Head collision prevention alarm probability determination method and device, electronic equipment and storage medium

Abstract

The invention provides an anti-head-collision alarm probability determination method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a vehicle front row image and a vehicle rear row image; identifying a first head which is located in a first target area and of which a matched human body is located in an area on the aisle in the front row image of the vehicle, and obtaining a first target head external connection frame; determining a first original alarm probability according to a size comparison relationship between the absolute head area of the first target head external frame and a first preset head area threshold value; identifying a second head which is located in a second target area in the vehicle rear row image and of which a matched human body is located in an area below the aisle, and obtaining at least one second target head external connection frame; determining at least one second original alarm probability according to a size comparison relationship between the absolute head area of each second target head external frame and a second preset head area threshold value; and determining that the target alarm probability is the maximum value in the first original alarm probability and the at least one second original alarm probability.

Description

Technical Field

[0001] The present disclosure relates to the technical field of determining the probability of an anti-collision head alarm in a vehicle, and in particular to a method, device, electronic device, and storage medium for determining the probability of an anti-collision head alarm. Background Art

[0002] In related anti-head collision technology, taking the scenario of preventing a person's head from colliding with an obstacle (entertainment screen) in a vehicle as an example: a Time of Flight (TOF) camera is usually used to obtain depth information between the person's head and the camera. The distance between the person's head and the camera indicated by the depth information is used to determine whether the person's head has entered the target area where there is a risk of head collision, and then the probability of an anti-head collision alarm is determined based on the judgment result.

[0003] However, the entertainment screen is usually set between the front seats and the second row of seats, and the TOF camera is usually set twenty centimeters behind the entertainment screen; therefore, the TOF camera cannot capture the depth image of the area in the aisle, resulting in the anti-collision head technology using the TOF camera being unable to determine the anti-collision head alarm probability of users in the aisle area, and further resulting in the inability to accurately determine the corresponding alarm probability of users in the aisle area. Summary of the Invention

[0004] The present disclosure provides a method, device, electronic device and storage medium for determining an anti-collision head alarm probability.

[0005] According to a first aspect of the present disclosure, a method for determining an anti-collision head alarm probability is provided, comprising:

[0006] Acquire a front-row image and a rear-seat image of the vehicle; the front-seat image of the vehicle refers to an image acquired by a non-depth image collector disposed in front of the front seats, and the rear-seat image of the vehicle refers to an image acquired by a non-depth image collector disposed behind an obstacle;

[0007] Identify a first human head located in a first target area in the front row image of the vehicle, and a matched human body located in an upper aisle area, to obtain a first target head circumscribed frame; the first target area is a risk area in the upper aisle area where a human head may collide with an obstacle; the aisle area includes an upper aisle area and a lower aisle area;

[0008] determining a first original alarm probability based on a comparison between an absolute head area of ​​the first target head bounding box and a first preset head area threshold;

[0009] Identifying a second human head located in a second target area in the rear-seat image of the vehicle and a matching human body located in the area below the aisle, and obtaining at least one second target human head circumscribed frame; the second target area is a risk area in the area below the aisle where the human head may collide with an obstacle;

[0010] determining at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head circumscribed frames and a second preset head area threshold;

[0011] The target alarm probability is determined to be the maximum value of the first original alarm probability and the at least one second original alarm probability.

[0012] In some embodiments of the present disclosure, identifying a first human head located in a first target area in the front row image of the vehicle and a matching human body located in an aisle area, and obtaining a first target human head bounding box includes:

[0013] Identifying a first human body and a first human head in the image of the front row of the vehicle, and obtaining a first human body circumference frame of each first human body and a first head circumference frame of each first human head in the image of the front row of the vehicle;

[0014] determining an intersection-over-union (IoU) ratio between each of the first human body circumference frames and each of the first head circumference frames, and matching the first human body circumference frames with the first head circumference frames whose IoU ratios are not less than a preset IoU threshold to obtain a matching result;

[0015] Determine, based on the position coordinates of the first human body external bounding box in the vehicle, the first human body external bounding box of the position coordinates in the front seat;

[0016] Filtering out a first head circumference frame that matches the first head circumference frame of the front seat from the first head circumference frames of the matching results, obtaining a first pre-selected head circumference frame;

[0017] According to the position coordinates of the first pre-selected head circumscribed frame, a first target head circumscribed frame having the position coordinates in the target area is determined.

[0018] In some embodiments of the present disclosure, determining a first original alarm probability based on a comparison between the absolute head area of ​​the first target head bounding box and a first preset head area threshold includes:

[0019] Obtaining a first initial alarm probability based on a comparison between the absolute head area of ​​the first target head bounding box and a first preset head area threshold;

[0020] Identify human bodies and human heads in the rear-seat image of the vehicle, and obtain the number of human bodies in the area under the aisle and the number of human heads matching the human bodies in the area under the aisle;

[0021] Determining a first alarm probability correction value based on a matching relationship between the number of human bodies and the number of heads in the area below the aisle;

[0022] The first initial alarm probability is corrected using the first alarm probability correction value to obtain a first original alarm probability.

[0023] In some embodiments of the present disclosure, the first preset head area threshold includes a front row first preset head area threshold and a front row second preset head area threshold, the front row first preset head area threshold being greater than the front row second preset head area threshold;

[0024] Obtaining a first initial alarm probability based on a comparison between the absolute head area of ​​the first target head circumscribed frame and a first preset head area threshold includes:

[0025] Multiplying the length and width of the first target head circumference frame to obtain the absolute head area of ​​the first target head circumference frame;

[0026] Determining whether the absolute head area of ​​the first target person's head circumscribed frame is greater than the first preset head area threshold of the front row;

[0027] If the judgment result is no, then

[0028] Subtracting the first preset head area threshold in the front row from the absolute head area of ​​the first target head circumscribed frame to obtain a first difference value;

[0029] Subtracting the first preset head area threshold of the front row from the second preset head area threshold of the front row to obtain a second difference;

[0030] taking the ratio of the first difference to the second difference as a first ratio;

[0031] The first ratio is added to the front-row preset alarm probability correction coefficient to obtain a first initial alarm probability.

[0032] In some embodiments of the present disclosure, a first alarm probability correction value is determined based on a matching relationship between the number of human bodies and the number of heads in the area below the aisle;

[0033] When the number of human bodies in the area below the aisle is 0, determining the alarm probability correction value to be a first correction value;

[0034] When the number of human bodies in the area under the aisle is 1 and the number of heads matching the human bodies in the area under the aisle is 0, the alarm probability correction value is determined to be a second correction value; the first correction value is smaller than the second correction value.

