CN112232136B - Vehicle safety belt detection method and device, electronic equipment and storage medium - Google Patents
Vehicle safety belt detection method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN112232136B CN112232136B CN202011000546.9A CN202011000546A CN112232136B CN 112232136 B CN112232136 B CN 112232136B CN 202011000546 A CN202011000546 A CN 202011000546A CN 112232136 B CN112232136 B CN 112232136B
- Authority
- CN
- China
- Prior art keywords
- line segment
- safety belt
- belt
- fitting
- seat belt
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a vehicle safety belt detection method and device, electronic equipment and a storage medium. The method comprises the following steps: a safety belt detection flow; the safety belt detection process comprises the following steps: extracting a region of interest in an image to be detected by using safety belt information, wherein the safety belt information comprises region information of the safety belt possibly appearing in the image; detecting a straight line segment in the region of interest; and judging whether the safety belt in the image to be detected is in a wearing state according to the linear line segment. The invention utilizes the images collected in the vehicle, and can remind personnel of wearing the safety belt by identifying whether the safety belt is correctly worn through image analysis, thereby greatly reducing the probability of not wearing the safety belt and reducing the corresponding manpower investment.
Description
Technical Field
The invention belongs to the field of automobiles, and particularly relates to a method and a device for detecting a safety belt of a vehicle, electronic equipment and a storage medium.
Background
With the improvement of living standard of residents, automobiles are becoming more popular in life. The meaning of standardizing the wearing of safety belts is becoming increasingly important. The seat belt may reduce the risk of mortality of the front passenger by 50% as estimated by the highway traffic safety agency (NHTSA). While the safety belt does in some cases also cause serious injury or death, nearly all safety professionals consider that tying the belt can significantly increase the chance of survival in an accident.
At present, schemes for detecting whether a person wears a safety belt in a vehicle are mainly divided into two types. A method for distinguishing the area of the safety belt includes such steps as printing ink on the safety belt, and supplementing light to make the grey value of the area higher than that of other areas. The method can modify the essential characteristics of the safety belt and is contrary to the wish of the vehicle owner. The utility model provides a realize the safety belt location through objects such as steering wheel, license plate, carry out sharp detection to the safety belt, judge the straight line number of safety belt, and then carry out the judgement of safety belt. According to the method, objects such as a steering wheel and a license plate are needed in the calibration process, so that images in the vehicle are needed to be shot from outside the vehicle in a long distance, the method is more suitable for traffic monitoring camera scenes, is not suitable for vehicle-mounted scenes, is low in detection accuracy, and is easy to generate misjudgment.
Disclosure of Invention
The invention aims to overcome the defects that in the prior art, when whether a safety belt is worn or not is detected by utilizing an image recognition technology, the safety belt is required to be positioned by objects such as a steering wheel and a license plate, so that the safety belt is not suitable for a vehicle-mounted scene and misjudgment is easy to occur due to low detection accuracy, and provides a vehicle safety belt detection method and device, electronic equipment and a storage medium which are more suitable for the vehicle-mounted scene.
The invention solves the technical problems through the following technical scheme:
the invention provides a vehicle safety belt detection method, which comprises a safety belt detection flow;
the safety belt detection process comprises the following steps:
extracting a region of interest in an image to be detected by using safety belt information, wherein the safety belt information comprises region information of the safety belt possibly appearing in the image;
detecting a straight line segment in the region of interest;
and judging whether the safety belt in the image to be detected is in a wearing state according to the linear line segment.
Preferably, the vehicle safety belt detection method further comprises a safety belt information calibration flow;
the safety belt information calibration process comprises the following steps:
and calibrating the safety belt information by using a reference image, wherein the reference image and the image to be detected are acquired at the same position.
Preferably, the belt information further includes coordinates of a fixed point on the belt shoulder strap;
calibrating the seat belt information using a reference image, comprising:
displaying a real-time video image on a display, wherein prompt information is marked in the real-time video image and used for prompting the positionable position of a shoulder belt of the safety belt and the possible occurrence area of the safety belt;
Collecting an image as a reference image after a person in the vehicle sits;
judging whether the reference image contains a fixed point on a belt shoulder strap or not:
if not, taking the upper intersection point of the straight line of the belt shoulder strap and the possible area of the belt as the upper fixing point of the belt shoulder strap;
if so, the area where the safety belt is likely to appear is marked by taking the fixing point on the safety belt shoulder belt as one boundary point of the area where the safety belt is likely to appear.
Preferably, the area where the seat belt may appear is a rectangular area, and the area information includes upper left and lower right corner coordinates of the rectangular area, or upper right and lower left corner coordinates of the rectangular area.
Preferably, the safety belt detection process further includes, after detecting the straight line segment in the region of interest:
selecting a straight line segment meeting preset conditions from the detected straight line segments as a seed straight line;
and using the seed straight line as a straight line segment for judging whether the safety belt is detected in the image to be detected later, or performing line segment fitting at least once by using the seed straight line, and using the line segment obtained after fitting as a straight line segment for judging whether the safety belt is detected in the image to be detected later.
Preferably, the safety belt information further comprises a safety belt preset angle, wherein the safety belt preset angle is an included angle between the safety belt and the positive direction of the x axis in the wearing state, one of the preset conditions is that the included angle between the straight line segment and the positive direction of the x axis is within a preset angle threshold value range, and the angle threshold value range is the safety belt preset angle +/-preset angle threshold value;
and/or, one of the preset conditions is that the length of the straight line segment is greater than a preset length threshold value.
Preferably, the safety belt detection process further includes, after detecting the straight line segment in the region of interest:
marking the end point coordinates of the linear line segment;
and calculating and storing the length of the linear line segment, the included angle between the linear line segment and the positive direction of the x-axis and the distance between the linear line segment and the pole.
Preferably, if at least one line segment fitting is performed by using the seed straight line, the at least one line segment fitting includes:
fitting conditions include: the distance between the end points of the two linear line segments to be fitted, the angle difference of the two linear line segments to be fitted and the difference of the distances between the two linear line segments to be fitted and the pole; when calculating the distance between the endpoints, selecting four endpoints of two linear line segments to be fitted to calculate one by one, and selecting the minimum value to represent the distance between the endpoints;
The first fitting comprises searching for a straight line segment fitted with the seed straight line from all detected straight line segments, wherein the threshold value of the distance between the endpoints is set to be a fixed value or the arithmetic square root of the product of the lengths of the two straight line segments to be fitted;
and/or, re-fitting includes finding a fitted line segment in the result of the primary fitting, wherein the threshold value of the distance between the end points is set to the arithmetic square root of the product of the lengths of the two linear line segments to be fitted or a fixed value;
in the first fitting and the second fitting, the fitting method includes: if the two linear line segments to be fitted meet the fitting condition, selecting the leftmost endpoint of the two linear line segments to be fitted as the left endpoint of the new linear line segment, the rightmost endpoint as the right endpoint of the new linear line segment, and recalculating the length of the new linear line segment, the included angle with the positive direction of the x axis and the distance from the linear line to the pole according to the left endpoint and the right endpoint.
