32 results about "Infant crying" patented technology

The present invention provides an intelligent bracelet for monitoring an infant. The intelligent bracelet is connected to a cloud server and an intelligent terminal and comprises a bracelet body which is worn at the wrist of the infant, a controller which is arranged on the bracelet body and is used for collecting the state information of the infant, thus the state information of the infant is subjected to corresponding interpretation processing so as to know the current condition of the infant. The intelligent bracelet for monitoring an infant is convenient to carry, parents can be helped to judge infant crying, the interactive communication between parents and the infant is increased, the emotional state of the infant can be timely and accurately understood, and the infant is scientifically and reasonably look after and nursed.

The invention provides a crying sound identification system and a crying sound identification method. The crying sound identification system comprises a crying sound database, a sound receiving unit,a processing unit, an output unit, an input unit and a pacifying device. The crying sound database is used for storing crying sound samples, each crying sound sample is corresponding to one crying sound reason; the sound receiving unit acquires crying sound signals; the processing unit judges crying sound reasons corresponding to the crying sound signals based on the crying sound samples. When itis impossible to judge crying sound reasons corresponding to crying sound signals based on the crying sound samples, the processing unit provides prompt information via the output unit. When the prompt information is provided and the processing unit detects input of any crying sound reasons via the input unit, the processing unit matches the crying sound signals and the input crying sound reasonsand stores the crying sound signals into the crying sound database as crying sound samples; the processing unit further starts the pacifying device based on the crying sound reasons. The system and the method can identify the reasons for baby crying from crying sound of babies and achieve better crying sound identification effect.

The invention discloses an infant crying emotion recognition method based on an improved long-short-time memory network. The infant crying data set speech is detected by endpoints and divided into frames, the temporal correlation characteristics of the infant crying data set speech are extracted, and a processing algorithm of the long-short time memory network is established according to the temporal correlation characteristics of different lengths. Then, the strategy of combining attention mechanism with depth of temporal sequence is introduced into the forgetting gate of the long-short memory network. The experimental results show that this method can not only greatly reduce the model parameters, but also show significant recognition performance on the recorded infant emotion database, and the recognition efficiency is high, which has a good application prospect.

The embodiment of the invention discloses an infant crying prompting method. The method comprises the steps of: S1: receiving environment voice frequency sent by a sound collection device in real time; S2: performing matching of the environment voice frequency and an infant crying model through training to determine whether there is infant crying or not in the environment voice frequency, and if yes, sending a prompting instruction to a corresponding command execution terminal to allow the command execution terminal to execute prompting operation corresponding to a prompting command. Comparedwith a traditional infant crying prompting method, the infant crying prompting method adds a voice frequency matching process to match the environment voice frequency with the infant crying model through training to ensure that the prompting command is only sent to the command execution terminal when infant crying is generated, allow the command execution terminal to execute the prompting operation corresponding to the prompting command, improve the alarm accuracy of the infant crying prompting device and solve the technical problem that the false alarm rate of an infant crying prompting device is reduced.

The invention discloses a recognition and warning method and system for crying sounds of infants. The method comprises: acquiring a crying sound signal and a motion image of an infant in real time; carrying out pretreatment on the crying sound signal of the infant, carrying out calculation to obtain an inverted Mel-frequency cepstral coefficient of the crying sound signal of the infant, carrying out discretization on the inverted Mel-frequency cepstral coefficient to obtain a discrete value of the inverted Mel-frequency cepstral coefficient; according to a preset infant crying sound identification model established based on a Bayesian classification algorithm, classifying the discrete value of the inverted Mel-frequency cepstral coefficient to identify a crying sound state corresponding tothe crying sound signal of the infant; checking the crying sound state obtained by identification of the crying sound signal of the infant by using an infant motion image obtained in real time; and sending out a corresponding warning signal. According to the invention, the calculation method is simple; the calculation efficiency is high; the recognition accuracy is high; and the accuracy and precision of crying sound state recognition are improved. The recognition and warning method and system can be widely applied in the baby care field.

