CN109840565A - A kind of blink detection method based on eye contour feature point aspect ratio - Google Patents
A kind of blink detection method based on eye contour feature point aspect ratio Download PDFInfo
- Publication number
- CN109840565A CN109840565A CN201910100968.4A CN201910100968A CN109840565A CN 109840565 A CN109840565 A CN 109840565A CN 201910100968 A CN201910100968 A CN 201910100968A CN 109840565 A CN109840565 A CN 109840565A
- Authority
- CN
- China
- Prior art keywords
- eye
- aspect ratio
- frame
- face image
- image set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 claims abstract description 14
- 238000013528 artificial neural network Methods 0.000 claims abstract description 5
- 210000000744 eyelid Anatomy 0.000 claims description 17
- 230000004397 blinking Effects 0.000 claims description 10
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 230000001815 facial effect Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000000193 eyeblink Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000001629 suppression Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010924 continuous production Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000005314 correlation function Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 210000000887 face Anatomy 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of blink detection methods based on eye contour feature point aspect ratio, this method comprises: obtaining video flowing, single frames face image set is extracted from the video flowing；The face rectangle frame extracted in facial image is concentrated from the single frames facial image using deep neural network, and obtains face rectangle frame coordinate；The eye key point coordinate that the single frames facial image is concentrated is acquired according to the face rectangle frame coordinate；The eyes aspect ratio of the single frames face image set is calculated, according to the eye key point coordinate to judge whether to blink according to the eyes aspect ratio.The present invention determines whether to blink by calculating eyes aspect ratio, the influence to blink judgement such as light intensity, personage's posture can be overcome, there is stronger robustness, and this method calculates simply, precision is high, high-efficient, can preferably be applied to the fields such as vivo identification, fatigue driving monitoring.
Description
Technical Field
The invention relates to the technical field of digital image processing and computer vision, in particular to a blink detection method based on the aspect ratio of eye contour feature points.
Background
At present, the security field is concerned with, wherein living body identification and fatigue driving monitoring are concerned with, and in the aspects of living body identification and fatigue driving monitoring, the functions of using blink detection to realize the functions belong to mainstream at present, and the method has the characteristics of no need of cooperation, realization of intellectualization and the like.
At present, a plurality of blink detection algorithms exist, but the accuracy rate of the blink detection algorithms needs to be improved. For example, based on blink detection of Adaboost, the Adaboost algorithm is an iterative algorithm, and its core idea is to train general classifiers (weak classifiers) with different classification capabilities for the same training set, and then add these weak classifiers together to form a stronger final classifier. However, the iteration times of the AdaBoost algorithm are not easy to set, and the AdaBoost algorithm has the defects that the classification precision of a classifier is reduced, the training comparison consumes time, and the AdaBoost algorithm is sensitive to abnormal values due to the fact that determination and unbalanced distribution of a training data set are required to be performed in a cross-validation mode. These disadvantages directly affect the blink detection effect and reduce the detection efficiency. For example, an ASM + Canny method is adopted to perform blink detection, that is, an ASM (Active Shape Model) algorithm is used to detect the human eye area, a Canny operator is used to calculate the edge contour of the human eye, and blinking is determined according to the up-down distance of the edge contour of the human eye, but the method has high requirements on light. Under weak light, the positioning effect of the ASM on human eyes and the detection effect of Canny on human eye edges are not good.
Disclosure of Invention
At least one of the objectives of the present invention is to overcome the above problems in the prior art, and to provide a blink detection method based on the aspect ratio of the eye contour feature points, which has better robustness by calculating the aspect ratio of both eyes to determine whether to blink.
In order to achieve the above object, the present invention adopts the following aspects.