[0035] In some embodiments of the present disclosure, determining at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head circumscribed frames and a second preset head area threshold includes:

[0036] Determining whether the occlusion area in the rear image of the vehicle is smaller than a preset occlusion area threshold;

[0037] If the judgment result is yes, each of the second target head circumference frames is traversed, and at least one second initial alarm probability is obtained based on the comparison relationship between the absolute head area of ​​each of the second target head circumference frames and the second preset head area threshold.

[0038] In some embodiments of the present disclosure, determining whether the occlusion area in the rear-seat image of the vehicle is smaller than a preset occlusion area threshold includes:

[0039] Obtaining a pixel value of each pixel in the rear image of the vehicle;

[0040] Comparing the pixel value of the pixel point with a preset pixel value, and determining the number of pixel points in the rear-row image of the current frame whose pixel value is greater than the preset pixel value;

[0041] Determine whether the number of pixel points whose pixel values ​​are greater than the preset pixel value is less than the preset number of pixel points.

[0042] In some embodiments of the present disclosure, the method for determining the probability of a head collision warning provided by the present disclosure includes: identifying the area in front of the first target area in the front row image of the vehicle; and

[0043] Acquire multiple training images, wherein the training images are images of heads of people in the aisle at the boundary position of the risk area;

[0044] Recognize the human heads in the training image to obtain a plurality of human head bounding boxes in the training image; wherein one human head corresponds to one human head bounding box;

[0045] Determine the center point position of each of the head circumference frames to obtain the center point positions of multiple head circumference frames;

[0046] A closed area formed by connecting lines of the plurality of center points is determined as a first target area.

[0047] According to a second aspect of the present disclosure, a device for determining an anti-collision head alarm probability is provided, comprising:

[0048] An acquisition unit is configured to acquire a front-row image and a rear-row image of the vehicle; the front-row image of the vehicle refers to an image acquired by a non-depth image collector disposed in front of the front seats, and the rear-seat image of the vehicle refers to an image acquired by a non-depth image collector disposed behind an obstacle;

[0049] a first recognition unit, configured to recognize a first human head located in a first target area in the image of the front row of the vehicle, and a matched human body located in an upper aisle area, and obtain a first target human head bounding box; the first target area is a risk area in the upper aisle area where a human head may collide with an obstacle; the aisle area includes an upper aisle area and a lower aisle area;

[0050] a first determining unit, configured to determine a first original alarm probability based on a comparison between an absolute head area of ​​the first target head bounding box and a first preset head area threshold;

[0051] a second recognition unit configured to recognize a second human head located in a second target area in the rear-seat image of the vehicle and whose matched human body is located in the area below the aisle, and obtain at least one second target human head circumscribed frame; the second target area being a risk area in the area below the aisle where the human head may collide with an obstacle;

[0052] a second determining unit, configured to determine at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head circumscribed frames and a second preset head area threshold;

[0053] The third determining unit is configured to determine a target alarm probability as a maximum value between the first original alarm probability and the at least one second original alarm probability.

[0054] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0055] at least one processor; and

[0056] a memory communicatively connected to at least one processor; wherein,

[0057] The memory stores instructions that can be executed by at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of the first aspect.

[0058] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the method of the aforementioned first aspect.

[0059] The present disclosure provides a method, device, electronic device and storage medium for determining the probability of anti-collision head alarm. The method comprises: acquiring a front row image and a rear row image of a vehicle; the front row image of the vehicle refers to an image acquired by a non-depth image collector set in front of the front seat, and the rear row image of the vehicle refers to an image acquired by a non-depth image collector set behind an obstacle; identifying a first human head in the front row image of the vehicle that is located in a first target area and a matching human body located in an upper aisle area, and obtaining a first target human head external bounding box; the first target area refers to a risk area where a human head in the upper aisle area may collide with an obstacle; the aisle area includes an upper aisle area and a lower aisle area; according to the absolute head area of ​​the first target human head external bounding box A first original alarm probability is determined by comparing the size of the head area with the first preset head area threshold; a second human head located in the second target area in the rear image of the vehicle and the matching human body located in the area under the aisle is identified to obtain at least one second target head external bounding box; the second target area refers to the risk area in the area under the aisle where the human head may hit an obstacle; at least one second original alarm probability is determined based on the size comparison between the absolute head area of ​​each of the second target head external bounding boxes and the second preset head area threshold; the target alarm probability is determined to be the maximum value of the first original alarm probability and the at least one second original alarm probability.

[0060] According to the solution disclosed in the present invention, by respectively determining the first original alarm probability of the area above the aisle in the front row image and at least one second original alarm probability of the area below the aisle in the back row image, and determining the target alarm probability as the maximum value of the first original alarm probability and the at least one second original alarm probability, it is possible to determine the anti-head collision alarm probability of users in the area above the aisle, thereby ensuring accurate determination of the alarm probability corresponding to users in the aisle area.

[0061] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0063] Figure 1 A flowchart of a method for determining the probability of an anti-collision head alarm provided by an embodiment of the present disclosure;

[0064] Figure 2 A schematic diagram of the seat distribution structure of a vehicle cabin provided by an embodiment of the present disclosure;

[0065] Figure 3 A schematic diagram of the structure of key points of the human body provided by an embodiment of the present disclosure;

[0066] Figure 4 A flowchart of a method for determining a first target head circumscribed frame provided by an embodiment of the present disclosure;

[0067] Figure 5 A flowchart of a method for obtaining a first original alarm probability provided by an embodiment of the present disclosure;

[0068] Figure 6 A flowchart of a method for obtaining a first initial alarm probability provided by an embodiment of the present disclosure;

[0069] Figure 7 A flowchart of a method for determining a target area provided by an embodiment of the present disclosure;

[0070] Figure 8 A schematic diagram of the structure of a device for determining the probability of an anti-collision head alarm provided by an embodiment of the present disclosure;

[0071] Figure 9 A schematic block diagram of an exemplary electronic device provided for an embodiment of the present disclosure. DETAILED DESCRIPTION

[0072] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0073] A method for determining the probability of an anti-collision head alarm provided in an embodiment of the present disclosure can be applied to vehicles equipped with infrared (IR) cameras. The executor of the method can be a processor set on the vehicle side or in the cloud, such as the processor of the anti-collision head alarm module in the vehicle, or the vehicle controller.

[0074] like Figure 1 As shown, the method for determining the anti-collision head alarm probability provided by the embodiment of the present disclosure includes the following steps:

[0075] Step 101, obtaining a front row image and a rear row image of a vehicle;

[0076] In one embodiment, the vehicle front row image refers to an image captured by a non-depth image collector disposed in front of the front seats.

