Safety belt detection method, device, equipment, medium and product

Abstract

The invention relates to the technical field of vehicles, and provides a safety belt detection method, device and equipment, a medium and a product. The safety belt detection method comprises the steps of obtaining a passenger image collected by a depth camera; calculating the area of a safety belt area according to pixel information and depth information of the safety belt area in the passenger image; based on the area of the safety belt area, a detection result representing the safety belt wearing state is obtained. According to the safety belt detection scheme, whether the passenger wears the safety belt accurately or not can be effectively detected, the situation that the safety belt is not worn in place due to excessive wearing, insufficient wearing and the like is recognized, and it is ensured that the safety belt effectively plays a safety protection role.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and in particular to a safety belt detection method, device, equipment, medium and product. Background Art

[0002] Seat belts are active safety equipment of vehicles. When the vehicle collides or other accidents occur, seat belts can effectively reduce the impact and inertia of the impact force and the damage to the occupants, and play a vital role in protecting the safety of the occupants.

[0003] In order to ensure that seat belts can effectively play a role in safety protection, how to detect whether passengers have worn seat belts accurately and identify situations where seat belts are not worn properly is a problem that needs to be solved urgently.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] In view of this, the present invention provides a seat belt detection method, device, equipment, medium and product, which can effectively detect whether the occupant is wearing the seat belt correctly, identify the situation of excessive wearing, insufficient wearing and other improper wearing of the seat belt, and ensure that the seat belt effectively plays a safety protection role.

[0006] According to one aspect of the present invention, there is provided a seat belt detection method, comprising: obtaining an image of an occupant captured by a depth camera; calculating the area of ​​the seat belt area in the occupant image based on pixel information and depth information of the seat belt area; and obtaining a detection result characterizing the wearing status of the seat belt based on the area of ​​the seat belt area.

[0007] In some embodiments, the pixel information and depth information of the seat belt area include the number of pixel units in the seat belt area, the coordinates of each pixel unit and the depth of each pixel unit; the calculating the area of ​​the seat belt area includes: calculating the depth size of each pixel unit in the seat belt area according to the depth relationship between the pixel units in the seat belt area; calculating the pixel size of each pixel unit in the seat belt area based on the depth size and coordinates of each pixel unit in the seat belt area according to the conversion relationship between the pixel coordinate system and the world coordinate system; calculating the area of ​​the seat belt area according to the number of pixel units in the seat belt area and the pixel size of each pixel unit.

[0008] In some embodiments, calculating the depth dimension of each pixel unit of the seat belt area includes: obtaining a pixel unit sequence of the seat belt area; calculating the depth difference between adjacent pixel units of the pixel unit sequence according to the depth of each pixel unit of the seat belt area, and obtaining the depth dimension of each pixel unit of the seat belt area.

[0009] In some embodiments, the calculating the pixel size of each pixel unit of the safety belt area includes: DeltaX i =(u i -cx)*DeltaZ i / fx; DeltaY i =(v i -cy)*DeltaZ i / fy; where (DeltaX i ,DeltaY i ) is the pixel size of the i-th pixel unit in the safety belt area, DeltaZ i is the depth dimension of the i-th pixel unit, (u i ,v i ) are the coordinates of the i-th pixel unit, (fx, fy) are the focal length of the depth camera, and (cx, cy) are the coordinates of the center point of the depth camera.

[0010] In some embodiments, calculating the area of ​​the safety belt zone includes:

[0011] Wherein, Sseatbelt is the area of ​​the seat belt area, and pixelNums is the number of pixel units in the seat belt area.

[0012] In some embodiments, after obtaining the occupant image captured by the depth camera, the method further includes: processing the occupant image through a semantic segmentation neural network to obtain pixel information and depth information of the seat belt area in the occupant image.

[0013] In some embodiments, obtaining a detection result characterizing a seat belt wearing state based on the area of ​​the seat belt region includes: obtaining a detection result characterizing a seat belt wearing state based on a relationship between the area of ​​the seat belt region and a threshold range.

[0014] In some embodiments, obtaining a detection result characterizing a seat belt wearing status based on the area of ​​the seat belt region includes: inputting the area of ​​the seat belt region and the occupant image into a convolutional neural network to obtain a detection result characterizing a seat belt wearing status.