Preferably, determining whether the safety belt is detected in the image according to the straight line segment includes:
calculating characteristic parameters of the linear line segments, normalizing the characteristic parameters when the characteristic parameters are more than 1, and calculating scores of the linear line segments;
Judging whether a straight line segment with the score exceeding a score threshold exists, if so, judging that the safety belt is in a wearing state, and if not, judging that the safety belt is in an unworn state.
Preferably, the safety belt information further comprises fixed point coordinates on the safety belt shoulder belt, and the characteristic parameter comprises a distance between an end point of the straight line segment, which is close to one side of the fixed point on the safety belt shoulder belt, and the fixed point coordinates on the safety belt shoulder belt;
and/or, the characteristic parameter comprises the length of the straight line segment.
The invention also provides a vehicle safety belt detection device, which comprises: a seat belt detection module;
the seat belt detection module includes:
the device comprises a region extraction unit, a detection unit and a detection unit, wherein the region extraction unit is used for extracting a region of interest in an image to be detected by utilizing safety belt information, and the safety belt information comprises region information possibly appearing in the image by the safety belt;
a line segment detection unit for detecting a straight line segment in the region of interest;
and the safety belt judging unit is used for judging whether the safety belt in the image to be detected is in a wearing state according to the linear line segment.
Preferably, the vehicle safety belt detection device further comprises a safety belt information calibration module;
The safety belt information calibration module is used for calibrating the safety belt information by utilizing a reference image, and the reference image and the image to be detected are collected at the same position.
Preferably, the belt information further includes coordinates of a fixed point on the belt shoulder strap;
the safety belt information calibration module is used for displaying a real-time video image on a display, wherein prompt information is marked in the real-time video image and used for prompting the positionable position of a safety belt shoulder belt and the possible occurrence area of the safety belt;
collecting an image as a reference image after a person in the vehicle sits;
judging whether the reference image contains a fixed point on a belt shoulder strap or not:
if not, taking the upper intersection point of the straight line of the belt shoulder strap and the possible area of the belt as the upper fixing point of the belt shoulder strap;
if so, the area where the safety belt is likely to appear is marked by taking the fixing point on the safety belt shoulder belt as one boundary point of the area where the safety belt is likely to appear.
Preferably, the area where the seat belt may appear is a rectangular area, and the area information includes upper left and lower right corner coordinates of the rectangular area, or upper right and lower left corner coordinates of the rectangular area.
Preferably, the seat belt detection unit further includes:
a seed selecting unit, configured to select a straight line segment meeting a preset condition from the detected straight line segments as a seed straight line after detecting the straight line segment in the region of interest;
the safety belt judging unit uses the seed straight line as a straight line segment for judging whether the safety belt is detected in the image to be detected or not, or the line segment fitting unit uses the seed straight line for carrying out line segment fitting at least once, and uses the line segment obtained after fitting as a straight line segment for judging whether the safety belt is detected in the image to be detected or not by the safety belt judging unit.
Preferably, the safety belt information further comprises a safety belt preset angle, wherein the safety belt preset angle is an included angle between the safety belt and the positive direction of the x axis in the wearing state, one of the preset conditions is that the included angle between the straight line segment and the positive direction of the x axis is within a preset angle threshold value range, and the angle threshold value range is the safety belt preset angle +/-preset angle threshold value;
and/or, one of the preset conditions is that the length of the straight line segment is greater than a preset length threshold value.
Preferably, the seat belt detection module further includes:
And the preprocessing unit is used for marking the end point coordinates of the linear line segment after detecting the linear line segment in the region of interest, and calculating and storing the length of the linear line segment, the included angle between the linear line segment and the positive direction of the x-axis and the distance between the linear line segment and the pole.
Preferably, if the line segment fitting unit performs at least one line segment fitting by using the seed straight line, the at least one line segment fitting includes:
fitting conditions include: the distance between the end points of the two linear line segments to be fitted, the angle difference of the two linear line segments to be fitted and the difference of the distances between the two linear line segments to be fitted and the pole; when calculating the distance between the endpoints, selecting four endpoints of two linear line segments to be fitted to calculate one by one, and selecting the minimum value to represent the distance between the endpoints;
the first fitting comprises searching for a straight line segment fitted with the seed straight line from all detected straight line segments, wherein the threshold value of the distance between the endpoints is set to be a fixed value or the arithmetic square root of the product of the lengths of the two straight line segments to be fitted;
and/or, re-fitting includes finding a fitted line segment in the result of the primary fitting, wherein the threshold value of the distance between the end points is set to the arithmetic square root of the product of the lengths of the two linear line segments to be fitted or a fixed value;
In the first fitting and the second fitting, the fitting method includes: if the two linear line segments to be fitted meet the fitting condition, selecting the leftmost endpoint of the two linear line segments to be fitted as the left endpoint of the new linear line segment, the rightmost endpoint as the right endpoint of the new linear line segment, and recalculating the length of the new linear line segment, the included angle with the positive direction of the x axis and the distance from the linear line to the pole according to the left endpoint and the right endpoint.
Preferably, the safety belt judging unit is configured to calculate a characteristic parameter of the straight line segment, normalize the characteristic parameter when the characteristic parameter is more than 1, and calculate a score of the straight line segment; judging whether a straight line segment with the score exceeding a score threshold exists, if so, judging that the safety belt is in a wearing state, and if not, judging that the safety belt is in an unworn state.
Preferably, the safety belt information further comprises fixed point coordinates on the safety belt shoulder belt, and the characteristic parameter comprises a distance between an end point of the straight line segment, which is close to one side of the fixed point on the safety belt shoulder belt, and the fixed point coordinates on the safety belt shoulder belt;
and/or, the characteristic parameter comprises the length of the straight line segment.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the seat belt detection method as described above when executing the program.
The present invention also provides a computer-readable storage medium having stored thereon a computer program, characterized in that the program when executed by a processor implements the steps of the seat belt detection method as described above.
On the basis of conforming to the common knowledge in the field, the above preferred conditions can be arbitrarily combined to obtain the preferred examples of the invention.