[0077] In one embodiment, the vehicle rear image refers to an image captured by a non-depth image collector disposed behind an obstacle.

[0078] In one embodiment, the front-row image collector is a non-depth image collector. Compared with a non-depth image collector, when processing the depth image collected by the depth image collector, it is usually necessary to convert the depth information in the depth image into the distance information between the target (human body, human head) in the depth image and the depth image collector, which consumes a lot of computing resources.

[0079] In one embodiment, the front row image collector is used to collect vehicle front row images, which include images of the front seats, second row seats and aisle. Therefore, the front row image collector can be set in front of or behind the rearview mirror inside the vehicle, or in an upper position of the vehicle.

[0080] In one embodiment, the seat distribution diagram in the vehicle cabin is as follows: Figure 2 As shown, the front row seats refer to seats 0 and 1, the second row seats refer to seats 2 and 3, the third row seats refer to seats 4 and 5, and the aisle seat refers to seat 7.

[0081] In one embodiment, the aisle includes an upper aisle area and a lower aisle area. The upper aisle area is captured by the front row image collectors, and the lower aisle area is captured by the rear row image collectors.

[0082] In one embodiment, since the front row image collector can be set in front of or behind the rearview mirror inside the vehicle, or in an upper position of the vehicle, the camera of the front row image collector is at a top-down angle. If the user in the front row seat stretches his head back to the obstacle (entertainment screen), the head corresponding to the user in the front row seat will be captured by the front row image collector.

[0083] In one embodiment, the rear row image collector is used to capture images of the second row of seats, the area under the aisle, and the third row of seats in the vehicle.

[0084] In one embodiment, the rear image collector may be installed twenty centimeters in front of the obstacle.

[0085] Step 102: identifying a first human head located in a first target area in the image of the front row of the vehicle, and a matching human body located in an area above the aisle, and obtaining a first target human head circumscribed frame;

[0086] In one embodiment, the first target area refers to a risk area in the aisle where a person's head may hit an obstacle.

[0087] In one embodiment, the aisle area includes an upper aisle area and a lower aisle area.

[0088] In one embodiment, a human head classification detection network can be used to identify the human body position and head position in the front row image of the vehicle, and match the human body and head; further, based on the human body position and head position, the first human head located in the first target area in the front row image of the vehicle and the matched human body located in the aisle area is determined to obtain the first target head external frame.

[0089] In one embodiment, if Figure 3 As shown, the human body key point detection network can be used to identify the human body, obtain the human body key points, and match the human head key points in the human body key points with the human head, thereby achieving matching between the human body and the human head.

[0090] In one embodiment, since the head collision prevention alarm probability determined by the present disclosure is the head collision prevention alarm probability of users in the aisle area, the target area is a risk area in the aisle area where users may collide with obstacles.

[0091] In one embodiment, the obstruction is typically an entertainment screen disposed between the front row of seats and the second row of seats.

[0092] Step 103: determining a first original alarm probability based on a comparison between the absolute head area of ​​the first target head bounding box and a first preset head area threshold;

[0093] In one embodiment, the absolute head area of ​​the first target person's head circumference frame refers to the area of ​​the first target person's head circumference frame.

[0094] In one embodiment, the first target head circumference frame may be a rectangle, and the absolute head area of ​​the first target head circumference frame refers to the product of the length and width of the first target head circumference frame.

[0095] In one embodiment, the first target head circumference frame may be an ellipse, and the absolute head area of ​​the first target head circumference frame refers to the product of the major semi-axis, the minor semi-axis and pi of the target head circumference frame.

[0096] In one embodiment, because the front-row image of the vehicle is not captured by a depth image collector, the size of the head bounding box corresponding to the head in the front-row image of the vehicle is highly correlated with the distance between the head and the obstacle. Specifically, the larger the head bounding box, the smaller the distance between the head and the front-row image collector, and the greater the distance between the head and the obstacle. Conversely, the smaller the head bounding box, the larger the distance between the head and the front-row image collector, and the smaller the distance between the head and the obstacle.

[0097] In one embodiment, the first preset head area threshold refers to the area value of the corresponding head circumscribed frame on the front row image of the vehicle when the head is at the boundary of the risk area.

[0098] In one embodiment, the first preset head area threshold may also refer to the area value of the head circumscribed frame corresponding to the front row image of the vehicle when the head contacts an obstacle.

[0099] In one embodiment, the area value of the external bounding box of the human head corresponding to the image of the front row of the vehicle when the human head is at the boundary of the risk area and the area value of the external bounding box of the human head corresponding to the image of the front row of the vehicle when the human head contacts an obstacle can also be used simultaneously to jointly determine the first original alarm probability.

[0100] Step 104: identifying a second human head located in a second target area in the vehicle rear row image, whose matched human body is located in the area below the aisle, and obtaining at least one second target human head circumscribed frame;

[0101] In one embodiment, the second target area refers to a risk area under the aisle where a person's head may hit an obstacle.

[0102] In one embodiment, a human head classification detection network can be used to identify the human body position and head position in the rear image of the vehicle, and match the human body and head; further, based on the human body position and head position, a second head located in the second target area in the rear image of the vehicle and the matched human body located in the aisle area is determined to obtain a second target head external frame.

[0103] Step 105: determining at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head bounding boxes and a second preset head area threshold;

[0104] In one embodiment, the absolute head area of ​​the second target person's head circumference frame refers to the area of ​​the second target person's head circumference frame.

[0105] In one embodiment, the shape of the second target head circumference frame may be the same as or different from the shape of the first target head circumference frame.

[0106] Specifically,

[0107] In one embodiment, the second target head circumference frame may be a rectangle, and the absolute head area of ​​the second target head circumference frame refers to the product of the length and width of the second target head circumference frame.

[0108] In one embodiment, the second target head circumference frame may be an ellipse, and the absolute head area of ​​the second target head circumference frame refers to the product of the major semi-axis, the minor semi-axis and pi of the target head circumference frame.

[0109] In one embodiment, the second preset head area threshold refers to the area value of the corresponding head circumscribed frame on the rear image of the vehicle when the head is at the boundary of the risk area.

[0110] In one embodiment, the second preset head area threshold may also refer to the area value of the head circumscribed frame corresponding to the rear image of the vehicle when the head contacts an obstacle.

[0111] In one embodiment, the area value of the external bounding box of the human head corresponding to the image of the rear row of the vehicle when the human head is at the boundary of the risk area and the area value of the external bounding box of the human head corresponding to the image of the rear row of the vehicle when the human head contacts an obstacle can also be used simultaneously to jointly determine the second original alarm probability.