A blink detection method based on aspect ratio of eye contour feature points comprises the following steps:
step 101, acquiring video stream data, and extracting a single-frame human face image set from the video stream;
102, extracting a face rectangular frame in a face image from the single-frame face image set by using a deep neural network, and acquiring coordinates of the face rectangular frame;
103, obtaining eye key point coordinates in the single-frame face image set according to the face rectangular frame coordinates;
and 104, calculating the eye aspect ratio of the single-frame human face image set according to the eye key point coordinates, and judging whether to blink according to the eye aspect ratio.
Preferably, the eye contour feature point aspect ratio-based blink detection method is characterized in that the number of eye key points of the human face is 12, each eye has 6 key points, the key points are respectively located on the upper eyelid, the lower eyelid, the left canthus and the right canthus of the two eyes of the human face, and the coordinates of the key points of each eye are represented by (x, y).
Preferably, the eye aspect ratio is an average value of the aspect ratio of key points of the left eye of the face and the aspect ratio of key points of the right eye of the face.
Preferably, the eye contour feature point aspect ratio-based blink detection method includes:
wherein,for the aspect ratio of key points for the left eye of the human face,the average value of the distance between the upper eyelid and the lower eyelid of the left eye of the human face, | | p1-p4| | | is the distance between the canthi of the left eye of the human face;the aspect ratio of the key points of the right eye of the human face,for distance of upper and lower eyelids of right eye of human faceAnd the average value, | | p7-p10| | | is the distance between the human right eye corners.
Preferably, the step 104 of the blink detection method based on the aspect ratio of the eye contour feature points specifically includes:
judging whether the frame number of the single-frame face image set with the calculated eye aspect ratio is greater than a threshold value, and if the frame number of the face image is greater than the threshold value, judging that the single-frame face image set is a continuous single-frame face image; and analyzing the eye aspect ratio of the continuous single-frame face image, drawing a corresponding eye aspect ratio curve graph, and judging as a blinking event when the curve graph drops rapidly.
Preferably, the step 104 is a blink detection method based on aspect ratio of eye contour feature points, and further includes: and when the frame number of the single-frame face image set is less than or equal to the threshold value, judging that the single-frame face image set is a discontinuous single-frame face image set, returning to the step 101, and re-extracting the single-frame face image set in the video stream.
In summary, due to the adoption of the technical scheme, the invention at least has the following beneficial effects:
whether blinking is judged by calculating the aspect ratio of the eyes, the influence of light intensity, human posture and the like on blinking judgment can be overcome, the method has strong robustness, is simple to calculate, high in accuracy of blink detection and high in efficiency, and can be better applied to the fields of living body recognition, fatigue driving monitoring and the like.
Drawings
Fig. 1 is a flowchart of a blink detection method based on aspect ratios of eye contour feature points according to an exemplary embodiment of the invention.
Fig. 2 is a schematic diagram of eye (left eye) keypoints according to an exemplary embodiment of the invention.
Fig. 3 is an eye aspect ratio fluctuation diagram according to an exemplary embodiment of the present invention, in which the horizontal axis is a time axis and the vertical axis is an eye aspect ratio average (BS).
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and embodiments, so that the objects, technical solutions and advantages of the present invention will be more clearly understood. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Fig. 1 illustrates a blink detection method based on aspect ratios of eye contour feature points according to an exemplary embodiment of the present invention. The method of this embodiment mainly includes:
step 101, acquiring video stream data, and extracting a single-frame human face image set from the video stream;
specifically, video stream data in a camera is acquired (at the time of blink detection, living body detection or face verification, the camera is turned on, corresponding video data is acquired, and in a corresponding device, such as a computer, a pc, a terminal, a pad and the like, a video exists in a streaming form, which can be decomposed into images of a single frame), then a corresponding image of the single frame is extracted from the video stream by using a correlation function in an opencv library in python, and when a person does a matching action, the action lasts for a period of time, so that a plurality of images of the single frame are extracted from the video stream.
102, extracting a face rectangular frame in a face image from the single-frame face image set by using a deep neural network, and acquiring coordinates of the face rectangular frame;
specifically, the single-frame image set obtained in step 101 is used to extract a face rectangle frame in the face image through a neural network. Of course, the extracted single frame image may have no human face, and is only a picture acquired by the camera. And if no human face is detected in the single frame image, detecting whether the next frame image has a human face. In this embodiment, we use a deep network called MTCNN to detect faces, MTCNN is a three-layer cascaded convolutional neural network, and MTCNN is divided into three steps: and carrying out multi-scale transformation on the image to obtain an image pyramid and obtain multi-scale information of the image. The network has three layers: 1. p-net (Proposal network): candidate frames and their bounding box regression vector sets are obtained mainly using a full convolutional network. These candidate vectors are then evaluated and calibrated. Finally, non-maximization suppression is used to remove a large number of repeated candidate regions. 2. R-net (refine network): all candidate regions are sent to R-Net, a full connection layer (FC) is added to the layer, more detailed processing can be carried out, a large number of candidate regions which do not meet the requirements are eliminated, calibration is carried out through bounding box regression, and combination is carried out through non-maximization suppression (NMS). This step is similar to the second step, which outputs more human facial features. For the face detection in step 102, only the rectangular frame of the position of the face needs to be acquired, so that in the subsequent processing, corresponding key point coordinates can be acquired according to the rectangular frame.
103, obtaining eye key point coordinates in the single-frame face image set according to the face rectangular frame coordinates;
specifically, fig. 2 shows a schematic diagram of a face left-eye keypoint of an exemplary embodiment of the present invention. It is assumed that each eye of the human face has 6 key points, and a total of 12 key points, and the six key points of each eye are respectively located on the upper eyelid, the lower eyelid, the left canthus and the right canthus (as shown in fig. 2), and each key point is represented by (x, y) coordinates. The left eye angle of the left eye of the person facing the lens is recorded as p1, and then the left key points of the left eye are marked clockwise, so that the eye key points of p1-p 6 of the left eye of the person are obtained. And marking the left canthus of the right eye facing the lens by a person as p7 clockwise around the rest key points of the right eye of the person to obtain the key points of the right eye of p7-p 12. Thus, the coordinate values of the key points p1-p 6 for the left eye and the coordinate values of the key points p7-p 12 for the right eye are obtained.
And 105, calculating the eye aspect ratio of the single-frame human face image set according to the eye key point coordinates, and judging whether to blink according to the eye aspect ratio.
Specifically, the eye aspect ratio of the single-frame face image set is obtained by the average value of the aspect ratio of the left eye key points and the aspect ratio of the right eye key points of the face:
wherein,for the left-eye keypoint aspect ratio,is the average value of the distance between the upper eyelid and the lower eyelid of the left eye, | | p1-p4| | | is the distance between the canthi of the left eye;is the right eye key point aspect ratio,is the average value of the distance between the upper eyelid and the lower eyelid of the right eye, and | p7-p10| is the distance between the corners of the eyes of the right eye.
In the actual detection, whether blinking is a continuous process is judged through the aspect ratio, or blinking is a continuous motion, eyes are opened, closed and opened, the aspect ratio of each frame of image is calculated, then an event that the aspect ratio is changed from large to small and then becomes large is found, whether blinking is judged cannot be simply calculated on the aspect ratio of a certain frame of image, and the whole process is analyzed, a continuous aspect ratio change curve shown in the figure three needs to be obtained, and the condition that blinking is judged if the aspect ratio is small as long as eyes are closed is prevented.
Specifically, the aspect ratio is calculated in real time, a fixed threshold is set empirically, and the blink is determined to be the case where the aspect ratio is larger than the threshold → smaller than the threshold → larger than the threshold. Therefore, in this example, it is first determined whether the number of frames of the single-frame face image set whose eye aspect ratio has been calculated is greater than a threshold, and if the number of frames of the face image is greater than the threshold, it is determined that the single-frame face image set is a continuous single-frame face image; and analyzing the eye aspect ratio of the continuous single-frame face image, drawing an eye aspect ratio curve chart shown in figure 3, and judging that the eye blinks when the eye aspect ratio drops rapidly. Further, the eye aspect ratio is maintained at a rough level when a person opens his eyes, and there may be a little difference between the individual persons, but a sharp drop occurs in the eye aspect ratio when the person blinks, so that the rapid drop event can be determined as a blinking event. And when the frame number of the single-frame face image set is less than or equal to the threshold value, judging that the single-frame face image set is a discontinuous single-frame face image set, returning to the step 101, re-extracting the video stream, and acquiring the single-frame face image set in the video stream.
In the embodiment, whether the eye blinks or not is judged by calculating the aspect ratio of the eyes, the influence of light intensity, human posture and the like on blink judgment can be overcome, the robustness is high, the calculation is simple, the accuracy and the efficiency of blink detection are high, and the method can be better applied to the fields of living body identification, fatigue driving monitoring and the like.
The foregoing is merely a detailed description of specific embodiments of the invention and is not intended to limit the invention. Various alterations, modifications and improvements will occur to those skilled in the art without departing from the spirit and scope of the invention.
Claims (6)
1. A blink detection method based on aspect ratio of eye contour feature points is characterized by comprising the following steps:
step 101, acquiring video stream data, and extracting a single-frame human face image set from the video stream;
102, extracting a face rectangular frame in a face image from the single-frame face image set by using a deep neural network, and acquiring coordinates of the face rectangular frame;
103, obtaining eye key point coordinates in the single-frame face image set according to the face rectangular frame coordinates;
and 104, calculating the eye aspect ratio of the single-frame human face image set according to the eye key point coordinates, and judging whether to blink according to the eye aspect ratio.
2. The method according to claim 1, wherein the number of eye key points of the human face is 12, 6 key points are provided for each eye, and are respectively located on the upper eyelid, the lower eyelid, the left canthus and the right canthus of the two eyes of the human face, and the coordinates of each eye key point are expressed by (x, y).
3. The method of claim 1, wherein the eye aspect ratio is an average of a human face left eye keypoint aspect ratio and a human face right eye keypoint aspect ratio.
4. The method of claim 3, wherein the eye aspect ratio is:
wherein,for the aspect ratio of key points for the left eye of the human face,the average value of the distance between the upper eyelid and the lower eyelid of the left eye of the human face, | | p1-p4| | | is the distance between the canthi of the left eye of the human face;the aspect ratio of the key points of the right eye of the human face,the average value of the distance between the upper eyelid and the lower eyelid of the right eye of the human face, and | p7-p10| is the distance between the corners of the right eye of the human face.
5. The method according to claim 1, wherein the step 104 specifically comprises:
judging whether the frame number of the single-frame face image set with the calculated eye aspect ratio is greater than a threshold value, and if the frame number of the face image is greater than the threshold value, judging that the single-frame face image set is a continuous single-frame face image; and analyzing the eye aspect ratio of the continuous single-frame face image, drawing a corresponding eye aspect ratio curve graph, and judging as a blinking event when the curve graph drops rapidly.