The invention has the positive progress effects that: according to the invention, objects such as a steering wheel and a license plate are not needed in the process of calibrating the safety belt information, and whether the safety belt is correctly worn or not can be reminded to wear the safety belt by utilizing the images acquired in the vehicle through image analysis and recognition, so that the probability of not wearing the safety belt is greatly reduced, and the corresponding manpower investment is reduced.
Drawings
FIG. 1 is a schematic view of a vehicle seat belt;
fig. 2 (a) is a schematic diagram showing a situation of identifying safety belt information in the vehicle seat belt detection method according to embodiment 1 of the present invention;
Fig. 2 (b) is a schematic diagram showing another case of identifying safety belt information in the vehicle seat belt detection method according to embodiment 1 of the present invention;
fig. 3 is a flowchart of a seat belt detection flow in the vehicle seat belt detection method of embodiment 1 of the present invention;
fig. 4 is a flowchart of preprocessing in a seat belt detection flow in the vehicle seat belt detection method of embodiment 1 of the present invention;
fig. 5 is a specific flowchart of step 205 in the vehicle seat belt detection method according to embodiment 1 of the present invention;
fig. 6 is a schematic block diagram of a vehicular seat belt detecting device of embodiment 2 of the invention;
FIG. 7 is a schematic diagram of an intelligent monitoring system;
fig. 8 is a schematic structural diagram of an electronic device according to embodiment 3 of the present invention.
Detailed Description
The invention is further illustrated by means of the following examples, which are not intended to limit the scope of the invention.
Example 1
The present embodiment provides a vehicle seat belt detection method for detecting whether a seat belt is in a wearing state. The safety belt according to this embodiment is a conventional car-mounted safety belt, and as shown in fig. 1, mainly includes a shoulder belt 11, a waist belt 12, a shoulder belt upper fixing point 13 (i.e., a position where the upper end of the shoulder belt is fixed), a winder 14, a buckle plug 15, a buckle socket 16, and a release button 17. It should be noted that fig. 1 only shows a common seat belt structure, and the seat belt to which the present invention is applied is not limited thereto, and other seat belts having similar structures and similar principles are also applicable to the present invention. The method comprises a safety belt information calibration flow and a safety belt detection flow.
The safety belt information calibration flow is used for calibrating safety belt information. In this embodiment, the seatbelt information includes area information in which the seatbelt may appear. The area in which the identified seat belt may appear may be a rectangular area, and the area information in which the seat belt may appear may include the upper left corner coordinates (x Left upper part ，y Left upper part ) And lower right angular position (x Lower right ，y Lower right ) Or the upper right and lower left corner coordinates of a rectangular region.
The belt information may also include the coordinates (x Fixed point ，y Fixed point ) And a preset angle theta of the safety belt Presetting . Fixed point coordinates (x) on the belt shoulder strap Fixed point ，y Fixed point ) Is the coordinates of any point on the fixed point on the belt shoulder strap. Preset angle theta of safety belt Presetting Wearing for safety beltThe included angle between the shoulder belt and the positive direction of the x-axis in the state (mainly the included angle between the shoulder belt and the positive direction of the x-axis in the wearing state of the safety belt).
In this embodiment, the safety belt information calibration process specifically may include:
and calibrating safety belt information by using the reference image, wherein the reference image and an image to be detected aimed at in a follow-up safety belt detection process are acquired at the same position.
The calibration of the safety belt information by using the reference image can specifically include:
displaying a real-time video image on a display, wherein prompt information is marked in the real-time video image, and the prompt information is used for prompting the positionable position of a shoulder belt of the safety belt and the possible occurrence area of the safety belt;
collecting an image as a reference image after a person in the vehicle sits;
judging whether the reference image contains a fixed point on the belt shoulder strap:
if not, taking the upper intersection point of the straight line of the belt shoulder strap and the possible area of the belt as the upper fixing point of the belt shoulder strap;
if so, the area where the safety belt is likely to appear is marked by taking the fixing point on the safety belt shoulder belt as one boundary point of the area where the safety belt is likely to appear.
The reference image can be acquired by the camera in the vehicle, and is more suitable for detecting the safety belt of the vehicle in the vehicle scene.
The calibration example is shown in fig. 2 (a) and 2 (b), wherein a solid large rectangular frame 101 represents the shooting range of the camera, a dotted rectangular frame 102 represents the possible area of the safety belt, two parallel solid lines represent the safety belt 103, and the angle between the virtual straight lines and the solid large rectangular frame 101 represents the preset safety belt angle θ Presetting The solid small rectangular box 104 represents the anchor point on the belt shoulder strap.
As shown in fig. 2 (a), the reference image 101 acquired by the camera cannot see the position of the fixed point on the shoulder strap of the safety belt, so that the range of the dotted rectangular frame 102 can be enlarged to ensure the calibration to be correct. As shown in fig. 2 (b), the reference image 101 acquired by the camera can clearly see the position of the fixed point 104 on the shoulder belt, and the position of the dotted rectangular frame 102 can be reduced to increase the running speed.
For the same vehicle, the belt information will not change after the camera is fixed. That is, the above-mentioned calibration procedure of the safety belt information is only required to be executed once after the camera is fixed. The seat belt information may be stored for retrieval by the seat belt detection process. In some embodiments, the above-mentioned belt calibration procedure may be omitted, and the belt information is set by means of a preset value or an empirical value.
As shown in fig. 3, the seat belt detection flow includes:
step 201: and extracting a region of interest (ROI) in the image to be detected by using the region information, wherein the image to be detected and the reference image adopt the same shooting angle (a camera can be fixed). In order to improve the image detection efficiency, the region of interest is extracted as the object of subsequent image processing, so that the complexity of the algorithm can be reduced. Using the upper left angular position (x) of the region where the seat belt is likely to appear in the seat belt information Left upper part ，y Left upper part ) And lower right angular position (x Lower right ，y Lower right ) And extracting the region of interest, wherein the specific operation is to set the gray value of the pixel point of the non-region of interest in the original image to zero and keep the gray value of the pixel point of the region of interest. In practical application, the method can also select to directly intercept the region of interest for operation, or can select to set the gray values of the pixel points of the non-region of interest to be any same pixel value.
Step 202: a straight line segment in a region of interest is detected. In order to effectively extract the straight line segments in the image, a straight line detection algorithm can be adopted to detect the straight line segments in the ROI image. In order to simplify the subsequent calculation process, the detected line segments are preprocessed here. As shown in fig. 4, the preprocessing process may include:
step 2021: the end point coordinates of the straight line segments are marked. Specifically, the end point of the straight line segment located at the left in the image is set as (x 1, y 1), and the end point located at the right is set as (x 2, y 2).
