Dummy chest injury prediction method, device, equipment and storage medium

Abstract

This application relates to a method, device, equipment, and storage medium for predicting chest injuries in dummy vehicles, specifically in the field of automobile collision dummy testing. The method comprises: obtaining collision test data and the collision velocity corresponding to the collision test data; wherein the collision test data is the rib displacement data of the dummy after a collision test at the collision velocity; performing an injury degree analysis on the collision test data and the collision velocity based on an injury prediction model to obtain a target injury degree; and determining a predicted chest injury value for the dummy based on the dummy's preset age and target injury degree. This method enables accurate prediction of chest injury conditions.

Description

Technical Field

[0001] The present application relates to the field of automobile collision dummy testing, and in particular to a dummy chest injury prediction method, apparatus, device and storage medium. Background Art

[0002] The vehicle's occupant protection performance is one of the indicators for evaluating vehicle performance, and the occupant's chest injury is one of the indicators for evaluating the occupant's injury situation.

[0003] At present, the damage value of the chest of a car collision is mainly obtained through real-car collision tests or simulation tests. The chest structure of the THOR dummy used in the collision test has seven ribs on each side. However, due to the limitation of the internal space of the THOR dummy's chest cavity, it is impossible to install a displacement sensor on each rib. Displacement sensors are only installed on the third rib on the left, the third rib on the right, the sixth rib on the left, and the sixth rib on the right. During the collision test, only the rib displacement curves of the third rib on the left, the third rib on the right, the sixth rib on the left, and the sixth rib on the right of the THOR dummy's chest can be collected. The displacement deformation of the other 10 ribs of the THOR dummy's chest cannot be collected. In addition, the currently collected THOR dummy chest rib displacement curves are the displacement changes of the ribs on the chest of an adult male dummy.

[0004] Therefore, the current collision test cannot accurately reflect the chest and rib injury status of the THOR dummy. Summary of the Invention

[0005] The present application provides a dummy chest injury prediction method, apparatus, device and storage medium, which can accurately predict the chest rib injury status.

[0006] To achieve the above objectives, this application adopts the following technical solutions:

[0007] In a first aspect, the present application provides a method for predicting dummy chest injuries, comprising:

[0008] Obtaining collision test data and a collision speed corresponding to the collision test data; wherein the collision test data is rib displacement data of a dummy after a collision test at the collision speed;

[0009] Based on the damage prediction model, the damage degree is analyzed on the collision test data and collision speed to obtain the target damage degree;

[0010] Based on the preset age and target injury level of the dummy, the predicted chest injury value of the dummy is determined.

[0011] In one embodiment, determining a chest injury prediction value corresponding to the dummy based on a preset age and target injury degree corresponding to the dummy includes:

[0012] The preset age and target injury severity of the dummy are input into the accelerated failure time model to determine the predicted chest injury value of the dummy; wherein the accelerated failure time model follows a Weibull distribution.

[0013] In one embodiment, based on the damage prediction model, a damage degree analysis is performed on the collision test data and the collision speed to obtain a target damage degree, including:

[0014] The collision test data and collision velocity are input into the damage prediction model to obtain first prediction data; wherein the first prediction data includes predicted rib displacement deformation curves of multiple groups of ribs; based on the first prediction data, feature decomposition is performed to obtain weights corresponding to the first prediction data; based on the first prediction data and the weights, a target damage degree is obtained.

[0015] In one embodiment, obtaining a target damage degree based on the first prediction data and the weight includes:

[0016] Based on the first prediction data and the weight, a first damage degree and a second damage degree are determined; and the sum of the first damage degree and the second damage degree is determined as the target damage degree.

[0017] In one embodiment, determining the first damage degree and the second damage degree based on the first prediction data and the weight includes:

[0018] Based on the first prediction data, the sum curve of rib displacement deformation, the difference curve of rib displacement deformation, the standard deviation of the sum curve of rib displacement deformation and the standard deviation of the difference curve of rib displacement deformation are determined; based on the sum curve, the difference curve, the standard deviation of the sum curve, the standard deviation of the difference curve and the weight, the first damage degree and the second damage degree are determined.

[0019] In one embodiment, the training process of the damage prediction model includes:

[0020] Build Model; Based on the simulation training set, The model is trained to obtain a damage prediction model; wherein the simulation training set includes the rib displacement deformation simulation curves of the seven ribs on the left side of the chest and the simulation curves of the rib displacement deformation of the seven ribs on the right side of the chest.

[0021] In one embodiment, after obtaining the damage prediction model, the following steps are included:

[0022] The damage prediction model is verified based on the image data corresponding to the collision test to obtain a damage prediction model with a preset accuracy.

