技术领域technical field
本发明属于视频图像处理领域，特别涉及真实场景下的非接触式心率检测方法。The invention belongs to the field of video image processing, in particular to a non-contact heart rate detection method in a real scene.
背景技术Background technique
随着计算机技术的发展和普及，越来越多的计算机理论涉及到医学领域，应用于医疗诊断和日常健康监测等各个环节，为医学的进步发挥了强大的辅助作用。心率是反映人体健康状况的指标之一，也是判断心血管疾病最基本的生理指标之一。传统上的临床心率检测手段需要利用心电图机的十二导联线接触人体的多个部位，操作繁杂，自动化程度不高，对使用者有较高的专业知识要求，不适用于普通场景下的心率检测。With the development and popularization of computer technology, more and more computer theories are involved in the field of medicine, and are applied in various links such as medical diagnosis and daily health monitoring, playing a powerful auxiliary role in the progress of medicine. Heart rate is one of the indicators reflecting the health status of the human body, and it is also one of the most basic physiological indicators for judging cardiovascular diseases. The traditional clinical heart rate detection method needs to use the 12-lead wire of the electrocardiograph to contact multiple parts of the human body. The operation is complicated, the degree of automation is not high, and the user has high professional knowledge requirements. Heart rate detection.
光学体积描记术(Photoplethysmography，PPG)是使用计算机视觉技术来进行心率检测最基本的方法，它通过发光二极管向皮下组织发射红光，红光被皮下的毛细血管网中的血红蛋白吸收，反射或透射到另一端的光敏晶体管，其信号经过处理后与动脉血液中的血红蛋白数量呈正相关，通过测量反射光强度，描记血液容积脉冲(Blood volumepulse，BVP)信号后，可以直接计算心率。傅明哲等人最早提出利用普通网络摄像头的非接触式心率检测方法。该方法利用独立成分分析(Independent Component Analysis,ICA)将三个平均的颜色踪迹分离为三个基源信号，通过分析第二个基源信号的功率谱估计心率。然而，ICA输出的第二个基源并不总能表示PPG信号，后来傅明哲等人对他们的方法进行了改进，加入了去趋势化和最高功率谱尖峰的信号选择操作。戚刚等人最近又提出了一种人脸旋转校正算法，对视频采集中的微弱面部晃动进行校正，消除了一部分人脸运动产生的噪声，使心率的测量的准确性进一步提高，适用于非合作场景下的心率测量。Photoplethysmography (PPG) is the most basic method of heart rate detection using computer vision technology. It emits red light to the subcutaneous tissue through light-emitting diodes. The red light is absorbed, reflected or transmitted by hemoglobin in the subcutaneous capillary network. To the phototransistor at the other end, its signal is positively correlated with the amount of hemoglobin in the arterial blood after processing, and the heart rate can be directly calculated by measuring the reflected light intensity and tracing the blood volume pulse (BVP) signal. Fu Mingzhe and others first proposed a non-contact heart rate detection method using an ordinary network camera. The method uses Independent Component Analysis (ICA) to separate the three averaged color traces into three base source signals, and estimates the heart rate by analyzing the power spectrum of the second base source signal. However, the second base source output by ICA does not always represent the PPG signal, and later Mingzhe Fu et al. improved their method by adding detrending and signal selection operations for the highest power spectrum peak. Qi Gang et al. recently proposed a face rotation correction algorithm, which corrects the weak facial shaking in video acquisition, eliminates part of the noise generated by face movement, and further improves the accuracy of heart rate measurement. It is suitable for non-human Heart rate measurement in collaborative scenarios.