The embodiment of the invention provides a speech detection method, a camera and a smart home care system. The speech detection method comprises the steps of: obtaining target audio information, wherein the target audio information is audio information collected by a audio sensor built in a video surveillance device; performing analysis of the target audio information, and obtaining audio featurescorresponding to the target audio information; determining whether the audio features corresponding to the target audio information satisfy a preset infant crying condition or not; and when the audiofeatures corresponding to the target audio information satisfy the preset infant crying condition, outputting a prompting message, wherein the prompting message at least comprises a prompting contentconfigured to prompt an infant crying condition generated at present. According to the embodiment of the invention, a device in a family is employed to perform detection of infant crying and emit a prompting to a guardian so that a family consumption cost is not greatly increased.

The invention discloses an emotion detection method based on baby expression recognition and cry. The method comprises the steps: step 1, capturing face image information of a baby, and collecting anaudio data signal of the baby; step 2, when the baby is in a crying state, executing step 5; step 3, processing the captured face image; step 4, training the facial expression image obtained in the step 3 through a convolutional neural network to obtain facial expression features; step 5, windowing the infant crying voice signal obtained in the step 1 to generate a spectrogram corresponding to theinfant crying voice signal; step 6, training a classifier through a multi-layer perceptron and a support vector machine; step 7, inputting the facial expression features obtained in the step 4 into aclassifier for classification; and step 8, training an emotion detection model, and inputting the spectrogram in the step 5 into the trained emotion detection model to determine the emotion corresponding to the baby cry signal. According to the invention, the accuracy of infant emotion analysis is effectively improved.

The invention relates to the infant monitoring technical field, and particularly discloses an infant soothing method, an infant stroller and computer storage medium. The abovementioned method comprises steps that a sound signal is received; a distance between the infant stroller and external equipment is determined; whether the sound signal includes sounds of infant crying is identified; when yes,whether the distance is greater than a preset distance is determined; when the distance is greater than the preset distance, playing equipment is controlled to play multimedia files. According to theinvention, when the sounds of infant crying is detected, and guardians are far away from the infant, and cannot soothe the infant directly, the multimedia files are played automatically to soothe theinfant.

The invention relates to a baby crying detection method and device based on audio and video fusion. The infant crying detection method based on audio and video fusion comprises the following steps: acquiring an audio and video file to be detected, and splitting the audio and video file to obtain audio data and video data; extracting a Mel spectrogram of the audio data, and extracting abstract features of the Mel spectrogram by using a ResNet neural network to obtain voice features; extracting grayscale images of continuous frames in the video data, and extracting abstract features of the grayscale images by using a 3DCNN network to obtain visual features; fusing the voice features and the visual features to obtain mixed features; and inputting the mixed features into a trained ResNet residual network to obtain a crying detection result. According to the infant crying detection method and device based on audio and video fusion, voice and visual information are combined for automatic recognition of infant crying, and the accuracy of infant crying detection is improved.

The application discloses an active implantable human communication infant teaching model, including a simulated infant model, a stimulation module, a human communication module, a control module anda drive module. The human communication module and the control module are set in the simulated infant model. The stimulation module can generate an electrical signal. The human communication module can receive the electrical signal generated by the stimulation module and transmit the electrical signal to the control module. The control module can make the drive module opened or closed. The drive module can drive the simulated infant model to generate an infant crying behavior. A weak electrical signal transmitted through the human communication module can be detected through the control module, and thus, data communication with the human body as a medium can be realized. The physical simulation of electromagnetic properties of human tissues is realized by using simulated human tissues in the signal pathway. The signal pathway of the simulated infant model simulates the corresponding parts of the infant's body to realize the infant crying and struggling modes, so as to achieve the basicfunction of medical nursing operation simulation training.

The invention discloses an auxiliary fixing device for magnetic resonance examination of an infant , which relates to the technical field of medical equipment. The device is characterized in that the device comprises a first wrapping piece, a second wrapping piece, a third wrapping piece, a pillow body, an inflation valve, a first binding belt and a second binding belt, wherein two groups of binding belts are arranged on the outer side wall of the pillow body at intervals; the inner wall faces of the first wrapping piece, the second wrapping piece and the third wrapping piece are all detachably connected with heat preservation mechanisms. According to the device, the first wrapping piece, the second wrapping piece, the third wrapping piece and the pillow body are internally inflated so that the device can be tightly attached to an infant to be wrapped and fixed, the infant can be conveniently wrapped with a sense of safety, the infant is not prone to being squeezed, the comfort of the infant wrapped by the wrapping body can be improved, magnetic resonance examination of the infant can be quietly completed, the magnetic resonance examination effect is prevented from being affected by infant crying or abnormal movement, and the infant magnetic resonance examination quality is improved. Meanwhile, the situation that adverse reactions are easily generated when a sedative is adopted for auxiliary inspection at present can be avoided.

The invention discloses a cloud automatic camera monitoring system and method facilitating infant nursing. The system comprises a cloud server, a wireless camera and a wireless terminal. The wirelesscamera and the wireless terminal are both connected with the cloud server; according to the cloud automatic monitoring system and method facilitating infant nursing provided by the invention, live video analysis is provided, meaningful data is analyzed by capturing and sending video streaming media, and data analysis and storage are carried out on a cloud server; the infant crying can be detectedwith high precision, the sleeping condition can be effectively detected, and meanwhile the safety of user data is guaranteed. And the time and energy for parents to take care of infants are reduced.

The invention discloses a sachet used for relieving infant crying and awake-staying at night. The sachet is a product obtained by packaging a traditional Chinese medicine composition with cotton cloth or nonwoven fabric, wherein the traditional Chinese medicine composition comprises the following materials in parts by weight: 50-70 parts of wormwood flowers, 2-6 parts of baeckea frutescens essential oil, 10-20 parts of anbian sweet osmanthus, and 10-20 parts of fructus amomi. A production method of the sachet comprises the following steps: (1), respectively picking the wormwood flowers, the anbian sweet osmanthus and the fructus amomi, removing the materials which are bitten by insects and mildewed, respectively drying until moisture content is lower than 13%, weighing materials according to the formula amount, uniformly mixing, and crushing until all materials pass through a 100-mesh sieve mesh; (2), extracting volatile oil from the baeckea frutescens flowers by a steam distillation method; (3), adding the powder obtained in the step (1) into the volatile oil obtained in the step (2), uniformly stirring, and packaging by the cotton cloth or nonwoven fabric to obtain the Chinese herbal medicine sachet which is used for relieving infant crying and awake-staying at night, dispelling mosquitoes and removing bad smell, keeping faint scent and removing foul smell.

The invention provides an infant cry analysis method and device, a model training method and electronic equipment. The infant cry analysis method comprises the steps pf: acquiring infant cry information; inputting the infant cry information into a trained neural network model, and analyzing acoustic features and rhythm features of the infant cry information to obtain classification probabilities of the acoustic features and the rhythm features, wherein the neural network model is obtained by training cry information marked based on the acoustic features and the rhythm features as a training sample; and determining an analysis result based on the classification probabilities of the acoustic features and the rhythm features. According to the method, while the acoustic features are analyzed, and the rhythm feature analysis is added, so that the infant crying rhythm can be identified while the audio, tone, sound energy and other features of the infant can be identified, the infant crying state can be analyzed in multiple dimensions, and the infant crying reason can be accurately determined.

A method for correcting infant crying identification includes the following steps: a detecting step provides an audio unit to detect a sound around an infant to generate a plurality of audio samples. A converting step provides a processing unit to convert the audio samples to generate a plurality of audio spectrograms. An extracting step provides a common model to extract the audio spectrograms to generate a plurality of infant crying features. An incremental training step provides an incremental model to train the infant crying features to generate an identification result. A judging step provides the processing unit to judge whether the identification result is correct according to a real result of the infant. When the identification result is different from the real result, an incorrect result is generated. A correcting step provides the processing unit to correct the incremental model according to the incorrect result.