[0015] In some embodiments, the detection result representing the wearing status of the seat belt includes any of the following items: the seat belt is fully worn; the seat belt is over-worn; the seat belt is under-worn.

[0016] According to another aspect of the present invention, there is provided a seat belt detection device for implementing the seat belt detection method described in any of the above embodiments, the seat belt detection device comprising: an occupant image acquisition module, configured to obtain an occupant image captured by a depth camera; a seat belt area acquisition module, configured to calculate the area of ​​the seat belt area based on pixel information and depth information of the seat belt area in the occupant image; and a detection result acquisition module, configured to obtain a detection result characterizing the wearing status of the seat belt based on the area of ​​the seat belt area.

[0017] According to another aspect of the present invention, there is provided an electronic device, comprising: a processor; a memory, wherein the memory stores executable instructions; wherein when the executable instructions are executed by the processor, the seat belt detection method described in any of the above embodiments is implemented.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided for storing a program, wherein the program, when executed by a processor, implements the seat belt detection method described in any of the above embodiments.

[0019] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the seat belt detection method described in any of the above embodiments is implemented.

[0020] The beneficial effects of the present invention compared with the prior art include at least:

[0021] By collecting the image of the occupant through the depth camera, and using the high-precision three-dimensional information acquisition capability of the depth camera, the pixel information and depth information of the seat belt in different wearing states can be collected; the seat belts in different wearing states include seat belts that are completely worn on the appropriate body parts of the occupant, seat belts that are only partially exposed due to being blocked by objects such as the occupant / seat, seat belts that are excessively pulled out and hung on other objects, etc.; the different wearing states of these seat belts correspond to the accurate wearing of the seat belts and the inadequate wearing of the seat belts, excessive wearing of the seat belts, etc. According to the pixel information of the XY plane and the depth information of the Z axis of the seat belt area in the occupant image, based on the conversion relationship between the pixel coordinate system and the world coordinate system, the area of ​​the seat belt area corresponding to the actual scene can be calculated. Furthermore, according to the corresponding relationship between the area of ​​the seat belt area and the different wearing states of the seat belt, the detection result representing the wearing state of the seat belt is obtained.

[0022] Therefore, the seat belt detection scheme of the present invention can effectively detect whether the occupant is wearing the seat belt correctly, identify the situation of over-wearing, under-wearing and other improper wearing of the seat belt, and ensure that the seat belt effectively plays a safety protection role.

[0023] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present invention, and together with the specification are used to explain the principles of the present invention. Obviously, the accompanying drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0025] Figure 1 A schematic diagram showing the steps of a seat belt detection method according to an embodiment of the present invention;

[0026] Figure 2 A schematic diagram showing the steps of calculating the area of ​​a safety belt zone in an embodiment of the present invention;

[0027] Figure 3 A schematic diagram of a module of a seat belt detection device according to an embodiment of the present invention is shown;

[0028] Figure 4 A schematic structural diagram of an electronic device in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0029] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments described herein. Instead, these embodiments are provided to make the present invention more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art.

[0030] The accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0031] In addition, the processes shown in the drawings are only exemplary and do not necessarily include all the steps. For example, some steps may be decomposed, some steps may be combined or partially combined, and the actual execution order may be changed according to the actual situation.

[0032] It should be noted that, in the absence of conflict, the embodiments of the present invention and features in different embodiments may be combined with each other.

[0033] Figure 1 The main steps of the seat belt testing method are shown in Figure 1. Figure 1 As shown, the seat belt detection method provided by the embodiment of the present invention includes the following steps.

[0034] S110, obtaining an image of the occupant captured by a depth camera.

[0035] The depth camera can be installed in front of the cabin or other suitable positions to collect images of the occupants in the cabin. The depth camera can use a time-of-flight (ToF) camera. The ToF camera can emit continuous light pulses (such as near-infrared light pulses) to the occupant position area, and use sensors to receive light pulses reflected by objects. By detecting the flight (round-trip) time of the light pulses, the distance of the pixel unit corresponding to the object relative to the ToF camera, that is, the depth, is obtained, and the occupant image is generated accordingly. Of course, the depth camera can also use a structured light camera, a binocular stereo vision camera, etc., and is not limited to a ToF camera.

