Facial expression evaluation method, system, device and medium based on visual perception

Abstract

The present invention discloses a facial expression evaluation method, system, device and medium based on visual perception, which relates to the field of facial evaluation technology, and includes the steps of: using 3D visual perception technology to obtain various reflective marks of the facial image of the object to be tested; capturing and analyzing the motion trajectory, speed and area of ​​facial muscles before and after contraction through reflective marks; obtaining the percentage through the area occupied by each expression feature point part when exerting force in various reflective marks, and obtaining the overall score through the sum of all percentages; using the overall score for grading, and performing facial evaluation through the grading results. The present invention determines the difference of the face before and after muscle contraction according to the motion trajectory and speed, and uses the occupied area to obtain the percentage of facial expression features in the different places, so as to achieve the supplement of the speed and motion trajectory defects through the area occupied by the healthy and affected sides, and improve the evaluation accuracy.

Description

Technical Field

[0001] The present invention relates to the field of facial evaluation technology, and in particular to a facial expression evaluation method, system, device and medium based on visual perception. Background Art

[0002] Neurological diseases often quietly attack the facial muscles of patients, making their expressions unnatural and stiff, or even completely losing their original liveliness and vitality. This disease not only seriously affects the daily quality of patients, but also brings them a heavy psychological burden. By studying the facial expressions of patients, we can better evaluate the facial state of patients and facilitate the observation of their symptoms.

[0003] Facial expression evaluation has long been a thorny problem. After acquiring the image or video information of facial expressions, traditional facial expression recognition methods first use some feature extraction methods to process the facial feature data to obtain the necessary facial expression feature information, and then use the facial expression feature information to evaluate facial expressions based on visual perception.

[0004] With the development of machine learning and computer vision technology, the accuracy and efficiency of facial expression recognition have been continuously improved. However, in the existing evaluation methods, due to the different expressions on different human bodies, the acquired expression feature points have large differences in speed and trajectory. These differences lead to large errors in the evaluation of facial expressions, bringing great uncertainty. Summary of the invention

[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art and provide a facial expression evaluation method, system, device and storage medium based on visual perception, so as to solve the problem in the prior art that due to the different expressions on different human bodies, the acquired expression feature points have large differences in speed and trajectory, and these differences lead to large errors in the evaluation of facial expressions, which brings great uncertainty.

[0006] The present invention specifically provides the following technical solutions:

[0007] A facial expression evaluation method based on visual perception comprises the following steps:

[0008] Using 3D visual perception technology to obtain three-dimensional structural information of the face of the object being measured in real time, and obtaining a real-time facial image through the three-dimensional structural information, wherein the facial image includes reflective marks pre-pasted at various positions on the face of the object being measured;

[0009] Obtain the position changes of the reflective marks in the real-time facial images before and after the muscle contraction and in the continuous facial images, and obtain the motion trajectory, speed and occupied area of ​​the facial expression features at the marked points through the position changes of the reflective marks;

[0010] The difference between the face of the tested subject before and after the muscle contraction is evaluated by the motion trajectory and speed, and the percentage of the facial expression feature at each marked point is obtained by the ratio of the area of ​​the affected side to the area of ​​the healthy side;

[0011] The overall facial state score is obtained by summing up all the percentages, and the facial expressions are graded using the overall facial state score, and the state of the facial expressions is evaluated based on the grading results.

[0012] Preferably, the method of obtaining the percentage of facial expression features at each marked point by the ratio of the area of ​​the affected side to the area of ​​the healthy side comprises the following steps:

[0013] The left and right dividing line of the forehead is set as the midline of the human body, and the forehead contraction rate a is obtained by the overall contraction area A of the bilateral foreheads and the contraction area A1 of the healthy side when exerting force; the specific expression is:

[0014] Forehead shrinkage rate .

[0015] The percentage of closed eyes is obtained by looking straight ahead with both eyes when relaxed, with the healthy side leakage area B and the affected side leakage area B1. ; The specific expression is:

[0016] Percentage of eyes closed .

[0017] The forced eye opening percentage c is obtained by using the healthy eye area C and the affected eye area C1 when the eyes are opened forcefully; the specific expression is:

[0018] Percentage of eyes opened .

[0019] Take two points on each side of the nose as the straight-line distance D on the healthy side and the straight-line distance D1 on the affected side to measure the amplitude of the nose-lifting movement on the left and right sides, and obtain the nose-lifting percentage d; the specific expression is:

[0020] Percentage of nose lift .