以上方法均可以在一定场景下通过非接触式视频法较为准确的测量出心率值，但是计算过程均过于漫长，对于受测人员而言，过高的时间代价将使非接触式检测方法带来的便捷性相抵消。尤其是戚刚等人的人脸旋转校正算法，虽然消除了一定的运动噪声，但与此同时，增加了算法的数据运算量，使得心率检测的检测时间和健壮性大大降低，在实际场景中，不具备较好的用户体验和可操作性。The above methods can measure the heart rate value more accurately through the non-contact video method in certain scenarios, but the calculation process is too long. For the tested personnel, the high time cost will make the non-contact detection method bring offset by the convenience. In particular, the face rotation correction algorithm by Qi Gang et al., although it eliminates certain motion noise, at the same time, it increases the amount of data calculation of the algorithm, which greatly reduces the detection time and robustness of heart rate detection. , does not have a good user experience and operability.
发明内容Contents of the invention
本发明的目的是通过以下方案来实现的：一种基于视频的快速心率检测方法，其特征在于，包括如下步骤：The object of the present invention is achieved through the following scheme: a kind of fast heart rate detection method based on video, it is characterized in that, comprises the steps:
1)视频采集：将一个普通USB摄像头置于人脸正前方1m位置，摄像头与计算机进行连接，使用OpenCV/Qt程序来控制摄像头进行视频采集。采集过程中会同步进行人脸检测，确保所采集的视频帧序列中包含人脸信息，在心率检测期间，USB保持开启状态，不间断采集视频。1) Video capture: Place an ordinary USB camera at a position 1m in front of the face, connect the camera to the computer, and use the OpenCV/Qt program to control the camera for video capture. During the collection process, face detection will be performed synchronously to ensure that the collected video frame sequence contains face information. During the heart rate detection period, the USB will remain on and the video will be collected continuously.
2)数据结构预处理：将摄像头采集的帧序列输入到一个双向循环的链表中，接收视频帧序列的同时，设置一个指向当前操作节点的指针，指针向前方向的节点，用作数据待采集节点；指针向后方向的节点，用作运算节点。链表长度足够长，保证视频帧的采集和运算不会发生访问冲突。双向循环链表的特性保证了数据采集时间上的连续性，以及多线程同时访问共享数据的安全性。2) Data structure preprocessing: input the frame sequence collected by the camera into a two-way circular linked list, while receiving the video frame sequence, set a pointer to the current operation node, and the node in the forward direction of the pointer is used as the data to be collected Node; the node in the backward direction of the pointer, used as an operation node. The length of the linked list is long enough to ensure that the acquisition and operation of video frames will not cause access conflicts. The characteristics of the bidirectional circular linked list guarantee the continuity of data collection time and the security of multi-threaded access to shared data at the same time.
3)多线程的并行心率计算：在数据预处理执行完毕之后，心率检测线程被唤醒，进入执行状态，与视频采集线程并行运行。视频采集线程持续将数据输入到共享数据区域，心率运算线程循环计算心率，两者并行运算，检测的总时间大大降低。3) Multi-threaded parallel heart rate calculation: After the data preprocessing is completed, the heart rate detection thread is awakened, enters the execution state, and runs in parallel with the video acquisition thread. The video acquisition thread continues to input data into the shared data area, and the heart rate calculation thread calculates the heart rate in a loop. The two operate in parallel, and the total detection time is greatly reduced.
本发明与现有的技术相比，具有的有益效果是：Compared with the prior art, the present invention has the beneficial effects of:
1)本发明针对接触式心率测量的操作复杂和肢体束缚问题，提出一种基于视频的快速心率检测方法。该技术无需利用电极或者传感器接触人体，只借助普通摄像头即可自动监测心率。有效地提高了检测效率和受测者的使用体验，适用于长时间的心率监测和疾病预防。1) The present invention proposes a video-based fast heart rate detection method for the complex operation and limb restraint problems of contact heart rate measurement. This technology does not need to use electrodes or sensors to touch the human body, and can automatically monitor heart rate with the help of ordinary cameras. It effectively improves the detection efficiency and the use experience of the subjects, and is suitable for long-term heart rate monitoring and disease prevention.
2)本发明采用了多线程设计结构，引入了并行运算的思想，有效的提高了运算速度，使非接触式心率检测技术具有更好的使用体验，提高了该技术的实际意义。2) The present invention adopts a multi-thread design structure, introduces the idea of parallel computing, effectively improves the computing speed, makes the non-contact heart rate detection technology have a better user experience, and improves the practical significance of the technology.
