技术领域technical field
本发明涉及传感器技术应用领域，特别是一种基于柔性传感器的人体坐姿识别方法及其系统。The invention relates to the application field of sensor technology, in particular to a flexible sensor-based human sitting posture recognition method and system thereof.
背景技术Background technique
坐姿状态是学生群体、办公人群绝大部分时间所处的状态，而不良坐姿会对人们的身体健康造成很大的影响，青少年在学习和生活当中，往往不能很好地注意保持正确身姿，而家长和老师又不能每时每刻在身边提醒，时间一长就形成近视和驼背；对于上班族来说，随着生活节奏加快以及工作压力日益增加，越来越多的成年人需要长时间面对电脑，由于坐姿不正确，常常会引起职业性肌肉骨骼疾患和下背痛等症状，不良坐姿对健康造成的负面影响，不仅会严重影响患者的身心健康和工作能力，也为医疗保健带来严重的经济负担，因此，针对人体坐姿识别的研究对人们的身体健康有着非常重要的意义。Sitting posture is the state where students and office workers spend most of their time, and bad sitting posture will have a great impact on people's health. Teenagers often fail to pay attention to maintaining correct posture in their studies and lives. And parents and teachers can't remind them all the time, myopia and hunchback will form over time; for office workers, with the accelerated pace of life and increasing work pressure, more and more adults need to work for a long time Facing the computer, improper sitting posture often causes symptoms such as occupational musculoskeletal disorders and low back pain. The negative impact of bad sitting posture on health will not only seriously affect the physical and mental health of patients and work ability, but also bring new benefits to medical care. Therefore, the research on human sitting posture recognition is of great significance to people's health.
传统的方法主要是基于计算机视觉等技术，通过图像和视频信息获取人体坐姿信息，此类识别方法一般需要配有照相机、摄像头、摄像机等硬件设备，这类识别装置往往价格昂贵，易受光照、人体着装、成像设备的精度等因素影响，对于用户坐姿的数据采集存在一定的误差，导致最终的坐姿识别结果不太准确。The traditional method is mainly based on technologies such as computer vision, and obtains the information of human sitting posture through image and video information. This kind of recognition method generally needs to be equipped with hardware equipment such as cameras, cameras, and video cameras. Such recognition devices are often expensive and susceptible to light, Affected by factors such as human body clothing and the accuracy of imaging equipment, there are certain errors in the data collection of the user's sitting posture, resulting in inaccurate final sitting posture recognition results.
发明内容Contents of the invention
为解决上述问题，本发明的目的在于提供一种基于柔性传感器的人体坐姿识别方法及其系统，通过柔性传感器来检测人体背部的弯曲数据，结合神经网络对坐姿进行辨识，识别较为准确，同时用户可以通过智能终端接收自身的坐姿数据。In order to solve the above problems, the object of the present invention is to provide a flexible sensor-based human body sitting posture recognition method and system thereof. The flexible sensor is used to detect the bending data of the human body's back, and the sitting posture is recognized in combination with a neural network. The recognition is relatively accurate. At the same time, the user You can receive your own sitting posture data through the smart terminal.
本发明解决其问题所采用的技术方案是：一种基于柔性传感器的人体坐姿识别方法，包括以下步骤：The technical scheme adopted by the present invention to solve its problem is: a kind of human body sitting position recognition method based on flexible sensor, comprises the following steps:
A、柔性传感器采集用户背部的弯曲数据；A. The flexible sensor collects the bending data of the user's back;
B、对用户背部的弯曲数据进行预处理；B. Preprocessing the bending data of the user's back;
C、对处理后的弯曲数据进行特征提取并组成特征向量；C. Carry out feature extraction to the processed bending data and form a feature vector;
D、将特征向量分为训练集和测试集，构建误差反向传播的人工神经网络模型，利用训练集以及测试集对误差反向传播的人工神经网络进行训练；D, divide feature vector into training set and test set, construct the artificial neural network model of error backpropagation, utilize the training set and test set to train the artificial neural network of error backpropagation;
E、利用已训练好的误差反向传播的人工神经网络对待识别用户背部的弯曲数据的特征向量进行识别，并得到识别结果；E. Use the trained artificial neural network of error backpropagation to identify the feature vector of the bending data on the back of the user to be identified, and obtain the identification result;
F、将识别结果传输到智能终端。F. Transmit the recognition result to the smart terminal.