Step 2022: and calculating and storing the length of the linear line segment, the included angle theta between the linear line segment and the positive direction of the x axis and the distance rho between the linear line segment and the pole. In practical application, any straight line detection algorithm, such as hough transform, LSD and the like, may be used, the line segment preprocessing portion may not be performed, or other relevant information capable of representing the line segment may be calculated, so as to facilitate the subsequent processing.
Step 203: and selecting a straight line segment meeting a preset condition from the detected straight line segments as a seed straight line. The preset conditions of this embodiment may include that the included angle between the straight line segment and the positive x-axis direction is within a preset angle threshold range, and the length of the line segment is greater than a preset length threshold. The angle threshold range is a seat belt preset angle + -preset angle threshold. The method can reduce the calculation amount of subsequent processing, quicken the detection speed, reduce occupied resources and facilitate the subsequent straight line fitting flow. In practical application, other preset conditions can be set according to the actual needs of the user, relevant characteristic parameters are used, and a proper threshold value is set.
Step 204: and (5) performing line segment fitting by using the seed straight line. In this example, two fits were specifically performed. The method comprises the following steps:
the fitting conditions comprise three conditions, wherein the first condition is the distance delta distance between the end points of two linear line segments to be fitted, the second condition is the angle difference delta theta of the two linear line segments to be fitted, and the third condition is the difference delta rho between the distances of the two linear line segments to be fitted to the poles. When calculating the distance delta between the endpoints, four endpoints of two linear line segments to be fitted are selected one by one to calculate, and the minimum value is selected to represent the distance delta between the endpoints, namely, the distances between the leftmost endpoint of the first linear line segment and the leftmost endpoint and the rightmost endpoint of the second linear line segment, and the distances between the rightmost endpoint of the first linear line segment and the leftmost endpoint and the rightmost endpoint of the second linear line segment are calculated respectively, and the minimum distance is selected as the distance between the two linear line segments.
The first fitting includes finding a straight line segment fitted with the seed straight line among all the detected straight line segments. The fitting conditions for the initial fitting may be set to be relatively severe. One case is that when the threshold value of the set Δdistance is relatively large, the corresponding Δθ and Δρ should be relatively small, and the other case is that when the threshold value of the set Δdistance is relatively small, the corresponding Δθ and Δρ should be relatively large. In practical application, both cases can be considered, and only one of the cases can be emphasized.
Re-fitting involves finding the fitted line segment in the result of the primary fitting. The fitting conditions for the re-fitting may be set relatively relaxed, i.e. the thresholds are different from the initial fitting. The threshold value of the distance delta between the endpoints can be variable, such as being set as the arithmetic square root of the product of the lengths of two line segments to be fitted, so as to realize the self-adaption of the line segment fitting.
In the first fitting and the second fitting, the fitting method includes: if the two linear line segments to be fitted meet the fitting condition, selecting the leftmost endpoint of the two linear line segments to be fitted as the left endpoint of the new linear line segment, the rightmost endpoint as the right endpoint of the new linear line segment, and recalculating the length of the new linear line segment, the included angle theta between the right endpoint and the positive direction of the x axis and the distance rho between the straight line and the pole according to the left endpoint and the right endpoint.
The foregoing merely provides a quadratic fitting method, and in other embodiments, other fitting may be performed according to circumstances (e.g. whether the straight line segment detected by the straight line detecting portion is long enough), for example:
only one fitting is performed, namely, a straight line segment which is fitted with the seed straight line is searched in all detected straight line segments, wherein the fitting condition can set a threshold value of the distance between endpoints to be a fixed value or a variable value (such as the arithmetic square root of the product of the lengths of two to-be-fitted line segments), and the specific threshold values of the angle difference delta theta of the two to-be-fitted straight line segments and the difference delta rho of the distances from the straight line segments to poles can be set according to the situation;
two fitting modes are performed, and the difference between the two fitting modes is that when the two fitting modes are performed for the first time, the fitting conditions set the threshold value of the distance between the endpoints to be variable (such as the arithmetic square root of the product of the lengths of the two line segments to be fitted), and when the two fitting modes are performed for the second time, the fitting conditions set the threshold value of the distance between the endpoints to be fixed values, and the specific threshold values of the angle difference delta theta of the two line segments to be fitted and the difference delta rho of the distances from the line segments to the poles can be set according to the situation;
Multiple fits of more than two times are performed, and the threshold value of the distance between the endpoints at each fit can be selected between a fixed value and a variable value.
Other fitting conditions may also be used for the above fitting. Even in some embodiments, it may be determined directly whether the safety belt is detected in the image to be detected according to the seed straight line without performing straight line fitting.
Step 205: and judging whether the safety belt in the image to be detected is in a wearing state according to the linear line segment obtained after fitting. As shown in fig. 5, specifically, the method may include:
step 2051: calculating characteristic parameters of the linear line segments, normalizing the characteristic parameters when the characteristic parameters are more than 1, and calculating scores of the linear line segments;
step 2052: judging whether a straight line segment with the score exceeding a score threshold exists, if so, judging that the safety belt is in a wearing state, and if not, judging that the safety belt is in an unworn state.
In this embodiment, the characteristic parameters include a distance from an end point of the straight line segment to a fixed point coordinate on the belt shoulder strap and a length of the straight line segment. In the present embodiment, the characteristics of the webbing in the picture are defined, and it is considered that the closer the detected end point of the straight line segment is to the coordinates (x Fixed point ,y Fixed point ) In this case, the more likely the straight line segment is a seat belt, and the more likely the detected straight line segment is a seat belt, the longer the detected straight line segment is. According to the two features, the coordinates (x Fixed point ,y Fixed point ) The distance of the safety belt is normalized with the length of the detected straight line segment, so that the judgment of the safety belt can be carried out by adopting only a single threshold value, and the safety belt is considered asIf the detected straight line has a line segment exceeding the score threshold, the detection of the seat belt is considered, and if none of the straight lines in the image exceeds the score threshold, the detection of the seat belt is considered.
The specific normalization mode is that
Score＝score1*w1+score2*w2
Wherein Score represents the whole fraction of the detected straight line segment, score1 represents the coordinates (x Fixed point ,y Fixed point ) The score of distance is designed as a function related to distance, and score2 represents the score of length of the detected straight line segment and is designed as a function related to length. w1 and w2 represent the weights of the two parts score1 and score2, respectively. In practical application, the weight of each part can be adjusted, and other characteristic parameters capable of representing the straight line can be selected for normalization. Wherein. The distance calculation can be carried out by selecting Euclidean distance or other methods capable of representing the distance.