3)采用了双向循环链表的数据结构来处理数据，将链表分为数据采集区域和运算区域两个部分，使得并行运算成为可能，使多线程能够异步、并行的访问数据。3) The data structure of the bidirectional circular linked list is used to process the data, and the linked list is divided into two parts: the data acquisition area and the operation area, which makes parallel operation possible and enables multiple threads to access data asynchronously and in parallel.
附图说明Description of drawings
图1是本发明的算法流程图Fig. 1 is the algorithm flowchart of the present invention
图2是本发明的设备安装效果Fig. 2 is the equipment installation effect of the present invention
图3是双向循环链表的结构示意图Figure 3 is a schematic diagram of the structure of a two-way circular linked list
图4是本发明提取的血液容积脉搏波形Fig. 4 is the blood volume pulse waveform extracted by the present invention
具体实施方式detailed description
以下将结合附图1至4对本发明做进一步的说明，但不应以此来限制本发明的保护范围。为了方便说明并且理解本发明的技术方案，以下说明所使用的方位词均以附图所展示的方位为准。The present invention will be further described below in conjunction with accompanying drawings 1 to 4, but this should not limit the protection scope of the present invention. For the convenience of description and understanding of the technical solution of the present invention, the orientation words used in the following description are all subject to the orientation shown in the drawings.
步骤1，将一个普通USB摄像头置于人脸正前方0.5m位置，摄像头与计算机进行连接，操作者使用OpenCV/Qt程序来控制摄像头进行视频采集，如图2所示。采集过程中会进行人脸检测，确保所采集的视频帧序列中包含人脸信息，在心率检测期间，USB保持开启状态，不间断采集视频。摄像头的设置参数为，640*480分辨率，30fps帧率以及RGB色域。Step 1, place an ordinary USB camera at a position 0.5m in front of the face, connect the camera to the computer, and the operator uses the OpenCV/Qt program to control the camera for video acquisition, as shown in Figure 2. During the collection process, face detection will be performed to ensure that the collected video frame sequence contains face information. During the heart rate detection period, the USB will remain on and the video will be collected continuously. The setting parameters of the camera are 640*480 resolution, 30fps frame rate and RGB color gamut.
步骤2，将摄像头采集的帧序列存入一个双向循环链表的节点当中，该链表节点个数设置为300，节点结构设计为两个分别指向前后的指针域、OpenCV存储图像的Mat格式的数据域、用于指示当前节点位置的标志域。循环接收连续的视频帧的同时，有一个指针指向当前操作节点，指针向前方向为待存入的视频帧节点，用作数据存入；指针向后方向为已存入视频帧节点，用作下一步的运算，链表长度足够长，保证视频帧存入不会影响到视频帧的计算，具体结构示意图如图3所示。双向循环链表的特性保证了数据录入的连续性，以及多线程运算的线程互斥和对临界数据访问的数据安全。Step 2, store the frame sequence captured by the camera into the nodes of a bidirectional circular linked list, the number of nodes in the linked list is set to 300, and the node structure is designed as two pointer fields pointing to the front and back respectively, and a data field in the Mat format of the OpenCV stored image , the flag field used to indicate the current node location. While cyclically receiving continuous video frames, there is a pointer pointing to the current operation node, the forward direction of the pointer is the video frame node to be stored, used for data storage; the backward direction of the pointer is the stored video frame node, used for In the next step of calculation, the length of the linked list is long enough to ensure that the storage of video frames will not affect the calculation of video frames. The specific structure diagram is shown in Figure 3. The characteristics of the two-way circular linked list ensure the continuity of data entry, as well as the thread mutual exclusion of multi-threaded operations and the data security of critical data access.
步骤3，通过布尔型控制变量来控制视频采集和心率运算的先后顺序，实现线程异步和线程同步，具体算法步骤为：Step 3, control the order of video acquisition and heart rate calculation through Boolean control variables, and realize thread asynchrony and thread synchronization. The specific algorithm steps are:
1)视频采集初始化阶段：链表中不存在视频帧，视频采集线程启动，心率运算进入循环等待状态。1) Video acquisition initialization stage: there is no video frame in the linked list, the video acquisition thread starts, and the heart rate calculation enters a cyclic waiting state.