6. The method of claim 1, wherein the step 104 further comprises: and when the frame number of the single-frame face image set is less than or equal to the threshold value, judging that the single-frame face image set is a discontinuous single-frame face image set, returning to the step 101, and re-extracting the single-frame face image set in the video stream.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910100968.4A CN109840565A (en) | 2019-01-31 | 2019-01-31 | A kind of blink detection method based on eye contour feature point aspect ratio |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910100968.4A CN109840565A (en) | 2019-01-31 | 2019-01-31 | A kind of blink detection method based on eye contour feature point aspect ratio |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109840565A true CN109840565A (en) | 2019-06-04 |
Family
ID=66884371
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910100968.4A Pending CN109840565A (en) | 2019-01-31 | 2019-01-31 | A kind of blink detection method based on eye contour feature point aspect ratio |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109840565A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110221699A (en) * | 2019-06-13 | 2019-09-10 | 北京师范大学珠海分校 | A kind of eye movement Activity recognition method of front camera video source |
| CN110348394A (en) * | 2019-07-15 | 2019-10-18 | 广东名阳信息科技有限公司 | A method of detection video static object |
| CN110705454A (en) * | 2019-09-29 | 2020-01-17 | 东华大学 | Face recognition method with living body detection function |
| CN111126347A (en) * | 2020-01-06 | 2020-05-08 | 腾讯科技（深圳）有限公司 | Human eye state recognition method and device, terminal and readable storage medium |
| CN111178152A (en) * | 2019-12-09 | 2020-05-19 | 上海理工大学 | Attention detection reminding device based on three-dimensional head modeling |
| CN111179551A (en) * | 2019-12-17 | 2020-05-19 | 西安工程大学 | Real-time monitoring method for dangerous chemical transport driver |
| CN111611939A (en) * | 2020-05-22 | 2020-09-01 | 云知声智能科技股份有限公司 | Eye fatigue state detection method and device |
| CN111645695A (en) * | 2020-06-28 | 2020-09-11 | 北京百度网讯科技有限公司 | Fatigue driving detection method and device, computer equipment and storage medium |
| CN112256132A (en) * | 2020-10-28 | 2021-01-22 | 南京工程学院 | Man-machine interaction system for gradually-frozen person design |
| CN113283318A (en) * | 2021-05-13 | 2021-08-20 | Oppo广东移动通信有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN113345177A (en) * | 2021-05-11 | 2021-09-03 | 深圳市海恒智能科技有限公司 | Self-service book borrowing and returning system based on face recognition |
| CN113362450A (en) * | 2021-06-02 | 2021-09-07 | 聚好看科技股份有限公司 | Three-dimensional reconstruction method, device and system |
| CN113627300A (en) * | 2021-08-02 | 2021-11-09 | 中电福富信息科技有限公司 | Face recognition and living body detection method based on deep learning |
| CN113642364A (en) * | 2020-05-11 | 2021-11-12 | 华为技术有限公司 | Face image processing method, device and equipment and computer readable storage medium |
| CN114267080A (en) * | 2021-12-30 | 2022-04-01 | 淮阴工学院 | Non-difference blink identification method based on angle change |
| CN114529972A (en) * | 2022-02-22 | 2022-05-24 | 山西医科大学第一医院 | Autonomous call processing method and system for amyotrophic lateral sclerosis patient |
| CN114863545A (en) * | 2022-07-05 | 2022-08-05 | 之江实验室 | Automatic blink detection method and device based on DeepLabCut |
| CN115937958A (en) * | 2022-12-01 | 2023-04-07 | 北京惠朗时代科技有限公司 | Blink detection method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160041614A1 (en) * | 2014-08-06 | 2016-02-11 | Samsung Display Co., Ltd. | System and method of inducing user eye blink |
| US20160104036A1 (en) * | 2014-10-13 | 2016-04-14 | Utechzone Co., Ltd. | Method and apparatus for detecting blink |
| CN107784288A (en) * | 2017-10-30 | 2018-03-09 | 华南理工大学 | A kind of iteration positioning formula method for detecting human face based on deep neural network |
