Method, device and equipment for testing and evaluating driver fatigue state monitoring system
By combining subjective and objective detection methods, creating test scenarios, acquiring data, determining feature indicators, and establishing and training detection models, the problems of individual differences and high costs in driver fatigue detection in existing technologies are solved, achieving non-invasive and accurate fatigue state monitoring.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-23
- Publication Date
- 2026-03-31
AI Technical Summary
Existing driver fatigue detection technologies suffer from several problems. Subjective detection methods are prone to deliberate concealment, cognitive bias, individual differences, and poor timeliness, while objective detection methods are affected by individual differences, invasiveness, and lighting conditions, and the equipment costs are high.
By combining subjective and objective detection methods, test scenarios are created to obtain test data, determine characteristic indicators for driver fatigue detection, establish and train a driver fatigue detection model, and perform prediction and optimization, thereby achieving non-invasive and low-cost testing and evaluation.
The test of a non-invasive, low-cost driver fatigue monitoring system was successfully completed, and the test results were accurate, thus improving driving safety.
Smart Images
Figure CN121764802A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of vehicle testing technology, specifically to a test and evaluation method, device, and equipment for a driver fatigue monitoring system. Background Technology
[0002] Currently, there are numerous driver fatigue detection technologies, mainly divided into two categories: subjective detection methods and objective detection methods. Subjective detection methods primarily rely on driver self-assessment and the Pearson Fatigue Rating Scale. Objective detection methods mainly focus on three aspects: detection based on driver physiological signals, detection based on vehicle motion information, and detection based on facial feature information.
[0003] However, existing driver fatigue detection technologies have the following problems: 1. Subjective detection methods are prone to problems such as deliberate concealment, cognitive bias, catering to expectations, individual differences, and poor timeliness of detection results; 2. While objective detection methods based on driver physiological signals have high accuracy and reliability, they are accompanied by strong individual differences and invasive detection problems, and the detection equipment is expensive. Detection based on vehicle motion information is non-invasive and has good timeliness, but it is greatly affected by the driver's operation and driving method. Detection based on facial feature information is non-invasive, but changes in driving scenarios and different lighting conditions will seriously affect its detection accuracy. Summary of the Invention
[0004] This application provides a test and evaluation method, device, and equipment for a driver fatigue monitoring system. It uses a combination of subjective and objective testing to test and evaluate the driver fatigue monitoring system, which is low-cost, non-invasive, and provides accurate test results.
[0005] In a first aspect, embodiments of this application provide a method for testing and evaluating a driver fatigue monitoring system, the method comprising: Create test scenarios to test the vehicle's driver fatigue monitoring system, and obtain fatigue test results and test data information during the test process; Based on the test data, determine the characteristic indicators for driver fatigue detection used to evaluate the performance of the driver fatigue monitoring system; A driving fatigue detection model is established and trained based on the driving fatigue detection feature indicators. The trained driving fatigue detection model is then used to predict the fatigue state in the test scenario. By comparing the prediction results of the driver fatigue detection model with the fatigue test results, the driver fatigue state monitoring system can be evaluated and optimized.
[0006] In conjunction with the first aspect, in one implementation method, The test data includes driver status information, vehicle operating status information, and subjective evaluation information of driver fatigue. The vehicle driver status information is the driver's facial appearance status information; The vehicle operating status information includes steering wheel angle, vehicle speed, vehicle lateral deviation speed, and vehicle lateral position offset. The subjective evaluation information on driver fatigue is based on the Karolinska Sleepiness Scale, which is a subjective scoring evaluation of the driver's fatigue level in the test scenario.
[0007] In conjunction with the first aspect, in one implementation, determining the driver fatigue detection characteristic indicators for evaluating the performance of the driver fatigue monitoring system based on the test data information specifically includes: The test data is preprocessed, and features are extracted from the preprocessed driver status information and vehicle operation status information to obtain driver status feature data and vehicle operation status feature data. Based on the changes in driver state characteristic data and vehicle operation state characteristic data with subjective evaluation information of driver fatigue, characteristic data that reflect different levels of driver fatigue are selected as driving fatigue detection characteristic indicators for evaluating the performance of the driver fatigue state monitoring system.