[0112] Step 106: Determine the target alarm probability as the maximum value of the first original alarm probability and the at least one second original alarm probability.

[0113] In one embodiment, since the area in the aisle is very small and there is at most only one user, there is only one first original alarm probability.

[0114] In one embodiment, the area under the aisle is relatively large and may contain multiple users, thus corresponding to multiple second original alarm probabilities.

[0115] In one embodiment, the target alarm probability is the maximum value of a first original alarm probability and a plurality of second original alarm probabilities.

[0116] The present disclosure provides a method, device, electronic device and storage medium for determining the probability of anti-collision head alarm. The method comprises: acquiring a front row image and a rear row image of a vehicle; the front row image of the vehicle refers to an image acquired by a non-depth image collector set in front of the front seat, and the rear row image of the vehicle refers to an image acquired by a non-depth image collector set behind an obstacle; identifying a first human head in the front row image of the vehicle that is located in a first target area and a matching human body located in an upper aisle area, and obtaining a first target human head external bounding box; the first target area refers to a risk area where a human head in the upper aisle area may collide with an obstacle; the aisle area includes an upper aisle area and a lower aisle area; according to the absolute head area of ​​the first target human head external bounding box A first original alarm probability is determined by comparing the size of the head area with the first preset head area threshold; a second human head located in the second target area in the rear image of the vehicle and the matching human body located in the area under the aisle is identified to obtain at least one second target head external bounding box; the second target area refers to the risk area in the area under the aisle where the human head may hit an obstacle; at least one second original alarm probability is determined based on the size comparison between the absolute head area of ​​each of the second target head external bounding boxes and the second preset head area threshold; the target alarm probability is determined to be the maximum value of the first original alarm probability and the at least one second original alarm probability.

[0117] According to the solution disclosed in the present invention, by respectively determining the first original alarm probability of the area above the aisle in the front row image and at least one second original alarm probability of the area below the aisle in the back row image, and determining the target alarm probability as the maximum value of the first original alarm probability and the at least one second original alarm probability, it is possible to determine the anti-head collision alarm probability of users in the area above the aisle, thereby ensuring accurate determination of the alarm probability corresponding to users in the aisle area.

[0118] In one embodiment, Identifying a first human head located in a first target area in the front row image of the vehicle, and a matching human body located in an area above the aisle, and obtaining an external bounding box of the first target human head, including:

[0119] Step 401, identifying first human bodies and first human heads in the vehicle front row image, and obtaining a first human body circumference frame of each first human body and a first head circumference frame of each first human head in the vehicle front row image;

[0120] In one embodiment, a human head classification detection network can be used to identify the first human body and the first head in the front row image of the vehicle, and obtain the first human body circumference frame of each first human body and the first head circumference frame of each first head in the front row image of the vehicle.

[0121] In one embodiment, the first human body circumference frame and the first head circumference frame are typically represented by their corresponding position coordinates. For example, if the first human body circumference frame is a rectangle, the position coordinates of the upper left vertex and the lower right vertex of the rectangle are typically used to represent the corresponding first human body circumference frame.

[0122] In one embodiment, the human head classification detection network can also determine the confidence of the first human body external frame of each first human body and the first head external frame of each first human head.

[0123] In one embodiment, the confidence level is used to indicate the credibility of the human body bounding box or the human head bounding box.

[0124] In one embodiment, a confidence threshold may be preset, and the first human body circumference frame and the first head circumference frame with a confidence lower than the preset confidence threshold are deleted, and only the credible first human body circumference frame and the first head circumference frame remain.

[0125] Step 402: Determine an IoU between each of the first human body bounding boxes and each of the first head bounding boxes, and match the first human body bounding boxes and the first head bounding boxes whose IoU ratios are not less than a preset IoU threshold to obtain a matching result.

[0126] In one embodiment, no matter the user's posture is sitting, standing or other postures (bending over, lowering the head, turning the head, etc.), the first human body external frame and the first head external frame generally have an overlapping area.

[0127] In one embodiment, the overlapping area of ​​the first human body external frame and the first head external frame is defined as the intersection area, and the total area occupied by the first human body external frame and the first head external frame is defined as the union area; the intersection-to-union ratio refers to the ratio of the intersection area to the union area.

[0128] In one embodiment, a larger IoU indicates greater overlap between the first person's circumscribed frame and the first head's circumscribed frame, i.e., a greater probability that the first person and the first head belong to the same user. Therefore, based on a large number of calibrations, the IoU of the first person's circumscribed frame and the first head's circumscribed frame for the same user in different postures can be used as a preset IoU threshold.

[0129] In one embodiment, in the matching result, the first human body and the first head corresponding to the first human body external bounding box and the first head external bounding box that can be matched belong to the same user.

[0130] Step 403, determining the first human body external frame of the position coordinates in the front seat according to the position coordinates of the first human body external frame in the vehicle;

[0131] In one embodiment, the position coordinates of the first human body external frame in the vehicle refer to the position coordinates of the center point of the first human body external frame in the vehicle.

[0132] In one embodiment, if the center point of the first person's external bounding box is at the front seat in the vehicle, it is determined that the first person's external bounding box is at the front seat.

[0133] Step 404 , selecting a first head external bounding box that matches the first head external bounding box at the front row seat from the first head external bounding boxes in the matching results, to obtain a first pre-selected head external bounding box;

[0134] In one embodiment, the first pre-selected head circumference frame refers to a first head circumference frame of a first head of a user in a front seat.

[0135] Step 405 : Determine a first target head bounding box whose position coordinates are in the target area according to the position coordinates of the first pre-selected head bounding box.

[0136] In one embodiment, it is necessary to determine whether to issue an anti-head collision alarm only when the first head external frame is located in the target area.

[0137] In one embodiment, when the first head circumference frame is not located in the target area, the probability of the head collision prevention alarm is 0.

[0138] In one embodiment, determining a first original alarm probability based on a comparison between the absolute head area of ​​the first target head bounding box and a first preset head area threshold includes:

[0139] Step 501: Obtain a first initial alarm probability based on a comparison between the absolute head area of ​​the first target head bounding box and a first preset head area threshold;

[0140] In one embodiment, when the absolute head area of ​​the first target person's head circumference frame is smaller than the first preset head area threshold, it indicates that the first target person's head is far away from the obstacle, and therefore there is no need to issue an anti-head collision alarm.

[0141] In one embodiment, when the absolute head area of ​​the first target person's head circumference frame is not less than a first preset head area threshold, it indicates that the first target person's head is close to an obstacle and an anti-collision alarm needs to be issued.