进一步，所述步骤B中对用户背部的弯曲数据进行预处理，利用算术平均法对用户背部的弯曲数据进行过滤处理。Further, in the step B, preprocessing is performed on the bending data of the user's back, and the arithmetic mean method is used to filter the bending data of the user's back.
进一步，所述步骤C中对处理后的弯曲数据进行特征提取并组成特征向量，提取处理后的弯曲数据的时域特征，并将时域特征组成特征向量，其中时域特征包括均值、标准值、最大值和最小值。Further, in the step C, feature extraction is performed on the processed bending data and a feature vector is formed, the time domain features of the processed bending data are extracted, and the time domain features are formed into a feature vector, wherein the time domain features include mean value, standard value , maximum and minimum values.
进一步，所述步骤D中将特征向量分为训练集和测试集，其中将特征向量以4：1的比例分成训练集和测试集。Further, in the step D, the feature vector is divided into a training set and a test set, wherein the feature vector is divided into a training set and a test set at a ratio of 4:1.
进一步，所述步骤D中构建误差反向传播的人工神经网络模型，其中误差反向传播的人工神经网络包括三层，第一层为具有4个神经元的输入层，第二层为具有4个神经元的隐含层，第三层为具有3个神经元的输出层。Further, in the step D, construct the artificial neural network model of error backpropagation, wherein the artificial neural network of error backpropagation includes three layers, the first layer is an input layer with 4 neurons, and the second layer is an input layer with 4 neurons. The hidden layer has 3 neurons, and the third layer is the output layer with 3 neurons.
一种基于柔性传感器的人体坐姿识别系统，包括用于采集用户背部的弯曲数据的测量单元以及对弯曲数据进行处理和识别的识别单元，所述测量单元包括用于采集用户背部的弯曲数据的柔性传感器、将柔性传感器采集的弯曲数据转换成电压分配值的微控制器以及将识别结果发送到用户智能终端的无线通信单元，所述识别单元包括对弯曲数据进行预处理的预处理单元、对处理后的数据进行特征提取的特征提取单元以及设置有误差反向传播的人工神经网络模型的分类识别单元，所述柔性传感器、微控制器、预处理单元以及特征提取单元依次连接，所述分类识别单元对特征提取单元输出的特征向量进行识别，并将识别结果传输到所述微控制器中，所述微控制器通过无线通信单元发送识别结果。A human sitting posture recognition system based on a flexible sensor, comprising a measurement unit for collecting bending data of the user's back and an identification unit for processing and identifying the bending data, the measuring unit includes a flexible sensor for collecting the bending data of the user's back sensor, a microcontroller that converts the bending data collected by the flexible sensor into a voltage distribution value, and a wireless communication unit that sends the recognition result to the user's smart terminal. The recognition unit includes a preprocessing unit that preprocesses the bending data, a processing A feature extraction unit for feature extraction of the final data and a classification recognition unit provided with an artificial neural network model of error backpropagation, the flexible sensor, microcontroller, preprocessing unit and feature extraction unit are connected in sequence, and the classification recognition The unit identifies the feature vector output by the feature extraction unit, and transmits the identification result to the microcontroller, and the microcontroller sends the identification result through the wireless communication unit.
进一步，所述柔性传感器上设置有将其固定在用户背部的运动贴布，所述柔性传感器包括由硅酮-镍纳米复合材料构成的应变片以及分压电阻，所述应变片与分压电阻分别连接到微控制器。Further, the flexible sensor is provided with a sports patch to fix it on the back of the user, and the flexible sensor includes a strain gauge made of silicone-nickel nanocomposite material and a voltage dividing resistor, and the strain gauge and the voltage dividing resistor connected to the microcontroller respectively.
进一步，所述测量单元还包括便携式壳体，所述微控制器设置于便携式壳体内，所述便携式壳体上设置有用于将其固定在用户衣服或腰带上的夹子或别针。Further, the measuring unit further includes a portable housing, the microcontroller is disposed in the portable housing, and the portable housing is provided with clips or pins for fixing it on the user's clothes or belt.