The method of the embodiment is divided into a safety belt information calibration flow and a safety belt detection flow. The safety belt information calibration process is only executed once, and the safety belt detection process is executed at each time of starting the vehicle or other appointed time and appointed scene. The safety belt information calibration flow is more suitable for vehicle-mounted scene application by utilizing images acquired in the vehicle without using objects such as a steering wheel, a license plate and the like in the process of calibrating safety belt information. The safety belt detection flow utilizes the information calibrated by the safety belt information calibration flow to carry out detection judgment, thereby improving the detection accuracy, reducing the calculated amount and accelerating the detection speed. The method can remind personnel of wearing the safety belt by analyzing and identifying whether the safety belt is worn correctly or not through the image, the probability of not wearing the safety belt is greatly reduced, and the corresponding manpower investment is reduced.
Example 2
The present embodiment provides a vehicle seat belt detection device for detecting whether a seat belt is in a wearing state. The safety belt detected in this embodiment is a conventional vehicle-mounted safety belt, and may be as shown in fig. 1. As shown in fig. 6, the apparatus includes: a seat belt information calibration module 301 and a seat belt detection module 302.
The seat belt information calibration module 301 is configured to calibrate seat belt information. In this embodiment, the seat belt information includes area information in which the seat belt may appear. According to the actual need or the subsequent safety belt detection flow, the safety belt information can also comprise fixed point coordinates on the safety belt shoulder belt and a preset safety belt angle. The coordinates of the fixed point on the belt shoulder strap are the coordinates of any point on the fixed point on the belt shoulder strap. The preset angle of the safety belt is an included angle between the safety belt and the positive direction of the x axis in the wearing state (mainly the included angle between the shoulder belt and the positive direction of the x axis in the wearing state of the safety belt).
In this embodiment, the belt information calibration module 301 may be specifically configured to:
the belt information is calibrated using a reference image that is acquired in the same location as the image to be detected for which the subsequent belt detection module 302 is directed.
The calibration of the safety belt information by using the reference image can specifically include:
displaying a real-time video image on a display, wherein prompt information is marked in the real-time video image, and the prompt information is used for prompting the positionable position of a shoulder belt of the safety belt and the possible occurrence area of the safety belt;
collecting an image as a reference image after a person in the vehicle sits;
Judging whether the reference image contains a fixed point on the belt shoulder strap:
if not, taking the upper intersection point of the straight line of the belt shoulder strap and the possible area of the belt as the upper fixing point of the belt shoulder strap;
if so, the area where the safety belt is likely to appear is marked by taking the fixing point on the safety belt shoulder belt as one boundary point of the area where the safety belt is likely to appear.
The reference image can be acquired by the camera in the vehicle, and is more suitable for detecting the safety belt of the vehicle in the vehicle scene.
For the same vehicle, the belt information will not change after the camera is fixed. That is, the above-mentioned belt information calibration module 301 only needs to be called once after the camera is fixed. The seat belt information may be stored for retrieval by the seat belt detection module 302. In some embodiments, the above-mentioned belt calibration module 301 may be omitted, and the belt information is set by a preset value or an empirical value.
The seat belt detection module 302 includes: a region extraction unit 3021, a line segment detection unit 3022, a preprocessing unit 3023, a seed selection unit 3024, a line segment fitting unit 3025, and a seat belt judgment unit 3026.
The region extraction unit 3021 is configured to extract a region of interest in an image to be detected using the region information, where the image to be detected and the reference image adopt the same photographing angle.
The line segment detection unit 3022 is configured to detect a straight line segment in the region of interest.
The preprocessing unit 3023 is configured to mark coordinates of an end point of a straight line segment after detecting the straight line segment in the region of interest, and calculate and store a segment length of the straight line segment, an angle with respect to the positive x-axis direction, and a distance from the straight line segment to a pole.
The seed selecting unit 3024 is configured to, after detecting the line segments in the region of interest, select, as the seed line, a line segment that meets a preset condition from the detected line segments. The preset conditions of this embodiment may include that the included angle between the straight line segment and the positive x-axis direction is within a preset angle threshold range, and the length of the line segment is greater than a preset length threshold, where the angle threshold range is a preset angle ± a preset angle threshold of the seat belt.
The line segment fitting unit 3025 is used for performing line segment fitting by using the seed straight line; in this example, two fits were specifically performed. The method comprises the following steps:
the fitting conditions comprise three conditions, wherein the first condition is the distance delta distance between the end points of two linear line segments to be fitted, the second condition is the angle difference delta theta of the two linear line segments to be fitted, and the third condition is the difference delta rho between the distances of the two linear line segments to be fitted to the poles. When calculating the distance delta between the endpoints, four endpoints of two line segments to be fitted are selected one by one for calculation, and the minimum value is selected to represent the distance delta between the endpoints, namely, the distances between the leftmost endpoint of the first line segment and the leftmost endpoint and the rightmost endpoint of the second line segment, and the distances between the rightmost endpoint of the first line segment and the leftmost endpoint and the rightmost endpoint of the second line segment are calculated respectively, and the minimum distance is selected as the distance between the two line segments.
The first fitting includes finding a straight line segment fitted with the seed straight line among all the detected straight line segments. The fitting conditions for the initial fitting may be set to be relatively severe. One case is that when the threshold value of the set Δdistance is relatively large, the corresponding Δθ and Δρ should be relatively small, and the other case is that when the threshold value of the set Δdistance is relatively small, the corresponding Δθ and Δρ should be relatively large. In practical application, both cases can be considered, and only one of the cases can be emphasized.
Re-fitting involves finding the fitted line segment in the result of the primary fitting. The fitting conditions for the re-fitting may be set relatively relaxed, i.e. the thresholds are different from the initial fitting. The threshold value of the distance delta between the endpoints can be variable, such as being set as the arithmetic square root of the product of the lengths of two line segments to be fitted, so as to realize the self-adaption of the line segment fitting.
In the first fitting and the second fitting, the fitting method includes: if the two linear line segments to be fitted meet the fitting condition, selecting the leftmost endpoint of the two linear line segments to be fitted as the left endpoint of the new linear line segment, the rightmost endpoint as the right endpoint of the new linear line segment, and recalculating the length of the new linear line segment, the included angle theta between the right endpoint and the positive direction of the x axis and the distance rho between the straight line and the pole according to the left endpoint and the right endpoint.