[0023] In a second aspect, the present application provides a dummy chest injury prediction device, comprising:

[0024] An acquisition module is used to acquire collision test data and a collision velocity corresponding to the collision test data; wherein the collision test data is rib displacement data of a dummy after a collision test at the collision velocity;

[0025] The analysis module is used to analyze the damage degree of the collision test data and the collision speed based on the damage prediction model to obtain the target damage degree;

[0026] The prediction module is used to determine the chest injury prediction value corresponding to the dummy based on the preset age and target injury degree corresponding to the dummy.

[0027] In a third aspect, the present application provides a computing device, including a memory and a processor;

[0028] One or more computer programs are stored in the memory, and the one or more computer programs include instructions; when the instructions are executed by the processor, the computing device executes the method as described in any one of the first aspects.

[0029] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program for executing the method as described in any one of the first aspects.

[0030] In a fifth aspect, the present application provides a computer program product, which includes one or more computer instructions. When the computer instructions are executed by a computer, the computer executes the method as described in any one of the first aspects.

[0031] It can be seen from the above technical solution that this application has at least the following beneficial effects:

[0032] In this application, by obtaining collision test data and the collision speed corresponding to the collision test data, a data basis is provided for predicting the chest injury prediction value; further, based on the injury prediction model, the collision test data and collision speed can be analyzed for the degree of injury to obtain the target injury degree, providing a basis for evaluating the degree of injury; furthermore, based on the preset age and target injury degree corresponding to the dummy, the chest injury prediction value corresponding to the dummy can be determined, ultimately achieving an accurate prediction of the chest injury status. By introducing an injury prediction model, this solution provides a way to accurately determine the target injury degree; furthermore, by introducing a preset age, it takes into account all age groups, ultimately achieving an accurate prediction of the chest injury status.

[0033] It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in this application does not imply that all features and advantages can be realized in any single embodiment. On the contrary, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution or beneficial effect is included in at least one embodiment. Therefore, the description of a technical feature, technical solution or beneficial effect in this specification does not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions and beneficial effects described in the present embodiment can also be combined in any appropriate manner. Those skilled in the art will understand that the embodiment can be implemented without one or more specific technical features, technical solutions or beneficial effects of a specific embodiment. In other embodiments, additional technical features and beneficial effects can also be identified in specific embodiments that do not embody all embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a diagram of the application environment of a dummy chest injury prediction method provided in an embodiment of the present application;

[0035] Figure 2 A schematic flow chart of a method for predicting chest injury of a dummy provided in an embodiment of the present application;

[0036] Figure 3 This is a structural block diagram of a dummy chest injury prediction device provided in an embodiment of the present application;

[0037] Figure 4 This is a diagram of the internal structure of a computer device provided in an embodiment of the application. DETAILED DESCRIPTION

[0038] The terms "first", "second" and "third" in this application specification and the accompanying drawings are used to distinguish different objects rather than to limit a specific order.

[0039] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0040] To make the description of the following embodiments clear and concise, a brief introduction to the related technologies is first given:

[0041] The vehicle's occupant protection performance is one of the indicators for evaluating vehicle performance, and the occupant's chest injury is one of the indicators for evaluating the occupant's injury situation.

[0042] Currently, damage to the chest area of ​​a car in a collision is primarily acquired through actual vehicle crash tests or simulations. The THOR (Test Device for Human Occupant Restraint) dummy's chest structure has seven ribs on each side. However, due to space limitations within the THOR dummy's chest cavity, displacement sensors cannot be installed on every rib. Displacement sensors are only installed on the third rib on the left, third rib on the right, sixth rib on the left, and sixth rib on the right. During the crash test, only the rib displacement curves of the third rib on the left, third rib on the right, sixth rib on the left, and sixth rib on the right of the THOR dummy's chest are collected. The displacement deformation of the other ten ribs on the THOR dummy's chest cannot be collected. Furthermore, the currently collected THOR dummy chest rib displacement curves represent the rib displacement changes of an adult male dummy.

[0043] Therefore, current collision tests cannot accurately reflect the chest and rib injury status of the THOR dummy.

[0044] In order to make the technical solution of this application clearer and easier to understand, the application scenarios of the technical solution of this application are introduced below with reference to the accompanying drawings. Figure 1 As shown in the figure, this figure is a schematic diagram of an application scenario provided by an embodiment of the present application.

[0045] In this application scenario, the terminal 102 can collect collision test data and collision speed, and send the collected collision test data and collision speed to the server 104 through the communication network. The server 104 calculates and analyzes them to obtain a chest injury prediction value, and sends the calculated chest injury prediction value to the terminal 102 through the communication network. The terminal 102 presents the prediction result for relevant technical personnel to view.

[0046] In order to make the technical solution of this application clearer and easier to understand, the following describes a method for predicting chest injuries of a dummy provided by an embodiment of this application in combination with the above application scenarios. Figure 2 As shown in FIG, this figure is a flow chart of a method for predicting dummy chest injuries provided in an embodiment of the present application.

[0047] S201: Obtain collision test data and a collision speed corresponding to the collision test data.

[0048] Among them, the collision test data may include the rib displacement data (such as rib displacement curve, etc.) of the dummy after the collision test at the collision speed and the test condition data (such as frontal collision, side collision and dummy model, etc.).

[0049] It should be noted that in the case of the THOR dummy model, that is, in the THOR dummy collision test, due to the limited internal space of the THOR dummy's chest cavity, it is impossible to install a displacement sensor on each rib, so displacement sensors can only be installed on the left and right sides of the third and sixth ribs; furthermore, during the THOR dummy collision test, only the rib displacement curves of the third rib on the left side, the sixth rib on the left side, the third rib on the right side and the sixth rib on the right side of the THOR dummy's chest can be collected, and the displacement deformation of the other 10 ribs on the THOR dummy's chest cannot be collected.

[0050] For example, the collision test data can be obtained through public research databases, literature or professional testing institutions, and then the rib displacement deformation curve of the third rib on the left side of the THOR dummy can be extracted from the collision test data. , Rib displacement deformation curve of the sixth rib on the left , Rib displacement deformation curve of the third rib on the right , Rib displacement deformation curve of the sixth rib on the right , and the collision velocity v.

[0051] It should be noted that the rib displacement deformation curve describes the relationship between the rib displacement deformation and time t.

[0052] S202: Based on the damage prediction model, perform damage degree analysis on the collision test data and collision speed to obtain a target damage degree.

[0053] The damage prediction model is based on The target damage degree can be used to evaluate the degree of chest damage of the dummy in a collision test.

[0054] For example, the collision test data (such as the rib displacement deformation curve of the third rib on the left) can be , Rib displacement deformation curve of the sixth rib on the left , Rib displacement deformation curve of the third rib on the right , Rib displacement deformation curve of the sixth rib on the right ) and collision speed v are input into the damage prediction model to obtain the rib displacement deformation curve including the left first rib corresponding to the THOR dummy , Rib displacement deformation curve of the second rib on the left , Rib displacement deformation curve of the third rib on the left , Rib displacement deformation curve of the fourth rib on the left , Rib displacement deformation curve of the fifth rib on the left , Rib displacement deformation curve of the sixth rib on the left , Rib displacement deformation curve of the seventh rib on the left , Rib displacement deformation curve of the first rib on the right , Rib displacement deformation curve of the second rib on the right , Rib displacement deformation curve of the third rib on the right , Rib displacement deformation curve of the fourth rib on the right , Rib displacement deformation curve of the fifth rib on the right , Rib displacement deformation curve of the sixth rib on the right , Rib displacement deformation curve of the seventh rib on the right The first prediction data.

[0055] Furthermore, based on the formula (where i ranges from [1,7], represents the rib displacement deformation curve of the i-th rib on the left side, The rib displacement deformation curve of the i-th rib on the right side is represented by: The sum curve of the rib displacement deformation of the i-th group of ribs is represented by the sum of the rib displacement deformation curve of the i-th rib on the left and the rib displacement deformation curve of the i-th rib on the right, where t represents time. The rib displacement deformation curves of the seven ribs on the left and the seven ribs on the right in the first prediction data are calculated to obtain the sum curve of the rib displacement deformation of the first group of ribs. , the sum curve of the displacement deformation of the second group of ribs , the sum curve of the displacement deformation of the third group of ribs , The sum curve of the displacement deformation of the fourth group of ribs , The sum curve of the displacement deformation of the fifth group of ribs , The sum curve of the displacement deformation of the sixth group of ribs , The sum curve of the displacement and deformation of the seventh group of ribs ,For example = , = , = , = , = , = , = ;as well as

[0056] Can be based on the formula (in, The rib displacement deformation difference curve of the i-th group of ribs is represented by the difference between the rib displacement deformation curve of the i-th rib on the left and the rib displacement deformation curve of the i-th rib on the right). The rib displacement deformation curves of the seven ribs on the left and the seven ribs on the right in the first prediction data are calculated to obtain the difference curve of the rib displacement deformation of the first group of ribs. , the difference curve of the second group of rib displacement deformation , the difference curve of the displacement deformation of the third group of ribs , the difference curve of the fourth group of rib displacement deformation , the difference curve of the fifth group of rib displacement deformation , Difference curve of rib displacement deformation of the sixth group , Difference curve of rib displacement deformation of the seventh group ,For example = , = , = , = , = , = , = .

[0057] Furthermore, based on the formula (Among them, totm i represents the average value of the rib displacement deformation sum value curve of the i-th group of ribs, and t represents the integral variable - time), performing mean calculation on the displacement deformation sum value curve of each group of ribs in the seven groups of ribs of the THOR dummy, and obtaining the average value totm1 of the displacement deformation sum value curve of the first group of ribs, the average value totm2 of the displacement deformation sum value curve of the second group of ribs, the average value totm3 of the displacement deformation sum value curve of the third group of ribs, the average value totm4 of the displacement deformation sum value curve of the fourth group of ribs, the average value totm5 of the displacement deformation sum value curve of the fifth group of ribs, the average value totm6 of the displacement deformation sum value curve of the sixth group of ribs, and the average value totm7 of the displacement deformation sum value curve of the seventh group of ribs; and

[0058] Can be based on the formula (Among them, diffm irepresents the average value of the rib displacement deformation difference curve of the i-th group of ribs, t represents the integral variable-time), the average value of the rib displacement deformation difference curve of each group of the seven groups of ribs of the THOR dummy is calculated, and the average value of the rib displacement deformation difference curve of the first group is difm1, the average value of the rib displacement deformation difference curve of the second group is difm2, the average value of the rib displacement deformation difference curve of the third group is difm3, the average value of the rib displacement deformation difference curve of the fourth group is difm4, the average value of the rib displacement deformation difference curve of the fifth group is difm5, the average value of the rib displacement deformation difference curve of the sixth group is difm6, and the average value of the rib displacement deformation difference curve of the seventh group is difm7;

[0059] Furthermore, based on the formula (in, i represents the standard deviation of the rib displacement deformation and value curve of the i-th group of ribs), the standard deviation of the sum of the displacement deformation and value curve of each group of ribs in the seven groups of THOR dummy ribs is calculated to obtain the standard deviation of the sum of the displacement deformation and value curve of the first group of ribs 1. Standard deviation of the second group of rib displacement and deformation curves 2. Standard deviation of the third group of rib displacement and deformation curves 3. Standard deviation of the fourth group of rib displacement and deformation curves 4. Standard deviation of the fifth group of rib displacement and deformation curves 5. Standard deviation of the sixth group of rib displacement and deformation curves 6. Standard deviation of the seventh group of rib displacement and deformation curves 7; and

[0060] Can be based on the formula (in, i represents the standard deviation of the rib displacement deformation difference curve of the i-th group of ribs), the standard deviation of the rib displacement deformation difference curve of each group of the seven groups of ribs of the THOR dummy is calculated to obtain the standard deviation of the rib displacement deformation difference curve of the first group 1. Standard deviation of the second group of rib displacement and deformation difference curves 2. Standard deviation of the third group of rib displacement and deformation difference curves 3. Standard deviation of the fourth group of rib displacement and deformation difference curves 4. Standard deviation of the fifth group of rib displacement and deformation difference curves 5. Standard deviation of the sixth group of rib displacement and deformation difference curves 6. Standard deviation of the seventh group of rib displacement and deformation difference curves 7.

[0061] Furthermore, the sum curve of the displacement deformation of the first group of ribs can be Extract the maximum value from the sum curve , from the sum curve of the displacement deformation of the second group of ribs Extract the maximum value from the sum curve , from the sum curve of the displacement deformation of the third group of ribs Extract the maximum value from the sum curve , from the sum curve of the displacement deformation of the fourth group of ribs Extract the maximum value from the sum curve , from the sum curve of displacement deformation of the fifth group of ribs Extract the maximum value from the sum curve , from the sum curve of the displacement deformation of the sixth group of ribs Extract the maximum value from the sum curve , from the sum curve of the displacement deformation of the seventh group of ribs Extract the maximum value from the sum curve , and the difference curve of displacement deformation of the first group of ribs Extract the maximum value of the difference curve , the difference curve of the second group of rib displacement deformation Extract the maximum value of the difference curve , the difference curve of the displacement deformation of the third group of ribs Extract the maximum value of the difference curve , the difference curve of the fourth group of rib displacement deformation Extract the maximum value of the difference curve , the difference curve of the fifth group of rib displacement deformation Extract the maximum value of the difference curve , Difference curve of rib displacement deformation of the sixth group Extract the maximum value of the difference curve , Difference curve of rib displacement deformation of the seventh group Extract the maximum value of the difference curve , and combine the extracted maximum values ​​into a matrix , and based on Construct the covariance matrix ∑ and perform eigendecomposition on the covariance matrix ∑ to finally obtain the sum weight corresponding to the first set of rib displacement deformation and value curves , the weight corresponding to the second set of rib displacement deformation and value curves , the third group of rib displacement deformation and value curves corresponding to the weight , the sum weight corresponding to the fourth group of rib displacement deformation and value curves , the weight corresponding to the fifth group of rib displacement deformation and value curves , the corresponding weights of the sixth group of rib displacement deformation and value curves , the weight corresponding to the seventh group of rib displacement deformation and value curves , and the difference weight corresponding to the first set of rib displacement deformation difference curves , the difference weight corresponding to the second group of rib displacement deformation difference curves , the difference weight corresponding to the difference curve of the third group of rib displacement deformation , the difference weight corresponding to the fourth group of rib displacement deformation difference curves , the difference weight corresponding to the fifth group of rib displacement deformation difference curves , the difference weight corresponding to the sixth group of rib displacement deformation difference curves , the difference weight corresponding to the seventh group of rib displacement deformation difference curves .

[0062] Furthermore, the first damage degree can be determined based on the maximum value of the sum curve of the rib displacement deformation of each of the seven groups, the standard deviation of the sum curve, and the sum weight. The specific formula can be as follows:

[0063]

[0064] in, is the first degree of injury, i is the group number of the ribs, n is the total number of rib groups (e.g. n=7), is the maximum value in the sum curve of the displacement and deformation of the ribs in group i, is the weight corresponding to the sum curve of the displacement deformation and value of the ribs in group i, is the standard deviation of the rib displacement and deformation value curve of group i.

[0065] The second damage degree can be determined based on the maximum value of the difference curve of the displacement and deformation of each of the seven groups of ribs, the standard deviation of the difference curve, and the difference weight. The specific formula can be as follows:

[0066]

[0067] in, is the second degree of injury, i is the group number of the ribs, n is the total number of rib groups (e.g. n=7), is the maximum value in the difference curve of rib displacement deformation of group i, is the difference weight corresponding to the difference curve of rib displacement deformation of group i, is the standard deviation of the difference curve of rib displacement deformation in group i.

[0068] Finally, the first damage degree can be and the second degree of damage The sum of is determined as the target damage degree, which can be expressed as:

[0069]

[0070] in, The target damage degree.

[0071] S203: Determine a chest injury prediction value corresponding to the dummy based on a preset age and target injury degree corresponding to the dummy.

[0072] Among them, the preset age can represent the age of the dummy, which can be set in advance, and the value can be set according to needs to achieve the prediction of chest injury conditions of occupants of different age groups; the chest injury prediction value can represent the chest injury condition of the dummy in the collision test.

[0073] Exemplarily, the preset age and target injury severity of the dummy are input into an accelerated failure time model to determine a predicted chest injury value for the dummy, wherein the accelerated failure time model follows a Weibull distribution.

[0074] Optionally, the accelerated failure time model analyzes and calculates the preset age and target injury degree, and obtains the chest injury prediction value corresponding to the determined dummy, which can be specifically expressed as:

[0075]

[0076] in, is the predicted value of chest injury, age is the preset age, is the target damage degree, λ is the scale parameter, which can be taken as 1 / 0.302; is the intercept term corresponding to time, which can be taken as 2.868; is the age coefficient, which can be -0.0181.

[0077] The above-mentioned dummy chest injury prediction method provides a data basis for predicting the chest injury prediction value by obtaining collision test data and the collision speed corresponding to the collision test data. Furthermore, based on the injury prediction model, the collision test data and collision speed can be analyzed for the degree of injury to obtain the target injury degree, providing a basis for evaluating the degree of injury. Furthermore, based on the preset age and target injury degree corresponding to the dummy, the chest injury prediction value corresponding to the dummy can be determined, ultimately achieving an accurate prediction of the chest injury status. This solution provides a way to accurately determine the target injury degree by introducing an injury prediction model. Furthermore, by introducing a preset age, it takes into account all age groups, ultimately achieving an accurate prediction of the chest injury status.

[0078] Based on the above embodiment, the present embodiment provides a detailed explanation of S202. Specifically, the present embodiment involves the training process of the damage prediction model, which specifically includes:

[0079] Build Model; Based on the simulation training set, The model is trained to obtain a damage prediction model.

[0080] The simulation training set includes the rib displacement deformation simulation curves of the seven ribs on the left side of the chest and the simulation curves of the rib displacement deformation of the seven ribs on the right side of the chest.

[0081] Optional, build The process of the model can be expressed as follows: the convolution layer can be used to perform two Conv3×3 convolutions, and then the activation function ReLU is used for activation processing, and then a 2×2 maximum pooling operation is used for downsampling to obtain the first feature image, so as to reduce the size of the curve image and increase the number of feature channels; further, the first feature image can be repeatedly processed once according to the above method (convolution processing-activation function processing-maximum pooling processing) to obtain the second feature image, further reducing the size of the curve image and increasing the number of feature channels; further, the second feature image can be repeatedly processed according to the above method (convolution processing-activation function processing-maximum pooling processing) to obtain the second feature image, further reducing the size of the curve image and increasing the number of feature channels; Processing-maximum pooling processing) is repeated once to obtain a third feature image; further, the third feature image can be repeatedly processed once according to the above method (convolution processing-activation function processing-maximum pooling processing) to obtain a fourth feature image, so that the size of the processed curve image is minimized and the number of feature channels is maximized; further, the fourth feature image can be subjected to a Conv2×2 transposed convolution process through a convolution layer and then up-sampled to obtain a fifth feature image to increase the size of the feature image, and then the fifth feature image and the third feature image are spliced ​​by skip connection to obtain a spliced ​​feature image to improve the curve image. The segmentation accuracy of the image is improved, and then the spliced ​​feature image is processed by Conv3×3 convolution twice through the convolution layer, and then activated by the activation function ReLU to obtain the sixth feature image; further, the sixth feature image can be up-sampled by Conv2×2 transposed convolution once through the convolution layer to obtain the seventh feature image to increase the size of the feature image, and then the seventh feature image and the second feature image are spliced ​​by jump connection to obtain the eighth feature image to improve the segmentation accuracy of the curve image, and then the eighth feature image is processed by Conv3×3 convolution twice through the convolution layer, and then activated by the activation function R eLU is activated to obtain the ninth feature image; further, the ninth feature image is up-sampled by performing a Conv2×2 transposed convolution process on the convolution layer to obtain the tenth feature image; then the tenth feature image and the first feature image are jump-connected and spliced ​​to obtain the eleventh feature image; further, the eleventh feature image is then subjected to two Conv3×3 convolution processes through the convolution layer, and then activated by the activation function ReLU to obtain the twelfth feature image; finally, the twelfth feature image is subjected to a Conv1×1 convolution process through the convolution layer to complete the construction of the U-Net model.

[0082] For example, the displacement deformation curve of the third rib on the left, the displacement deformation curve of the sixth rib on the left, the displacement deformation curve of the third rib on the right, the displacement deformation curve of the sixth rib on the right, and the collision speed v in the simulation training set can be input into the constructed Model, by The model analyzes and calculates it to obtain the predicted displacement deformation curves of seven ribs on each side of the left and right sides; further, the predicted displacement deformation curves of the seven ribs on the left, the displacement deformation curves of the seven ribs on the right, the displacement deformation curves of the seven ribs on the left in the simulation training set, and the displacement deformation curves of the seven ribs on the right in the simulation training set can be calculated for loss, and the adjustment can be made according to the results of the loss calculation. The model parameters in the model are adjusted until the preset loss threshold is reached. Training of the model, i.e. the damage prediction model.

[0083] Furthermore, after obtaining the damage prediction model, the method may further include: verifying the damage prediction model based on image data corresponding to the collision test to obtain a damage prediction model with a preset accuracy.

[0084] For example, three miniature high-speed cameras (with a recording frequency of 1000 Hz) can be installed at the upper, middle, and lower positions of the thoracic spine of the THOR dummy, respectively. After marking the tip of each rib, a sliding table collision test is performed to collect video / image data of the movement of the seven ribs on the left side and the seven ribs on the right side of the THOR dummy's chest during the test. The video / image data are analyzed and processed to draw the displacement deformation curves of the seven ribs on the left side and the seven ribs on the right side of the THOR dummy's chest; further, the displacement deformation curves of the third rib on the left side, the third rib on the right side, and the sixth rib on the left side of the THOR dummy's chest analyzed from the video / image data can be used to generate the displacement deformation curves of the seven ribs on the left side and the seven ribs on the right side of the THOR dummy's chest. The rib displacement deformation curves of the first rib and the sixth rib on the right and the collision test speed are input into the damage prediction model, and the predicted displacement deformation curves of the seven ribs on the left and the seven ribs on the right of the THOR dummy are output. The predicted value of the THOR dummy chest rib displacement deformation curve analyzed from the slide collision test video data is compared with the test value of the rib displacement deformation curve predicted by the U-Net model, and the loss calculation is performed to determine the loss between the predicted value of the rib displacement deformation curve and the test value of the rib displacement deformation curve; if the loss exceeds the preset accuracy value, the damage prediction model is trained again to ultimately ensure the prediction accuracy of the damage prediction model.

[0085] In the embodiment of the present application, by constructing Model, and based on the simulation training set, The model is fully trained, laying the foundation for the prediction accuracy of the damage prediction model.

[0086] Combined with the above Figures 1 to 2 The dummy chest injury prediction method provided in the embodiment of the present application is introduced in detail. The device and equipment provided in the embodiment of the present application will be introduced in conjunction with the accompanying drawings.

[0087] like Figure 3 As shown in FIG. 3 , this figure is a schematic diagram of a dummy chest injury prediction device provided in an embodiment of the present application. The device 300 includes: an acquisition module 301, an analysis module 302, and a prediction module 303, wherein:

[0088] An acquisition module 301 is configured to acquire collision test data and a collision velocity corresponding to the collision test data; wherein the collision test data is rib displacement data of a dummy after a collision test at the collision velocity;

[0089] An analysis module 302 is configured to perform damage degree analysis on the collision test data and the collision speed based on a damage prediction model to obtain a target damage degree;

[0090] The prediction module 303 is configured to determine a chest injury prediction value corresponding to the dummy based on a preset age and target injury degree corresponding to the dummy.

[0091] In one embodiment, the prediction module 303 is specifically configured to:

[0092] The preset age and target injury severity of the dummy are input into the accelerated failure time model to determine the predicted chest injury value of the dummy; wherein the accelerated failure time model follows a Weibull distribution.

[0093] In one embodiment, the analysis module 302 is specifically configured to:

[0094] The collision test data and collision velocity are input into the damage prediction model to obtain first prediction data; wherein the first prediction data includes predicted rib displacement deformation curves of multiple groups of ribs; based on the first prediction data, feature decomposition is performed to obtain weights corresponding to the first prediction data; based on the first prediction data and the weights, a target damage degree is obtained.

[0095] In one embodiment, the analysis module 302 is specifically configured to:

[0096] Based on the first prediction data and the weight, a first damage degree and a second damage degree are determined; and the sum of the first damage degree and the second damage degree is determined as the target damage degree.

[0097] In one embodiment, the analysis module 302 is specifically configured to:

[0098] Determining, based on the first prediction data, a sum curve of rib displacement deformation, a difference curve of rib displacement deformation, a standard deviation of the sum curve of rib displacement deformation, and a standard deviation of the difference curve of rib displacement deformation;

[0099] A first damage degree and a second damage degree are determined based on the sum curve, the difference curve, the standard deviation of the sum curve, the standard deviation of the difference curve, and the weight.

[0100] In one embodiment, the dummy chest injury prediction device 300 further includes:

[0101] Training module for building Model; Based on the simulation training set, The model is trained to obtain a damage prediction model; wherein the simulation training set includes the rib displacement deformation simulation curves of the seven ribs on the left side of the chest and the simulation curves of the rib displacement deformation of the seven ribs on the right side of the chest.

[0102] In one embodiment, the dummy chest injury prediction device 300 further includes:

[0103] The verification module is used to verify the damage prediction model based on the image data corresponding to the collision test to obtain a damage prediction model with a preset accuracy.

[0104] The dummy chest injury prediction device 300 according to the embodiment of the present application may correspond to the method described in the embodiment of the present application, and the above-mentioned other operations and / or functions of each module / unit of the dummy chest injury prediction device 300 are respectively to achieve Figure 2 For the sake of brevity, the corresponding processes of the various methods in the illustrated embodiments are not described again here.

[0105] The embodiment of the present application further provides a computing device, which can be a local computing device or an application server.

[0106] like Figure 4 As shown, this figure is a schematic diagram of a computing device provided by an embodiment of the present application, and the computing device 400 includes a bus 401, a processor 402, a communication interface 403 and a memory 404. The processor 402, the memory 404 and the communication interface 403 communicate with each other via the bus 401.

[0107] The bus 401 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0108] The processor 402 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0109] The communication interface 403 is used for external communication. For example, the communication interface 403 can be used to communicate with the terminal 102. The communication interface 403 is used to send the chest injury prediction value to the terminal 102 so that the terminal 102 can display the dummy chest injury prediction result.

[0110] Memory 404 may include volatile memory, such as random access memory (RAM). Memory 404 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0111] The memory 404 stores executable codes, and the processor 402 executes the executable codes to perform the aforementioned dummy chest injury prediction method.

[0112] Specifically, in the implementation Figure 3 In the case of the embodiment shown, and Figure 3 When each module or unit of the dummy chest injury prediction device described in the embodiment is implemented by software, Figure 3 The software or program code required for the functions of each module / unit in the system may be partially or completely stored in the memory 404. The processor 402 executes the program code corresponding to each unit stored in the memory 404 to perform the aforementioned dummy chest injury prediction method.

[0113] Embodiments of the present application also provide a computer-readable storage medium. The computer-readable storage medium can be any available medium capable of being stored by a computing device, or a data storage device such as a data center that contains one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, hard disk, or magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the aforementioned dummy chest injury prediction method.

[0114] The present application also provides a computer program product comprising one or more computer instructions that, when loaded and executed on a computing device, fully or partially generate the process or function described in the present application.

[0115] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0116] When the computer program product is executed by a computer, the computer performs any of the aforementioned methods for predicting dummy chest injuries. The computer program product may be a software installation package, and when any of the aforementioned methods for predicting dummy chest injuries is needed, the computer program product may be downloaded and executed on the computer.

[0117] The descriptions of the processes or structures corresponding to the above figures have different emphases. For parts that are not described in detail in a certain process or structure, please refer to the relevant descriptions of other processes or structures.

[0118] The above description is only a specific implementation method of the present application, but the protection scope of the present application is not limited thereto. Any changes or replacements within the technical scope disclosed in the present application should be included in the protection scope of the present application.

Claims

1. A method for predicting dummy chest injuries, characterized in that: The method comprises: Obtaining collision test data and a collision speed corresponding to the collision test data; wherein the collision test data is rib displacement data of a dummy after a collision test at the collision speed; Inputting the collision test data and the collision speed into a damage prediction model to obtain first prediction data; wherein the first prediction data includes predicted rib displacement deformation curves of multiple groups of ribs; Performing feature decomposition based on the first prediction data to obtain a weight corresponding to the first prediction data; Obtaining a target damage degree based on the first prediction data and the weight; The preset age and target injury degree corresponding to the dummy are input into an accelerated failure time model to determine a predicted chest injury value corresponding to the dummy; wherein the accelerated failure time model obeys a Weibull distribution.

2. The method according to claim 1, characterized in that Obtaining a target damage degree based on the first prediction data and the weight includes: determining a first damage degree and a second damage degree based on the first prediction data and the weight; The sum of the first damage degree and the second damage degree is determined as a target damage degree.

3. The method according to claim 2, characterized in that The determining of the first damage degree and the second damage degree based on the first prediction data and the weight includes: Determining, based on the first prediction data, a sum curve of rib displacement deformation, a difference curve of rib displacement deformation, a standard deviation of the sum curve of rib displacement deformation, and a standard deviation of the difference curve of rib displacement deformation; A first damage degree and a second damage degree are determined based on the sum curve, the difference curve, the standard deviation of the sum curve, the standard deviation of the difference curve, and the weight.

4. The method according to claim 1, wherein The training process of the damage prediction model includes: Build Model; Based on the simulation training set, The model is trained to obtain the damage prediction model; wherein the simulation training set includes the rib displacement deformation simulation curves of the seven ribs on the left side of the chest and the simulation curves of the rib displacement deformation of the seven ribs on the right side of the chest.

5. The method according to claim 4, characterized in that After obtaining the damage prediction model, the method includes: The damage prediction model is verified based on the image data corresponding to the collision test to obtain a damage prediction model with a preset accuracy.

6. A dummy chest injury prediction device, characterized in that: The device comprises: an acquisition module, configured to acquire collision test data and a collision velocity corresponding to the collision test data; wherein the collision test data is rib displacement data of a dummy after a collision test at the collision velocity; An analysis module, configured to perform damage degree analysis on the collision test data and the collision speed based on a damage prediction model to obtain a target damage degree; a prediction module, configured to determine a chest injury prediction value corresponding to the dummy based on a preset age corresponding to the dummy and the target injury degree; The analysis module is further configured to input the collision test data and the collision speed into a damage prediction model to obtain first prediction data; wherein the first prediction data includes predicted rib displacement and deformation curves of multiple groups of ribs; perform eigendecomposition based on the first prediction data to obtain weights corresponding to the first prediction data; and obtain a target damage degree based on the first prediction data and the weights; The prediction module is further used to input the preset age corresponding to the dummy and the target injury degree into the accelerated failure time model to determine the chest injury prediction value corresponding to the dummy; wherein the accelerated failure time model obeys the Weibull distribution.

7. A computing device, characterized in that including memory and processor; One or more computer programs are stored in the memory, and the one or more computer programs include instructions; when the instructions are executed by the processor, the computing device executes the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1 to 5.

Images