| CN108269333A (en) * | 2018-01-08 | 2018-07-10 | 平安科技（深圳）有限公司 | Face identification method, application server and computer readable storage medium |
| CN109191791A (en) * | 2018-10-30 | 2019-01-11 | 罗普特（厦门）科技集团有限公司 | A kind of fatigue detection method and device merging multiple features |
-
2019
- 2019-01-31 CN CN201910100968.4A patent/CN109840565A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160041614A1 (en) * | 2014-08-06 | 2016-02-11 | Samsung Display Co., Ltd. | System and method of inducing user eye blink |
| US20160104036A1 (en) * | 2014-10-13 | 2016-04-14 | Utechzone Co., Ltd. | Method and apparatus for detecting blink |
| CN107784288A (en) * | 2017-10-30 | 2018-03-09 | 华南理工大学 | A kind of iteration positioning formula method for detecting human face based on deep neural network |
| CN108269333A (en) * | 2018-01-08 | 2018-07-10 | 平安科技（深圳）有限公司 | Face identification method, application server and computer readable storage medium |
| CN109191791A (en) * | 2018-10-30 | 2019-01-11 | 罗普特（厦门）科技集团有限公司 | A kind of fatigue detection method and device merging multiple features |
Non-Patent Citations (3)
| Title |
|---|
| TEREZA SOUKUPOV´ A 等: "Real-Time Eye Blink Detection using Facial Landmarks", 《21ST COMPUTER VISION WINTER WORKSHOP》 * |
| 柳龙飞 等: "基于人脸特征点分析的疲劳驾驶实时检测方法", 《电视技术》 * |
| 黄洁媛 等: "基于卷积神经网络的人眼状态检测", 《扬州大学学报(自然科学版)》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110221699A (en) * | 2019-06-13 | 2019-09-10 | 北京师范大学珠海分校 | A kind of eye movement Activity recognition method of front camera video source |
| CN110221699B (en) * | 2019-06-13 | 2022-03-25 | 北京师范大学珠海分校 | Eye movement behavior identification method of front-facing camera video source |
| CN110348394A (en) * | 2019-07-15 | 2019-10-18 | 广东名阳信息科技有限公司 | A method of detection video static object |
| CN110705454A (en) * | 2019-09-29 | 2020-01-17 | 东华大学 | Face recognition method with living body detection function |
| CN111178152A (en) * | 2019-12-09 | 2020-05-19 | 上海理工大学 | Attention detection reminding device based on three-dimensional head modeling |
| CN111179551A (en) * | 2019-12-17 | 2020-05-19 | 西安工程大学 | Real-time monitoring method for dangerous chemical transport driver |
| CN111126347B (en) * | 2020-01-06 | 2024-02-20 | 腾讯科技（深圳）有限公司 | Human eye state identification method, device, terminal and readable storage medium |
| CN111126347A (en) * | 2020-01-06 | 2020-05-08 | 腾讯科技（深圳）有限公司 | Human eye state recognition method and device, terminal and readable storage medium |
| WO2021227988A1 (en) * | 2020-05-11 | 2021-11-18 | 华为技术有限公司 | Face image processing method, apparatus and device, and computer readable storage medium |
| CN113642364B (en) * | 2020-05-11 | 2024-04-12 | 华为技术有限公司 | Face image processing method, device, equipment and computer readable storage medium |
| CN113642364A (en) * | 2020-05-11 | 2021-11-12 | 华为技术有限公司 | Face image processing method, device and equipment and computer readable storage medium |
| CN111611939A (en) * | 2020-05-22 | 2020-09-01 | 云知声智能科技股份有限公司 | Eye fatigue state detection method and device |
| CN111645695A (en) * | 2020-06-28 | 2020-09-11 | 北京百度网讯科技有限公司 | Fatigue driving detection method and device, computer equipment and storage medium |
| CN111645695B (en) * | 2020-06-28 | 2022-08-09 | 北京百度网讯科技有限公司 | Fatigue driving detection method and device, computer equipment and storage medium |
| CN112256132A (en) * | 2020-10-28 | 2021-01-22 | 南京工程学院 | Man-machine interaction system for gradually-frozen person design |
| CN113345177A (en) * | 2021-05-11 | 2021-09-03 | 深圳市海恒智能科技有限公司 | Self-service book borrowing and returning system based on face recognition |
| CN113283318B (en) * | 2021-05-13 | 2024-07-16 | Oppo广东移动通信有限公司 | Image processing method, device, electronic equipment and storage medium |
| CN113283318A (en) * | 2021-05-13 | 2021-08-20 | Oppo广东移动通信有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN113362450A (en) * | 2021-06-02 | 2021-09-07 | 聚好看科技股份有限公司 | Three-dimensional reconstruction method, device and system |