[0008] In conjunction with the first aspect, in one implementation method, For feature extraction of vehicle driver status information, edge contour method or facial behavior analysis tool method are used; The edge contour method involves extracting the edge contour of the driver's facial image to obtain the driver's facial contour, performing partitioned projection on the facial contour, locating the positions of the eyes and mouth, extracting feature point data of the eyes and mouth, and obtaining driver state feature data, which includes blinking frequency, eyelid opening, head angle, and mouth corner opening. The facial behavior analysis tool is used to extract multiple facial features of the driver to obtain facial feature information, and then obtain driver state feature data based on the facial feature information.
[0009] In conjunction with the first aspect, in one implementation method, feature extraction of vehicle operating status information specifically includes: The vehicle operating status information is processed to obtain vehicle operating status feature data, which includes the mean and variance of vehicle speed, the mean and variance of steering wheel angular velocity, steering wheel angular velocity, vehicle lateral deviation speed, and vehicle lateral position deviation.
[0010] In conjunction with the first aspect, in one implementation, the step of establishing and training a driving fatigue detection model based on the driving fatigue detection feature indicators, and using the trained driving fatigue detection model to predict the fatigue state of the test scenario, specifically includes: Based on the characteristic indicators of driver fatigue detection, multiple driver fatigue detection models were established, and the established driver fatigue detection models were trained and validated. Based on the verification results, the driving fatigue detection model with the highest accuracy was improved and optimized to obtain the optimal driving fatigue detection model. The optimal driving fatigue detection model is used to predict the driver fatigue state in the test scenario.
[0011] In conjunction with the first aspect, in one implementation, the improvement and optimization of the driving fatigue detection model with the highest accuracy is specifically achieved by adjusting the model's loss function or network architecture.
[0012] In conjunction with the first aspect, in one implementation, the evaluation and optimization of the driver fatigue monitoring system based on the comparison between the prediction results of the driver fatigue detection model and the fatigue test results specifically includes: The prediction results of the driver fatigue detection model are compared with the fatigue test results: If the two are consistent, no action is taken; If the two are inconsistent, the driver fatigue monitoring system of the vehicle will be optimized by combining the subjective evaluation information of driver fatigue.
[0013] Secondly, embodiments of this application provide a test and evaluation device for a driver fatigue monitoring system, the driver fatigue monitoring system test and evaluation device comprising: The testing module is used to create test scenarios to test the vehicle's driver fatigue monitoring system, obtain fatigue test results and test data information during the test process; The determination module is used to determine the driving fatigue detection characteristic indicators for evaluating the performance of the driver fatigue state monitoring system based on the test data information. The prediction module is used to establish and train a driving fatigue detection model based on the driving fatigue detection feature indicators, and to use the trained driving fatigue detection model to predict the fatigue state of the test scenario. The comparison module is used to evaluate and optimize the driver fatigue monitoring system by comparing the prediction results of the driver fatigue detection model with the fatigue test results.
[0014] Thirdly, embodiments of this application provide a driver fatigue monitoring system testing and evaluation device, which includes a processor, a memory, and a driver fatigue monitoring system testing and evaluation program stored in the memory and executable by the processor. When the driver fatigue monitoring system testing and evaluation program is executed by the processor, it implements the steps of the driver fatigue monitoring system testing and evaluation method described above.
[0015] The beneficial effects of the technical solutions provided in this application include: By creating test scenarios to test a vehicle's driver fatigue monitoring system, fatigue test results and test data information during the testing process are obtained. Then, based on the test data information, driver fatigue detection feature indicators are determined to evaluate the performance of the driver fatigue monitoring system. A driver fatigue detection model is then established and trained based on the driver fatigue detection feature indicators. The trained driver fatigue detection model is used to predict the fatigue state in the test scenarios. Finally, by comparing the prediction results of the driver fatigue detection model with the fatigue test results, the driver fatigue monitoring system is evaluated and optimized. The driver fatigue monitoring system is tested and evaluated using a combination of subjective and objective detection methods, which is low-cost, non-invasive, and provides accurate test results. Attached Figure Description
[0016] Figure 1 This is a flowchart illustrating the test and evaluation method for the driver fatigue monitoring system of this application; Figure 2 This is a schematic diagram of the functional modules of the driver fatigue monitoring system test and evaluation device of this application; Figure 3 This is a schematic diagram of the hardware structure of the driver fatigue monitoring system test and evaluation equipment in this application. Detailed Implementation
[0017] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.