The foregoing merely provides a quadratic fitting method, and in other embodiments, other fitting may be performed according to circumstances (e.g. whether the straight line segment detected by the straight line detecting portion is long enough), for example:
only one fitting is performed, namely, a straight line segment which is fitted with the seed straight line is searched in all detected straight line segments, wherein the fitting condition can set a threshold value of the distance between endpoints to be a fixed value or a variable value (such as the arithmetic square root of the product of the lengths of two to-be-fitted line segments), and the specific threshold values of the angle difference delta theta of the two to-be-fitted straight line segments and the difference delta rho of the distances from the straight line segments to poles can be set according to the situation;
two fitting modes are performed, and the difference between the two fitting modes is that when the two fitting modes are performed for the first time, the fitting conditions set the threshold value of the distance between the endpoints to be variable (such as the arithmetic square root of the product of the lengths of the two line segments to be fitted), and when the two fitting modes are performed for the second time, the fitting conditions set the threshold value of the distance between the endpoints to be fixed values, and the specific threshold values of the angle difference delta theta of the two line segments to be fitted and the difference delta rho of the distances from the line segments to the poles can be set according to the situation;
Multiple fits of more than two times are performed, and the threshold value of the distance between the endpoints at each fit can be selected between a fixed value and a variable value.
Other fitting conditions may also be used for the above fitting. Even in some embodiments, it may be determined directly whether the safety belt is detected in the image to be detected according to the seed straight line without performing straight line fitting.
The belt judgment unit 3026 is configured to judge whether the belt in the image to be detected is in a wearing state according to the straight line segment. Specifically, the characteristic parameters of the straight line segment can be calculated, when the characteristic parameters are more than 1, the characteristic parameters are normalized, and the score of the straight line segment is calculated; judging whether a straight line segment with the score exceeding a score threshold exists, if so, judging that the safety belt is in a wearing state, and if not, judging that the safety belt is in an unworn state. In this embodiment, the characteristic parameters include a distance from an end point of the straight line segment to a fixed point coordinate on the belt shoulder strap and a length of the straight line segment.
The device of the present embodiment includes a seat belt information calibration module 301 and a seat belt detection module 302. The belt information calibration module 301 is executed only once, and the belt detection module 302 may execute at each vehicle start or other designated time and designated scene. The safety belt information calibration module 301 is more suitable for vehicle-mounted scene application by utilizing images acquired in a vehicle without using objects such as a steering wheel, a license plate and the like in the process of calibrating safety belt information. The safety belt detection module 302 performs detection and judgment by using the information calibrated by the safety belt information calibration flow, so that the detection accuracy is improved, the calculated amount is reduced, and the detection speed is accelerated. Whether the safety belt is correctly worn or not is recognized through image analysis on the whole, personnel can be reminded of wearing the safety belt, probability of not wearing the safety belt is greatly reduced, and corresponding manpower investment is reduced.
The device of the embodiment can be integrated in a vehicle-mounted intelligent monitoring system for monitoring and feeding back whether the driver wears the safety belt, and the position of the safety belt detection device in the system is shown in fig. 7. The video acquisition equipment can be provided with an off-vehicle camera and is used for acquiring a reference image and an image to be detected. The face recognition module and other detection modules can utilize the existing equipment on the vehicle, the detection result of the safety belt detection device can be provided for the state feedback module, and if the safety belt is not worn, voice prompt can be carried out through voice prompt equipment. The state feedback module can also upload whether the safety belt is worn to the monitoring center communication module so as to facilitate remote monitoring of the behavior of the driver.
Example 3
Fig. 8 is a schematic structural diagram of an electronic device according to embodiment 3 of the present invention. The electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, which when executed implements a vehicle seat belt detection method of embodiment 1. The electronic device 40 shown in fig. 8 is merely an example and should not be construed as limiting the functionality and scope of use of embodiments of the present invention.
As shown in fig. 8, the electronic device 40 may be embodied in the form of a general purpose computing device, which may be a server device, for example. Components of electronic device 40 may include, but are not limited to: the at least one processor 41, the at least one memory 42, a bus 43 connecting the different system components, including the memory 42 and the processor 41.
The bus 43 includes a data bus, an address bus, and a control bus.
The processor 41 executes various functional applications and data processing such as the vehicle seat belt detection method provided in embodiment 1 of the present invention by running a computer program stored in the memory 42.
The electronic device 40 may also communicate with one or more external devices 44 (e.g., keyboard, pointing device, etc.). Such communication may be through an input/output (I/O) interface 45. Also, model-generating device 40 may 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, via network adapter 46. As shown in fig. 8, the network adapter 46 communicates with the other modules of the model-generating device 40 via the bus 43. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in connection with the model-generating device 40, including, but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, data backup storage systems, and the like.
It should be noted that although several units/modules or sub-units/modules of an electronic device are mentioned in the above detailed description, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more units/modules described above may be embodied in one unit/module in accordance with embodiments of the present invention. Conversely, the features and functions of one unit/module described above may be further divided into ones that are embodied by a plurality of units/modules.
Example 4
The present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of a vehicle seat belt detection method provided in embodiment 1.
More specifically, among others, readable storage media may be employed including, but not limited to: portable disk, hard disk, random access memory, read only memory, erasable programmable read only memory, optical storage device, magnetic storage device, or any suitable combination of the foregoing.
In a possible embodiment, the invention may also be implemented in the form of a program product comprising program code for causing a terminal device to carry out the steps of implementing any one of the vehicle seat belt detection methods described in embodiment 1, when said program product is run on the terminal device.
Wherein the program code for carrying out the invention may be written in any combination of one or more programming languages, which program code may execute entirely on the user device, partly on the user device, as a stand-alone software package, partly on the user device and partly on the remote device or entirely on the remote device.
Claims (18)
1. The vehicle safety belt detection method is characterized by comprising a safety belt detection flow;
the safety belt detection process comprises the following steps:
extracting a region of interest in an image to be detected by using safety belt information, wherein the safety belt information comprises region information of the safety belt possibly appearing in the image;
detecting a straight line segment in the region of interest;
selecting a straight line segment meeting preset conditions from the straight line segments as a seed straight line;
at least one line segment fitting is carried out by utilizing the seed straight line, and a line segment obtained after fitting is used for judging whether the safety belt in the image to be detected is in a wearing state or not;
the step of judging whether the safety belt in the image to be detected is in a wearing state by using the line segments obtained after fitting comprises the following steps:
calculating characteristic parameters of each line segment obtained after fitting, normalizing the characteristic parameters when the characteristic parameters are more than 1, and calculating the score of the line segment obtained after fitting;
Judging whether a line segment obtained after fitting with the score exceeding a score threshold exists or not, if so, judging that the safety belt is in a wearing state, and if not, judging that the safety belt is in an unworn state.
2. The vehicle seat belt detection method according to claim 1, characterized in that the vehicle seat belt detection method further includes a seat belt information calibration process;
the safety belt information calibration process comprises the following steps:
and calibrating the safety belt information by using a reference image, wherein the reference image and the image to be detected are acquired at the same position.