2)视频采集循环阶段：链表数据预处理填装完毕之后，满足心率计算的条件，心率运算线程从等待状态唤醒至运行状态，视频采集和心率运算并行执行，实现连续、快速的心率检测。2) Video acquisition cycle stage: After the preprocessing of linked list data is filled and the conditions for heart rate calculation are met, the heart rate calculation thread wakes up from the waiting state to the running state, and video acquisition and heart rate calculation are executed in parallel to achieve continuous and rapid heart rate detection.
步骤4，开始连续心率检测，具体包括：Step 4, start continuous heart rate detection, including:
1)数据输入：获取链表的当前指针，从指针当前节点向后取连续S帧作为计算数据，同时，对该序列做逆置变换，其第i帧(i＝1,2,3…S)的输入按照公式(1)执行：1) Data input: Obtain the current pointer of the linked list, take consecutive S frames backward from the current node of the pointer as the calculation data, and at the same time, perform an inverse transformation on the sequence, the i-th frame (i=1,2,3...S) The input of is performed according to the formula (1):
frames[i]＝List[S-i] (1)frames[i]=List[S-i] (1)
其中，frames[i]表示第i帧图像，List[S-i]表示所取的链表子序列对应节点；Among them, frames[i] represents the i-th frame image, and List[S-i] represents the corresponding node of the linked list subsequence taken;
2)ROI区域的选定：利用OpenCV提供的人脸识别算法，对每一帧进行人脸检测，得到只包含人脸信息的区域矢量Rect(x,y,w,h)，作为ROI(Region Of Interest，感兴趣区域)区域，其中Rect表示一个矩阵区域，x和y表示矩阵的起始顶点，w和h表示矩阵的长和宽。2) Selection of ROI area: Use the face recognition algorithm provided by OpenCV to detect the face of each frame, and obtain the area vector Rect(x,y,w,h) containing only face information as the ROI(Region Of Interest, region of interest) region, where Rect represents a matrix region, x and y represent the starting vertex of the matrix, and w and h represent the length and width of the matrix.
3)色域转换：将人脸视频的色彩空间由RGB转换为YIQ，使图像的亮度信息和色度信息分开，便于亮度和色度的单独处理，其中YIQ是NTSC(National Television StandardsCommittee)彩色电视系统使用的色彩空间，其中Y通道存储图像的亮度(Luminance)信息，I和Q通道存储图像的色度(chrominance)信息，I表示从橙色到青色的颜色变化，Q表示从紫色到黄绿色的颜色变化，RGB和YIQ的转换关系如下式所示：3) Color gamut conversion: Convert the color space of face video from RGB to YIQ, so that the brightness information and chrominance information of the image are separated, which is convenient for the separate processing of brightness and chrominance, wherein YIQ is NTSC (National Television Standards Committee) color TV The color space used by the system, in which the Y channel stores the luminance (Luminance) information of the image, the I and Q channels store the chrominance (chrominance) information of the image, I indicates the color change from orange to cyan, and Q indicates the color change from purple to yellow-green The color change, the conversion relationship between RGB and YIQ is shown in the following formula:
4)空间滤波：利用一种多尺度的图像高斯金字塔分解方法对视频进行空间分解，它通过对每帧图像进行连续地高斯平滑和降采样，得到多个层次的子带集合，计算步骤如下：4) Spatial filtering: A multi-scale image Gaussian pyramid decomposition method is used to spatially decompose the video. It performs continuous Gaussian smoothing and downsampling on each frame of image to obtain a multi-level subband set. The calculation steps are as follows:
①输入第j(j＝1,2,3…)帧图像作为第0层，计算分解层数L；① Input the jth (j=1,2,3...) frame image as the 0th layer, and calculate the number of decomposition layers L;
②对前一层图像先进行高斯滤波，后降采样，图像尺寸变为原来的1/4，记为次一层②Gaussian filtering is performed on the image of the previous layer, and then down-sampling, the image size becomes 1/4 of the original, which is recorded as the next layer
③将步骤②迭代执行L-1次，得到第L层子带图像；③ Step ② is iteratively executed L-1 times to obtain the L-th layer subband image;
④j＝j+1，循环执行以上步骤，输出子带序列；④j=j+1, execute the above steps in a loop, and output the subband sequence;
5)时域滤波：利用一种基于光流法的理想带通滤波对金字塔分解后的帧序列进行理想带通滤波，计算步骤如下：5) Time-domain filtering: use an ideal band-pass filter based on the optical flow method to perform ideal band-pass filtering on the frame sequence after pyramid decomposition, and the calculation steps are as follows:
①输入整个图像帧序列，该序列共S帧，大小为w*h；① Input the entire image frame sequence, the sequence has a total of S frames, and the size is w*h;
②取每一帧的第(x，y)点像素(x＝1,2,3…w，y＝1,2,3…h)在1～S帧的值作为该点的光流序列perFrame[x][y]；②Take the value of the (x, y)th pixel (x=1,2,3...w, y=1,2,3...h) in frames 1 to S of each frame as the optical flow sequence perFrame of the point [x][y];
③对perFrame[x][y]进行快速傅里叶变换，得到(x，y)点在视频帧序列中的时域分量，对该分量进行通频带为0.83～2.00Hz的理想带通滤波，并进行傅里叶逆变换，得到第i帧图像(x，y)像素的纯净的血液容积脉搏BVP信号BVP[x][y][i]；③ Perform fast Fourier transform on perFrame[x][y] to obtain the time-domain component of point (x, y) in the video frame sequence, and perform ideal band-pass filtering on this component with a passband of 0.83-2.00Hz, And perform inverse Fourier transform to obtain the pure blood volume pulse BVP signal BVP[x][y][i] of the i-th frame image (x, y) pixel;
④x＝x+1，y＝y+1，循环执行以上步骤，输出所有像素的BVP信号；④x=x+1, y=y+1, perform the above steps in a loop, and output the BVP signals of all pixels;
6)功率谱计算心率：通过对BVP[x][y][i]进行计算，得出心率值，计算步骤如下：6) Calculation of heart rate by power spectrum: By calculating BVP[x][y][i], the heart rate value is obtained. The calculation steps are as follows:
①将第i帧(i＝1,2,3…S)的所有像素点的BVP值进行求和，得到一维BVP信号量B[i]；①Sum the BVP values of all pixels in the i-th frame (i=1,2,3...S) to obtain a one-dimensional BVP semaphore B[i];
②对B[i]进行快速傅里叶变换，得到其功率谱PBvp：② Perform fast Fourier transform on B[i] to obtain its power spectrum PBvp :
F(t)＝fft(B(t)) (3)F(t)=fft(B(t)) (3)
PBvp(t)＝|F(t)|2 (4)PBvp (t)＝|F(t)|2 (4)
其中，fft是快速傅里叶变换函数。where fft is the fast Fourier transform function.
③心率值HR的计算：③ Calculation of heart rate value HR:
T＝max{PBvp(t)} (5)T=max{PBvp (t)} (5)
其中，fps为视频的帧率。Among them,fps is the frame rate of the video.
至此，一次基于人脸视频处理的心率检测基本完成。So far, a heart rate detection based on face video processing is basically completed.
根据上述说明书的揭示和教导，本发明所属领域的技术人员还可以对上述实施方式进行变更和修改。因此，本发明并不局限于上面揭示和描述的具体实施方式，对本发明的一些修改和变更也应当落入本发明的权利要求的保护范围内。此外，尽管本说明书中使用了一些特定的术语，但这些术语只是为了方便说明，并不对本发明构成任何限制。According to the disclosure and teaching of the above-mentioned specification, those skilled in the art to which the present invention belongs can also make changes and modifications to the above-mentioned embodiment. Therefore, the present invention is not limited to the specific embodiments disclosed and described above, and some modifications and changes to the present invention should also fall within the protection scope of the claims of the present invention. In addition, although some specific terms are used in this specification, these terms are only for convenience of description and do not constitute any limitation to the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710543119.7ACN107330945A (en) | 2017-07-05 | 2017-07-05 | A kind of examing heartbeat fastly method based on video |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710543119.7ACN107330945A (en) | 2017-07-05 | 2017-07-05 | A kind of examing heartbeat fastly method based on video |
| Publication Number | Publication Date |
|---|---|
| CN107330945Atrue CN107330945A (en) | 2017-11-07 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710543119.7APendingCN107330945A (en) | 2017-07-05 | 2017-07-05 | A kind of examing heartbeat fastly method based on video |
| Country | Link |
|---|---|
| CN (1) | CN107330945A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109770885A (en)* | 2019-03-25 | 2019-05-21 | 合肥工业大学 | A fast heart rate detection method based on preview frame |
| CN110501363A (en)* | 2019-08-30 | 2019-11-26 | 中国科学院长春应用化学研究所 | X-ray diffractometer with multi-thread sample test method |
| CN112233813A (en)* | 2020-09-28 | 2021-01-15 | 兰州大学 | A non-contact non-invasive heart rate and respiration measurement method and system based on PPG |
| CN112395464A (en)* | 2020-11-18 | 2021-02-23 | 扬州船用电子仪器研究所(中国船舶重工集团公司第七二三研究所) | Electronic information intermediate frequency data processing method based on openmp acceleration |
| CN112954423A (en)* | 2020-04-08 | 2021-06-11 | 深圳市明源云客电子商务有限公司 | Animation playing method, device and equipment |
| CN113052115A (en)* | 2021-04-06 | 2021-06-29 | 合肥工业大学 | Unmanned aerial vehicle airborne vital sign detection method based on video method |
| CN116999044A (en)* | 2023-09-07 | 2023-11-07 | 南京云思创智信息科技有限公司 | Real-time motion full-connection bidirectional consistent optical flow field heart rate signal extraction method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100334919C (en)* | 2005-03-25 | 2007-08-29 | 北京北方烽火科技有限公司 | Managing method for user contex of multi-object positioning in GMLC |
| CN103077057A (en)* | 2012-12-31 | 2013-05-01 | 浙江创佳数字技术有限公司 | Android intelligent set-top box-based Loader method |
| US20160132732A1 (en)* | 2014-11-10 | 2016-05-12 | Utah State University | Remote Heart Rate Estimation |
| CN105989357A (en)* | 2016-01-18 | 2016-10-05 | 合肥工业大学 | Human face video processing-based heart rate detection method |
| CN106778695A (en)* | 2017-01-19 | 2017-05-31 | 北京理工大学 | A kind of many people's examing heartbeat fastly methods based on video |
| CN106725411A (en)* | 2017-02-15 | 2017-05-31 | 成都皓图智能科技有限责任公司 | A kind of contactless heartbeat detecting method accelerated based on GPU |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100334919C (en)* | 2005-03-25 | 2007-08-29 | 北京北方烽火科技有限公司 | Managing method for user contex of multi-object positioning in GMLC |
| CN103077057A (en)* | 2012-12-31 | 2013-05-01 | 浙江创佳数字技术有限公司 | Android intelligent set-top box-based Loader method |
| US20160132732A1 (en)* | 2014-11-10 | 2016-05-12 | Utah State University | Remote Heart Rate Estimation |
| CN105989357A (en)* | 2016-01-18 | 2016-10-05 | 合肥工业大学 | Human face video processing-based heart rate detection method |
| CN106778695A (en)* | 2017-01-19 | 2017-05-31 | 北京理工大学 | A kind of many people's examing heartbeat fastly methods based on video |
| CN106725411A (en)* | 2017-02-15 | 2017-05-31 | 成都皓图智能科技有限责任公司 | A kind of contactless heartbeat detecting method accelerated based on GPU |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109770885A (en)* | 2019-03-25 | 2019-05-21 | 合肥工业大学 | A fast heart rate detection method based on preview frame |
| CN110501363A (en)* | 2019-08-30 | 2019-11-26 | 中国科学院长春应用化学研究所 | X-ray diffractometer with multi-thread sample test method |
| CN112954423A (en)* | 2020-04-08 | 2021-06-11 | 深圳市明源云客电子商务有限公司 | Animation playing method, device and equipment |
| CN112233813A (en)* | 2020-09-28 | 2021-01-15 | 兰州大学 | A non-contact non-invasive heart rate and respiration measurement method and system based on PPG |
| CN112395464A (en)* | 2020-11-18 | 2021-02-23 | 扬州船用电子仪器研究所(中国船舶重工集团公司第七二三研究所) | Electronic information intermediate frequency data processing method based on openmp acceleration |
| CN113052115A (en)* | 2021-04-06 | 2021-06-29 | 合肥工业大学 | Unmanned aerial vehicle airborne vital sign detection method based on video method |
| CN116999044A (en)* | 2023-09-07 | 2023-11-07 | 南京云思创智信息科技有限公司 | Real-time motion full-connection bidirectional consistent optical flow field heart rate signal extraction method |
| CN116999044B (en)* | 2023-09-07 | 2024-04-16 | 南京云思创智信息科技有限公司 | Real-time motion full-connection bidirectional consistent optical flow field heart rate signal extraction method |
| Publication | Publication Date | Title |
|---|---|---|
| CN107330945A (en) | A kind of examing heartbeat fastly method based on video | |
| Casado et al. | Face2PPG: An unsupervised pipeline for blood volume pulse extraction from faces | |
| Karthick et al. | Analysis of vital signs using remote photoplethysmography (RPPG) | |
| Qi et al. | Video-based human heart rate measurement using joint blind source separation | |
| Alnaggar et al. | Video-based real-time monitoring for heart rate and respiration rate | |
| McDuff et al. | Fusing partial camera signals for noncontact pulse rate variability measurement | |
| Cheng et al. | Illumination variation-resistant video-based heart rate measurement using joint blind source separation and ensemble empirical mode decomposition | |
| Iozzia et al. | Relationships between heart-rate variability and pulse-rate variability obtained from video-PPG signal using ZCA | |
| Zhang et al. | Illumination variation-resistant video-based heart rate monitoring using LAB color space | |
| CN103826532A (en) | Remote monitoring of vital signs | |
| CN105266772A (en) | Measuring method for physiological parameter | |
| Przybyło | A deep learning approach for remote heart rate estimation | |
| CN115553777B (en) | Non-contact mental stress detection method and system | |
| CN111345803A (en) | Heart rate variability measuring method based on mobile device camera | |
| Zhang et al. | Simultaneous detection of blink and heart rate using multi-channel ICA from smart phone videos | |
| Qi et al. | Non-contact driver cardiac physiological monitoring using video data | |
| Blackford et al. | Measuring pulse rate variability using long-range, non-contact imaging photoplethysmography | |
| Blackford et al. | Long-range non-contact imaging photoplethysmography: cardiac pulse wave sensing at a distance | |
| Das et al. | Time–frequency learning framework for rPPG signal estimation using Scalogram-based feature map of facial video data | |
| Wu et al. | Anti-jamming heart rate estimation using a spatial–temporal fusion network | |
| Nikolaiev et al. | Non-contact video-based remote photoplethysmography for human stress detection | |
| Li et al. | ECG signal reconstruction based on facial videos via combined explicit and implicit supervision | |
| Hwang et al. | Phase-shifted remote photoplethysmography for estimating heart rate and blood pressure from facial video | |
| Wiede et al. | Signal fusion based on intensity and motion variations for remote heart rate determination | |
| Ben Salah et al. | Contactless heart rate estimation from facial video using skin detection and multi-resolution analysis |
| 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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | Application publication date:20171107 | |
| WD01 | Invention patent application deemed withdrawn after publication |