[0142] Step 502: identifying human bodies and human heads in the rear-seat image of the vehicle, and obtaining the number of human bodies in the area below the aisle and the number of human heads matching the human bodies in the area below the aisle;

[0143] In one embodiment, the number of human bodies in the area below the aisle may be one or more, which is not limited in the present disclosure.

[0144] In one embodiment, the number of all human bodies and the number of human heads matching the human bodies in the area below the aisle are obtained.

[0145] Step 503: determining a first alarm probability correction value based on a matching relationship between the number of human bodies and the number of heads in the area below the aisle;

[0146] In one embodiment, the first alarm probability correction value is used to correct the first original alarm probability.

[0147] In one embodiment, when the number of human bodies and heads in the area under the aisle is both 0, it means that the human bodies and heads in the area under the aisle are stretched forward too far, resulting in the rear image collector being unable to capture them, so the alarm probability needs to be increased.

[0148] In one embodiment, when the number of human bodies in the area under the aisle is not less than 1 and the number of heads is 0, it means that the human bodies and heads in the area under the aisle are not completely matched, that is, there is at least one real head that extends forward too far, so the alarm probability needs to be increased.

[0149] In one embodiment, when the number of human bodies and the number of heads in the area under the aisle are equal, it means that the human bodies and heads in the area under the aisle can be completely matched, that is, the human bodies and heads are all within the acquisition range of the rear row image collector, so the alarm probability needs to be reduced.

[0150] Step 504: Use the first alarm probability correction value to correct the first initial alarm probability to obtain a first original alarm probability.

[0151] In one embodiment, the first alarm probability correction value is added to the first initial alarm probability to obtain the first original alarm probability.

[0152] In one embodiment, the first initial alarm probability is corrected using the alarm probability correction values ​​in the aforementioned several cases to obtain the first original alarm probability.

[0153] In one embodiment, the accuracy of the anti-head collision alarm can be improved by correcting the first original alarm probability.

[0154] In one embodiment, the first preset head area threshold includes a front first preset head area threshold and a front second preset head area threshold, and the front first preset head area threshold is greater than the front second preset head area threshold.

[0155] In one embodiment, the first preset head area threshold in the front row refers to the area value of the corresponding head circumscribed frame on the front row image of the vehicle when the head is at the boundary of the risk area.

[0156] In one embodiment, the second preset head area threshold in the front row refers to the area value of the corresponding head circumscribed frame on the front row image of the vehicle when the head contacts an obstacle.

[0157] In one embodiment, because the front-row image of the vehicle is not captured by a depth image collector, the area of ​​the first head's bounding box corresponding to the first head in the front-row image of the vehicle is highly correlated with the distance between the first head and the obstacle. Specifically, the larger the area of ​​the first head's bounding box, the smaller the distance between the first head and the front-row image collector, and the larger the distance between the first head and the obstacle. Conversely, the smaller the area of ​​the first head's bounding box, the larger the distance between the first head and the front-row image collector, and the smaller the distance between the first head and the obstacle.

[0158] Obtaining a first initial alarm probability based on a comparison between the absolute head area of ​​the first target head circumscribed frame and a first preset head area threshold includes:

[0159] Step 601: multiply the length and width of the first target head bounding box to obtain the absolute head area of ​​the first target head bounding box;

[0160] In one embodiment, the first target head circumference frame may be a rectangle, and the absolute head area of ​​the first target head circumference frame refers to the product of the length and width of the target head circumference frame.

[0161] In one embodiment, the first target head circumference frame may be an ellipse, and the absolute head area of ​​the first target head circumference frame refers to the product of the major semi-axis, the minor semi-axis and pi of the first target head circumference frame.

[0162] Step 602 , determining whether the absolute head area of ​​the first target head bounding box is greater than the first preset head area threshold of the front row;

[0163] In one embodiment, if the judgment result is yes, that is, when the absolute head area is greater than the first preset head area threshold in the front row, it means that the distance between the first target person's head and the obstacle is relatively far, and there is no need to issue an anti-collision alarm. In this case, the original alarm probability is 0.

[0164] Step 603 , if the judgment result is negative, subtracting the first preset head area threshold in the front row from the absolute head area of ​​the first target head circumscribed frame to obtain a first difference value;

[0165] In one embodiment, if the judgment result is no, that is, when the absolute head area is not greater than the first preset head area threshold in the front row, it means that the distance between the first target person's head and the obstacle is relatively close, and it is necessary to further determine the first original alarm probability by combining the size comparison relationship between the absolute head area and the first preset head area threshold in the front row and the second preset head area threshold in the front row.

[0166] Step 604: Subtract the first preset head area threshold of the front row from the second preset head area threshold of the front row to obtain a second difference;

[0167] In one embodiment, the difference between the first preset head area threshold in the front row and the second preset head area threshold in the front row is used to indicate a distance interval within which a person's head may collide with an obstacle.

[0168] Step 605: taking the ratio of the first difference to the second difference as a first ratio;

[0169] Step 606: sum the first ratio and the front-row preset alarm probability correction coefficient to obtain a first initial alarm probability;

[0170] In one embodiment, the front row preset alarm probability correction coefficient refers to the alarm probability when the anti-collision alarm is issued. Generally, the alarm probability when the anti-collision alarm is not issued is 0, and the alarm probability when the anti-collision alarm is issued is 1.

[0171] In one embodiment, a proportional coefficient of the first ratio may also be preset, correspondingly:

[0172] The first ratio is added to the front-row preset alarm probability correction coefficient to obtain a first initial alarm probability, including:

[0173] multiplying the first ratio by a proportional coefficient to obtain a first product;

[0174] The first product is added to the front row preset alarm probability correction coefficient to obtain a first initial alarm probability; wherein the sum of the proportional coefficient and the front row preset alarm probability correction coefficient is equal to the alarm probability when the anti-collision head alarm is issued, that is, equal to 1.

[0175] In one embodiment, the first alarm probability correction value is determined based on a matching relationship between the number of human bodies and the number of heads in the area below the aisle;

[0176] When the number of human bodies in the area below the aisle is 0, determining the alarm probability correction value to be a first correction value;

[0177] When the number of human bodies in the area under the aisle is 1 and the number of heads matching the human bodies in the area under the aisle is 0, the alarm probability correction value is determined to be a second correction value; the first correction value is smaller than the second correction value.

[0178] In one embodiment, when the number of human bodies in the area under the aisle is 0, it means that the human bodies in the area under the aisle are extending forward too far, resulting in the inability of the rear image collector to capture them. Therefore, the alarm probability needs to be increased. For example, the first correction value can be 0.2. The specific value of the first correction value is not limited in this disclosure.

[0179] In one embodiment, when the number of human bodies in the area under the aisle is greater than the number of human heads, it means that the human bodies and human heads in the area under the aisle are not completely matched, that is, there is at least one real human head that extends forward too far, so the alarm probability needs to be increased. For example, the second correction value can be 0.3. The specific value of the second correction value is not limited in the present disclosure.

[0180] In one embodiment, determining at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head circumscribed frames and a second preset head area threshold includes:

[0181] Determining whether the occlusion area in the rear image of the vehicle is smaller than a preset occlusion area threshold;

[0182] In one embodiment, the blocked area refers to an area where the absolute head area of ​​the second target person's head circumference frame overlaps with the area that can be photographed by the rear-row image collector.

[0183] In one embodiment, the preset occlusion area threshold is half of the total area of ​​the rear row image.

[0184] In one embodiment, if the occlusion area in the rear row of images of the current frame is smaller than a preset occlusion area threshold, it can be determined that the video collector is not highly occluded at the current moment.

[0185] In one embodiment, if the occlusion area in the rear image of the current frame is larger than a preset occlusion area threshold, it can be determined that the video collector is highly occluded at the current moment, and the human head is very close to the entertainment screen, and an anti-head collision alarm is issued.

[0186] If the judgment result is yes, each of the second target head circumference frames is traversed, and at least one second initial alarm probability is obtained based on the comparison relationship between the absolute head area of ​​each of the second target head circumference frames and the second preset head area threshold.

[0187] In one embodiment, the implementation process of obtaining at least one second initial alarm probability based on the comparison relationship between the absolute head area of ​​each of the second target head external frame and the second preset head area threshold is as described in the above-mentioned steps 601 to 606.

[0188] In one embodiment, determining whether the blocked area in the rear-seat image of the vehicle is smaller than a preset blocked area threshold includes:

[0189] Obtaining a pixel value of each pixel in the rear image of the vehicle;

[0190] Comparing the pixel value of the pixel point with a preset pixel value, and determining the number of pixel points in the rear-row image of the current frame whose pixel value is greater than the preset pixel value;

[0191] In one embodiment, the preset pixel value may be any pixel value between 0 and 127. When the pixel value is 0, it indicates that the video collector is completely blocked and the rear image is black.

[0192] Determine whether the number of pixel points whose pixel values ​​are greater than the preset pixel value is less than the preset number of pixel points.

[0193] In one embodiment, the preset number of pixels may be half of the total number of pixels, or may be one-third of the total number of pixels, which is not limited in the present disclosure.

[0194] In one embodiment, if the number of pixel points whose pixel values ​​are greater than the preset pixel value is less than the preset number of pixel points, it means that the rear image collector is not blocked by the height of the human head. At this time, the human head is far away from the obstacle or is on one side of the obstacle, so there is no need to issue an anti-head collision alarm.

[0195] In one embodiment, if the number of pixel points whose pixel values ​​are greater than the preset pixel value is not less than the preset number of pixel points, it means that the rear image collector is highly blocked by the human head. At this time, the distance between the human head and the obstacle is close, so an anti-head collision alarm needs to be issued.

[0196] In one embodiment, the method for determining the probability of a head collision warning is as follows:

[0197] Step 701: Acquire multiple training images, where the training images are images of heads of people in an aisle at the boundary of the risk area.

[0198] In one embodiment, the boundaries of the risk area are typically defined by product managers.

[0199] In one embodiment, the product manager can designate relevant personnel to perform target area calibration in the aisle area. Only one relevant person can be in the aisle area at the same time, stretching their head to the boundary of the risk area, and the training image is collected by the front image collector in the car.

[0200] Based on this, in one embodiment, a training image only contains a human head.

[0201] In one embodiment, the boundary position of the risk area refers to the position where the anti-collision head alarm is triggered.

[0202] Step 702: Recognize the human head in the training image to obtain multiple human head bounding boxes in the training image; wherein each human head corresponds to a human head bounding box;

[0203] In one embodiment, a human head classification detection network is used to identify the human head in each training image to obtain a bounding box of the human head in each training image.

[0204] Step 703: Determine the center point position of each of the head circumference frames to obtain the center point positions of multiple head circumference frames;

[0205] In one embodiment, the center point position is used to indicate the center coordinate point of the circumscribed frame of the head.

[0206] In one embodiment, the center point position of the head circumference frame may be obtained through the position of the head circumference frame.

[0207] Step 704: Determine a closed area formed by connecting lines of multiple center points as a first target area.

[0208] In one embodiment, the size of the first target area is determined by the size of an area enclosed by all head positions at the boundary positions of the risk area in all training images.

[0209] Figure 8 This is a structural diagram of a device for determining the probability of an anti-collision head alarm provided by an embodiment of the present disclosure, such as Figure 8 As shown, the anti-collision head alarm probability determination device 800 includes:

[0210] An acquisition unit 801 is configured to acquire a front-row image and a rear-row image of the vehicle; the front-row image of the vehicle refers to an image acquired by a non-depth image collector disposed in front of the front seats, and the rear-seat image of the vehicle refers to an image acquired by a non-depth image collector disposed behind an obstacle;

[0211] A first recognition unit 802 is configured to recognize a first human head located in a first target area in the image of the front row of the vehicle, and a matching human head located in an upper aisle area, and obtain a first target head bounding box; the first target area is a risk area in the upper aisle area where a human head may collide with an obstacle; the aisle area includes an upper aisle area and a lower aisle area;

[0212] A first determining unit 803 is configured to determine a first original alarm probability based on a comparison between an absolute head area of ​​the first target head bounding box and a first preset head area threshold;

[0213] A second recognition unit 804 is configured to recognize a second human head located in a second target area in the rear-seat image of the vehicle and whose matched human body is located in the area below the aisle, and obtain at least one second target human head bounding box; the second target area is a risk area in the area below the aisle where the human head may collide with an obstacle;

[0214] A second determining unit 805 is configured to determine at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head bounding boxes and a second preset head area threshold;

[0215] The third determining unit 806 is configured to determine a target alarm probability as a maximum value between the first original alarm probability and the at least one second original alarm probability.

[0216] In one embodiment, the first identification unit 802 is specifically configured to:

[0217] Identifying a first human body and a first human head in the image of the front row of the vehicle, and obtaining a first human body circumference frame of each first human body and a first head circumference frame of each first human head in the image of the front row of the vehicle;

[0218] determining an intersection-over-union (IoU) ratio between each of the first human body circumference frames and each of the first head circumference frames, and matching the first human body circumference frames with the first head circumference frames whose IoU ratios are not less than a preset IoU threshold to obtain a matching result;

[0219] Determine, based on the position coordinates of the first human body external bounding box in the vehicle, the first human body external bounding box of the position coordinates in the front seat;

[0220] Filtering out a first head circumference frame that matches the first head circumference frame of the front seat from the first head circumference frames of the matching results, obtaining a first pre-selected head circumference frame;

[0221] According to the position coordinates of the first pre-selected head circumscribed frame, a first target head circumscribed frame having the position coordinates in the target area is determined.

[0222] In one embodiment, the first determining unit 803 is specifically configured to:

[0223] Obtaining a first initial alarm probability based on a comparison between the absolute head area of ​​the first target head bounding box and a first preset head area threshold;

[0224] Identify human bodies and human heads in the rear-seat image of the vehicle, and obtain the number of human bodies in the area under the aisle and the number of human heads matching the human bodies in the area under the aisle;

[0225] Determining a first alarm probability correction value based on a matching relationship between the number of human bodies and the number of heads in the area below the aisle;

[0226] The first initial alarm probability is corrected using the first alarm probability correction value to obtain a first original alarm probability.

[0227] In one embodiment, the first preset head area threshold includes a front row first preset head area threshold and a front row second preset head area threshold, wherein the front row first preset head area threshold is greater than the front row second preset head area threshold;

[0228] In one embodiment, the first determining unit 803 is specifically configured to:

[0229] Multiplying the length and width of the first target head circumference frame to obtain the absolute head area of ​​the first target head circumference frame;

[0230] Determine whether the absolute head area of ​​the first target head frame is greater than the first preset head area threshold of the front row; if the judgment result is no, then

[0231] Subtracting the first preset head area threshold in the front row from the absolute head area of ​​the first target head circumscribed frame to obtain a first difference value;

[0232] Subtracting the first preset head area threshold of the front row from the second preset head area threshold of the front row to obtain a second difference;

[0233] taking the ratio of the first difference to the second difference as a first ratio;

[0234] The first ratio is added to the front-row preset alarm probability correction coefficient to obtain a first initial alarm probability.

[0235] In one embodiment, the first determining unit 803 is specifically configured to:

[0236] When the number of human bodies in the area below the aisle is 0, determining the alarm probability correction value to be a first correction value;

[0237] When the number of human bodies in the area under the aisle is 1 and the number of heads matching the human bodies in the area under the aisle is 0, the alarm probability correction value is determined to be a second correction value; the first correction value is smaller than the second correction value.

[0238] In one embodiment, the second determining unit 805 is specifically configured to:

[0239] Determining whether the occlusion area in the rear image of the vehicle is smaller than a preset occlusion area threshold;

[0240] If the judgment result is yes, each of the second target head circumference frames is traversed, and at least one second initial alarm probability is obtained based on the comparison relationship between the absolute head area of ​​each of the second target head circumference frames and the second preset head area threshold.

[0241] In one embodiment, the second determining unit 805 is specifically configured to:

[0242] Obtaining a pixel value of each pixel in the rear image of the vehicle;

[0243] Comparing the pixel value of the pixel point with a preset pixel value, and determining the number of pixel points in the rear-row image of the current frame whose pixel value is greater than the preset pixel value;

[0244] Determine whether the number of pixel points whose pixel values ​​are greater than the preset pixel value is less than the preset number of pixel points.

[0245] In one embodiment, the anti-collision head alarm probability determination device 800 further includes a first target area determination unit, which is configured to:

[0246] Acquire multiple training images, wherein the training images are images of heads of people in the aisle at the boundary position of the risk area;

[0247] Recognize the human heads in the training image to obtain a plurality of human head bounding boxes in the training image; wherein one human head corresponds to one human head bounding box;

[0248] Determine the center point position of each of the head circumference frames to obtain the center point positions of multiple head circumference frames;

[0249] A closed area formed by connecting lines of the plurality of center points is determined as a first target area.

[0250] It should be noted that the above explanation of the method embodiment is also applicable to the device of this embodiment, and the principles are the same, which is not limited in this embodiment.

[0251] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device and a non-transitory computer-readable storage medium storing computer instructions.

[0252] Specifically, an embodiment of the present disclosure provides an electronic device, including:

[0253] at least one processor; and

[0254] a memory communicatively connected to at least one processor; wherein,

[0255] The memory stores instructions that can be executed by at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the aforementioned method for determining the probability of anti-collision head alarm.

[0256] An embodiment of the present disclosure provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the steps of the aforementioned method for determining the probability of an anti-collision head alarm.

[0257] Figure 9 A schematic block diagram of an example electronic device 900 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, in-vehicle devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0258] like Figure 9 As shown, the electronic device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in a ROM (Read-Only Memory) 902 or a computer program loaded from a storage unit 908 into a RAM (Random Access Memory) 903. In the RAM 903, various programs and data required for the operation of the device 900 can also be stored. The computing unit 801, ROM 902, and RAM 903 are connected to each other via a bus 904. An I / O (Input / Output) interface 905 is also connected to the bus 904.

[0259] Various components in the device 900 are connected to the I / O interface 905, including an input unit 904, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, an optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the device 900 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0260] The computing unit 901 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a CPU (Central Processing Unit), a GPU (Graphic Processing Unit), various specialized AI (Artificial Intelligence) computing chips, various computing units that run machine learning model algorithms, a DSP (Digital Signal Processor), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the various methods and processes described above, such as the method for determining the probability of a head collision warning. For example, in some embodiments, the method for determining the probability of a head collision warning warning can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the method described above can be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to execute the aforementioned anti-collision head alarm probability determination method in any other appropriate manner (for example, by means of firmware).

[0261] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, FPGAs (Field Programmable Gate Arrays), ASICs (Application-Specific Integrated Circuits), ASSPs (Application-Specific Standard Products), SOCs (System on Chips), CPLDs (Complex Programmable Logic Devices), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0262] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0263] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, RAM, ROM, EPROM (Electrically Programmable Read-Only-Memory) or flash memory, optical fiber, CD-ROM (Compact Disc Read-Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0264] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0265] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A method for determining the probability of an anti-collision head alarm, characterized in that: include: Acquire a front image and a rear image of the vehicle; The vehicle front row image refers to an image collected by a non-depth image collector arranged in front of the front seat, and the vehicle rear row image refers to an image collected by a non-depth image collector arranged behind an obstacle; Identifying a first human head located in a first target area in the image of the front row of the vehicle, and a matching human body located in an area above the aisle, to obtain a first target human head circumscribed frame; The first target area refers to the risk area in the aisle upper area where a person's head may hit an obstacle; the aisle area includes the aisle upper area and the aisle lower area; determining a first original alarm probability based on a comparison between an absolute head area of ​​the first target head bounding box and a first preset head area threshold; Identifying a second human head located in a second target area in the rear-seat image of the vehicle, whose matched human body is located in the area below the aisle, and obtaining at least one second target human head circumscribed frame; The second target area refers to a risk area under the aisle where a person's head may hit an obstacle; determining at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head circumscribed frames and a second preset head area threshold; The target alarm probability is determined to be the maximum value of the first original alarm probability and the at least one second original alarm probability.

