from the upper face end to the tongue region, and (iii) the side face of the woodenplate, at least the face from the upper face end to a small hole region.

Device comprises a partial recess (22) in the upper part of a screening cover with the exception of a tongue, recess providing an opening (20) for the passage of the elastic contacts, the vertical distance between the plane of the elastic contacts and tongue allowing the passage of the chip card (12), which is positioned and held within the recess.

The invention discloses an intestinal tract disease early warning method based on tongue diagnosis, and the method comprises an obtaining step of obtaining the current image information which comprises a tongue image; an extraction step of extracting a tongue image in the current image information; and a comparison step of comparing the tongue image with an intestinal tract disease recognition model to judge whether a tongue region has cracks or not, and if yes, sending alarm information. The invention also provides an identification model construction method. According to the intestinal tractdisease early warning method based on tongue diagnosis, the acquired image is compared with the intestinal tract disease recognition model to judge the physical condition of the user; according to the intestinal tract disease recognition model, the confrontation training is carried out on a real graph and a manually simulated tongue crack picture, so that the intestinal tract disease recognitionmodel is more robust, and the problem of a large amount of misjudgment is avoided.

The invention discloses a cantilever piece of a magnetic head folded piece combination, which comprises a flexible member provided with a tongue region and an external trigger board, wherein, the external trigger board is provided with a pair of rigid arms and a pair of elastic arms; the pair of elastic arms lead the flexible member to have favorable flexibility and is helpful to a magnetic head to have enough displacement; the pair of rigid arms provide enough rigidity to the flexible member, thus not only preventing the magnetic head folded piece combination provided with the external trigger board from generating cantilever piece deformation and emboss separation successfully, but also contributing to the improvement of the shakeproof performance of a magnetic head folded piece group. Moreover, the structure of the cantilever piece provides good static property to the magnetic head; therefore, the dynamic property of the magnetic head folded piece combination is enhanced. The invention also discloses the magnetic head folded piece combination provided with the cantilever piece and a manufacturing method thereof as well as a disk driving unit provided with the magnetic head folded piece combination.

The invention relates to the field of artificial intelligence, and discloses a tongue greasy coating detection method and device, equipment and a storage medium, which are used for improving the recognition accuracy of a greasy coating tongue image. The tongue greasy coating detection method comprises the following steps: receiving a tongue image to be detected, and obtaining an initial tongue area in the tongue image; carrying out super-pixel segmentation on the initial tongue region to obtain a plurality of super-pixel tongue regions; according to the pixel value of each superpixel tongue area, carrying out the color feature identification on each superpixel tongue area to obtain an alternative greasy coating area; according to a relationship between each pixel value and an adjacent pixel value in the alternative greasy moss region, carrying out texture feature identification on the alternative greasy moss region to obtain a target greasy moss region; and when the proportion of the target greasy coating area in the initial tongue area is greater than a preset greasy coating ratio, determining that the tongue image to be detected is a greasy coating tongue image. In addition, the invention also relates to a block chain technology, and the greasy coating tongue image can be stored in a block chain node.

The invention discloses a tongue image data analysis method based on a strong supervision algorithm and a deep learning network. According to the method, a YOLO network technology and a Bilinear network technology are combined together; a bounding box of the tongue region is predicted through a YOLO network; then, the image output by the YOLO network is processed by using a Bilinear network; the prediction of the tongue image characteristics is output; finally, the Bilinear network is transplanted into a mobile phone (Iphone and Android) and a server, and the mobile phone is connected with theserver. And the characteristics predicted by the Bilinear network are compared with a built-in symptom database to finish judgment of the tongue image characteristics and symptoms, and finally healthsuggestions and suggestions for the tongue image patient are proposed at the user terminal. The result provides a certain health consultation effect for the user, can help doctors to know and analyzethe condition of the patient to a great extent, and can also be applied and guided in the future life of the patient.

The invention discloses a tongue region detection method and system based on deep learning. The method comprises the steps: carrying out the marking of an obtained image data set comprising a tongue,and carrying out the preprocessing of the marked image data set, so as to obtain a first image data set; setting proportions of a plurality of fixed reference frames, and performing clustering in a k-means clustering mode to obtain a plurality of clustering center reference frames; performing training based on a DarkNet network structure, determining an output layer dimension according to the plurality of clustering center reference boxes, and performing training according to the first image data set to determine a first detection model; utilizing the first detection model to detect an image data set which does not contain a tongue part so as to obtain a false detection image data set of false detection; and adjusting the network structure of the first detection model, modifying the dimension of an output layer, and retraining by using the first image data set and the false detection image data set to determine a tongue detection model for tongue region detection.

The invention provides a tongue quality and coated tongue segmentation method and device. Based on feature points in a face image, the method comprises the following steps of determining a target region (including an oral cavity region and a tongue region) in the face image; segmenting the target area based on the color component of the target area in the HSV space to obtain a tongue area, based on the color components of the tongue area in the CIELAB space, acquiring the tongue quality area and the coated tongue area from the tongue area through segmentation, so that the tongue quality area and the coated tongue area are segmented from the face image through a layered progressive mode based on the feature points and the color components of different color spaces, and a foundation is laidfor automatic tongue diagnosis.

The invention relates to the technical field of artificial intelligence, in particular to a tongue tremor degree evaluation system based on artificial intelligence. The system comprises: obtaining a mouth key point and a tongue key point through a key point detection unit; acquiring a depth image sequence and an RGB image sequence of the tongue in a set time through an image acquisition unit; extracting an area sequence of the tongue region and a pixel value sequence of the tongue region in a difference image sequence of the tongue region obtained by a frame difference method through a featureextraction unit; correcting the area sequence and the pixel value sequence through a feature correction unit; and inputting the corrected area sequence and pixel value sequence into a time sequence classification network through a tremor degree evaluation unit to obtain the tremor degree of the tongue. The area sequence and the pixel value sequence in the difference image sequence are corrected through the change of the distance between the face of the to-be-detected person and the camera, so that on one hand, the error of the difference image analysis result can be eliminated, and on the other hand, the tongue tremor degree evaluation result can be more accurate.