3. The vehicle seat belt detection method according to claim 2, wherein the seat belt information further includes coordinates of a fixed point on a seat belt shoulder strap;
calibrating the seat belt information using a reference image, comprising:
displaying a real-time video image on a display, wherein prompt information is marked in the real-time video image and used for prompting the positionable position of a shoulder belt of the safety belt and the possible occurrence area of the safety belt;
collecting an image as a reference image after a person in the vehicle sits;
judging whether the reference image contains a fixed point on a belt shoulder strap or not:
if not, taking the upper intersection point of the straight line of the belt shoulder strap and the possible area of the belt as the upper fixing point of the belt shoulder strap;
If so, the area where the safety belt is likely to appear is marked by taking the fixing point on the safety belt shoulder belt as one boundary point of the area where the safety belt is likely to appear.
4. A vehicle seat belt detection method as claimed in any one of claims 1 to 3, wherein the area in which the seat belt may appear is a rectangular area, and the area information includes upper left and lower right coordinates of the rectangular area, or upper right and lower left coordinates of the rectangular area.
5. The vehicle seat belt detection method according to claim 1, wherein the seat belt information further includes a seat belt preset angle, the seat belt preset angle is an angle between the seat belt and the positive x-axis direction in a wearing state, one of the preset conditions is that an angle between a straight line segment and the positive x-axis direction is within a preset angle threshold value, and the angle threshold value range is a preset angle threshold value of the seat belt preset angle ± preset angle threshold value;
and/or, one of the preset conditions is that the length of the straight line segment is greater than a preset length threshold value.
6. The vehicle seat belt detection method according to claim 5, wherein the seat belt detection process further includes, after detecting the straight line segment in the region of interest:
Marking the end point coordinates of the linear line segment;
and calculating and storing the length of the linear line segment, the included angle between the linear line segment and the positive direction of the x-axis and the distance between the linear line segment and the pole.
7. The vehicle seat belt detection method according to claim 6, wherein if at least one line segment fitting is performed using the seed straight line, the at least one line segment fitting includes:
fitting conditions include: the distance between the end points of the two linear line segments to be fitted, the angle difference of the two linear line segments to be fitted and the difference of the distances between the two linear line segments to be fitted and the pole; when calculating the distance between the endpoints, selecting four endpoints of two linear line segments to be fitted to calculate one by one, and selecting the minimum value to represent the distance between the endpoints;
the first fitting comprises searching for a straight line segment fitted with the seed straight line from all detected straight line segments, wherein the threshold value of the distance between the endpoints is set to be a fixed value or the arithmetic square root of the product of the lengths of the two straight line segments to be fitted;
and/or, re-fitting includes finding a fitted line segment in the result of the primary fitting, wherein the threshold value of the distance between the end points is set to the arithmetic square root of the product of the lengths of the two linear line segments to be fitted or a fixed value;
In the first fitting and the second fitting, the fitting method includes: if the two linear line segments to be fitted meet the fitting condition, selecting the leftmost endpoint of the two linear line segments to be fitted as the left endpoint of the new linear line segment, the rightmost endpoint as the right endpoint of the new linear line segment, and recalculating the length of the new linear line segment, the included angle with the positive direction of the x axis and the distance from the linear line to the pole according to the left endpoint and the right endpoint.
8. The vehicle seat belt detection method according to claim 1, wherein the seat belt information further includes a fixed point coordinate on a seat belt shoulder belt, and the characteristic parameter includes a distance between an end point of the straight line segment on a side near the fixed point on the seat belt shoulder belt and the fixed point coordinate on the seat belt shoulder belt;
and/or, the characteristic parameter comprises the length of the straight line segment.
9. A vehicle seat belt detection device, characterized by comprising: a seat belt detection module;
the seat belt detection module includes:
the device comprises a region extraction unit, a detection unit and a detection unit, wherein the region extraction unit is used for extracting a region of interest in an image to be detected by utilizing safety belt information, and the safety belt information comprises region information possibly appearing in the image by the safety belt;
A line segment detection unit for detecting a straight line segment in the region of interest;
a seed selecting unit, configured to select a straight line segment meeting a preset condition from the detected straight line segments as a seed straight line after detecting the straight line segment in the region of interest;
the line segment fitting unit is used for performing line segment fitting at least once by utilizing the seed straight line, and using the line segment obtained after fitting as a judgment whether the safety belt in the image to be detected is in a wearing state or not;
the safety belt judging unit is used for judging whether the safety belt in the image to be detected is in a wearing state according to the line segment obtained after fitting;
the safety belt judging unit is specifically used for calculating characteristic parameters of each line segment obtained after fitting, normalizing the characteristic parameters when the characteristic parameters are more than 1, and calculating the score of the straight line segment; judging whether a line segment obtained after fitting with the score exceeding a score threshold exists or not, if so, judging that the safety belt is in a wearing state, and if not, judging that the safety belt is in an unworn state.
10. The vehicle seat belt detection apparatus according to claim 9, wherein the vehicle seat belt detection apparatus further comprises a seat belt information calibration module;
The safety belt information calibration module is used for calibrating the safety belt information by utilizing a reference image, and the reference image and the image to be detected are collected at the same position.
11. The vehicle seat belt detection apparatus according to claim 10, wherein the seat belt information further includes coordinates of a fixed point on a seat belt shoulder strap;
the safety belt information calibration module is used for displaying a real-time video image on a display, wherein prompt information is marked in the real-time video image and used for prompting the positionable position of a safety belt shoulder belt and the possible occurrence area of the safety belt;
collecting an image as a reference image after a person in the vehicle sits;
judging whether the reference image contains a fixed point on a belt shoulder strap or not:
if not, taking the upper intersection point of the straight line of the belt shoulder strap and the possible area of the belt as the upper fixing point of the belt shoulder strap;
if so, the area where the safety belt is likely to appear is marked by taking the fixing point on the safety belt shoulder belt as one boundary point of the area where the safety belt is likely to appear.
12. The vehicular webbing detection apparatus according to any one of claims 9-11, wherein the area in which the webbing is likely to appear is a rectangular area, and the area information includes upper left-hand and lower right-hand coordinates of the rectangular area, or upper right-hand and lower left-hand coordinates of the rectangular area.
13. The vehicle seat belt detection device according to claim 9, wherein the seat belt information further includes a seat belt preset angle, the seat belt preset angle being an angle between the seat belt and the positive x-axis direction in the wearing state, one of the preset conditions being that an angle between a straight line segment and the positive x-axis direction is within a preset angle threshold value, the angle threshold value being a preset angle threshold value of the seat belt preset angle ± preset angle threshold value;
and/or, one of the preset conditions is that the length of the straight line segment is greater than a preset length threshold value.