2. The method according to claim 1, characterized in that The step of identifying a first human head located in a first target area in the front row image of the vehicle and having a matching human body located in an area above the aisle, and obtaining a first target human head external frame, includes: Identifying a first human body and a first human head in the image of the front row of the vehicle, and obtaining a first human body circumference frame of each first human body and a first head circumference frame of each first human head in the image of the front row of the vehicle; determining an intersection-over-union (IoU) ratio between each of the first human body circumference frames and each of the first head circumference frames, and matching the first human body circumference frames with the first head circumference frames whose IoU ratios are not less than a preset IoU threshold to obtain a matching result; Determine, based on the position coordinates of the first human body external bounding box in the vehicle, the first human body external bounding box of the position coordinates in the front seat; Filtering out a first head circumference frame that matches the first head circumference frame of the front seat from the first head circumference frames of the matching results, obtaining a first pre-selected head circumference frame; According to the position coordinates of the first pre-selected head circumscribed frame, a first target head circumscribed frame having the position coordinates in the target area is determined.

3. The method according to claim 1, characterized in that The determining of a first original alarm probability based on a comparison between the absolute head area of ​​the first target head circumscribed frame and a first preset head area threshold includes: Obtaining a first initial alarm probability based on a comparison between the absolute head area of ​​the first target head bounding box and a first preset head area threshold; Identify human bodies and human heads in the rear-seat image of the vehicle, and obtain the number of human bodies in the area under the aisle and the number of human heads matching the human bodies in the area under the aisle; Determining a first alarm probability correction value based on a matching relationship between the number of human bodies and the number of heads in the area below the aisle; The first initial alarm probability is corrected using the first alarm probability correction value to obtain a first original alarm probability.

4. The method according to claim 3, characterized in that The first preset head area threshold comprises a front row first preset head area threshold and a front row second preset head area threshold, wherein the front row first preset head area threshold is greater than the front row second preset head area threshold; Obtaining a first initial alarm probability based on a comparison between the absolute head area of ​​the first target head circumscribed frame and a first preset head area threshold includes: Multiplying the length and width of the first target head circumference frame to obtain the absolute head area of ​​the first target head circumference frame; Determining whether the absolute head area of ​​the first target person's head circumscribed frame is greater than the first preset head area threshold of the front row; If the judgment result is no, then Subtracting the first preset head area threshold in the front row from the absolute head area of ​​the first target head circumscribed frame to obtain a first difference value; Subtracting the first preset head area threshold of the front row from the second preset head area threshold of the front row to obtain a second difference; taking the ratio of the first difference to the second difference as a first ratio; The first ratio is added to the front-row preset alarm probability correction coefficient to obtain a first initial alarm probability.

5. The method according to claim 3, characterized in that determining a first alarm probability correction value according to a matching relationship between the number of human bodies and the number of heads in the area below the aisle; When the number of human bodies in the area below the aisle is 0, determining the alarm probability correction value to be a first correction value; When the number of human bodies in the area under the aisle is 1 and the number of heads matching the human bodies in the area under the aisle is 0, the alarm probability correction value is determined to be a second correction value; the first correction value is smaller than the second correction value.

6. The method according to claim 1, wherein The determining of at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head circumscribed frames and a second preset head area threshold comprises: Determining whether the occlusion area in the rear image of the vehicle is smaller than a preset occlusion area threshold; If the judgment result is yes, each of the second target head circumference frames is traversed, and at least one second initial alarm probability is obtained based on the comparison relationship between the absolute head area of ​​each of the second target head circumference frames and the second preset head area threshold.

7. The method according to claim 6, characterized in that The determining whether the blocked area in the rear image of the vehicle is smaller than a preset blocked area threshold includes: Obtaining a pixel value of each pixel in the rear image of the vehicle; Comparing the pixel value of the pixel point with a preset pixel value, and determining the number of pixel points in the rear-row image of the current frame whose pixel value is greater than the preset pixel value; Determine whether the number of pixel points whose pixel values ​​are greater than the preset pixel value is less than the preset number of pixel points.

8. The method according to claim 1, characterized in that The method of identifying the area in front of the first target area in the front row image of the vehicle includes: Acquire multiple training images, wherein the training images are images of heads of people in the aisle at the boundary position of the risk area; Recognize the human heads in the training image to obtain a plurality of human head bounding boxes in the training image; wherein one human head corresponds to one human head bounding box; Determine the center point position of each of the head circumference frames to obtain the center point positions of multiple head circumference frames; A closed area formed by connecting lines of the plurality of center points is determined as a first target area.

9. A device for determining the probability of an anti-collision head alarm, characterized in that: include: An acquisition unit, configured to acquire a front row image and a rear row image of the vehicle; The vehicle front row image refers to an image collected by a non-depth image collector arranged in front of the front seat, and the vehicle rear row image refers to an image collected by a non-depth image collector arranged behind an obstacle; a first recognition unit, configured to recognize a first human head located in a first target area in the image of the front row of the vehicle, and a matching human body located in an area above the aisle, and obtain a first target human head circumscribed frame; The first target area refers to the risk area in the aisle upper area where a person's head may hit an obstacle; the aisle area includes the aisle upper area and the aisle lower area; a first determining unit, configured to determine a first original alarm probability based on a comparison between an absolute head area of ​​the first target head bounding box and a first preset head area threshold; a second recognition unit, configured to recognize a second human head located in a second target area in the image of the rear row of the vehicle, whose matched human body is located in the area below the aisle, and obtain at least one second target human head circumscribed frame; The second target area refers to a risk area under the aisle where a person's head may hit an obstacle; a second determining unit, configured to determine at least one second original alarm probability based on a comparison between the absolute head area of ​​each of the second target head circumscribed frames and a second preset head area threshold; The third determining unit is configured to determine a target alarm probability as a maximum value between the first original alarm probability and the at least one second original alarm probability.

10. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

11. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 8.

Images