进一步，所述无线通信单元为蓝牙通信模块。Further, the wireless communication unit is a Bluetooth communication module.
本发明的有益效果是：本发明采用的一种基于柔性传感器的人体坐姿识别方法，通过对用户背部的弯曲数据进行采集，并将采集的弯曲数据进行预处理和特征提取，然后利用提取到的特征向量对误差反向传播的人工神经网络进行训练，训练完成的误差反向传播的人工神经网络即可以对待识别用户背部的弯曲数据进行识别，并将识别结果发送到用户的智能终端上，使用户可以了解到自己当前的坐姿情况；The beneficial effects of the present invention are: a human body sitting posture recognition method based on a flexible sensor adopted in the present invention collects the bending data of the user's back, preprocesses the collected bending data and extracts features, and then uses the extracted The eigenvectors train the artificial neural network of error backpropagation, and the trained artificial neural network of error backpropagation can recognize the bending data of the user's back to be recognized, and send the recognition result to the user's smart terminal, so that Users can know their current sitting posture;
本发明的一种基于柔性传感器的人体坐姿识别系统，采用柔性传感器对用户背部的弯曲数据进行采集，结构简单，识别精确，可以根据用户背部弯曲的不同角度输出不同的值到微控制器中，最后在通过识别单元对弯曲数据进行识别得到识别结果，然后通过无线通信单元将识别结果发送到用户智能终端，使用户可以随时了解到自身的坐姿情况，采用柔性传感器作为坐姿的检测装置，相比于传统的通过压力传感器检测压力值而言，结构更加简单。A human body sitting posture recognition system based on a flexible sensor of the present invention uses a flexible sensor to collect the bending data of the user's back, has a simple structure and accurate recognition, and can output different values to the microcontroller according to different angles of the user's back bending, Finally, the recognition unit is used to identify the bending data to obtain the recognition result, and then the recognition result is sent to the user's smart terminal through the wireless communication unit, so that the user can know his own sitting posture at any time. The flexible sensor is used as the sitting posture detection device. Compared with the traditional way of detecting the pressure value through the pressure sensor, the structure is simpler.
附图说明Description of drawings
下面结合附图和实例对本发明作进一步说明。The present invention will be further described below in conjunction with accompanying drawing and example.
图1是本发明一种基于柔性传感器的人体坐姿识别方法的流程框图；Fig. 1 is a kind of flow chart of the human body sitting posture recognition method based on flexible sensor of the present invention;
图2是本发明一种基于柔性传感器的人体坐姿识别系统的原理框图；Fig. 2 is a functional block diagram of a human sitting posture recognition system based on a flexible sensor of the present invention;
图3是误差反向传播的人工神经网络的网络模型结构图。Fig. 3 is a network model structure diagram of an artificial neural network for error backpropagation.
具体实施方式Detailed ways
参照图1，本发明的一种基于柔性传感器的人体坐姿识别方法，包括以下步骤：A、柔性传感器11采集用户背部的弯曲数据；B、对用户背部的弯曲数据进行预处理；C、对处理后的弯曲数据进行特征提取并组成特征向量；D、将特征向量分为训练集和测试集，构建误差反向传播的人工神经网络模型，利用训练集以及测试集对误差反向传播的人工神经网络进行训练；E、利用已训练好的误差反向传播的人工神经网络对待识别用户背部的弯曲数据的特征向量进行识别，并得到识别结果；F、将识别结果传输到智能终端。With reference to Fig. 1, a kind of human body sitting posture recognition method based on flexible sensor of the present invention comprises the following steps: A, flexible sensor 11 collects the bending data of user's back; B, the bending data of user's back is preprocessed; C, processing The final bending data is subjected to feature extraction and composed of feature vectors; D. Divide feature vectors into training sets and test sets, construct artificial neural network models for error backpropagation, and use training sets and test sets to artificial neural network models for error backpropagation The network is trained; E, the artificial neural network using the trained error backpropagation is used to identify the feature vector of the bending data of the user's back to be identified, and the identification result is obtained; F, the identification result is transmitted to the intelligent terminal.
本发明在采集用户的坐姿数据时，和传统的通过压力检测器检测各个检测点的压力值不同，而是采集用户背部的弯曲数据作为坐姿数据，由于各个用户的体重、身材不同，所以传统的使用压力检测器检测就存在一定的误差，不能准确且全方位的检测，检测的数据和坐姿数据也很难对应起来，而本发明将用户背部的弯曲数据作为坐姿数据是较为合理的，背部弯曲的不同程度可以转换成不同的坐姿数据，在进行后续的计算和识别中也较为方便。When the present invention collects the user's sitting posture data, it is different from the traditional pressure value of each detection point detected by the pressure detector, but collects the bending data of the user's back as the sitting posture data. Because the weight and body of each user are different, the traditional There are certain errors in detection using a pressure detector, which cannot be accurately and comprehensively detected, and it is difficult to correspond between the detected data and the sitting posture data. However, it is more reasonable for the present invention to use the bending data of the user's back as the sitting posture data. Different degrees of sitting posture can be converted into different sitting posture data, which is also more convenient for subsequent calculation and identification.
具体地，在对背部的弯曲数据进行预处理中，利用的是算术平均法进行的过滤处理，由于人体坐姿信号在某一特定范围内上下浮动，并且受到的信号干扰随机性较强，所以采用算术平均法对坐姿信号进行过滤，本发明的采样频率设定为50Hz，每种坐姿维持60秒，所以在每一次采样中，每种坐姿都有3000个采样点，对于任意一个采样点，对其连续的100个采样点进行算术平均运算，用计算得到的值代替该点的采样值。Specifically, in the preprocessing of the back bending data, the arithmetic mean method is used for filtering processing. Since the human sitting posture signal fluctuates up and down within a certain range, and the received signal interference is relatively random, the The arithmetic mean method filters the sitting posture signal, and the sampling frequency of the present invention is set to 50Hz, and every kind of sitting posture maintains 60 seconds, so in every sampling, every kind of sitting posture has 3000 sampling points, for any sampling point, to The arithmetic average operation is performed on 100 continuous sampling points, and the sampling value of this point is replaced by the calculated value.
具体地，提取处理后的弯曲数据的时域特征，并将时域特征组成特征向量，其中时域特征包括均值、标准值、最大值和最小值。Specifically, the time-domain features of the processed bending data are extracted, and the time-domain features are composed into a feature vector, wherein the time-domain features include mean value, standard value, maximum value and minimum value.
具体地，在对误差反向传播的人工神经网络模型进行训练前，首先将特征向量分为训练集和测试集，其中将特征向量以4：1的比例分成训练集和测试集，训练集用于对误差反向传播的人工神经网络模型进行训练，测试集用于对训练好的误差反向传播的人工神经网络模型进行测试。Specifically, before training the artificial neural network model of error backpropagation, the feature vector is firstly divided into a training set and a test set, wherein the feature vector is divided into a training set and a test set at a ratio of 4:1, and the training set is used The artificial neural network model for error backpropagation is used for training, and the test set is used for testing the trained artificial neural network model for error backpropagation.
本发明所采用的误差反向传播的人工神经网络，包括三层，第一层为具有4个神经元的输入层，第二层为具有4个神经元的隐含层，第三层为具有3个神经元的输出层，如图3所示，其中201为输入层、202为隐含层、203为输出层，整个网络的工作过程分为两个部分：第一部分正向传播过程输入信息从输入层经隐含层逐层计算各神经元的输出值；第二部分反向传播过程输出误差逐层向前计算出隐含层各神经元的误差，并用此误差不断地来更新网络的其他参数，使得神经网络可以自动调整参数，不断完善自身结构，提高识别的准确度。The artificial neural network of error backpropagation adopted in the present invention comprises three layers, the first layer is an input layer with 4 neurons, the second layer is a hidden layer with 4 neurons, and the third layer is an input layer with 4 neurons. The output layer of 3 neurons is shown in Figure 3, where 201 is the input layer, 202 is the hidden layer, and 203 is the output layer. The working process of the entire network is divided into two parts: the first part is the forward propagation process input information The output value of each neuron is calculated layer by layer from the input layer through the hidden layer; the second part of the backpropagation process outputs the error layer by layer to calculate the error of each neuron in the hidden layer, and uses this error to continuously update the network. Other parameters enable the neural network to automatically adjust parameters, continuously improve its own structure, and improve the accuracy of recognition.
参照图2，本发明的一种基于柔性传感器的人体坐姿识别系统，包括测量单元1以及识别单元2，测量单元1用于采集用户背部的弯曲数据，然后将弯曲数据传输到识别单元2，识别单元2进行识别后将识别结果传输回测量单元1中，由测量单元1对识别结果进行传输。Referring to Fig. 2, a human sitting posture recognition system based on a flexible sensor of the present invention includes a measurement unit 1 and a recognition unit 2, the measurement unit 1 is used to collect the bending data of the user's back, and then transmits the bending data to the recognition unit 2 for recognition After the identification, the unit 2 transmits the identification result back to the measurement unit 1, and the measurement unit 1 transmits the identification result.
具体地，测量单元1包括柔性传感器11、微控制器12以及无线通信单元13，柔性传感器11由应变片111和感应电路组成，应变片111会由于用户背部的弯曲而发生形变，所以在应变片111上分配的电压就会改变，然后微控制器12可以在其模拟输入端口获得介于0～1023之间的模拟值，在本发明中用其代表用户背部的弯曲数据，从而得到用户的坐姿数据，而无线通信单元13用于将最终的识别结果传输到用户的智能终端上，供用户对自身的坐姿进行在线的查看，以便用户了解自己的坐姿并进行调整。Specifically, the measurement unit 1 includes a flexible sensor 11, a microcontroller 12, and a wireless communication unit 13. The flexible sensor 11 is composed of a strain gauge 111 and a sensing circuit. The strain gauge 111 will be deformed due to the bending of the user's back, so the strain gauge The voltage distributed on 111 will change, and then the microcontroller 12 can obtain an analog value between 0 and 1023 at its analog input port, which is used in the present invention to represent the bending data of the user's back, thereby obtaining the user's sitting posture data, and the wireless communication unit 13 is used to transmit the final recognition result to the user's smart terminal, for the user to check his own sitting posture online, so that the user can understand his own sitting posture and make adjustments.
具体地，感应电路为一个分压电阻112，分压电阻112与应变片111串联构成柔性传感器11，并且分压电阻112的阻值和应变片111在稳定状态下的阻值相等。Specifically, the sensing circuit is a voltage dividing resistor 112, which is connected in series with the strain gauge 111 to form the flexible sensor 11, and the resistance of the voltage dividing resistor 112 is equal to the resistance of the strain gauge 111 in a steady state.
在柔性传感器11上设置有运动贴布，可以将柔性传感器11固定在用户背部的衣服上，不会对用户造成影响以及不适感，本发明的应变片111主要由硅酮-镍纳米复合材料构成，这种材料的应变片111感测的形变范围较大，同时能够保持独特的反压阻性，并且成本较低，应变片111选择共聚酯硅酮材料作为衬底，并在其中添加NiNs镍纳米链和NCCF镀镍碳纤维，应变片111所受的形变越大，其阻值越小。The flexible sensor 11 is provided with a sports patch, which can fix the flexible sensor 11 on the clothes on the user's back without affecting or causing discomfort to the user. The strain gauge 111 of the present invention is mainly composed of silicone-nickel nanocomposite materials , the strain gauge 111 of this material has a large deformation range, while maintaining the unique reverse piezoresistivity, and the cost is low. The strain gauge 111 chooses copolyester silicone material as the substrate, and adds NiNs in it Nickel nano-strands and NCCF nickel-plated carbon fibers, the greater the deformation of the strain gauge 111, the smaller the resistance.
具体地，测量单元1还包括一个便携式壳体，将分压电阻112、微控制器12、无线通信单元13以及电源模块设置在便携式壳体内，然后在便携式壳体的外面设置夹子或者别针，从而使得便携式壳体可以固定在用户的衣服或者腰带上，避免对用户造成不适。Specifically, the measurement unit 1 also includes a portable housing, the voltage dividing resistor 112, the microcontroller 12, the wireless communication unit 13 and the power supply module are arranged in the portable housing, and then clips or pins are set outside the portable housing, thereby This enables the portable housing to be fixed on the user's clothes or belt to avoid discomfort to the user.
具体地，本发明的微控制器12是通过Arduino Uno实现，微控制器12Arduino Uno的每个模拟输入端口A0～A5的分辨率为10位即读取的模拟值介于0～1023之间。Specifically, the microcontroller 12 of the present invention is implemented by an Arduino Uno, and the resolution of each analog input port A0-A5 of the microcontroller 12Arduino Uno is 10 bits, that is, the read analog value is between 0-1023.
而无线通信单元13为蓝牙通信模块，通过蓝牙进行无线数据传输，将最终的坐姿数据的识别结果传输到用户的智能终端，本发明的蓝牙通信模块采用的是HC-06蓝牙设备，体积小巧，传输距离能够达到10米，增加了本发明的可靠性。And wireless communication unit 13 is bluetooth communication module, carries out wireless data transmission by bluetooth, the identification result of final sitting position data is transmitted to the intelligent terminal of user, what bluetooth communication module of the present invention adopts is HC-06 bluetooth equipment, and volume is small and exquisite, The transmission distance can reach 10 meters, which increases the reliability of the present invention.
识别单元2包括预处理单元21、特征提取单元22以及分类识别单元23，预处理单元21对微控制器12传输的弯曲数据进行过滤处理，减少干扰，然后通过特征提取单元22提取处理后的弯曲数据的时域特征，并组成特征向量，最后在利用设置了误差反向传播的人工神经网络模型的分类识别单元23对特征向量进行识别即可得到坐姿数据的识别结果，然后将识别结果传输到微控制器12中，微控制器12通过无线通信单元13将识别结果传输到用户的智能终端。The recognition unit 2 includes a preprocessing unit 21, a feature extraction unit 22, and a classification recognition unit 23. The preprocessing unit 21 filters the bending data transmitted by the microcontroller 12 to reduce interference, and then extracts the processed bending data through the feature extraction unit 22. The time domain characteristic of data, and form feature vector, utilize the classification identification unit 23 of the artificial neural network model of error backpropagation to be identified at last feature vector and can obtain the recognition result of sitting position data, then recognition result is transmitted to In the microcontroller 12, the microcontroller 12 transmits the recognition result to the user's smart terminal through the wireless communication unit 13.
本发明中的误差反向传播的人工神经网络模型可以移植到用户的智能终端中，并通过开发APP为用户提供相关服务，可以根据分类结果，预先在APP中设定坐姿别的阈值和时长，一旦用户的坐姿数据小于阈值，且持续时间达到阈值的预设时长时，APP就会通过振动或者铃声的形式通知使用者纠正错误的坐姿，除此之外，智能终端还可以显示实时的坐姿状况，并通过大数据和云计算的相关技术，一方面为使用者提供更详细更个性化的建议，另一方面可以进行云连接，与好朋友分享和对比，建立激励机制，从而有助于用户在日常生活中自觉地保持良好的坐姿。The artificial neural network model of error backpropagation in the present invention can be transplanted into the user's intelligent terminal, and provide relevant services for the user by developing an APP, and can pre-set the threshold and duration of different sitting postures in the APP according to the classification results, Once the user's sitting posture data is less than the threshold and the duration reaches the threshold preset time, the APP will notify the user to correct the wrong sitting posture through vibration or ringtone. In addition, the smart terminal can also display the real-time sitting posture status , and through the related technologies of big data and cloud computing, on the one hand, it can provide users with more detailed and personalized suggestions; Consciously maintain a good sitting posture in daily life.
以上所述，只是本发明的较佳实施例而已，本发明并不局限于上述实施方式，只要其以相同的手段达到本发明的技术效果，都应属于本发明的保护范围。The above descriptions are only preferred embodiments of the present invention, and the present invention is not limited to the above-mentioned embodiments, as long as they achieve the technical effects of the present invention by the same means, they should all belong to the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810188887.XACN108549834A (en) | 2018-03-08 | 2018-03-08 | A kind of human body sitting posture recognition methods and its system based on flexible sensor |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810188887.XACN108549834A (en) | 2018-03-08 | 2018-03-08 | A kind of human body sitting posture recognition methods and its system based on flexible sensor |
| Publication Number | Publication Date |
|---|---|
| CN108549834Atrue CN108549834A (en) | 2018-09-18 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810188887.XAPendingCN108549834A (en) | 2018-03-08 | 2018-03-08 | A kind of human body sitting posture recognition methods and its system based on flexible sensor |
| Country | Link |
|---|---|
| CN (1) | CN108549834A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109063697A (en)* | 2018-10-11 | 2018-12-21 | 浙江工业大学 | A kind of human body sitting posture detection method based on channel state information |
| CN109820373A (en)* | 2019-03-28 | 2019-05-31 | 重庆邮电大学 | Adaptive adjustment method of sitting posture based on intelligent seat |
| CN110135447A (en)* | 2018-10-30 | 2019-08-16 | 初速度（苏州）科技有限公司 | The system for adjusting personnel's sitting posture in vehicle according to the personal information of identification |
| CN112190258A (en)* | 2020-09-30 | 2021-01-08 | 珠海格力电器股份有限公司 | Seat angle adjusting method and device, storage medium and electronic equipment |
| CN113855004A (en)* | 2021-09-30 | 2021-12-31 | 山东科技大学 | Preparation method and application of flexible pressure sensor |
| CN114323368A (en)* | 2021-12-09 | 2022-04-12 | 深圳先进技术研究院 | Flexible intelligent sitting posture monitoring system based on hip pressure |
| CN115127704A (en)* | 2022-06-24 | 2022-09-30 | 永艺家具股份有限公司 | Sitting posture identification method based on flexible film pressure sensor |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011250915A (en)* | 2010-06-01 | 2011-12-15 | Sharp Corp | Posture recognition device |
| CN104720821A (en)* | 2015-04-01 | 2015-06-24 | 深圳柔微传感科技有限公司 | Method and smart clothing for achieving real-time posture monitoring |
| CN105761454A (en)* | 2016-03-22 | 2016-07-13 | 宁波元鼎电子科技有限公司 | Intelligent sitting posture corrector |
| CN106530640A (en)* | 2016-12-16 | 2017-03-22 | 深圳柔微传感科技有限公司 | Intelligent children garment, gesture data processing method and intelligent children wearable device |
| CN107038404A (en)* | 2016-10-19 | 2017-08-11 | 中国地质大学（武汉） | A kind of sitting posture detecting method and system based on BP neural network |
| CN107296427A (en)* | 2017-08-24 | 2017-10-27 | 云康新智能科技（深圳）有限公司 | A kind of cushion and sitting posture analysis method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011250915A (en)* | 2010-06-01 | 2011-12-15 | Sharp Corp | Posture recognition device |
| CN104720821A (en)* | 2015-04-01 | 2015-06-24 | 深圳柔微传感科技有限公司 | Method and smart clothing for achieving real-time posture monitoring |
| CN105761454A (en)* | 2016-03-22 | 2016-07-13 | 宁波元鼎电子科技有限公司 | Intelligent sitting posture corrector |
| CN107038404A (en)* | 2016-10-19 | 2017-08-11 | 中国地质大学（武汉） | A kind of sitting posture detecting method and system based on BP neural network |
| CN106530640A (en)* | 2016-12-16 | 2017-03-22 | 深圳柔微传感科技有限公司 | Intelligent children garment, gesture data processing method and intelligent children wearable device |
| CN107296427A (en)* | 2017-08-24 | 2017-10-27 | 云康新智能科技（深圳）有限公司 | A kind of cushion and sitting posture analysis method |
| Title |
|---|
| 何园: "基于神经网络的坐姿传感技术", 《中国优秀硕士学位论文全文数据库 信息科技辑（月刊）自动化技术》* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109063697A (en)* | 2018-10-11 | 2018-12-21 | 浙江工业大学 | A kind of human body sitting posture detection method based on channel state information |
| CN110135447A (en)* | 2018-10-30 | 2019-08-16 | 初速度（苏州）科技有限公司 | The system for adjusting personnel's sitting posture in vehicle according to the personal information of identification |
| CN110135447B (en)* | 2018-10-30 | 2021-08-24 | 初速度（苏州）科技有限公司 | System for adjusting sitting posture of personnel in vehicle according to identified personnel information |
| CN109820373A (en)* | 2019-03-28 | 2019-05-31 | 重庆邮电大学 | Adaptive adjustment method of sitting posture based on intelligent seat |
| CN112190258A (en)* | 2020-09-30 | 2021-01-08 | 珠海格力电器股份有限公司 | Seat angle adjusting method and device, storage medium and electronic equipment |
| CN112190258B (en)* | 2020-09-30 | 2021-07-20 | 珠海格力电器股份有限公司 | Seat angle adjusting method and device, storage medium and electronic equipment |
| CN113855004A (en)* | 2021-09-30 | 2021-12-31 | 山东科技大学 | Preparation method and application of flexible pressure sensor |
| CN113855004B (en)* | 2021-09-30 | 2024-10-11 | 山东科技大学 | A preparation method and application of flexible pressure sensor |
| CN114323368A (en)* | 2021-12-09 | 2022-04-12 | 深圳先进技术研究院 | Flexible intelligent sitting posture monitoring system based on hip pressure |
| CN115127704A (en)* | 2022-06-24 | 2022-09-30 | 永艺家具股份有限公司 | Sitting posture identification method based on flexible film pressure sensor |
| CN115127704B (en)* | 2022-06-24 | 2023-07-28 | 永艺家具股份有限公司 | A Sitting Posture Recognition Method Based on Flexible Thin Film Pressure Sensor |
| Publication | Publication Date | Title |
|---|---|---|
| CN108549834A (en) | A kind of human body sitting posture recognition methods and its system based on flexible sensor | |
| CN102292025B (en) | Determining energy expenditure of a user | |
| TWI393579B (en) | The state of the muscle movement state analysis system, methods and computer program products | |
| Pärkkä et al. | Personalization algorithm for real-time activity recognition using PDA, wireless motion bands, and binary decision tree | |
| US10335061B2 (en) | Methods and apparatus for detecting hand-to-mouth behavior | |
| US9261967B2 (en) | Gesture recognition system | |
| CN106999065A (en) | Wearable pain monitor using accelerometry | |
| CN104888444A (en) | Intelligent gloves, method and system for calorie consumption and hand posture recognition | |
| CN105833508A (en) | Intelligent gloves capable of measuring calorie consumption and monitoring hand posture identification as well as estimation method and system | |
| US20200060546A1 (en) | A System and Method for Monitoring Human Performance | |
| CN114323368A (en) | Flexible intelligent sitting posture monitoring system based on hip pressure | |
| CN109512188B (en) | Sitting posture detection method and device and seat | |
| CN103654798A (en) | Emotion monitoring and recording method and device | |
| CN110292386A (en) | A human sitting posture detection system and method based on video information and piezoelectric information | |
| CN108042106A (en) | Artificial intelligence deviation rectifying method for improving detection precision of human body sign non-invasive detection equipment | |
| CN113679344A (en) | Sleep monitor detection method and device and sleep monitor | |
| CN103340633A (en) | Sleeping posture identification method based on bioelectrical impedance | |
| CN110772246B (en) | A device and method for synchronously detecting bioelectrical signals and pressure signals | |
| CN106667127A (en) | Intelligent cushion capable of correcting sitting posture | |
| CN203290920U (en) | Portable emotion analysis meter | |
| CN207070264U (en) | A kind of heart rate earphone | |
| CN107799178B (en) | Health service method and system for remote monitoring | |
| CN108392193A (en) | A kind of wearable device and its working method of real-time monitor heart rate | |
| CN211534418U (en) | Human health state assessment equipment based on footprint characteristics | |
| CN112071426A (en) | Functional body fitness testing method, device and equipment |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20180918 | |
| RJ01 | Rejection of invention patent application after publication |