| CN113627300A (en) * | 2021-08-02 | 2021-11-09 | 中电福富信息科技有限公司 | Face recognition and living body detection method based on deep learning |
| CN114267080A (en) * | 2021-12-30 | 2022-04-01 | 淮阴工学院 | Non-difference blink identification method based on angle change |
| CN114529972A (en) * | 2022-02-22 | 2022-05-24 | 山西医科大学第一医院 | Autonomous call processing method and system for amyotrophic lateral sclerosis patient |
| CN114863545B (en) * | 2022-07-05 | 2022-10-21 | 之江实验室 | Automatic blink detection method and device based on deep LabCut |
| CN114863545A (en) * | 2022-07-05 | 2022-08-05 | 之江实验室 | Automatic blink detection method and device based on DeepLabCut |
| CN115937958A (en) * | 2022-12-01 | 2023-04-07 | 北京惠朗时代科技有限公司 | Blink detection method, device, equipment and storage medium |
| CN115937958B (en) * | 2022-12-01 | 2023-12-15 | 北京惠朗时代科技有限公司 | Blink detection method, blink detection device, blink detection equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109840565A (en) | A kind of blink detection method based on eye contour feature point aspect ratio | |
| Shao et al. | Deep convolutional dynamic texture learning with adaptive channel-discriminability for 3D mask face anti-spoofing | |
| CN107346422B (en) | Living body face recognition method based on blink detection | |
| JP5629803B2 (en) | Image processing apparatus, imaging apparatus, and image processing method | |
| CN103632132B (en) | Face detection and recognition method based on skin color segmentation and template matching | |
| JP4743823B2 (en) | Image processing apparatus, imaging apparatus, and image processing method | |
| CN101339607B (en) | Human face recognition method and system, human face recognition model training method and system | |
| CN102375970B (en) | A kind of identity identifying method based on face and authenticate device | |
| CN104123543B (en) | A kind of eye movement recognition methods based on recognition of face | |
| CN107368778A (en) | Method for catching, device and the storage device of human face expression | |
| CN107330371A (en) | Acquisition methods, device and the storage device of the countenance of 3D facial models | |
| CN102214309B (en) | Special human body recognition method based on head and shoulder model | |
| KR20170006355A (en) | Method of motion vector and feature vector based fake face detection and apparatus for the same | |
| CN106570447B (en) | Based on the matched human face photo sunglasses automatic removal method of grey level histogram | |
| CN109858375A (en) | Living body faces detection method, terminal and computer readable storage medium | |
| CN106650574A (en) | Face identification method based on PCANet | |
| CN111460950A (en) | Cognitive distraction method based on head-eye evidence fusion in natural driving conversation behavior | |
| CN109409298A (en) | A kind of Eye-controlling focus method based on video processing | |
| CN111291701A (en) | Sight tracking method based on image gradient and ellipse fitting algorithm | |
| CN107862298B (en) | Winking living body detection method based on infrared camera device | |
| CN109711309A (en) | A kind of method whether automatic identification portrait picture closes one's eyes | |
| CN111860394A (en) | Gesture estimation and gesture detection-based action living body recognition method | |
| CN112633217A (en) | Human face recognition living body detection method for calculating sight direction based on three-dimensional eyeball model | |
| TWI427545B (en) | Face recognition method based on sift features and head pose estimation | |
| Dahal et al. | Incorporating skin color for improved face detection and tracking system |
Legal Events
| 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: 20190604 |
|
| RJ01 | Rejection of invention patent application after publication |