48 results about "Mouth angle" patented technology

The invention discloses a method for positioning and tracking outer contour feature points of key face organs and a device thereof, comprising the following steps: first, the positions of canthi points and angulus oris points on the face are obtained; the medial axis of the eyes and the mouth is identified based on the positions; the outer contour feature points of the eyes and eyebrows are identified on the medial axis of the eyes, and the outer contour feature points of the mouth are identified on the medial axis of the mouth. Consistent positioning and tracking of outer contour feature points of key face organs are conducted on such a basis. The technical proposal provided by the embodiment of the invention solves the problem of inaccurate positioning of outer contour feature points of eyes, mouth and eyebrows when people makes various faces; based on the positioning of the feature points, the existing two-dimensional and three-dimensional face models can be real-time driven, and face expressions and motions such as frowning, blinking and mouth opening of people before a camera can be real-time simulated, thus creating various real and vivid face animations.

The present invention relates to detection method of human face characteristics of motion image, firstly, take picture of human face image to construct discipline collection, then proceed principal component analysis, Hough conversion etc. to make its position and size coinside with the eye in discipline collection iamge, then project onto space of above mentioned characteristic eye, finally, take the eye being selected with minimum error between original eye and its projection as the detection result, and use integral projection to obtain precision position of mouth, nostril and apex nasi. Compared with existent method, the detection speed raises 225 times and accuracy increased by 1.27%.

The invention discloses a robot head structure with a tongue expression. The robot head structure comprises a head formed by a head shell, a scalp and a head fixing base. The head is provided with a tongue assembly, a mouth assembly and an eye assembly. Each assembly is provided with a corresponding driving steering engine and an action driving component thereof. The tongue assembly comprises a tongue extensible assembly, a tongue vertical-tilting assembly and a tongue left-right-swing assembly. The mouth assembly is used for completing a mouth opening-closing action expression and / or a mouthcorner smile action expression. The eye assembly is used for completing eyeball left-right-moving action expression and / or an eyelid blinking action expression. The tongue assembly is used for completing extending-retracting, vertical tilting and left-right swing action expressions of a tongue. Rich expression actions including rich tongue expressions can be provided for a humanoid robot, and on the basis of providing using functions for people, rich expression experience and joyful mood feelings for communication are added.

The invention discloses a deep learning-based ship driver fatigue detection method and system. The method comprises the following steps: acquiring a current video frame image; carrying out face detection by an improved Retinaface face detection network, and marking five face key points of the left eye, the right eye, the nose tip, the left mouth corner and the right mouth corner at the same time; cutting adaptively eyes and mouth pictures according to the positions of the five face key points, and identifying the opening and closing states of the eyes and the mouth by an improved ShuffleNet v2 convolutional neural network; calculating PERCLOS parameters of the eyes and the mouth; and comprehensively judging whether a driver is fatigued or not through a random forest model by fusing the PERCLOS parameters of the eyes and the mouth. Face detection and key point positioning can be rapidly realized, manual picture information extraction is not needed, the opening and closing states of the eyes and the mouth can be automatically identified, and eye and mouth feature parameters are fused based on the idea of multi-feature fusion, so that the fatigue state of the ship driver can be rapidly and accurately detected.

The invention discloses a mouth angle coordinate average value change-based smiling face recognition method. The method comprises the following steps of: (1) face detection; (2) HOG feature extraction; (3) Gabor feature extraction; (4) a support vector machine; (5) a mouth angle coordinate method; according to the method, firstly, a Shi & Tomasi corner detection algorithm is adopted to carry out corner detection on the mouth, and then smiling face recognition is carried out through a method of calculating the coordinate average value of the mouth. Experimental results show that the correct rate of smiling face recognition based on the mouth angle coordinate method is 92% and is higher than that of a Gabor and HOG feature extraction method.

The invention relates to a networking spray fire-fighting device comprising a fire extinguishing agent refraction direction, an intelligent nose and mouth blade, a three-way valve, a constant temperature probe, a connecting head, a pressure gauge, a fire extinguishing agent tank, an outer ring of a bearing, an inner ring of the bearing, a connecting rod, an inner ring of a hook, an outer ring of the hook, an intelligent nose and mouth angle radian, a fire extinguishing agent outlet, a networking linkage port, a fire extinguishing agent refraction angle, a tank rotation direction and a hot airflow. The networking spray fire-fighting device is capable of effectively releasing and spraying a fire extinguishing agent by 360-degree omni-directional fast rotation, thus achieving an efficient fire-fighting effect.

The invention discloses a text synchronous driving three-dimensional facial expression and mouth shape realization method and system, and the method comprises the following steps: a preprocessing stage: S1, making a B l endshape change curve: making a curve in Maya; the curves comprise two types of mouth shape and expression starting and ending change curves; the mouth shape change curve comprises mouth shape curve groups of all counted adjacent phoneme sequence pairs, and for specific expressions, the mouth shape change curve comprises two change curves from neutral change to specific expression, namely expression start and from specific expression change to immediate expression end; s2, manufacturing a B < l > endshape: in the Maya, manufacturing the B < l > endshape; the mouth shape B l endshape comprises three groups of mouth shapes of immobile mouth corners, upward mouth corners and laugh mouth corners, each group comprises six mouth shapes of Fv, Open, PBM, ShCh, W and wi de, and the expression B l endshape is made. According to the method, the natural and vivid mouth shape and expression animation can be synchronously driven through the text, meanwhile, the mouth corner part can be naturally fused, and the intelligent interaction capability of the virtual human is improved.

The invention relates to the technical field of medical instruments, in particular to a radio frequency treatment guide plate for trigeminal nerve treatment, and aims to overcome the defects that in the prior art, the puncture direction and depth are determined by means of anatomical mark points of a human body, the accuracy is poor, the effect is unstable, complications are prone to occurring, and a patient is injured to a certain extent due to repeated fluoroscopy during an operation. According to the technical key points, the equipment comprises an occlusion plate which is arranged in a semicircular shape and corresponds to the tooth part of a patient in shape for occlusion and fixation to realize positioning; a positioning column provided with a positioning hole, wherein the positioning hole corresponds to the position of the nose tip of the patient so as to detect the positioning accuracy; an injection channel which is located on one side of the occlusion plate and is 20-30 mm away from the mouth corner, wherein the injection channel has the approach direction reaching the skull base oval hole and is used for enabling the injection needle to reach the trigeminal ganglion through guiding puncture of the injection channel; and a connecting part, wherein one end of the connecting part is connected with the injection channel, and the other end is connected with the joint part of the occlusion plate and the positioning column, so that the whole treatment guide plate is integrally formed.

The invention discloses a method for controlling virtual human mouth motion, which comprises the following steps: after a three-dimensional human face grid model is acquired, consulting the definitionfor FDP characteristic points of a mouth area in MPEG-4 to select the characteristic points at the zygomatic arch, two mouth corners, upper boundary of lower lip and mandible boundary as boundary points of a mouth functional area, and finally determining the mouth functional area in the three-dimensional grid model according to the characteristic point coordinates of the zygomatic arch; and adopting an oscillating guide-bar mechanism model to drive grid points of the mouth functional area to move, simulating a real mouth opening-closing process, and eliminating the overstretching phenomenon at the boundary by processing a peak of a peak buffer area. The method can adopt the same mathematical model to realize opposite motion processes such as opening and closing of the human mouth under continuous condition, the mathematical model is simple, the physical significance is definite, the control is easy, and the simulation effect is vivid.

The present invention relates to detection method of human face characteristics of motion image, firstly, take picture of human face image to construct discipline collection, then proceed principal component analysis, Hough conversion etc. to make its position and size coinside with the eye in discipline collection iamge, then project onto space of above mentioned characteristic eye, finally, take the eye being selected with minimum error between original eye and its projection as the detection result, and use integral projection to obtain precision position of mouth, nostril and apex nasi. Compared with existent method, the detection speed raises 225 times and accuracy increased by 1.27%.

The invention discloses an LED light spot improving method, which includes the following steps of a, selecting: selecting a support with the cup depth of 0.46+ / -0.02mm and the cup mouth angle of 135 degrees and selecting fluorescent powder with the grain size of 17-19 mu m and fluorescent powder with the grain size of 4-5 mu m for matching; and b, dispensing adhesive: dispensing yellow fluorescent adhesive to form a semi-spherical structure and form two surfaces approximately parallel to each other with a lens. By matching of the fluorescent powder with the grain size of 17-19 mu m and the fluorescent powder with the grain size of 4-5 mu m, luminous flux can be improved effectively by the LED light sport improving method, and further, the fluorescent powder with the grain size of 4-5 mu m substitutes for diffusion powder, the problems of large light spot difference and fast attenuation of products, caused by combination of three kinds of powder different in chemical property, are avoided. Besides, the luminous area is enlarged by dispensing the yellow fluorescent adhesive to form the semi-spherical structure.

The invention relates to a block for producing personalised root post flanges for dental implantology and to a method for producing a personalised root post flange for dental implantology, of the type that can be screwed onto a previously osseointegrated implant, normally in an incorrect position, and over which aesthetic material can be directly placed, providing the dental part with shape and colour, without having to apply a dental crown over the root post flange. Said block is characterised in that it has a structure that allows the selection of the mouth angle required for the channel used for the insertion of the screw for fixing to the implant.

The invention relates to an average face-based method. The face alignment method based on Platts transform comprises the following steps: carrying out face detection on an original input video image and an artificially photographed image by utilizing a multi-task cascade convolutional neural network, cutting and scaling an area where a positioned face is located, and further obtaining five key point coordinates of eyes, two mouth angles and a nose in a cut image with the same size; averaging the five feature points in all the cut pictures respectively, and calculating to obtain an average facemodel; carrying out Platts transformation on the to-be-aligned face image by taking the average face as a reference to obtain an aligned image; and identifying and verifying the aligned face image.

The invention relates to a manufacturing method of a comfortable full-mouth adsorption false tooth meeting an aesthetic function, and belongs to the field of false tooth manufacturing, wherein the method comprises the steps: drawing a central line, a mouth angle line and a lip base line on a temporary base plate loaded with a wax rim after occlusion mouth testing, and taking a bisector of jaw spacing after mouth testing; marking out an incision edge connecting line of the anterior incisor below the lip base line, and marking out a smile line above the lip base line; connecting the incisor edge line of the anterior incisor with the bisector of the jaw spacing after mouth testing to form a tooth arrangement plane curve on the same plane, wherein the mouth angle line is a positioning line between a three-tooth distal center and a four-tooth proximal center, the curve length from the mouth angle line to the central line is the basis for determining the width specification of the whole set of false tooth, the smile line is a base line for determining the tooth neck of the anterior incisor, and the distance from the smile line to the incisor edge connecting line of the anterior incisor is the basis for determining the height specification of the whole set of false tooth. Therefore, specification parameters of the whole set of false teeth and tooth arrangement positions of the false teeth can be determined, and the method is simple and easy to implement and has aesthetic characteristics on the premise of meeting false tooth functions.

The invention aims to provide a whole process that beverage is drunk in a safer, interesting and brand-new mode for people, and provides a primary-secondary distribution beverage bottle. According tothe primary-secondary distribution beverage bottle, one bottle body is only contains purified water, mineral water and other plain water which cannot chemically reacts with a plastic bottle, and the other bottle body contains solid beverage. When people drink, a bottle cover of the secondary bottle body is opened, the solid beverage is taken out, then a bottle cover of the primary bottle body is opened, the solid beverage is placed into the water in the primary bottle body, the bottle bodies are covered well with the covers respectively, the bottle is manually shaken to be uniform, after the solid beverage is sufficiently dissolved, people can drink the beverage, and the process same as that people usually make tea at home is a drinking-while-making operation process. By means of the bottle, the taste is more delicious, or after the bottle cover of the primary bottle is opened, the plain water can be directly drunk. In this way, the novel bottled drink bottle in which the liquid and the solid beverage are separately placed is achieved. According to the primary-secondary distribution beverage bottle, a bottle mouth angle of 45 degrees is more beneficial to drinking.

The invention relates to a mouth key point labeling method and apparatus, electronic equipment and a storage medium. The method comprises the steps of obtaining a first face image; determining a firstupper lip area and a first lower lip area on the first face image; determining a target mouth area set according to the first upper lip area and the first lower lip area; when the target mouth area set is empty, determining that outer mouth corner key points comprise a leftmost point and a rightmost point in a union set of the first upper lip area and the first lower lip area; when the target mouth area set is not empty, determining that the outer mouth corner key points comprise a leftmost point and a rightmost point in the target mouth area set; and labeling the mouth key points according to the outer mouth corner key points. According to the method, mouth key points can be automatically labeled, the labeling efficiency is improved, the cost is saved, and when mouth key points marked bythe method provided by the invention are used for training a model, the model training precision can be improved.

The invention discloses a method for positioning and tracking outer contour feature points of key face organs and a device thereof, comprising the following steps: first, the positions of canthi points and angulus oris points on the face are obtained; the medial axis of the eyes and the mouth is identified based on the positions; the outer contour feature points of the eyes and eyebrows are identified on the medial axis of the eyes, and the outer contour feature points of the mouth are identified on the medial axis of the mouth. Consistent positioning and tracking of outer contour feature points of key face organs are conducted on such a basis. The technical proposal provided by the embodiment of the invention solves the problem of inaccurate positioning of outer contour feature points ofeyes, mouth and eyebrows when people makes various faces; based on the positioning of the feature points, the existing two-dimensional and three-dimensional face models can be real-time driven, and face expressions and motions such as frowning, blinking and mouth opening of people before a camera can be real-time simulated, thus creating various real and vivid face animations.

The invention discloses a facial mask containing active peptide components. The facial mask containing active peptide components comprises a piece of thin oil paper, graphene fibers, meshes, conducting fibers, cotton fibers, eye ports, energy magnetic loops, peptide collagen, a mouth corner port, a nose port, a thin sponge slice, kaolin powder, biologic energy magnet powder and volcanic mud powder. The facial mask containing active peptide components disclosed by the invention is reasonable to make, promotion of blood circulation of five sense organs in a face is facilitated, further the peptide collagen applied to the face can be absorbed, the effects of resisting ageing and caring skin are achieved, the skin of the face is watery, tender and bright, face fat is convenient to absorb, andthe facial mask containing active peptide components is more suitable for skin care people having heavy face fat; and besides, a volcanic mud powder component is added, so that aged cutin can be softened and removed, and metabolism of skin and muscle can be accelerated.

The invention relates to the technical field of medical instruments, and discloses a therapeutic apparatus for improving oral expression skills of a Parkinson's disease patient. The therapeutic apparatus comprises a mask module and a host; the mask module comprises a mouth periphery part, side wings, a fixing belt, a lip lifting electrode sheet area, a lip lowering electrode sheet area and mouth corner moving devices. The two side wings are respectively arranged on the left and right sides of the mouth periphery part; two ends of the fixing belt are respectively connected with the two side wings; the lip lifting electrode sheet area and the lip lowering electrode sheet area are respectively arranged at the upper end and the lower end of the mouth periphery part and are electrically connected with the host; and the mouth corner moving devices electrically connected with the host are arranged on the two side wings and used for closing the mouth and making mouth corners move towards the two sides. According to the therapeutic apparatus for improving the oral expression skills of the Parkinson's disease patient provided by the invention, through accurate control of low-frequency electrical stimulation and adjustment of the vibration frequency and duration, the oral muscles of the patient are contracted and relaxed; the oral feeling of the patient is improved; and the purposes of relieving muscle stiffness and increasing or improving the oral expression skills are achieved.

The invention relates to a human face false detection filtering method and system, an electronic device and a storage medium, and the method comprises the steps: detecting human face key points in anRGB image through a human face key point detection model, and enabling the human face key points to comprise a left eye center, a right eye center, a nose center, a left mouth corner and a right mouthcorner; obtaining a module a of a vector a from the center of the nose to the center of the left eye, a module b of a vector b from the center of the nose to the center of the right eye, a module c of a vector c from the center of the nose to the left mouth corner and a module d of a vector d from the center of the nose to the right mouth corner; calculating the difference between a and b and thedifference between c and d to obtain values of a-b and c-d; and judging a face detection result of the RGB image by comparing the absolute value of the value of a-b and the absolute value of the value of c-d with a preset threshold value. According to the invention, the problems of large amount of detection network parameters used for face detection, incapability of being applied to the equipmentend and complex false detection filtering method in the prior art are solved, the detection speed and the detection precision are improved, and the method is simple and easy to operate.

The invention discloses a visual feeding function monitoring and rehabilitation supporting headgear for disabled old people, which comprises a training headgear, an annular patch, a swallowing organ monitoring module, an oral cavity training module and an anti-cough module, the swallowing organ monitoring module is used for judging the severity of disability according to the movement degree of each part of the oral cavity of the disabled old person in the feeding process, and the oral cavity training module is used for performing targeted training on each part according to specific data measured by the swallowing organ monitoring module. The cough prevention module is used for collecting generated saliva to prevent saliva from flowing out of mouth corners or choking disabled old people in the training process. By means of the swallowing organ monitoring module and the oral cavity training module, the severity of disability of all parts of the oral cavity of the disabled old people can be measured, and proper rehabilitation training can be conducted according to different damage degrees of swallowing organs.

The invention is applicable to the technical field of machine learning, and provides a method and device for recognizing facial expressions. The method includes: acquiring a face image meeting preset conditions; detecting the positions of N feature points of the face in the face image, where N is an integer greater than zero; according to the detected N feature points position, select the eyes, nose and / or corners of the mouth as key feature points to calibrate the face image; perform facial expression recognition on the calibrated face image through the trained deep convolutional neural network, and output Recognition results. Through the present invention, the accuracy rate of human facial expression recognition can be effectively improved, and a higher expression recognition rate can be achieved even for human faces on images with many occlusions, low image resolution, or side faces.

The embodiment of the invention provides a method for automatically detecting a three-dimensional facial mark, and relates to the technical field of machine learning and pattern recognition. The detection method comprises: roughly detecting a nose using a three-dimensional local shape descriptor; accurately positioning the tip of the nose by utilizing the characteristics such as the symmetry of the human face; determining areas where eyes and a mouth are located according to the distribution of the human facial features; using a convolutional neural network to minimize combination loss, and providing candidate objects of canthus and mouth corners; and maximizing the similarity between the candidate points and the real marks according to the features of the candidate points and the adjacent points thereof, and iteratively updating the candidate objects to realize accurate detection of canthus and mouth corners. According to the method, under the condition that only three-dimensional face data are used, high-precision automatic face mark detection is achieved, and the method is stable for facial expressions and postures and higher in practicability.

The invention provides a computer vision-based facial nerve disease rehabilitation condition static detection system, which comprises a face detection segmentation module for processing a front-view face image to obtain an eye image and a mouth image, an eye anomaly detection module for obtaining an eye anomaly degree based on an eye white area and a pupil area, a mouth anomaly detection module which is used for obtaining the mouth overall offset, the mouth deformation degree and the mouth corner depth difference, and a rehabilitation effect judgment module which is used for obtaining a plurality of illness state description change curves based on the eye anomaly degree, the mouth overall offset, the mouth deformation degree and the mouth corner depth difference and analyzing the rehabilitation situation of a patient according to the curves. According to the invention, abnormity judgment is carried out on the eyes and the mouth of the patient based on the single-frame image, and interference of asymmetry of the eyes of the normal human face on disease judgment can be avoided, so that the disease of the patient can be judged more accurately, and errors are small.

The invention relates to engineering equipment, in particular to a polishing device for a steel box girder T rib sliced mouth. The polishing device comprises a jig frame, a fixture, a travelling trolley, a grinder installation frame and a grinder; the jig frame comprises a stand column, a transverse beam and track beams; the track beams are connected with the upper end of the stand column; the transverse beam is transversally connected between the track beams and is provided with the fixture; the upper surfaces of the track beams are provided with slide rails; wheels of the travelling trolley are matched with the slide rails and travel along the slide rail paths; and the travelling trolley is provided with the grinder installation frame which is provided with the grinder. The invention can be connected with multiple transverse beams to enable a bulk of T ribs to be arranged on the fixture, a regulating rod is pushed to drive a motor grinder to move straightly along a media axis of the track beams, and meanwhile, a sliced mouth angle of the T rib is polished; therefore, a bulk of T rib sliced mouths can be polished during the traveling of the travelling trolley with high production efficiency. The invention has the advantages of simple structure, low construction cost, good rigidity, high planeness, simplified manual operation procedure, convenience and safety.

The invention relates to the field of oral examination, in particular to an auxiliary device for oral examination of children. The technical problem to be solved by the invention is to provide the auxiliary device for oral examination of children. The auxiliary device for oral examination of children comprises a backrest adjusting assembly, a cushion adjusting assembly, an expansion assembly, a disinfection assembly and the like. The disinfection assembly is connected with a second bottom frame. When the devicer is used, a child is automatically prevented from sliding down, potential safety hazards are greatly reduced, the expander is kept stable when the two mouth corners of the child are expanded, the expansion elasticity can be adjusted to adapt to children with different mouth sizes, meanwhile, the mouth corner on one side is automatically loosened, and then the mouth corner on the other side is deeply expanded, so that the deep teeth are exposed, the dilator is automatically cleaned and disinfected, and disinfectant fluid absorbed and hidden in the cotton rope is completely removed.

The invention relates to the technical field of baby and infant products, and discloses a self-service adult simulation direct drinking face fitting leakage-proof drink learning cup. The drink learning cup comprises a water cup body and a cup body, wherein the top of the cup body is provided with a water outlet, the bottom of the cup body is provided with a cleaning port, the water outlet is provided with a drink learning leakage-proof structure, the front face of the cup body is provided with two inflatable angulus oris pressing blocks, the side face of the cup body is provided with a takingdevice, and the left side of the cup body is fixedly connected with a left handle. The self-service adult simulation direct drinking face fitting leakage-proof drink learning cup has the advantages that through the cooperative use of the drink learning leakage-proof structure and the like, it can be gradually learned that water drinking with the cup or drinking straw cups all needs to suck water with the mouth to make the water enter the oral cavity, so that an infant can change the amount of the water flowing to the mouth by changing the angle of the cup inclination when drinking, thereby solving the problem of the drink learning cup how to help children learn drinking and transit to use a normal cup to drink.

The invention relates to the technical field of image data processing, and discloses a face brightness calculation method, apparatus and device, and a storage medium. The method comprises the steps ofextracting a head region from an image; extracting a face region from the head region, wherein the face area comprises a marked left mouth corner key point, a marked right mouth corner key point, a marked left forehead key point, a marked right forehead key point, a marked left cheek key point, a marked right cheek key point, a marked left upper jaw key point and a marked right upper jaw key point; judging the face orientation according to the midpoint of the face area, the left mouth corner key point and the right mouth corner key point; when the face orientation is a front face, extractinga hexagonal area defined by the left forehead key point, the right forehead key point, the left cheek key point, the right cheek key point, the left upper jaw key point and the right upper jaw key point from the face area; calculating the average brightness of the hexagonal area, wherein the accuracy of face brightness calculation can be improved.

The invention discloses a tongue body and tongue coating image recognition and separation method. The method comprises the following steps: step 1, collecting a tongue image; 2, performing color correction in different modes on the obtained tongue image pictures according to different acquisition devices so as to improve the accuracy of a subsequent image processing algorithm; step 3, positioning a tongue body region in the tongue image picture, including cutting the tongue image picture into sub-image blocks at will, training the sub-image blocks by using a trained convolutional neural network, and positioning the tongue body according to the left and right eye positions, the nose tip position and the left and right mouth corner positions of the human face to obtain a complete rectangular image containing the tongue body; step 4, realizing segmentation of the tongue body and the non-tongue body; and 5, according to different colors of the tongue coating and the tongue body, performing coating separation on the segmented tongue image. The method has the beneficial effects that color correction is carried out on the collected tongue body image, the influence of color cast on subsequent tongue body feature analysis is improved, and accurate separation of the tongue body and the moss is realized.