[0036] The occupant image includes pixel information on the XY plane and depth information on the Z axis, wherein the pixel information includes the number and coordinates of pixel units, and the depth information includes the depth of the pixel units.

[0037] By collecting images of occupants with a depth camera and utilizing the high-precision three-dimensional information acquisition capability of the depth camera, pixel information and depth information of seat belts in different wearing states can be collected; seat belts in different wearing states include seat belts that are completely worn on appropriate parts of the occupant's body, seat belts that are blocked by objects such as the occupant / seat and only partially exposed, seat belts that are excessively pulled out and hung on other objects, etc.; these different wearing states of seat belts correspond to accurate wearing of the seat belts, insufficient wearing of the seat belts, excessive wearing of the seat belts, and other inadequate wearing conditions.

[0038] Among them, take the three-point seat belt (including shoulder belt and waist belt) commonly used in vehicle seats as an example. The seat belt is fully worn on the appropriate body part of the occupant, that is, the seat belt is worn on the chest and hips of the occupant; the seat belt is blocked by the occupant / seat and other objects and only partially exposed, including: the shoulder belt of the seat belt is pressed by the occupant, the waist belt of the seat belt is blocked by the occupant, the shoulder belt and waist belt are blocked by the occupant, the shoulder belt is hung on the back of the seat, etc.; the seat belt is pulled out too much and hung on other objects, such as the shoulder belt is pulled out too much and fixed with a clip or wrapped around the back of the seat, etc.

[0039] Furthermore, in some implementations, the timing for the depth camera to capture the image of the occupant may be: in response to a seat seating signal and a seat belt locking signal, the depth camera is triggered to capture the image of the occupant. The seat seating signal may be detected by a sensor such as a pressure sensor installed on the seat and / or a camera (such as the depth camera) installed in the cockpit. The seat belt locking signal is detected by a sensor inside the seat belt system, indicating that the seat belt is pulled out and the locking tongue is inserted into the buckle; the seat belt locking detection function is usually provided by the seat belt system and is equipped in most vehicles. Based on the existing seat belt locking detection function of the vehicle, the present invention can further detect whether the occupant is wearing the seat belt accurately, and identify the situation of improper wearing caused by "pretending" wearing behaviors such as over-wearing and under-wearing of the seat belt.

[0040] The condition for triggering the depth camera to collect the occupant image may not include the seat belt locking signal, so the seat belt detection method of the present invention can also detect the situation where the seat belt is not locked.

[0041] In addition, when the depth camera is triggered to capture occupant images at least in response to a seat occupancy signal, the depth camera can be controlled to only capture occupant images in a target seat area (i.e., the area where the seats corresponding to the seat occupancy signal are located). For example, the depth camera can be controlled to only emit light pulses to the target seat area, so as to capture occupant images in a targeted manner and reduce interference.

[0042] In some implementations, the depth camera may also capture images of occupants in the entire seating area, without being limited to the above description.

[0043] S120, calculating the area of ​​the seat belt region according to the pixel information and depth information of the seat belt region in the occupant image.

[0044] The seat belt area can be obtained by cutting / segmenting the occupant image. According to the pixel information of the seat belt area in the XY plane and the depth information of the Z axis, based on the conversion relationship between the pixel coordinate system and the world coordinate system, the area of ​​the seat belt area corresponding to the actual scene can be calculated.

[0045] S130, obtaining a detection result representing a wearing state of the seat belt based on the area of ​​the seat belt region.

[0046] The detection results representing the wearing state of the seat belt may include any of the following items: the seat belt is fully worn; the seat belt is over-worn; the seat belt is under-worn. According to the corresponding relationship between the area of ​​the seat belt area and the different wearing states of the seat belt: when the seat belt is fully worn, the area of ​​the seat belt area is within the normal range; when the seat belt is over-worn, the area of ​​the seat belt area is greater than the normal range; when the seat belt is under-worn, the area of ​​the seat belt area is less than the normal range.

[0047] The normal range can be a set threshold, or can vary according to factors such as the type of occupant (such as adults, children) and the area of ​​the occupant area obtained based on the occupant image. The type of occupant and the area of ​​the occupant area can be obtained by processing the occupant image using a convolutional neural network; the area of ​​the occupant area can also be calculated in the same way as the area of ​​the seat belt area, that is, the area of ​​the occupant area is calculated based on the pixel information and depth information of the occupant area in the occupant image.

[0048] The detection result representing the wearing status of the seat belt is obtained based on the area of ​​the seat belt region, without the need for complex image analysis, thus saving computing resources and improving the detection speed.

[0049] Therefore, the seat belt detection scheme of the present invention can efficiently and accurately detect whether the occupant is wearing the seat belt correctly, identify situations such as over-wearing, under-wearing, etc., and ensure that the seat belt effectively plays a safety protection role.

[0050] Furthermore, when it is detected that the seat belt is not worn properly, a reminder can be issued through the corresponding seat, the vehicle's multimedia system, etc., to prompt the corresponding occupant to wear the seat belt correctly.

[0051] In some embodiments, after obtaining the occupant image captured by the depth camera, it also includes: processing the occupant image through a semantic segmentation neural network to obtain pixel information and depth information of the seat belt area in the occupant image.

[0052] After training, the semantic segmentation neural network can associate each pixel in the occupant image with a specific category label (specifically, the category label of the seat belt in this embodiment), so as to segment the image of the seat belt area from the occupant image. The output of the semantic segmentation neural network includes the image of the seat belt area, the pixel information of the seat belt area, and the depth information.

[0053] The semantic segmentation neural network can specifically adopt suitable network models such as fully convolutional neural network, SegNet, ENet, etc.

[0054] Figure 2 Illustrate the main steps in calculating the area of ​​the safety belt area; combine Figure 2 As shown, in some embodiments, the pixel information and depth information of the seat belt area include the number of pixel units in the seat belt area, the coordinates of each pixel unit, and the depth of each pixel unit; calculating the area of ​​the seat belt area includes:

[0055] S210, calculating the depth size of each pixel unit in the safety belt area according to the depth relationship between the pixel units in the safety belt area.

[0056] The depth dimension of a pixel unit refers to the actual depth of the pixel unit in the Z-axis direction corresponding to the actual scene, and can be obtained by subtracting the depths between the pixel units according to the depth relationship between the pixel units.

[0057] Specifically, calculating the depth size of each pixel unit in the seat belt area includes: obtaining a pixel unit sequence in the seat belt area; calculating the depth difference between adjacent pixel units in the pixel unit sequence according to the depth of each pixel unit in the seat belt area, and obtaining the depth size of each pixel unit in the seat belt area.

[0058] For example, the pixel units of the safety belt area are arranged in sequence to obtain a pixel unit sequence of "pixel unit a, pixel unit b, pixel unit c..." When calculating the depth difference between adjacent pixel units, the depth size of pixel unit a can be defaulted to 0 or a suitable value such as a unit depth size, and the depth sizes of the remaining pixel units are obtained by subtracting the depth of the previous pixel unit from the depth of the pixel unit. In this way, the depth size of each pixel unit in the safety belt area is obtained.

[0059] S220 , calculating the pixel size of each pixel unit in the safety belt area according to a conversion relationship between the pixel coordinate system and the world coordinate system and based on the depth size and coordinates of each pixel unit in the safety belt area.

[0060] The pixel size of a pixel unit refers to the actual size of the pixel unit in the XY plane corresponding to the actual scene.

[0061] Specifically, the pixel size of the pixel unit in the safety belt area can be calculated using the following formula:

[0062] DeltaX i =(u i -cx)*DeltaZ i / fx, DeltaY i =(v i -cy)*DeltaZ i / fy;

[0063] Among them, (DeltaX i ,DeltaY i ) is the pixel size of the i-th pixel unit in the safety belt area, DeltaX i is the size of the i-th pixel unit in the X-axis direction, DeltaY i is the size of the i-th pixel unit in the Y-axis direction, DeltaZ i is the depth dimension of the i-th pixel unit in the Z-axis direction; (u i ,v i ) is the coordinate of the i-th pixel unit based on the pixel coordinate system; (fx, fy) is the focal length of the depth camera, fx is the focal length in the X-axis direction, and fy is the focal length in the Y-axis direction; (cx, cy) is the center point coordinate of the depth camera, that is, the principal point coordinate. cx, cy, fx, and fy are fixed parameters of the depth camera and can be obtained through camera calibration.

[0064] S230: Calculate the area of ​​the seat belt area according to the number of pixel units in the seat belt area and the pixel size of each pixel unit.

[0065] Specifically, the area of ​​the safety belt area can be calculated using the following formula:

[0066]

[0067] Among them, Sseatbelt is the area of ​​the seat belt area, and pixelNums is the number of pixel units in the seat belt area.

[0068] In some embodiments, obtaining a detection result characterizing the wearing state of the seat belt based on the area of ​​the seat belt region includes: obtaining a detection result characterizing the wearing state of the seat belt based on a relationship between the area of ​​the seat belt region and a threshold range.

[0069] Specifically, when the area of ​​the seat belt area is within the range of the threshold interval, the detection result is that the seat belt is fully worn, that is, the seat belt is accurately worn; when the area of ​​the seat belt area exceeds the threshold interval (greater than or less than the threshold interval), the detection result of the seat belt not being properly worn is obtained. The detection result of the seat belt not being properly worn may further include excessive wearing of the seat belt (the area of ​​the seat belt area is greater than the threshold interval) and insufficient wearing of the seat belt (the area of ​​the seat belt area is less than the threshold interval).

[0070] In some embodiments, based on the area of ​​the seat belt region, a detection result characterizing the wearing status of the seat belt is obtained, including: inputting the area of ​​the seat belt region and the occupant image into a convolutional neural network to obtain a detection result characterizing the wearing status of the seat belt.

[0071] After training, the convolutional neural network can determine the area ratio of the seat belt area in the occupant area of ​​the occupant image based on the area of ​​the seat belt area and the occupant image, and thereby determine the detection result characterizing the seat belt wearing status.

[0072] Among them, the convolutional neural network can be trained to identify the occupant area in the occupant image through image feature extraction and analysis, and then obtain the area of ​​the occupant area, for example, by inferring the area of ​​the occupant area through the pixel information surrounded by the edge of the occupant area.

[0073] After training, the convolutional neural network can also obtain the occupant type (adult, child, etc.) by processing the occupant image, and then determine the detection result characterizing the seat belt wearing status based on the relationship between the area of ​​the seat belt area and the specific threshold interval corresponding to the occupant type (for example, when the occupant type is an adult, it corresponds to a larger threshold interval, and when the occupant type is a child, it corresponds to a smaller threshold interval).

[0074] The seat belt detection method described in the above-mentioned embodiments of the present invention can be implemented by an independent processing device installed in the vehicle, or the functional modules, processing units, program products, etc. that implement the above-mentioned seat belt detection method can be integrated into the existing processing device of the vehicle. It can also be implemented by a remote, cloud-based processing device and communicated with relevant components of the vehicle to realize the detection of the wearing status of the seat belt.

[0075] The embodiment of the present invention further provides a seat belt detection device, which can be used to implement the seat belt detection method described in any of the above embodiments. The features and principles of the seat belt detection method described in any of the above embodiments can be applied to the following seat belt detection device embodiments.

[0076] Figure 3 The main modules of the seat belt detection device are shown in figure. Figure 3 As shown, the seat belt detection device 300 provided by the embodiment of the present invention includes: an occupant image acquisition module 310, configured to obtain an occupant image captured by a depth camera; a seat belt area acquisition module 320, configured to calculate the area of ​​the seat belt area according to the pixel information and depth information of the seat belt area in the occupant image; a detection result acquisition module 330, configured to obtain a detection result representing the wearing status of the seat belt based on the area of ​​the seat belt area.

[0077] Furthermore, the seat belt detection device 300 may also include modules for implementing other process steps of the above-mentioned seat belt detection method embodiments. The specific principles of each module can refer to the description of the above-mentioned seat belt detection method embodiments, and will not be repeated here.

[0078] The seat belt detection device of the present invention can be independently deployed in the vehicle, or the relevant functional modules can be integrated into the existing processing device of the vehicle. It can also be deployed remotely, in the cloud and communicated with relevant components of the vehicle to realize the detection of the wearing status of the seat belt.

[0079] The seat belt detection device of the present invention can efficiently and accurately detect whether the occupant has correctly worn the seat belt, identify situations where the seat belt is over-worn, under-worn, or otherwise improperly worn, and ensure that the seat belt effectively plays a safety protection role.

[0080] An embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein executable instructions are stored in the memory, and when the executable instructions are executed by the processor, the seat belt detection method described in any of the above embodiments is implemented.

[0081] The electronic device of the present invention can be independently deployed in the vehicle, or integrated into the existing processing device of the vehicle, or can be deployed remotely, in the cloud and communicated with relevant components of the vehicle to detect the wearing status of the seat belt.

[0082] The electronic device of the present invention can efficiently and accurately detect whether the occupant has correctly worn the seat belt, identify situations where the seat belt is over-worn, under-worn, or otherwise improperly worn, and ensure that the seat belt effectively plays a safety protection role.

[0083] Figure 4 The main structure of the electronic equipment is shown in Figure 1. Figure 4 As shown, the electronic device 400 is in the form of a general computing device, and the components of the electronic device 400 include but are not limited to: at least one processor 410, at least one memory 420, a bus 430 connecting different platform components (including the memory 420 and the processor 410), etc.

[0084] The memory 420 stores program codes, which can be executed by the processor 410, so that the processor 410 performs the steps of the seat belt detection method described in any of the above embodiments. The memory 420 may include a readable medium in the form of a volatile storage unit, and the memory 420 may also include a program / utility with one or more program modules, such program modules include but are not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include the implementation of a network environment.

[0085] Bus 430 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0086] The electronic device 400 can also communicate with one or more external devices, and the external device can be one or more of the devices such as keyboard, pointing device, bluetooth device, etc. These external devices enable the user to communicate interactively with the electronic device 400. The electronic device 400 can also communicate with one or more other computing devices, and the computer device shown includes a router and a modem. This communication can be carried out through an input / output (I / O) interface. In addition, the electronic device 400 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and / or a public network, such as the Internet) through a network adapter.

[0087] An embodiment of the present invention further provides a computer-readable storage medium for storing a program, and when the program is executed, the steps of the seat belt detection method described in any of the above embodiments are implemented.

[0088] The storage medium of the present invention can be executed by a processing device deployed in the vehicle, or by a processing device deployed remotely or in the cloud and connected to the relevant components of the vehicle to realize the detection of the wearing status of the seat belt. When the storage medium of the present invention is executed by the processor, it can efficiently and accurately detect whether the occupant is wearing the seat belt accurately, identify the situation of excessive wearing, insufficient wearing and other improper wearing of the seat belt, and ensure that the seat belt effectively plays a safety protection role.

[0089] The storage medium may be a portable compact disk read-only memory and include program code, and may be run on a terminal device, such as a processing device of a vehicle. However, the storage medium of the present invention is not limited thereto, and may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0090] The storage medium may be any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. The readable signal medium may include a data signal propagated in a baseband or as a part of a carrier wave, wherein a readable program code is carried. The readable signal medium may also be any readable medium other than a readable storage medium, which may send, propagate or transmit a program for use by an instruction execution system, device or component or for use in combination with the instruction execution system, device or component.

[0091] The program code for performing the operation of the present invention can be written in any combination of one or more programming languages, and the program code can be executed entirely on the terminal device, partially on the terminal device, as an independent software package, partially on the terminal device, partially on the remote device, or completely on the remote device or server. In the case of a remote device, the remote device can be connected to the terminal device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external device, such as through the Internet using an Internet service provider.

[0092] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the seat belt detection method described in any of the above embodiments.

[0093] The computer program product of the present invention can be executed by a processing device deployed in the vehicle, or by a processing device deployed remotely or in the cloud and connected to the relevant components of the vehicle to realize the detection of the wearing status of the seat belt. When the computer program product of the present invention is executed by the processor, it can efficiently and accurately detect whether the occupant is wearing the seat belt accurately, identify the situation of excessive wearing, insufficient wearing and other improper wearing of the seat belt, and ensure that the seat belt effectively plays a safety protection role.

[0094] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A seat belt detection method, It is characterized in that include: Obtaining an occupant image captured by a depth camera; Calculating the area of ​​the seat belt region according to the pixel information and depth information of the seat belt region in the occupant image; Based on the area of ​​the seat belt region, a detection result representing the wearing state of the seat belt is obtained.

2. The seat belt detection method according to claim 1, It is characterized in that The pixel information and depth information of the safety belt area include the number of pixel units in the safety belt area, the coordinates of each pixel unit and the depth of each pixel unit; The calculating the area of ​​the safety belt area includes: Calculating the depth size of each pixel unit in the safety belt area according to the depth relationship between the pixel units in the safety belt area; Calculate the pixel size of each pixel unit in the safety belt area based on the depth size and coordinates of each pixel unit in the safety belt area according to the conversion relationship between the pixel coordinate system and the world coordinate system; The area of ​​the safety belt area is calculated according to the number of pixel units in the safety belt area and the pixel size of each pixel unit.

3. The seat belt detection method according to claim 2, It is characterized in that The step of calculating the depth dimension of each pixel unit of the safety belt area includes: Obtaining a pixel unit sequence of the safety belt area; According to the depth of each pixel unit in the safety belt area, the depth difference between adjacent pixel units in the pixel unit sequence is calculated to obtain the depth size of each pixel unit in the safety belt area.

4. The seat belt detection method according to claim 2, It is characterized in that The step of calculating the pixel size of each pixel unit of the safety belt area includes: DeltaX i =(u i -cx)*DeltaZ i / fx; DeltaY i =(v i -cy)*DeltaZ i / fy; Among them, (DeltaX i ,DeltaY i ) is the pixel size of the i-th pixel unit in the safety belt area, DeltaZ i is the depth dimension of the i-th pixel unit, (u i ,v i ) are the coordinates of the i-th pixel unit, (fx, fy) are the focal length of the depth camera, and (cx, cy) are the coordinates of the center point of the depth camera.

5. The seat belt detection method according to claim 2, It is characterized in that The calculating the area of ​​the safety belt region comprises: Wherein, Sseatbelt is the area of ​​the seat belt area, and pixelNums is the number of pixel units in the seat belt area.

6. The seat belt detection method according to claim 1, It is characterized in that After obtaining the passenger image captured by the depth camera, the method further includes: The occupant image is processed by a semantic segmentation neural network to obtain pixel information and depth information of the seat belt area in the occupant image.

7. The seat belt detection method according to claim 1, It is characterized in that The obtaining, based on the area of ​​the seat belt region, a detection result representing the wearing state of the seat belt comprises: According to the relationship between the area of ​​the seat belt region and the threshold interval, a detection result representing the wearing state of the seat belt is obtained.

8. The seat belt detection method according to claim 1, It is characterized in that The obtaining, based on the area of ​​the seat belt region, a detection result representing the wearing state of the seat belt comprises: The area of ​​the seat belt region and the occupant image are input into a convolutional neural network to obtain a detection result representing the wearing status of the seat belt.

9. The seat belt detection method according to claim 1, It is characterized in that The detection result representing the wearing state of the seat belt includes any of the following items: The seat belt is fully worn; Over-wearing of seat belts; Inadequate seat belt wearing.

10. A seat belt detection device, It is characterized in that Used to implement the seat belt detection method according to any one of claims 1 to 9, the seat belt detection device comprises: An occupant image acquisition module, configured to obtain an occupant image captured by a depth camera; A seat belt area acquisition module, configured to calculate the area of ​​the seat belt area according to the pixel information and depth information of the seat belt area in the occupant image; The detection result acquisition module is configured to obtain a detection result representing a wearing state of the seat belt based on the area of ​​the seat belt region.

11. An electronic device, It is characterized in that include: processor; A memory, wherein executable instructions are stored in the memory; Wherein, when the executable instruction is executed by the processor, the seat belt detection method described in any one of claims 1-9 is implemented.

12. A computer-readable storage medium for storing a program, It is characterized in that When the program is executed by a processor, the seat belt detection method according to any one of claims 1 to 9 is implemented.

13. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the seat belt detection method according to any one of claims 1 to 9 is implemented.

Images