[0018] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.
[0019] In a first aspect, embodiments of this application provide a test and evaluation method for a driver fatigue monitoring system.
[0020] In one embodiment, reference is made to Figure 1 , Figure 1 This is a flowchart illustrating the testing and evaluation method for the driver fatigue monitoring system described in this application. Figure 1 As shown, the testing and evaluation methods for driver fatigue monitoring systems include: S1: Create a test scenario to test the vehicle's driver fatigue monitoring system, obtain fatigue test results and test data information during the test process; that is, select a suitable driver and test environment, perform performance testing on the vehicle's driver fatigue monitoring system, obtain the identified fatigue test results, and test data information during the test process. S2: Based on the test data information, determine the characteristic indicators for driver fatigue detection used to evaluate the performance of the driver fatigue monitoring system; S3: Establish and train a driving fatigue detection model based on the driving fatigue detection feature indicators, and use the trained driving fatigue detection model to predict the fatigue state of the test scenario. S4: By comparing the prediction results of the driver fatigue detection model with the fatigue test results, the driver fatigue monitoring system is evaluated and optimized. Specifically, by comparing the prediction results of the driver fatigue detection model with the fatigue test results, the performance of the driver fatigue monitoring system is evaluated. Based on the evaluation results, reasonable suggestions are proposed to improve the performance of the driver fatigue monitoring system, thereby enhancing driving safety.
[0021] Furthermore, in one embodiment, the test data information includes vehicle driver status information, vehicle operating status information, and subjective evaluation information of driver fatigue.
[0022] The vehicle driver status information includes the driver's facial features, such as the head, eyes, and mouth. This facial information can be recorded and collected via a camera.
[0023] Vehicle operating status information includes steering wheel angle, vehicle speed, vehicle lateral deviation speed, and vehicle lateral position offset. This information can be measured and recorded using steering wheel sensors, inertial navigation systems, and data acquisition equipment.
[0024] The subjective evaluation information of driver fatigue is based on the Karolinska Sleepiness Scale, which is a subjective scoring evaluation of the driver's fatigue level in the test scenario. Drivers trained in KSS (Subjective Evaluation of Driver Fatigue Information) report their current perceived driver fatigue. The data is recorded by test staff at 5-minute intervals, or as needed. During the test, test staff collect and record the data while the driver drives naturally until they report being unable to continue driving.
[0025] Furthermore, in one embodiment, the driver fatigue detection characteristic indicators used to evaluate the performance of the driver fatigue monitoring system are determined based on the test data information, specifically including: S201: Preprocess the test data information, and extract features from the preprocessed vehicle driver status information and vehicle operation status information to obtain driver status feature data and vehicle operation status feature data. S202: Based on the changes in driver state characteristic data and vehicle operation state characteristic data with subjective evaluation information of driver fatigue, select characteristic data that reflect different levels of driver fatigue, and use them as characteristic indicators for evaluating the performance of the driver fatigue state monitoring system.
[0026] Specifically, the acquired test data undergoes data cleaning and preprocessing. Then, features are extracted from the preprocessed test data regarding the driver's state and vehicle operating status, yielding feature data. Next, by analyzing the changes in these feature data with subjective evaluations of driver fatigue, appropriate feature data are selected as driver fatigue detection indicators to evaluate the performance of the driver fatigue monitoring system. For example, blink frequency, eyelid opening, and head angle are chosen as driver fatigue detection indicators. Subsequently, the driver fatigue detection model automatically identifies the driver's fatigue state based on the actual data corresponding to the input driver fatigue detection indicator.
[0027] In this application, the feature extraction of the vehicle driver's state information is performed using either edge contouring or facial behavior analysis tools. The edge contour method involves extracting the edge contour of the driver's facial image to obtain the driver's facial contour, performing partitioned projection on the facial contour, locating the positions of the eyes and mouth, extracting feature point data of the eyes and mouth, and obtaining driver state feature data. The driver state feature data includes blinking frequency, eyelid opening (eyelid aspect ratio), head angle, and mouth corner opening (mouth aspect ratio).
[0028] The facial behavior analysis tool extracts multiple facial features of the driver to obtain facial feature information, and then obtains driver state feature data based on the facial feature information.
[0029] In this application, the feature extraction of vehicle operating status information specifically includes: The vehicle operating status information is processed to obtain vehicle operating status feature data, which includes the mean and variance of vehicle speed, the mean and variance of steering wheel angular velocity, steering wheel angular velocity, vehicle lateral deviation speed, and vehicle lateral position deviation.
[0030] Furthermore, in one embodiment, a driving fatigue detection model is established and trained based on the driving fatigue detection feature indicators. The trained driving fatigue detection model is then used to predict the fatigue state in the test scenario, specifically including: S301: Based on the characteristic indicators of driver fatigue detection, establish multiple driver fatigue detection models, and train and validate the established driver fatigue detection models; S302: Based on the verification results, the driving fatigue detection model with the highest accuracy is improved and optimized to obtain the optimal driving fatigue detection model; S303: Use the optimal driver fatigue detection model to predict the driver fatigue state in the test scenario.
[0031] Specifically, a driver fatigue detection model is established based on the characteristic indicators of driver fatigue detection. The established driver fatigue detection model can predict the driver's fatigue state based on the actual data corresponding to the input driver fatigue detection characteristic indicators. The established driver fatigue detection model is trained and validated, and the driver fatigue detection model with the highest accuracy is selected. The model is then improved and optimized to obtain the optimal driver fatigue detection model.
[0032] This application improves and optimizes the driver fatigue detection model with the highest accuracy. Specifically, the improvement and optimization involves adjusting the model's loss function or network architecture. That is, based on the model's detection results, and considering both detection accuracy and speed, the application improves both by designing a better model loss function or by refining the model's network structure, thereby reducing false positives and false negatives.
[0033] Furthermore, in one embodiment, the driver fatigue state monitoring system is evaluated and optimized by comparing the prediction results of the driver fatigue detection model with the fatigue test results, specifically including: The prediction results of the driver fatigue detection model are compared with the fatigue test results: If the two are consistent, no action is taken; If the two are inconsistent, the driver fatigue monitoring system should be optimized by combining the subjective evaluation information of driver fatigue, and reasonable suggestions should be made for improving the performance of the driver fatigue monitoring system.
[0034] Specifically, when making rationalization suggestions, based on the prediction results of the driver fatigue detection model and the monitoring results of the driver fatigue state monitoring system, and combined with the driver's own feelings, when it is found that the driver fatigue state monitoring system falsely detects fatigue, the system can be improved (such as adjusting the monitoring model or improving the fatigue monitoring threshold) to reduce false detections. When it is found that the system misses detections, the amount of training data input can be increased, and different forms of driver fatigue data can be added to train the system and reduce missed detections.
[0035] The driver fatigue monitoring system testing and evaluation method of this application embodiment creates a test scenario to test the vehicle's driver fatigue monitoring system, obtains fatigue test results and test data information during the test process, then determines driver fatigue detection feature indicators for evaluating the performance of the driver fatigue monitoring system based on the test data information, then establishes and trains a driver fatigue detection model based on the driver fatigue detection feature indicators, uses the trained driver fatigue detection model to predict the fatigue state of the test scenario, and then compares the prediction results of the driver fatigue detection model with the fatigue test results to achieve the evaluation and optimization of the driver fatigue monitoring system. This method uses a combination of subjective and objective detection to test and evaluate the driver fatigue monitoring system, which is low-cost, non-invasive, and provides accurate test results.
[0036] Secondly, embodiments of this application also provide a test and evaluation device for a driver fatigue monitoring system.
[0037] In one embodiment, reference is made to Figure 2 , Figure 2 This is a schematic diagram of the functional modules of the driver fatigue monitoring system test and evaluation device of this application. Figure 2 As shown, the driver fatigue monitoring system test and evaluation device includes: a test module, a determination module, a prediction module, and a comparison module.
[0038] The testing module is used to create test scenarios to test the driver fatigue monitoring system of a vehicle, obtain fatigue test results and test data information during the test process; the determination module is used to determine the driving fatigue detection feature indicators for evaluating the performance of the driver fatigue monitoring system based on the test data information; the prediction module is used to establish and train a driving fatigue detection model based on the driving fatigue detection feature indicators, and use the trained driving fatigue detection model to predict the fatigue state of the test scenario; the comparison module is used to evaluate and optimize the driver fatigue monitoring system by comparing the prediction results of the driving fatigue detection model with the fatigue test results.
[0039] Thirdly, embodiments of this application provide a test and evaluation device for a driver fatigue monitoring system. The test and evaluation device for a driver fatigue monitoring system can be a personal computer (PC), a laptop computer, a server, or other device with data processing capabilities.
[0040] Reference Figure 3 , Figure 3 This is a schematic diagram of the hardware structure of the driver fatigue monitoring system testing and evaluation equipment involved in the embodiments of this application. In this embodiment, the driver fatigue monitoring system testing and evaluation equipment may include a processor, a memory, a communication interface, and a communication bus.
[0041] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.
[0042] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting internal components of the driver fatigue monitoring system test and evaluation equipment, as well as interfaces used for interconnecting the driver fatigue monitoring system test and evaluation equipment with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.
[0043] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.
[0044] The processor can be a general-purpose processor, which can call the driver fatigue monitoring system test and evaluation program stored in the memory and execute the driver fatigue monitoring system test and evaluation method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the driver fatigue monitoring system test and evaluation program is called can refer to the various embodiments of the driver fatigue monitoring system test and evaluation method of this application, and will not be repeated here.
[0045] Those skilled in the art will understand that Figure 3 The hardware structure shown does not constitute a limitation of this application and may include more or fewer components than shown, or combine certain components, or have different component arrangements.
[0046] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.
[0047] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.
[0048] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.
[0049] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.
[0050] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.
[0051] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. A method of testing and evaluating a driver fatigue state monitoring system, characterized by, The driver fatigue state monitoring system test evaluation method comprises: A test scene is created to test the driver fatigue state monitoring system of a vehicle, to obtain fatigue test results and test data information during the test; Driver fatigue detection feature indexes for evaluating the performance of the driver fatigue state monitoring system are determined based on the test data information; A driver fatigue detection model is established and trained according to the driver fatigue detection feature indexes, and the trained driver fatigue detection model is used to predict the fatigue state of the test scene; The driver fatigue state monitoring system is evaluated and optimized according to the comparison between the prediction results of the driver fatigue detection model and the fatigue test results.
2. The driver fatigue state monitoring system test evaluation method according to claim 1, wherein: The test data information comprises vehicle driver state information, vehicle operating state information, and driver fatigue subjective evaluation information; The vehicle driver state information is driver facial state information; The vehicle operating state information comprises steering wheel angle, vehicle speed, vehicle lateral deviation speed, and vehicle lateral position deviation; The driver fatigue subjective evaluation information is subjective scoring and evaluation information of the fatigue state of the vehicle driver in the test scene based on the Karolinska Sleepiness Scale.
3. A test evaluation method of a driver fatigue state monitoring system according to claim 2, characterized in that, The driver fatigue detection feature indexes for evaluating the performance of the driver fatigue state monitoring system are determined based on the test data information, specifically comprising: The test data information is preprocessed, and the vehicle driver state information and the vehicle operating state information after preprocessing are subjected to feature extraction, to obtain driver state feature data and vehicle operating state feature data; According to the changes of the driver state feature data and the vehicle operating state feature data with the driver fatigue subjective evaluation information, feature data for reflecting different fatigue degrees of the driver are screened out as the driver fatigue detection feature indexes for evaluating the performance of the driver fatigue state monitoring system.
4. The driver fatigue state monitoring system test evaluation method according to claim 3, wherein: For feature extraction of the vehicle driver state information, an edge contour method or a facial behavior analysis tool method is used; The edge contour method comprises extracting the edge contour of the driver's face image to obtain the facial contour of the driver, performing zoned projection on the facial contour, positioning the eyes and mouth, extracting feature point data of the eyes and mouth, obtaining driver state feature data, and the driver state feature data comprises blink frequency, eyelid opening degree, head angle, and mouth opening degree; The facial behavior analysis tool method comprises extracting multiple facial features of the driver by a facial behavior analysis tool to obtain facial feature information, and obtaining driver state feature data according to the facial feature information.
5. The method of claim 3, wherein the method further comprises: determining a fatigue level of the driver based on the determined fatigue level of the driver and the determined fatigue level of the driver in the previous test; and providing a feedback to the driver based on the determined fatigue level of the driver. For feature extraction of the vehicle operating state information, specifically comprising: The vehicle operating state information is processed to obtain vehicle operating state feature data, and the vehicle operating state feature data comprises vehicle speed mean and variance, steering wheel angular velocity mean and variance, steering wheel angular velocity, vehicle lateral deviation speed, and vehicle lateral position deviation.
6. The method of claim 1, wherein the method further comprises: determining a fatigue level of the driver based on the determined eye movement of the driver; and providing a feedback to the driver based on the determined fatigue level of the driver. The driving fatigue detection model is established and trained according to the driving fatigue detection feature index, and the driving fatigue detection model after training is used to predict the fatigue state of the test scene, and specifically includes: Based on the driving fatigue detection feature index, a plurality of driving fatigue detection models are established, and the established driving fatigue detection models are trained and verified; According to the verification result, the driving fatigue detection model with the highest accuracy is improved and optimized to obtain the optimal driving fatigue detection model; The optimal driving fatigue detection model is used to predict the fatigue state of the test scene.
7. A method of testing and evaluating a driver fatigue monitoring system as claimed in claim 6, characterised in that: The driving fatigue detection model with the highest accuracy is improved and optimized, and for the improvement and optimization, the loss function or network architecture of the model is adjusted.
8. The test and evaluation method for a driver fatigue monitoring system as described in claim 1, characterized in that, The driving fatigue detection model prediction result is compared with the fatigue test result to realize the evaluation and optimization of the driver fatigue state monitoring system, and specifically includes: The driving fatigue detection model prediction result is compared with the fatigue test result: If they are consistent, no processing is performed; If they are inconsistent, the driver fatigue state monitoring system of the vehicle is optimized in combination with the subjective evaluation information of the driving fatigue.
9. A driver fatigue state monitoring system test evaluation device characterized by comprising: The driver fatigue state monitoring system test evaluation device includes: A test module for creating a test scene to test the driver fatigue state monitoring system of the vehicle to obtain fatigue test results and test data information during the test process; A determination module for determining a driving fatigue detection feature index for evaluating the performance of the driver fatigue state monitoring system based on the test data information; A prediction module for establishing and training a driving fatigue detection model according to the driving fatigue detection feature index, and using the trained driving fatigue detection model to predict the fatigue state of the test scene; A comparison module for comparing the driving fatigue detection model prediction result with the fatigue test result to realize the evaluation and optimization of the driver fatigue state monitoring system.
10. A driver fatigue state monitoring system test evaluation apparatus characterized by comprising: The driver fatigue state monitoring system test evaluation device includes a processor, a memory, and a driver fatigue state monitoring system test evaluation program stored on the memory and executable by the processor, wherein the driver fatigue state monitoring system test evaluation program is executed by the processor to realize the steps of the driver fatigue state monitoring system test evaluation method in any one of claims 1 to 8.