[0021] The percentage of tooth protrusion e is obtained by comparing the tooth leakage area E of the healthy side and the tooth leakage area E1 of the affected side when protruding teeth with force; the specific expression is:

[0022] Teeth baring percentage ×100%.

[0023] The inflation percentage f is obtained by the healthy cheek inflation volume F and the affected cheek inflation volume F1 when inflating vigorously; the specific expression is:

[0024] Air inflation percentage f ×100%.

[0025] The pouting percentage g is obtained by the bulging volume G of the healthy side lips and the bulging volume G1 of the affected side lips when pouting hard; the specific expression is:

[0026] Pout percentage g= ×100%.

[0027] Preferably, obtaining the overall score of the facial state by summing up all the percentages comprises the following steps:

[0028] Get forehead contraction rate a and eye closure percentage , Percentage of eyes opened with force , percentage of nose lift , the sum of the percentage of baring teeth e, the percentage of puffing air f and the percentage of pouting lips g;

[0029] The overall score of the facial state is obtained by the ratio of the sum of the percentages to the number of facial expression features at each marking point.

[0030] Preferably, the facial expression is graded using the overall facial state score, and the state of the facial expression is evaluated according to the grading result, comprising the following steps:

[0031] Setting a facial state threshold, and dividing the overall facial state score into regions by the facial state threshold to obtain a set facial expression level;

[0032] The overall score of the facial state to be evaluated is compared with the set facial expression level to obtain a grading result of the facial expression, and the facial state is evaluated based on the grading result of the facial expression.

[0033] Preferably, using 3D visual perception technology to obtain the three-dimensional structural information of the face of the object being measured in real time, and obtaining a real-time facial image through the three-dimensional structural information, comprises the following steps:

[0034] Paste several reflective markers at various locations on the face of the subject being tested;

[0035] The reflective markers attached to the face of the object being tested are photographed and tracked synchronously through 3D visual perception technology to obtain the three-dimensional structural information of the face of the object being tested;

[0036] A real-time facial image is acquired through the three-dimensional structure information.

[0037] The present invention provides a facial expression evaluation system based on visual perception, comprising:

[0038] A data collection module, used to obtain three-dimensional structural information of the face of the object under test in real time by using 3D visual perception technology, and obtain a real-time facial image through the three-dimensional structural information, wherein the facial image includes reflective marks pre-pasted at various positions on the face of the object under test;

[0039] A capture module is used to obtain the position changes of the reflective marks in the real-time facial images before and after the muscle contraction and in the continuous facial images, and obtain the motion trajectory, speed and occupied area of ​​the facial expression features at the marked points through the position changes of the reflective marks;

[0040] A score acquisition module, used to evaluate the difference of the subject's face before and after muscle contraction through the motion trajectory and speed, and to obtain the percentage of facial expression features at each marked point through the ratio of the area of ​​the affected side to the area of ​​the healthy side;

[0041] An evaluation module is used to obtain an overall score of the facial state by summing up all the percentages, and to grade the facial expressions using the overall score of the facial state, and to evaluate the state of the facial expressions according to the grading results.

[0042] The present invention provides a computer device, comprising a memory and a processor, wherein a program is stored in the memory, and when the program is executed by the processor, the processor executes the steps of the above-mentioned facial expression assessment method based on visual perception.

[0043] The present invention provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned facial expression evaluation method based on visual perception are implemented.

[0044] Compared with the prior art, the present invention has the following significant advantages:

[0045] The present invention records the motion trajectory, speed and occupied area of ​​facial expression features through reflective marks, so as to obtain more comprehensive feature data, determine the difference of the face before and after muscle contraction according to the motion trajectory and speed, and use the occupied area to obtain the percentage of facial expression features in the difference, and make up for the difference defects of speed and motion trajectory through the area occupied by the healthy side and the area occupied by the affected side, thus solving the problem of large errors in facial expression evaluation caused by the difference in speed and motion trajectory, and can obtain more comprehensive facial expression evaluation based on visual perception, thereby improving the evaluation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of the structure of the 3D visual perception technology provided by the present invention for obtaining the three-dimensional structural information of the face of the object under test in real time;

[0047] Figure 2 The present invention is a flowchart of a facial expression evaluation method based on visual perception. DETAILED DESCRIPTION

[0048] The following is a clear and complete description of the technical solutions of the embodiments of the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0049] The purpose of the present invention is to provide a facial expression evaluation method, system, device and storage medium based on visual perception, such as Figure 1 As shown in the figure, 3D visual perception technology is used to obtain the three-dimensional structural information of the patient's face, and computer vision technology is used to automatically identify and analyze facial expressions, extract key features such as the amplitude and speed of muscle movement, and convert them into digital data. Then, artificial intelligence technology is used to intelligently analyze these data, combine historical data and the latest research results, predict the patient's condition, and give an evaluation result. Specifically:

[0050] like Figure 2 As shown, the present invention proposes a facial expression evaluation method based on visual perception to solve the problems of the prior art, which specifically includes the following steps:

[0051] Step S1: Utilize 3D visual perception technology to obtain the three-dimensional structural information of the face of the object under test in real time, and obtain a real-time facial image through the three-dimensional structural information. The facial image includes reflective marks pre-pasted on various positions of the face of the object under test, and the mark point positions of various reflective marks specifically include the forehead, left and right eyes, left and right sides of the nose, both sides of the mouth, and cheeks.

[0052] Specifically:

[0053] Several markers are attached to various locations on the face of the subject to be measured. This allows accurate tracking of facial muscle movements during the capture process. The number and location of the markers should be selected based on specific application requirements.

[0054] Through 3D visual perception technology, the high-brightness reflective markers attached to the surface of the object to be tested are photographed synchronously at high speed, feature tracked, and the image is reconstructed in three dimensions to obtain the three-dimensional structural information of the face of the object to be tested. Real-time facial images are obtained through the three-dimensional structural information.

[0055] This method can accurately capture and analyze the movement trajectory and speed of facial muscles. Specifically, by calculating the displacement, speed, acceleration and other parameters of the marker points, the movement status and trend of facial muscles can be analyzed. These data can provide doctors with in-depth understanding and judgment of the patient's condition.

[0056] Among them, the facial feature point precise positioning technology is:

[0057] Algorithm selection: Common algorithms include the Active Shape Model (ASM) and the Active Appearance Model (AAM). The ASM algorithm builds a shape model by manually marking feature points in training samples, and performs iterative matching in combination with local features during the search process. The AAM algorithm further considers texture information and improves positioning accuracy by simultaneously modeling shape and texture.

[0058] Feature extraction: In the feature extraction stage, feature descriptors such as image gradient features, scale-invariant feature transform (SIFT) or histogram of oriented gradients (HOG) are usually used to capture key information of the face.

[0059] Optimization method: In order to improve the accuracy and robustness of positioning, multi-scale and multi-directional feature extraction and matching methods, as well as optimization strategies based on shape constraints and texture constraints are often used.

[0060] When acquiring facial images, you can also directly use a camera to capture images of the human face from multiple angles, and use multiple images to model the face. That is, use the Holistic method based on global information to focus on the structure and layout of the entire face; the Local feature method based on local information focuses on local features of the face, such as the position and shape of the eyes, nose, mouth, etc.; use a deep neural network to learn and recognize complex facial features, and use these features instead of the features obtained by reflective markers.

[0061] Step S2: Obtain the position change of the reflective mark in the real-time facial image before and after the muscle contraction and in the continuous facial image, and obtain the movement trajectory, speed and occupied area of ​​the facial expression feature at the marked point through the position change of the reflective mark.

[0062] The specific steps include:

[0063] The displacement, velocity, acceleration and area of ​​the marker point are obtained through the three-dimensional reconstructed image, and the displacement, velocity, acceleration and area are converted into digital data.

[0064] Step S3: Evaluate the differences in the face of the subject before and after muscle contraction through the motion trajectory and speed, and obtain the percentage of facial expression features at each marked point through the ratio of the area of ​​the affected side to the area of ​​the healthy side.

[0065] Specifically:

[0066] Obtaining respective recovery percentage scores according to the areas occupied by the expression feature points of various reflective marks when applying force includes the following steps:

[0067] The left and right dividing lines of the forehead are set as the midline of the human body, and the overall contraction area A of the forehead on both sides and the contraction area a of the healthy side are obtained when force is applied, and the contraction rate of the forehead is obtained; the specific expression is:

[0068] Forehead shrinkage rate .

[0069] The percentage of closed eyes is obtained by the leakage area B on the healthy side and B1 on the affected side when the left and right eyes are looking straight ahead when relaxed; the specific expression is:

[0070] Percentage of eyes closed .

[0071] The percentage of forced eye opening is obtained by the size of the bilateral eye area C healthy side and C1 affected side when the eyes are opened forcefully; the specific expression is:

[0072] Percentage of eyes opened .

[0073] Take two points on each side of the nose as the linear distance D healthy side and D1 affected side to measure the amplitude of the nose lifting movement on both sides, and obtain the percentage of nose lifting; the specific expression is:

[0074] Percentage of nose lift .

[0075] The percentage of tooth protrusion is obtained by measuring the size of the bilateral oral tooth leakage area E on the healthy side and E1 on the affected side when the teeth are protruded; the specific expression is:

[0076] Teeth baring percentage ×100%.

[0077] The inflation percentage is obtained by the volume of the cheeks on both sides when inflating hard, F on the healthy side and F1 on the affected side; the specific expression is:

[0078] Air inflation percentage f ×100%.

[0079] The pouting percentage is obtained by the volume of the lips on both sides when pouting hard, G on the healthy side and G1 on the affected side; the specific expression is:

[0080] Pout percentage g= ×100%.

[0081] This step also includes the following steps:

[0082] Based on the convolutional neural network (CNN), facial features are extracted through convolutional layers and pooling layers. Based on the recurrent neural network (RNN), the temporal information in facial expressions is captured through the recurrent layer.

[0083] Data preprocessing:

[0084] Preprocess the data of different modalities, including denoising, normalization, alignment and other operations, to ensure the quality and consistency of the data.

[0085] The facial features and time information of different modalities are fused, and the fused features are used to train the deep learning model, and the facial expressions are recognized by the trained deep learning model. The features of different modalities can be fused by cascade fusion, weight-based fusion, matrix decomposition and other methods. The specific method should be determined according to the characteristics of the data and the application requirements.

[0086] The fused features are used to train a machine learning model or a deep learning model to evaluate and classify facial expressions. During the training process, methods such as cross-validation and regularization can be used to improve the generalization and robustness of the model.

[0087] Training Data:

[0088] Facial expression classification and recognition requires a large amount of training data. This data is usually obtained by collecting facial images with different expressions and manually annotating them. The quality and quantity of training data have a significant impact on the performance of the model.

[0089] Optimization strategy:

[0090] In order to improve the performance of the model, optimization strategies such as data enhancement, transfer learning, and model compression are often used. In addition, multimodal data (such as voice, text, etc.) can also be combined for joint learning and reasoning.

[0091] Specific description of the areas to be assessed:

[0092] Brow lift: Assess the ability to lift the eyebrows, reflecting the function of the frontalis and orbicularis oculi muscles.

[0093] Eye closure: Assessing the ability of the eyelids to close, reflecting the function of the orbicularis oculi muscle and facial nerve.

[0094] Nose lift: Assess the ability of the nose to lift up, reflecting the function of the nasal muscles.

[0095] Teeth protrusion: Assess the upward movement of the corners of the mouth and the protrusion of the teeth, reflecting the function of the muscles at the corners of the mouth and the facial nerves.

[0096] Inflating: Assess the ability to inflate the cheeks, reflecting the function of the buccinator muscles.

[0097] Lip pout: Assess the ability of the lips to protrude, reflecting the function of the perioral muscles.

[0098] Forced eye opening: Assess the ability to open the eyes wide, further reflecting the function of the orbicularis oculi muscle and facial nerve.

[0099] By evaluating these key areas, the system can gain a more comprehensive understanding of the patient's facial expressions, provide doctors with a more accurate basis for diagnosis, and help patients better manage neurological diseases.

[0100] Step S4: Obtain the overall score of the facial state by summing up all percentages, and use the overall score of the facial state to grade the facial expression, and evaluate the state of the facial expression according to the grading result.

[0101] S4 is specifically:

[0102] Get forehead contraction rate a and eye closure percentage , Percentage of eyes opened with force , percentage of nose lift , the sum of the percentage of baring teeth e, the percentage of puffing air f and the percentage of pouting g.

[0103] The overall score of the facial state is obtained by the ratio of the sum of percentages to the number of facial expression features at each landmark.

[0104] The overall score is divided into multiple levels using the area, and the state of facial expression is evaluated by the level, including:

[0105] A facial state threshold is set, and the overall facial state score is divided into regions by the facial state threshold to obtain a set facial expression level.

[0106] The overall score of the facial state to be evaluated is compared with the set facial expression level to obtain the facial expression grading result, and the facial state is evaluated based on the facial expression grading result.

[0107] The percentages of each part are added together, and the higher the score, the better the recovery, with a full score of 100%. The classification is shown in Table 1:

[0108] Table 1 Classification table

[0109]

[0110] Based on the above method, the present invention provides a facial expression evaluation system based on visual perception, including: a data collection module, a capture module, a score acquisition module and an evaluation module.

[0111] Among them, the data collection module is used to use 3D visual perception technology to obtain the three-dimensional structural information of the face of the object under test in real time, and obtain a real-time facial image through the three-dimensional structural information, and the facial image contains reflective marks pre-pasted at various positions on the face of the object under test; the capture module is used to obtain the position changes of reflective marks in the real-time facial image before and after muscle contraction and in continuous facial images, and obtain the motion trajectory, speed and area occupied by the facial expression characteristics at the marked point through the position changes of the reflective marks; the score acquisition module is used to evaluate the differences in the face of the object under test before and after muscle contraction through the motion trajectory and speed, and obtain the percentage of facial expression characteristics at each marked point through the ratio of the area of ​​the affected side to the area of ​​the healthy side; the evaluation module is used to obtain the overall score of the facial state through the sum of all percentages, and use the overall score of the facial state to grade the facial expression, and evaluate the state of the facial expression through the grading results.

[0112] The present invention also provides a computer device, including a memory and a processor. The memory stores a program, and when the program is executed by the processor, the processor executes the steps of a facial expression assessment method based on visual perception.

[0113] According to the disclosed embodiments, a computing device may communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth communications, etc.), or with any device (e.g., routers, modems, etc.) that enables a computing device to communicate with one or more other computing devices.

[0114] The present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, a facial expression evaluation method based on visual perception is implemented.

[0115] According to the disclosed embodiments, the storage medium may be a non-volatile computer-readable storage medium, such as but not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, the storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0116] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0117] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A facial expression assessment method based on visual perception, characterized in that: The steps include: Using 3D visual perception technology to obtain three-dimensional structural information of the face of the object being measured in real time, and obtaining a real-time facial image through the three-dimensional structural information, wherein the facial image includes reflective marks pre-pasted at various positions on the face of the object being measured; Obtain the position changes of the reflective marks in the real-time facial images before and after the muscle contraction and in the continuous facial images, and obtain the motion trajectory, speed and occupied area of ​​the facial expression features at the marked points through the position changes of the reflective marks; The difference between the face of the tested subject before and after the muscle contraction is evaluated by the motion trajectory and speed, and the percentage of the facial expression feature at each marked point is obtained by the ratio of the area of ​​the affected side to the area of ​​the healthy side; Obtaining an overall facial state score by summing up all the percentages, and grading facial expressions using the overall facial state score, and evaluating the state of the facial expressions according to the grading results; The percentage of facial expression features at each marking point is obtained by the ratio of the affected side area to the healthy side area, including the following steps: The left and right dividing line of the forehead is set as the midline of the human body, and the forehead contraction rate a is obtained by the overall contraction area A of the bilateral foreheads and the contraction area A1 of the healthy side when exerting force; the specific expression is: Forehead shrinkage rate a=2A1÷A×100%; The percentage of closed eyes b is obtained by the healthy side leakage area B and the affected side leakage area B1 when the left and right eyes are looking straight ahead when relaxed; the specific expression is: Eye closure percentage b = (B-B1) ÷ B × 100%; The forced eye opening percentage c is obtained by using the healthy eye area C and the affected eye area C1 when the eyes are opened forcefully; the specific expression is: Percentage of forced eye opening c = C1 ÷ C × 100%; Take two points on each side of the nose as the straight-line distance D on the healthy side and the straight-line distance D1 on the affected side to measure the amplitude of the nose-lifting movement on the left and right sides, and obtain the nose-lifting percentage d; the specific expression is: The percentage of nose lift d = D1 ÷ D × 100%; The percentage of tooth protrusion e is obtained by comparing the tooth leakage area E of the healthy side and the tooth leakage area E1 of the affected side when protruding teeth with force; the specific expression is: The percentage of tooth baring is e=E1÷E×100%; The inflation percentage f is obtained by the healthy cheek inflation volume F and the affected cheek inflation volume F1 when inflating vigorously; the specific expression is: Air inflation percentage f = F1 ÷ F × 100%; The pouting percentage g is obtained by the bulging volume G of the healthy side lips and the bulging volume G1 of the affected side lips when pouting hard; The specific expression is: Pouting percentage g=G1÷G×100%.

2. A facial expression assessment method based on visual perception as claimed in claim 1, characterized in that: Obtaining the overall facial status score by summing up all the percentages includes the following steps: Obtain the sum of the forehead contraction rate a, eye closing percentage b, eye force opening percentage c, nose lifting percentage d, teeth baring percentage e, air blowing percentage f, and mouth pouting percentage g; The overall score of the facial state is obtained by the ratio of the sum of the percentages to the number of facial expression features at each marking point.

3. A facial expression assessment method based on visual perception as claimed in claim 1, characterized in that: The facial expression is graded using the overall facial state score, and the state of the facial expression is evaluated according to the grading result, including: Setting a facial state threshold, and dividing the overall facial state score into regions by the facial state threshold to obtain a set facial expression level; The overall score of the facial state to be evaluated is compared with the set facial expression level to obtain a grading result of the facial expression, and the facial state is evaluated based on the grading result of the facial expression.

4. A facial expression assessment method based on visual perception as claimed in claim 1, characterized in that: Using 3D visual perception technology to obtain the three-dimensional structural information of the face of the object being measured in real time, and obtaining a real-time facial image through the three-dimensional structural information, includes the following steps: Paste several reflective markers at various locations on the face of the subject being tested; The reflective markers attached to the face of the object being tested are photographed and tracked synchronously through 3D visual perception technology to obtain the three-dimensional structural information of the face of the object being tested; A real-time facial image is acquired through the three-dimensional structure information.

5. A facial expression evaluation system based on visual perception, characterized in that: include: A data collection module, used to obtain three-dimensional structural information of the face of the object under test in real time by using 3D visual perception technology, and obtain a real-time facial image through the three-dimensional structural information, wherein the facial image includes reflective marks pre-pasted at various positions on the face of the object under test; A capture module is used to obtain the position changes of the reflective marks in the real-time facial images before and after the muscle contraction and in the continuous facial images, and obtain the motion trajectory, speed and occupied area of ​​the facial expression features at the marked points through the position changes of the reflective marks; A score acquisition module, used to evaluate the difference of the subject's face before and after muscle contraction through the motion trajectory and speed, and to obtain the percentage of facial expression features at each marked point through the ratio of the area of ​​the affected side to the area of ​​the healthy side; An evaluation module, used to obtain an overall score of the facial state by summing up all the percentages, and to grade the facial expressions using the overall score of the facial state, and to evaluate the state of the facial expressions according to the grading results; The percentage of facial expression features at each marking point is obtained by the ratio of the affected side area to the healthy side area, including the following steps: The left and right dividing line of the forehead is set as the midline of the human body, and the forehead contraction rate a is obtained by the overall contraction area A of the bilateral foreheads and the contraction area A1 of the healthy side when exerting force; the specific expression is: Forehead shrinkage rate a=2A1÷A×100%; The percentage of closed eyes b is obtained by the healthy side leakage area B and the affected side leakage area B1 when the left and right eyes are looking straight ahead when relaxed; the specific expression is: Eye closure percentage b = (B-B1) ÷ B × 100%; The forced eye opening percentage c is obtained by using the healthy eye area C and the affected eye area C1 when the eyes are opened forcefully; the specific expression is: Percentage of forced eye opening c = C1 ÷ C × 100%; Take two points on each side of the nose as the straight-line distance D on the healthy side and the straight-line distance D1 on the affected side to measure the amplitude of the nose-lifting movement on the left and right sides, and obtain the nose-lifting percentage d; the specific expression is: The percentage of nose lift d = D1 ÷ D × 100%; The percentage of tooth protrusion e is obtained by comparing the tooth leakage area E of the healthy side and the tooth leakage area E1 of the affected side when protruding teeth with force; the specific expression is: The percentage of tooth baring is e=E1÷E×100%; The inflation percentage f is obtained by the healthy cheek inflation volume F and the affected cheek inflation volume F1 when inflating vigorously; the specific expression is: Air inflation percentage f = F1 ÷ F × 100%; The pouting percentage g is obtained by the bulging volume G of the healthy side lips and the bulging volume G1 of the affected side lips when pouting hard; The specific expression is: Pouting percentage g=G1÷G×100%.

6. A computer device, characterized in that: It comprises a memory and a processor, wherein a program is stored in the memory, and when the program is executed by the processor, the processor executes the steps of a facial expression assessment method based on visual perception as claimed in any one of claims 1 to 4.

7. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a facial expression assessment method based on visual perception according to any one of claims 1 to 4 are implemented.

Images