16 results about "Closed eyes" patented technology

A system and method for combining short-wave infrared (SWIR) imaging with image processing algorithms to measure pupil size dynamics in closed eye conditions, either in response to an external stimulus (such as light-induced PLR) or occurring spontaneously during sleep or anesthesia. The system, with deep learning-based analysis, can further estimate gaze direction during eye movements and achieve robust pupillometry. The system has at least one electromagnetic radiation source for radiating electromagnetic energy in a spectral range of 700 nm to 2,000 nm, and an imaging device for detecting electromagnetic energy and sending image data to a processor which performs algorithmic and deep- learning based analysis of image data to continuously track changes in pupil diameter and gaze direction over time.

This invention relates to the field of intelligent driving assistance technology, specifically disclosing a driver state monitoring system based on intelligent image recognition. Built on a Raspberry Pi embedded platform, it includes a video acquisition module, a core algorithm processing module, and a peripheral control module. The video acquisition module is compatible with USB or CSI cameras, acquiring and preprocessing driver facial images. The core algorithm processing module incorporates a lightweight facial key point detection model to extract facial features and identify dangerous behaviors such as closed eyes and head tilting, and features a feature calibration and fatigue grading mechanism. The peripheral control module connects to alarm and display components, triggering corresponding warning actions and can be expanded to include multiple alarm methods. The system can also track continuous driving time and expand remote monitoring functionality. This invention features low hardware cost, low power consumption, convenient deployment, fast edge-side inference response, and accurate monitoring. It is adaptable to multiple scenarios, requires no manual intervention, comprehensively prevents fatigue driving, and is easy to deploy on a large scale.

A method includes receiving, by a processor, a video of an illuminated face. Video frames in the video are associated with different time instants of a time series, respectively. The method includes determining a region of interest (ROI) that includes a closed eye on the face based on at least one frame in a plurality of video frames in the video. The method includes slicing each ROI based on slice parameters for selecting a vertical slice that includes a eye centroid of the closed eye, and selecting a horizontal slice that intersects the vertical slice at a pivot (P) located lp pixels from the eye centroid. The method includes determining an eyeball motion signal according to the time series, based on a measured value of a shadow intensity of the eye within at least one of the vertical slice or the horizontal slice at each of the different time instants.

The application relates to a line-of-sight calibration, motion tracking and precision testing method based on adaptive time sequence analysis prediction, which specifically comprises the following steps: through image color recognition technology, color blocks and calibration point positions in a screen are recognized in a scene image, the actual size of the screen is combined, and the screen position is roughly positioned; distortion calibration is carried out based on the calibration point information, the screen position is precisely positioned, and the mapping relationship between the scene image and the precisely positioned screen is determined; eye control type nine-point calibration is carried out, the mapping model of the eyeball physical information and the gazing space position of the tester is determined, the screen gazing point coordinates are obtained in combination with the mapping relationship; line-of-sight positioning precision quantitative testing is carried out; through an adaptive time sequence analysis prediction algorithm, the current line-of-sight positioning is optimized and the future gazing point position is predicted in combination with historical line-of-sight information, meanwhile, the parameters in the time sequence analysis prediction algorithm are adaptively adjusted according to real-time precision or the intensity of shaking; through an eye state detection algorithm, the state of the human eye in the open and closed eye state is detected, and eye movement control is carried out.

The invention discloses a pupil diameter automatic detection method based on infrared imaging. The pupil diameter automatic detection method comprises the steps that an eye image is collected; acquiring a grayscale image of the eye of the eye-closed patient by using infrared imaging equipment; preprocessing the image; performing size unification and normalization processing on the acquired infrared eye image to obtain a standardized input image; pupil segmentation; inputting the preprocessed image into an image segmentation network based on deep learning, and outputting a binary mask image of the pupil region by the network; acquiring a mask pixel area of the binarized mask image, and converting the mask pixel area into an equivalent diameter of a pupil region; and carrying out visual output on equivalent diameter data and other related image features of the pupil region in the adopted period. The method is high in speed and uniform in measurement standard, and human differences are reduced.

Vigilance level assessment device, comprehensive a portrait acquisition component for acquiring a portrait, which is an image taken of the face of an occupant of a vehicle, an eye opening degree calculation component that calculates the eye opening degree of the occupant based on the portrait, an open / closed eyes assessment component that evaluates whether the eye opening level is less than a predefined threshold or not, a smile assessment angle calculation component that, based on the portrait, determines the positions of both corners of the mouth and the midpoint of both inner corners of the eyes of the occupant, and calculates a smile assessment angle as an angle from a straight line connecting one corner of the mouth and the midpoint of both inner corners of the eyes, and a straight line connecting the other corner of the mouth and the midpoint of both inner corners of the eyes. a smile assessment component that assesses whether the occupant is smiling or not based on the smile assessment angle, and an alertness level assessment component that assesses a decrease in the inmate's alertness level when the eye opening level has become less than a threshold, and it has also been assessed that the inmate is not smiling.

The present application relates to the technical field of image processing, and discloses an image processing method, a device, a storage medium, and a program product. The image processing method comprises: if a portrait to be processed that meets a processing condition (for example, having a feature that affects image aesthetics, such as closed eyes) is present in an image to be processed, acquiring a facial feature of said portrait, and selecting, from an image library, a reference image that has high similarity to the facial feature of said portrait and compensates for said image defect (for example, a reference portrait having the same facial appearance as said portrait and having open eyes); and then performing image restoration on said image on the basis of the reference image to obtain a target image that have overcome said image defect, for example, replacing the part of said image that affects image aesthetics with the corresponding part in the reference image. In this way, the overall aesthetics of the image can be improved, facilitating the improvement of visual experience of a user viewing the target image.

The application discloses a meditation posture adjustment method based on a light neural network and an acoustic sense, and has the characteristics that the method specifically comprises the following steps: collecting real-time video data in a user's meditation process, detecting and extracting features of human body skeleton key points of an image by using a light neural network model, analyzing key point position information to calculate the deviation of a current posture from a standard meditation posture, matching corresponding voice action instructions according to the deviation type, playing voice guidance to guide the user to adjust the body action in real time, and cyclically monitoring the posture change and dynamically updating the voice feedback until the posture is corrected. The application effectively reduces the calculation complexity of the model to adapt to mobile terminal deployment, and innovatively uses voice instructions to replace visual feedback, solves the pain point that the user needs to open eyes to confirm the posture in the meditation process and destroys the immersion, realizes real-time and accurate posture correction in a closed eye state, and has good application prospect.

A method and a monitoring device determine a binary eye state of a person. In the process, a monotonically decreasing slope with at least one specified minimum height is determined in a time curve of an eyelid gap which is determined in a camera-based manner. A current transition threshold for a change between an open and a closed eye is determined as a specified percentage value of the distance between the average value of a number of starting points of decreasing slopes and an average value of a number of corresponding end points. The eye is detected as being closed if the current eyelid gap is smaller than the current transition threshold. A monotonically increasing slope in the eyelid gap curve which has at least one specified minimum height is determined starting from the last end point. The eye is detected as being open if an end of the current increasing slope has been reached.

The utility model discloses a BJD (ball joint doll) doll head shell structure comprising independent opening and closing eyelids and a detachable eye piece assembly, and belongs to the technical field of toy dolls, namely a ball joint doll. Holes are formed in the eyes of an existing BJD doll head shell without eyelids. A pair of pivot seats is arranged on the inner side of the face shell, the middle pivot seat is used for fixing inner pivots of the eyelids, the left pivot seat and the right pivot seat are used for fixing outer pivots of the eyelids, the eyelashes are located below the eyelids and clamped in eye frames of the face shell to limit turning of the eyelids, the turning angle is right between the eye closing state and the eye opening state, and the left eyelids and the right eyelids are independently controlled to be opened and closed around the axes. The eye sheet is fixed on the eye sheet plate, and the eye sheet plate is detachably mounted at the insertion position on the eye sheet baffle; the eye sheet baffle is inserted into the anti-rotation positioning groove through the positioning column and is inserted into the face shell; the eyelids of the BJD doll can be independently opened and closed, the eyelids can be quickly replaced, and the BJD doll is adaptive to diversified head shell designs.

The invention discloses an image processing method and device, electronic equipment and a computer readable medium, and relates to the technical field of computers.The method comprises the steps that eye opening and closing recognition operation is executed on multiple continuous to-be-processed images corresponding to a target user, and an eye closing image and an eye opening image of the target user are obtained; performing fuzzy detection and quality detection on the eye closing image and the eye opening image; and under the condition that the fuzzy detection result and the quality detection result of the eye closing image and the eye opening image are both passed, performing replacement operation on a human eye area of a to-be-adjusted face image of the target user by using a specified image to obtain a replacement image. Therefore, by performing fuzzy detection and quality detection on the eye closing image and the eye opening image obtained by recognition, the quality of subsequent replacement images can be ensured.

The invention discloses a dynamic weighting method for face quality evaluation based on task semantic driving, which comprises the following steps: performing multi-dimensional quality analysis on a to-be-evaluated face image, and extracting nine original scores including the number of faces, secondary copying, overexposure / too dark, blurring, occlusion / incomplete, non-crown-free, posture, eye closing and strabismus; the method comprises the following steps of: mapping a category label or a natural language description of an application scene (such as identification photo auditing, face payment and the like) into a high-dimensional semantic vector through a task semantic encoder; inputting the vector into a dynamic weight generation network, and adaptively outputting a weight coefficient corresponding to each quality dimension to realize sensitivity adjustment of alignment with a service target; and calculating a comprehensive quality score through weighted fusion and applying the score to business decision. According to the method, an end-to-end driving mechanism from task semantics to quality evaluation weight is constructed for the first time, so that a single universal model can flexibly adapt to various heterogeneous scenes, and manual weight configuration or repeated training of a special model is not needed.

The application discloses a kind of based on brain-heart coupling feature's children hyperactivity autism comorbid auxiliary evaluation system, comprising: signal acquisition module, for synchronous acquisition of the subject under the two resting states of open eye rest and closed eye rest Electroencephalogram and electrocardiogram signal;Signal pre-processing module is used to filter and artifact removal to the collected electroencephalogram, and R wave detection and marking are carried out to electrocardiogram;Heartbeat evoked potential feature extraction module is used to extract heartbeat evoked potential feature;Heartbeat evoked brain network feature extraction module is used to construct heartbeat evoked brain network based on the functional connection change in R wave before and after time window, and graph theory feature is extracted;Auxiliary evaluation module is used to output the auxiliary evaluation result of brain-heart coupling state based on heartbeat evoked potential feature and graph theory feature.Using the application, the objective auxiliary evaluation of children attention deficit hyperactivity disorder comorbid autism spectrum disorder related brain-heart coupling abnormality can be realized.

The invention relates to the field of rehabilitation training, in particular to a cognitive impairment rehabilitation training prompting instrument. The invention provides a cognitive impairment rehabilitation training prompting instrument which can fully and effectively enhance the memory of a patient and improve the cognition of the patient. A cognitive impairment rehabilitation training prompting instrument comprises a base, a shell and the like. A shell is placed on the base. During training, a patient closes eyes, when a nurse shouts out numbers, the patient puts hands into the shell, when the number shouts out by the nurse is counted, one spherical handle is pressed down, so that one character block moves downwards, and the nurse can see whether the number on the character block is correct or not through the groove in the shell; a patient carefully listens to digits shouted by nursing personnel to improve concentration, and then finds the corresponding spherical handle according to the shouted digits, so that the thinking power of the patient is exercised, and the effects of enhancing memory and improving whole cognition are fully and effectively achieved.

This video streaming service aims to provide high-quality still images (photographs) that are unaffected by the viewer's internet connection environment, and that do not include any misfires regardless of the shutter timing. [Solution] Because still images are output based on high-quality video data captured and edited, the shutter output will not be a low-resolution image like a screenshot. Furthermore, a still image for the shutter is prepared in advance by removing any mis-shots from multiple still images composed based on a predetermined frame rate, so that the still image written in response to the shutter operation does not include images with closed eyes or other imperfections. Therefore, even amateur viewers who do not possess the shooting skills of a professional photographer can obtain still images (photographs) without mis-shots, and only still images permitted by those managing the copyright of the subject images can be provided to the viewer.