14. The vehicle seat belt detection apparatus according to claim 13, wherein the seat belt detection module further comprises:
and the preprocessing unit is used for marking the end point coordinates of the linear line segment after detecting the linear line segment in the region of interest, and calculating and storing the length of the linear line segment, the included angle between the linear line segment and the positive direction of the x-axis and the distance between the linear line segment and the pole.
15. The vehicle seat belt detection apparatus according to claim 14, wherein the line segment fitting unit performs at least one line segment fitting using the seed straight line, the at least one line segment fitting including:
Fitting conditions include: the distance between the end points of the two linear line segments to be fitted, the angle difference of the two linear line segments to be fitted and the difference of the distances between the two linear line segments to be fitted and the pole; when calculating the distance between the endpoints, selecting four endpoints of two linear line segments to be fitted to calculate one by one, and selecting the minimum value to represent the distance between the endpoints;
the first fitting comprises searching for a straight line segment fitted with the seed straight line from all detected straight line segments, wherein the threshold value of the distance between the endpoints is set to be a fixed value or the arithmetic square root of the product of the lengths of the two straight line segments to be fitted;
and/or, re-fitting includes finding a fitted line segment in the result of the primary fitting, wherein the threshold value of the distance between the end points is set to the arithmetic square root of the product of the lengths of the two linear line segments to be fitted or a fixed value;
in the first fitting and the second fitting, the fitting method includes: if the two linear line segments to be fitted meet the fitting condition, selecting the leftmost endpoint of the two linear line segments to be fitted as the left endpoint of the new linear line segment, the rightmost endpoint as the right endpoint of the new linear line segment, and recalculating the length of the new linear line segment, the included angle with the positive direction of the x axis and the distance from the linear line to the pole according to the left endpoint and the right endpoint.
16. The vehicle seat belt detection apparatus according to claim 9, wherein the seat belt information further includes a fixed point coordinate on a seat belt shoulder belt, and the characteristic parameter includes a distance between an end point of the straight line segment on a side near the fixed point on the seat belt shoulder belt and the fixed point coordinate on the seat belt shoulder belt;
and/or, the characteristic parameter comprises the length of the straight line segment.
17. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the seat belt detection method of any one of claims 1 to 8 when the program is executed by the processor.
18. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the steps of the seat belt detection method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011000546.9A CN112232136B (en) | 2020-09-22 | 2020-09-22 | Vehicle safety belt detection method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011000546.9A CN112232136B (en) | 2020-09-22 | 2020-09-22 | Vehicle safety belt detection method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112232136A CN112232136A (en) | 2021-01-15 |
| CN112232136B true CN112232136B (en) | 2023-05-05 |
Family
ID=74107434
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011000546.9A Active CN112232136B (en) | 2020-09-22 | 2020-09-22 | Vehicle safety belt detection method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112232136B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809099A (en) * | 2014-12-31 | 2016-07-27 | 成都理想境界科技有限公司 | Safety belt detection method based on monitor image |
| CN107330367A (en) * | 2017-05-27 | 2017-11-07 | 湖北天业云商网络科技有限公司 | A kind of Safe belt detection method and system positioned based on trunk |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914830B (en) * | 2014-02-22 | 2017-02-01 | 小米科技有限责任公司 | Straight line detection method and device |
| US11046273B2 (en) * | 2019-01-22 | 2021-06-29 | GM Global Technology Operations LLC | Seat belt status determining system and method |
| CN110458093B (en) * | 2019-08-09 | 2024-04-09 | 径卫视觉科技（上海）有限公司 | Safety belt detection method based on driver monitoring system and corresponding equipment |
| CN111582077A (en) * | 2020-04-23 | 2020-08-25 | 广州亚美智造科技有限公司 | Safety belt wearing detection method and device based on artificial intelligence software technology |
-
2020
- 2020-09-22 CN CN202011000546.9A patent/CN112232136B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809099A (en) * | 2014-12-31 | 2016-07-27 | 成都理想境界科技有限公司 | Safety belt detection method based on monitor image |
| CN107330367A (en) * | 2017-05-27 | 2017-11-07 | 湖北天业云商网络科技有限公司 | A kind of Safe belt detection method and system positioned based on trunk |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112232136A (en) | 2021-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109460699B (en) | Driver safety belt wearing identification method based on deep learning | |
| US9180887B2 (en) | Driver identification based on face data | |
| CN112149761B (en) | Electric power intelligent construction site violation detection method based on YOLOv4 improved algorithm | |
| CN110826370B (en) | Method and device for identifying identity of person in vehicle, vehicle and storage medium | |
| EP2589218B1 (en) | Automatic detection of moving object by using stereo vision technique | |
| US20130004024A1 (en) | Method and system of identifying one or more features represented in a plurality of sensor acquired data sets | |
| CN110738150B (en) | Camera linkage snapshot method and device and computer storage medium | |
| García et al. | Driver monitoring based on low-cost 3-D sensors | |
| CN111553214B (en) | Method and system for detecting smoking behavior of driver | |
| CN112613471B (en) | Face living body detection method, device and computer readable storage medium | |
| CN112381870A (en) | Ship identification and navigational speed measurement system and method based on binocular vision | |
| CN113034378B (en) | Method for distinguishing electric automobile from fuel automobile | |
| CN109389060A (en) | A kind of vehicle week anti-collision warning method of view-based access control model | |
| CN111931644A (en) | Method, system and equipment for detecting number of people on vehicle and readable storage medium | |
| CN101320477A (en) | Human body tracing method and equipment thereof | |
| CN114092857A (en) | Gateway-based collection card image acquisition method, system, equipment and storage medium | |
| CN112232136B (en) | Vehicle safety belt detection method and device, electronic equipment and storage medium | |
| CN110569732B (en) | Safety belt detection method based on driver monitoring system and corresponding equipment | |
| CN107767366A (en) | A kind of transmission line of electricity approximating method and device | |
| CN111652137A (en) | Illegal vehicle detection method and device, computer equipment and storage medium | |
| CN111241918A (en) | Vehicle anti-tracking method and system based on face recognition | |
| CN113989886B (en) | Crewman identity verification method based on face recognition | |
| US10885334B2 (en) | Method and system for detecting object(s) adjacent to vehicle | |
| CN110633637B (en) | Auxiliary driving method based on Haar-Like feature extraction algorithm and gray value difference analysis | |
| CN109886213B (en) | Fatigue state determination method, electronic device, and computer-readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |