System and method for detecting proper seat belt use and distance - Patents.com

Abstract

A system and method for detecting seat belt positioning includes capturing a near infrared (NIR) image of an occupant by a camera, applying a median filter to the NIR image to remove glint, converting the NIR image to a black and white image, scanning across the black and white (B / W) image, detecting multiple transitions between black and white segments corresponding to stripes extending longitudinally along the length of the seat belt, and using detection of the multiple transitions to indicate seat belt detection. Converting the NIR image to a black and white image may include using a local binary threshold to determine whether a given pixel in the B / W image should be black or white based on whether a corresponding source pixel in the NIR image is brighter than an average of neighboring pixels within a predetermined distance from the corresponding source pixel.

Description

[Technical field]

[0001] (CROSS REFERENCE TO RELATED APPLICATIONS) This PCT international patent application claims the benefit of U.S. patent application Ser. No. 17 / 343,366, filed June 9, 2021, the contents of which are incorporated herein by reference in their entirety.

[0002] FIELD OF THEINVENTION The present invention relates generally to systems and methods for detecting proper seat belt use and distance to occupants using a vision system such as a near-infrared (NIR) camera. [Background technology]

[0003] Cameras and other image detection devices have been utilized to detect one or more objects. A control system in communication with these cameras can receive images captured by the cameras and process these images. The processing of these images can include detecting one or more objects found within the captured images. Based on these detected objects, the control system can perform some type of action in response to these detected changes.

[0004] Conventional systems for detecting seat belt use typically rely on seat belt buckle switches. However, these conventional systems cannot detect whether the seat belt is properly positioned or whether the seat belt buckle has been misaligned. Seat track sensors are typically used to determine the distance to an occupant in a vehicle. However, the use of such seat track sensors does not take into account the occupant's body position relative to the seat. Summary of the Invention [Means for solving the problem]

[0005] In one example, a method for detecting seat belt positioning includes capturing, by a camera, a near infrared (NIR) image of an occupant, The method also includes converting the NIR image to a black and white image, scanning across the black and white image to detect a number of transitions between black and white segments corresponding to stripes extending longitudinally along a length of the seat belt, and using detection of the number of transitions to indicate detection of the seat belt.

[0006] In another example, a system for detecting seat belt positioning includes a seat belt having a plurality of stripes extending longitudinally along a length of the seat belt, the plurality of stripes being arranged in an alternating pattern of light and dark in near infrared light. The system also includes a camera configured to capture a near infrared (NIR) image of a seat belted occupant, and a processor in communication with the camera and programmed to receive the NIR image of the seat belted occupant and determine a position of the seat belt based on detecting a transition corresponding to the alternating pattern of stripes.

[0007] Further objects, features, and advantages of the present invention will become readily apparent to those skilled in the art after studying the following description, taken in conjunction with the drawings and claims, which are annexed hereto and form a part of this specification. [Brief description of the drawings]

[0008] [Figure 1] 1 shows a conveyance device having a system for detecting proper seat belt use and detecting the distance to the seat belt.

[0009] [Diagram 2] 1 shows a view looking towards the front of a cabin of a vehicle having a system for detecting proper seat belt use and detecting the distance to the seat belt.

[0010] [Diagram 3]FIG. 1 shows a block diagram of a system for detecting proper seat belt use and detecting distance to the seat belt.

[0011] [Figure 4] A first example of improper seat belt positioning is shown.

[0012] [Diagram 5] A second example of improper seat belt positioning is shown.

[0013] [Figure 6] A third example of improper seat belt positioning is shown.

[0014] [Figure 7A] 1 illustrates a near infrared (NIR) image of a person fastening a seat belt, according to one embodiment of the present disclosure.

[0015] [Figure 7B] 7B illustrates a filtered image based on the NIR image of FIG. 7A in accordance with the present disclosure.

[0016] [Figure 7C] 7B shows a black / white image based on the NIR image of FIG. 7A in accordance with the present disclosure.

[0017] [Figure 7D] 7B shows an image based on the NIR image of FIG. 7A showing detection points according to the present disclosure.

[0018] [Figure 8A] A magnified portion of the NIR image is shown.

[0019] [Figure 8B] 8B shows a filtered image of a magnified portion of FIG. 8A.

[0020] [Figure 9] 1 shows a close-up NIR image of a seat belt with an asymmetric stripe pattern in accordance with the present disclosure.

[0021] [Figure 10] 7D illustrates a magnified portion of the black / white image of FIG. 7C including a seat belt and showing horizontal scan lines, in accordance with the present disclosure.

[0022] [Figure 11] 7D showing a magnified portion of the image of FIG. 7D, illustrating the detected angle of the seat belt, in accordance with the present disclosure.

[0023] [Figure 12] 1 shows a flow chart enumerating steps in a method for detecting seat belt positioning. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024] 1, a transport device 10 is shown having a seat belt detection system 12 for detecting proper seat belt use and / or detecting distance to a seat belt. In this example, the seat belt detection system 12 is integrated into the transport device 10. However, it should be understood that the seat belt detection system 12 may be a stand-alone system separate from the transport device 10. In some embodiments, the seat belt detection system 12 may use some or all of the components present in the transport device 10 for other systems and / or for other purposes, such as driver monitoring in an advanced driver assistance system (ADAS). Thus, the seat belt detection system 12 of the present disclosure may be implemented at very low additional cost.

[0025] With respect to the transport device 10, the transport device 10 is shown in FIG. 1 as a sedan-type automobile. However, it should be understood that the transport device 10 may be any type of transport device capable of transporting people or goods from one location to another. Thus, in addition to being a sedan-type automobile, the transport device 10 may be a pickup truck, a lorry, a trailer, a tractor, a mining transport device, and the like. It should also be understood that the transport device 10 is not limited to wheeled transport devices, but may also include non-wheeled transport devices, such as aircraft and watercraft. Again, the term transport device should be understood broadly to include any type of transport device capable of transporting people or goods from one location to another, and should not be limited to the examples specifically listed above.

[0026] Referring to FIG. 2, the cabin 14 of the transporter 10 is shown. As is well understood in the art, the cabin 14 is essentially the interior of the transporter 10 where occupants and / or items are located when the transporter is moving. The cabin 14 of the transporter may be defined by one or more pillars that structurally define the cabin 14. For example, in FIG. 2, an A-pillar 16A and a B-pillar 16B are shown. FIG. 1 further shows that there may be a third pillar or a C-pillar 16C. Of course, it should be understood that the transporter 10 may include any of a number of pillars to define the cabin 14. Furthermore, it should be understood that the transporter 10 may be designed to remove these pillars and essentially form an open-air cabin 14 as commonly found in automobiles with convertible tops.

[0027] Seats 18A and 18B are located within the passenger cabin 14. Seats 18A and 18B are configured to support occupants of the transport device 10. The transport device 10 may have any number of seats. It should further be understood that the transport device 10 may not have any seats at all.

[0028] The transport device 10 may have one or more cameras 20A-20F, which are located and mounted on the transport device 10 so as to have a view of at least a portion of the cabin 14 and function as part of a vision system. Thus, the cameras 20A-20F may have a view of the passengers sitting in the seats 18A and / or 18B. Here, the cameras 20A and 20C are located on the A-pillar 16A. The camera 20B is located on the rearview mirror 22. The camera 20D may be located on the dashboard 24 of the transport device 10. The cameras 20E and 20F may be focused on the driver and / or passengers and may be located adjacent to the transport device cluster 21 or the steering wheel 23, respectively. Of course, it should be understood that any one of several different cameras may be utilized. Thus, it should be understood that only one camera may be utilized or multiple cameras may be utilized. Furthermore, cameras 20A-20F may be located and mounted anywhere on vehicle 10 so long as they have a field of view of at least a portion of cabin 14.

[0029] The cameras 20A-20F may be any type of camera capable of capturing visual information. This visual information may be information in the visible spectrum, but also outside the visible spectrum, such as infrared or ultraviolet light. Here, the cameras 20A-20F are near-infrared (NIR) cameras capable of capturing images generated by reflection of near-infrared light. Near-infrared light may include any light in the near-infrared region of the electromagnetic spectrum (780 nm to 2500 nm). However, the seat belt detection system 12 of the present disclosure may be configured to use a specific wavelength or range of wavelengths within the near-infrared region.

[0030] The source of this near-infrared light may be a natural source, such as the sun, but may also be an artificial source, such as near-infrared light source 26. Near-infrared light source 26 may be mounted anywhere within cabin 14 of conveying device 10, so long as it is capable of projecting near-infrared light onto at least a portion of cabin 14. Here, near-infrared light source 26 is mounted on rearview mirror 22, but it should be understood that near-infrared light source 26 may be mounted anywhere within cabin 14. Additionally, while only one near-infrared light source 26 is shown, it should be understood that there may be two or more near-infrared light sources 26 located within cabin 14 of conveying device 10.

[0031] An output device 28 may also be located within the cabin 14 for relaying information to one or more occupants located within the cabin 14. Here, the output device 28 is shown to be a display device to convey visual information to one or more occupants located within the cabin 14. However, it should be understood that the output device 28 may be any output device capable of providing information to one or more occupants located within the cabin 14. Thus, for example, the output device may be an audio output device that provides audio information to one or more occupants located within the cabin 14 of the transport device 10. It should further be understood that the output device 28 may be a transport device subsystem that controls the functionality of the transport device.

[0032] Referring to FIG. 3, a more detailed diagram of the seat belt detection system 12 is shown. Here, the system 12 includes a control system 13 having a processor 30 in communication with a memory 32 that includes instructions 34 for performing any one of several different methods disclosed herein. The processor 30 may include a single standalone processor or may include two or more processors that may be distributed across multiple systems operating together. The memory 32 may be any type of memory capable of storing digital information. For example, the memory may be solid state memory, magnetic memory, optical memory, etc. Additionally, it should be understood that the memory 32 may be incorporated within the processor 30 or may be separate from the processor 30 as shown.

[0033] The processor 30 may also be in communication with the camera 20. The camera 20 may be the same as the cameras 20A-20F shown and described in FIG. 2. The camera 20 may be a near-infrared camera, similar to the cameras 20A-20F in FIG. 2. The camera 20 may include multiple physical devices, such as the cameras 20A-20F shown in FIG. 2. The camera 20 has a field of view 21.

[0034] The near-infrared light source 26 may also be in communication with the processor 30. When activated by the processor 30, the near-infrared light source 26 projects near-infrared light 36 onto an object 38, which can absorb or reflect the near-infrared light 40 towards the camera 20, which can capture images indicative of the absorbed or reflected near-infrared light 40. These images can then be provided to the processor 30.

[0035] The processor 30 may also be in communication with the output device 28. The output device 28 may include visual and / or audio output devices that may provide information to one or more passengers located within the passenger cabin 14 of FIG. 2. It should be further understood that the output device 28 may be a vehicle system, such as a safety system, that may take certain actions based on input received from the processor 30. For example, the processor 30 may instruct the output device 28 to limit or minimize the functionality of the vehicle 10 of FIG. 1. As described later in this specification, one of the functions that the seat belt detection system 12 may perform is to detect whether the passenger is properly fastening the safety belt. If the safety belt is not properly fastened, the processor 30 may instruct the output device 28 to limit the functionality of the vehicle 10, such that the vehicle 10 may only travel at a significantly reduced speed.

[0036] FIG. 4 illustrates a first example of improper seat belt positioning, showing an improperly adjusted seat belt 50 for an occupant 44 seated in seat 18A of the transport device 10. The improperly adjusted seat belt 50 in this example is loosely draped over the shoulder of the occupant 44. FIG. 5 illustrates a second example of improper seat belt positioning, showing the seat belt 50 passing under the armpit of the occupant 44. FIG. 6 illustrates a third example of improper seat belt positioning, showing the seat belt 50 passing around the back of the occupant 44. The seat belt detection system can detect other examples of improper seat belt positioning, such as a seat belt being missing or not worn by the occupant 44, even if the buckle is misidentified (e.g., by passing the seat belt behind the occupant 44 and buckling it or by placing a foreign object in the buckle latch).

[0037] FIG. 7A illustrates a near infrared (NIR) image of an occupant 44 wearing a seat belt 50, according to an aspect of the present disclosure. This may represent an image captured by the camera 20 and received by the processor 30. In some embodiments, the occupant 44 may be a driver of the transportation device 10. However, the occupant 44 may also be a passenger of the transportation device 10. FIG. 7B illustrates a filtered image based on the NIR image of FIG. 7A. FIG. 7C illustrates a black / white image based on the NIR image of FIG. 7A, and FIG. 7D illustrates an image based on the NIR image of FIG. 7A showing detection points, according to the present disclosure. Specifically, FIG. 7D illustrates the seat belt 50 passing through each of a first region of interest (ROI) 60 and a second ROI 62. The first ROI 60 may be located above the shoulder of the occupant 44, and the second ROI 62 may be located below and to the left of the first ROI. The second ROI 62 may correspond to a central torso region of the occupant 44. The ROIs 60, 62 may each have a fixed position within the field of view 21 of the camera 20. Alternatively, the system 12 may adjust the position of one or both of the ROIs 60, 62 based on the detected position of the occupant 44 within the field of view 21 of the camera 20.

[0038] FIG. 8A shows a magnified portion of the NIR image including a portion of the occupant 44, and FIG. 8B shows a filtered image of the magnified portion of FIG. 8A. FIGS. 8A-8B illustrate the removal of glint, which appears as a small bright area in the NIR image. The glint in this example is on an earring worn by the occupant 44. However, glint may arise from other sources, such as a loose thread or piece of lint or other material. The presence of glint on or near the seat belt 50 may, in some cases, interfere with subsequent processing steps and may reduce the effectiveness of the proper operation of the seat belt detection system 12.

[0039] FIG. 9 shows a close-up NIR image of a seat belt 50 having an asymmetric stripe pattern in accordance with the present disclosure. Specifically, the seat belt 50 includes a number of stripes 68a-68g extending longitudinally along the length of the seat belt 50. The stripes 68a-68g have alternating high and low intensities in the NIR image. The illustrated seat belt 50 includes seven stripes 68a-68g. However, the seat belt 50 may have a fewer or greater number of stripes 68a-68g. The stripes 68a-68g are arranged in an asymmetric pattern, with each of the stripes 68a-68g having a corresponding width Wa-Wg in a direction perpendicular to the length dimension of the seat belt 50. The sum of the widths Wa-Wg of the stripes 68a-68g is the overall width Wt of the seat belt 50. The stripes 68a-68g include two outermost stripes 68a, 68g that surround a number of inner stripes 68b-68f that define a total inner width Wti. The widths Wa-Wg of the stripes 68a-68g can have an irregular pattern, and such an irregular pattern can reduce the variation in which the system 12 makes a false positive detection. More specifically, the widths Wb-Wf of the inner stripes 68b-68f can have an irregular pattern to reduce the possibility of a false positive detection. A false positive detection may be when the camera 20 images something other than the seat belt 50 and detects that a pattern matches a pattern on the seat belt 50.

[0040] The stripes 68a-68g may be features of the material woven or otherwise constructed to form the seat belt 50. The material forming the stripes 68a-68g may extend throughout the entire seat belt 50, such that the stripes 68a-68g are visible on either of two opposing sides of the seat belt 50. Because the stripes 68a-68g include an asymmetric pattern, the orientation of the seat belt 50 may be determined based on an image of the pattern of the stripes 68a-68g. One or more twists in the seat belt 50 may be detected as a reversal of the asymmetric pattern of the stripes 68a-68g.

[0041] FIG. 10 illustrates an enlarged portion of the black / white image of FIG. 7C including the seat belt 50 and showing a horizontal scan line, in accordance with the present disclosure. FIG. 10 illustrates how the system 12 scans across a row (horizontal line) of pixels looking for a series of seven consecutive segments having alternating colors, corresponding to the stripes 68a-68g, separated by six transition points. The two outermost stripes 68a, 68g can have any width greater than 0 pixels. In other words, they can have any width resolvable by the camera 20. The outermost stripes 68a, 68g can serve to define the outer edge of the next adjacent stripes 68b, 68f. Thus, the seat belt 50 can include five inner stripes 68b-68f surrounded by the two outermost stripes 68a, 68f. The system 12 can recognize the seat belt 50 based on the relative widths of the inner stripes 68b-68f. Specifically, the inner stripes 68b-68f may have corresponding widths Wb-Wf that are different from one another to define a pattern of different width ratios. Each row of pixels shown in FIG. 10 may detect a seat belt based on the pattern shown at the bottom of the image. Specifically, the system 12 may detect a series of one or more black pixels (b>0) followed by a series of white pixels having a scaled width of one unit (w1). Continuing from left to right, the system 12 may further detect a black stripe 68c having a scaled width of one unit (b1), then a white stripe 68d having a scaled width of four units (w4), then a black stripe 68e having a scaled width of two units (b2), then a white stripe 68f having a scaled width of one unit (w1), and finally another series of one or more black pixels (b>0).

[0042] The ratio of the widths of the inner stripes 68b-68f shown in FIG. 10 is 1:1:4:2:1, however, the inner stripes 68b-68f may have different widths to define different ratios. If the detected inner stripes 68b-68f match the 1:1:4:2:1 ratio within a certain error tolerance, the horizontal position of the seat belt 50 in each given scan line may be indicated as a detection point. In some embodiments, the cumulative width of the ratio section of the detection event may be stored with the detection point. Thus, the system 12 may filter out noise in a given scan line that may possibly match the 1:1:4:2:1 ratio, but does not have a total inner width Wti that is within a tolerance of the total inner width Wti of the seat belt 50. Such noise is detected in scan lines adjacent or proximate to the given scan line. Using the relative ratio of the widths Wb-Wf of the corresponding inner stripes 68b-68f, the system 12 may detect the seat belt 50.

[0043] FIG. 11 shows an enlarged portion of the image of FIG. 7D , showing the detected angle 70 of the seat belt 50. In particular, FIG. 11 shows how the system 12 recognizes the seat belt 50 and records its position as a detection point 64. When multiple detection points 64 are accumulated, the angle 70 of the seat belt 50 can be determined. The angle 70 can represent the inclination of the seat belt 50 in the longitudinal direction, or the difference in distance from the camera between a higher portion of the seat belt 50 and a lower portion of the seat belt 50. The detection point 64 in the first ROI 60 above the shoulder of the occupant 44 represents the higher portion of the seat belt 50. Similarly, the detection point 64 in the second ROI 62 in the central region of the torso of the occupant 44 can represent the lower portion of the seat belt 50.

[0044] Once the angle 70 is determined, the actual width of the seat belt 50 at the detection point 64 can be used to determine a compensated pixel width. For example, at an angle of 73 degrees, the detection point 64, which has a pixel width of 100 pixels measured by the camera 20, is multiplied by the sine of the angle 70 (sin 73 degrees) to obtain a compensated pixel width of approximately 95.6 pixels. Using the known total inner width Wti of the seat belt 50 and details of the camera 20 (e.g., pixel arc length), the location of the detection point 64 on the seat belt 50 can be determined. This distance can provide a more accurate measurement of the position of the occupant 44 as opposed to traditional methods that rely on the position of the seat.

[0045] A method 100 for detecting seat belt positioning is shown in the flow chart of Figure 12. Method 100 includes, at step 102, capturing an image of occupant 44 by camera 20. Step 102 may include capturing an image in the near infrared (NIR) spectrum, which may include detecting reflected NIR light provided by near infrared light source 26. Step 102 may further include transmitting the image from camera 20 as a video stream or as one or more still images to control system 13 having processor 30 for further processing.

[0046] Method 100 also includes filtering the image to remove glints, at step 104. Processor 30 may perform step 104, which may include applying a median filter to the image. The median filter may preserve edges while smoothing abnormally bright or dark regions (i.e., glints) that may result from seat belt threads, bad pixels in camera 20, or other noise-inducing particles such as lint on seat belt 50. This step 104 reduces the number of false positive detections of black / white transitions, thereby improving the performance and reliability of method 100.

[0047] Method 100 also includes converting the filtered image to black and white (B / W) at step 106. The terms black and white may include any representation of a pixel in one of two binary states representing dark or light. Processor 30 may perform step 106, which may include using a local binary threshold to determine whether any given pixel in the B / W image should be black or white. Such a local binary threshold may compare a source pixel in the source image (i.e., the filtered image) to neighboring pixels within a predetermined distance of the pixel. If the source pixel is brighter than the average of the neighboring pixels, the corresponding pixel in the B / W image may be set to white, and if the source pixel is darker than the average of the neighboring pixels, the corresponding pixel in the B / W image may be set to black. In some embodiments, the predetermined distance may be about 100 pixels. In some embodiments, the predetermined distance may be equal to or approximately equal to the pixel width of the seat belt 50 when the seat belt 50 is in a nominal position relative to the camera (e.g., when used for a passenger 44 of medium build who is seated in the seat 18a in a middle position).

[0048] The method 100 also includes, at step 108, scanning the B / W image to detect black / white (or white / black) transitions and using the detection of these transitions to indicate a detection 64 of the seat belt 50. The processor 30 may perform step 108, which includes comparing the relative distances between the transitions to determine whether they correlate to a ratio of the widths of the inner stripes 68b-68f of the seat belt 50, and if so, marking the location as a detection 64. For example, the processor 30 may be programmed to scan across a horizontal line in the B / W image to detect a grouping of transitions spaced apart by a distance that matches the 1:1:4:2:1 ratio of the widths of the inner stripes 68b-68f shown in FIG. 10. Similarly, the processor 30 may be programmed to detect an inverse ratio of the widths of the inner stripes 68b-68f (i.e., transitions separated by a distance that matches a 1:2:4:1:1 pattern) to detect a seat belt 50 having an inverse orientation at a particular location.

[0049] The method 100 also includes calculating the angle of the seat belt 50 in step 110. The processor 30 may perform step 110, which may include determining an angle 70 of the seat belt 50 using a number of detection points 64, such as the position of the seat belt 50 in two regions of interest (ROIs) 60, 62. The angle 70 may represent the inclination of the seat belt 50 in the longitudinal direction, or the difference in distance from the camera between a higher portion of the seat belt 50 and a lower portion of the seat belt 50.

[0050] The method 100 also includes calculating a distance to the seat belt 50 in step 112. The processor 30 may perform step 112, which may include using a pixel width of the seat belt 50 within the ROI 60, 62. Step 112 may further use the previously determined angle of the seat belt 50 to calculate a distance to 50, such as a distance to the seat belt 50 in one or more of the ROIs 60, 62. For example, at an angle of 73 degrees, the processor 30 may determine a pixel width of the detection point 64 to be 100 pixels. The processor 30 may then multiply the pixel width of 100 pixels by the sine of the angle 70 (sin 73 degrees) to determine a compensated pixel width of approximately 95.6 pixels. Using the known total inner width Wti of the seat belt 50 and details of the camera 20 (e.g., pixel arc length), the processor 30 may calculate a distance between the camera and the detection point 64 of the seat belt 50. This distance can provide a more accurate measurement of the position of the occupant 44 as opposed to conventional methods that rely on seat position.

[0051] The method 100 also includes determining whether the seat belt 50 is properly positioned at step 114. The processor 30 may perform step 114, which may include using the angle of the seat belt 50 and / or the distance to the seat belt 50. For example, the processor 30 may calculate a measured angle of the seat belt 50 and compare the measured angle to a range of measurements corresponding to a proper position of the seat belt 50. Similarly, the processor 30 may calculate one or more distances to the seat belt 50 and one or more distances to a distance corresponding to a proper positioning of the seat belt 50. In some embodiments, the processor 30 may calculate a rate of change of the distance to determine whether the seat belt 50 is loose, for example, as shown in FIG. 4. The processor 30 may also designate the seat belt 50 as improperly positioned if the processor 30 is unable to identify the seat belt 50 in one or more regions of the image, such as one or more of the ROIs 60, 62. Examples of such improper positioning are shown in FIGS. 5 and 6.

[0052] The method 100 also includes determining whether the seat belt 50 is twisted at step 116. The processor 30 may perform step 116, which may include detecting one or more reversals of the asymmetric pattern on the seat belt. For example, the processor 30 may designate the seat belt as improperly positioned if it detects a number of twists in the seat belt 50 that exceeds a twist threshold.

[0053] The method 100 also includes generating a first signal if the system 12 determines that the seat belt 50 is properly positioned or generating a second signal if the system 12 determines that the seat belt 50 is improperly positioned at step 118. The processor 30 may perform step 118, which may include providing a declaration to the occupant 44, such as an audible or visual warning message. In another example, the processor 30 may send a signal to an output device 28, such as a safety system, to take action to limit or minimize functionality of the transporter 10. For example, the transporter 10 may be prevented from moving or from exceeding a very low speed until or unless the system 12 determines that the seat belt 50 is properly positioned.

[0054] By implementing the methods of the present disclosure, the seat belt detection system 12 can determine whether the occupant 44 is properly fastening the seat belt 50. The systems and methods of the present disclosure can provide increased confidence that the occupant 44 is properly fastening the seat belt 50.

[0055] Additionally, as previously discussed, the seat belt 50 may have a light absorbing and / or reflective material 50C located or disposed thereon. The cameras 20A-20F may capture images of the material 50C. As previously discussed, this material 50C may be a known pattern with pattern elements separated from one another by a known distance 52. The seat belt detection system 12 may then review these captured images from the cameras 20A-20F and determine whether the distance from the seat belt 50 to the cameras is an expected distance that generally indicates that the seat belt 50 is properly resting on the body 48 of the occupant 44. Furthermore, because this pattern is known, clothing that may be worn by the occupant 44 and that may reflect and / or absorb light, such as infrared, may be disregarded as it is highly unlikely that the clothing would have a pattern that matches the pattern of the stripes 68a-68g on the seat belt 50.

[0056] If it is determined that the occupant 44 is properly fastened to the seat belt 50, the seat belt detection system 12 may allow the transport device 10 to operate in a normal mode. However, if the seat belt detection system 12 indicates that the occupant 44 is not properly fastened to the seat belt 50, the control system 12 may take any one of several different actions. For example, the seat belt detection system 12 may indicate to the occupant 44 using the output device 28 to provide a visual and / or audio cue that the seat belt 50 is not properly fastened. Additionally, the output device 28 may communicate with any one of several different transport device systems to restrict operation of the transport device 10 until the seat belt 50 is properly fastened by the occupant 44.

[0057] The seat belt detection system 12 may also communicate with other control systems to improve system reliability. For example, the seat belt detection system 12 may also communicate with one or more sensors, such as a sensor that detects when the safety belt latch 61 or tongue is inserted into the safety belt buckle 63. If the seat belt detection system 12 determines that the safety belt buckle is properly latched and that the seat belt 50 is properly positioned across the body 48 of the occupant 44, the seat belt detection system 12 may determine with greater confidence that the seat belt 50 is properly utilized by the occupant 44.

[0058] In some embodiments, dedicated hardware implementations such as application specific integrated circuits, programmable logic arrays, and other hardware devices can be constructed to implement one or more steps of the methods described herein. Applications that may include the apparatus and systems of various embodiments can broadly include a variety of electronic and computer systems. One or more embodiments described herein may implement functionality using two or more specific interconnected hardware modules or devices, with associated control and data signals that can communicate between and through the modules, or as part of a specific application specific integrated circuit. Thus, the system encompasses software, firmware, and hardware implementations.

[0059] According to various embodiments of the present disclosure, the methods described herein may be implemented by a software program executable by a computer system. Furthermore, in exemplary and non-limiting embodiments, implementations may include distributed processing, component / object distributed processing, and parallel processing. Alternatively, a virtual computer system processing may be constructed to implement one or more of the methods or functions as described herein.

[0060] Additionally, the methods described herein may be embodied in a computer-readable medium. The term "computer-readable medium" includes a single medium or multiple media, such as a centralized or distributed database and / or associated caches and servers that store one or more sets of instructions. The term "computer-readable medium" is also intended to include any medium capable of storing, encoding, or carrying a set of instructions for execution by a processor, or causing a computer system to perform any one or more of the methods or operations disclosed herein.

[0061] As those skilled in the art will readily appreciate, the above description is intended as an illustration of the principles of the present invention, and is not intended to limit the scope or application of the present invention, in that the present invention is susceptible to modifications, variations, and alterations without departing from the spirit of the invention, as defined in the following claims.

Claims

1. 1. A method for detecting seat belt positioning, comprising: capturing a near infrared (NIR) image of an occupant with a camera; converting the NIR image into a black and white image; scanning across the black and white image to detect a number of transitions between black and white segments corresponding to stripes extending longitudinally along the length of the seat belt; using detection of the plurality of transitions to indicate detection of the seat belt; and calculating a distance from the camera to the seat belt using a pixel width of the seat belt and an angle of the seat belt based on detection of the seat belt in each of two different regions of interest in the NIR image; A method comprising:

2. The method of claim 1 , further comprising applying a median filter to the NIR image to remove glint before converting the NIR image to the black and white image.

3. 2. The method of claim 1, wherein converting the NIR image to the black-and-white image includes using a local binary threshold to determine whether a given pixel in the black-and-white image should be black or white based on whether a corresponding source pixel in the NIR image is brighter than an average of neighboring pixels within a predetermined distance from the corresponding source pixel.

4. The method of claim 3 , wherein the predetermined distance is approximately 100 pixels, or the predetermined distance is approximately equal to a pixel width of the seat belt.

5. The method of claim 1 , wherein using the detection of the plurality of transitions to indicate the detection of the seat belt further comprises comparing a relative distance between the transitions to a ratio of widths of the stripes.

6. The method of claim 1 , wherein the widths of the stripes define an asymmetric pattern.

7. determining the orientation of the seat belt based on an order of the plurality of transitions corresponding to the asymmetric pattern differing based on an orientation of the seat belt; 7. The method of claim 6, further comprising: determining a twist of the seat belt based on one or more changes in the orientation of the seat belt along a length of the seat belt.

8. Calculating the distance to the seat belt calculating the sine of the seat belt angle; multiplying the pixel width of the seat belt by the sine of the angle of the seat belt to determine a compensated pixel width; 2. The method of claim 1, further comprising: calculating the distance to the seat belt based on the compensated pixel width and a known inner width of the seat belt.

9. the stripes of the seat belt include a plurality of inner stripes surrounded by outermost stripes at each of two edges along the length of the seat belt; 9. The method of claim 8, wherein the known inner width of the seat belt is a sum of the widths of the inner stripes.

10. determining whether the seat belt is properly positioned using at least one of the angle of the seat belt or the distance from the camera to the seat belt; 10. The method of claim 1, further comprising: generating a signal in response to determining that the seat belt is properly positioned or in response to determining that the seat belt is not properly positioned.

11. A system for detecting seat belt positioning, a seat belt having a plurality of stripes extending longitudinally along a length of the seat belt, the plurality of stripes being arranged in an alternating light and dark pattern in the near infrared; a camera configured to capture a near infrared (NIR) image of the seat belted occupant; a processor in communication with the camera and programmed to receive the NIR image of the occupant wearing the seat belt and determine a position of the seat belt based on detecting transitions corresponding to the alternating pattern of stripes; Equipped with the processor is further programmed to calculate a distance from the camera to the seat belt using a pixel width of the seat belt and an angle of the seat belt based on detection of the seat belt in each of two different regions of interest in the NIR image. system.

12. 12. The system of claim 11, wherein the processor is further programmed to convert the NIR image to a black and white image using a local binary threshold to determine whether a given pixel in the black and white image should be black or white based on whether a corresponding source pixel in the NIR image is brighter than an average of neighboring pixels within a predetermined distance from the corresponding source pixel.

13. 13. The system of claim 12, wherein the predetermined distance is approximately 100 pixels, or the predetermined distance is approximately equal to a pixel width of the seat belt.

14. 12. The system of claim 11, wherein determining the position of the seat belt based on detecting the transitions corresponding to the alternating pattern of stripes further comprises comparing a relative distance between the transitions to a ratio of widths of the plurality of stripes.

15. The system of claim 11 , wherein the widths of the stripes define an asymmetric pattern.

16. Calculating the distance to the seat belt comprises: calculating the sine of the seat belt angle; multiplying the pixel width of the seat belt by the sine of the angle of the seat belt to determine a compensated pixel width; and calculating the distance to the seat belt based on the compensated pixel width and a known inner width of the seat belt.

17. The processor, determining whether the seat belt is properly positioned using at least one of the angle of the seat belt or the distance from the camera to the seat belt; 12. The system of claim 11, further programmed to: generate a signal in response to determining that the seat belt is properly positioned or in response to determining that the seat belt is not properly positioned.

Images