CN113255612A

CN113255612A - Preceding vehicle starting reminding method and system, electronic device and storage medium

Info

Publication number: CN113255612A
Application number: CN202110754724.5A
Authority: CN
Inventors: 范钧钧; 朱磊; 李成军; 张海强; 孟绍旭
Original assignee: Zhidao Wanglian Technology Shenzhen Co ltd; Zhidao Network Technology Beijing Co Ltd
Current assignee: Zhidao Wanglian Technology Shenzhen Co ltd; Zhidao Network Technology Beijing Co Ltd
Priority date: 2021-07-05
Filing date: 2021-07-05
Publication date: 2021-08-13

Abstract

The embodiment of the invention provides a preceding vehicle starting reminding method and system, electronic equipment and a storage medium, wherein the method comprises the following steps: continuously acquiring video frames based on a vehicle-mounted camera of a vehicle when the vehicle is in a parking state; the vehicle-mounted camera is used for acquiring images in front of the vehicle; determining a front vehicle detection frame and at least one detection feature point based on each video frame; wherein, the detection characteristic points are positioned in the detection frame of the front vehicle; and determining the front vehicle starting of the vehicle based on the front vehicle detection frame and the detection characteristic points corresponding to each video frame in the continuous video frames, and sending the reminding information of the front vehicle starting. The vehicle-mounted intelligent traffic reminding device can detect the state of the front vehicle in real time when the vehicle stops, timely remind when the front vehicle starts, and effectively improve traffic efficiency.

Description

Preceding vehicle starting reminding method and system, electronic device and storage medium

Technical Field

The invention relates to the technical field of intelligent driving, in particular to a method and a system for reminding the start of a preceding vehicle, an electronic device and a storage medium.

Background

When a vehicle is driven, if the vehicle is in traffic jam or red light at an intersection, the vehicle needs to stop for waiting. When a driver plays a mobile phone or is distracted while waiting, and a front vehicle cannot find the situation in time after starting and does not make a corresponding reaction in time, vehicles passing through the intersection become fewer, and congestion is aggravated.

In the prior art, by means of a starting reminding mode, under the condition that a current vehicle stops waiting, if the previous vehicle is started or the front vehicle is not in the vehicle, a system sends reminding information to remind a driver of making a corresponding response, and the problem that the road traffic rate is reduced due to untimely response of the driver is solved.

At present, the starting reminding mode in the prior art mainly comprises two modes. The first type is realized based on distance measuring sensors such as an ultrasonic radar, and has the problems of high cost, failure identification when the front is out of the vehicle or the distance is far away, and continuous alarm when the road is congested. The second type is realized based on an image recognition technology, and the problems of false detection, missed detection, local detection and false alarm when the front vehicle is not started exist.

Therefore, how to provide a method and a system for reminding the start of a preceding vehicle, an electronic device and a storage medium to remind the vehicle in time when the start of the preceding vehicle is detected so as to improve the traffic efficiency is a problem to be solved urgently.

Disclosure of Invention

Aiming at the defects in the prior art, the embodiment of the invention provides a method and a system for reminding the start of a preceding vehicle, an electronic device and a storage medium, and at least solves the technical problem that the detection result is inaccurate when the detection of whether the preceding vehicle starts or not.

The embodiment of the invention provides a preceding vehicle starting reminding method, which comprises the following steps:

continuously acquiring video frames based on a vehicle-mounted camera of a vehicle when the vehicle is in a parking state; the vehicle-mounted camera is used for acquiring images in front of the vehicle;

determining a front vehicle detection frame and at least one detection feature point based on each video frame; wherein, the detection characteristic points are positioned in the detection frame of the front vehicle;

and determining the front vehicle starting of the vehicle based on the front vehicle detection frame and the detection characteristic points corresponding to each video frame in the continuous video frames, and sending the reminding information of the front vehicle starting.

According to the preceding vehicle starting reminding method provided by the invention, the preceding vehicle detection frame and at least one detection characteristic point are determined based on the video frame, and the method specifically comprises the following steps:

determining a front vehicle detection frame based on the video frame;

determining at least three detection feature points based on the video frame; wherein, all the detection characteristic points are not positioned on the same straight line.

According to the preceding vehicle starting reminding method provided by the invention, the preceding vehicle starting of the vehicle is determined based on the preceding vehicle detection frame and the detection characteristic point corresponding to each video frame in the continuous video frames, and the preceding vehicle starting reminding information is sent, and the method specifically comprises the following steps:

determining a target video frame and a corresponding target front vehicle detection frame set according to the similar properties of the graph based on a front vehicle detection frame and a detection characteristic point corresponding to each video frame in the continuous video frames; the target front vehicle in the target video frame is the same vehicle;

and determining the front vehicle starting of the vehicle based on the target video frame and the corresponding target front vehicle detection frame set, and sending the front vehicle starting reminding information.

According to the preceding vehicle starting reminding method provided by the invention, based on the preceding vehicle detection frame and the detection characteristic point corresponding to each video frame in the continuous video frames, the target video frame and the corresponding target preceding vehicle detection frame set are determined according to the similarity of the images, and the method specifically comprises the following steps:

determining a first video frame and a first detection feature point; the method comprises the following steps that a front vehicle detection frame corresponding to a first video frame is missing; the first detection characteristic point is a detection characteristic point corresponding to the first video frame;

determining a first feature pattern based on the first detected feature points;

determining a first target front vehicle detection frame corresponding to the first video frame according to the similarity of the images based on the second front vehicle detection frame, the second characteristic image and the first characteristic image; the second characteristic pattern is matched with the first characteristic pattern; the second feature pattern is determined based on the second detected feature points; the second front vehicle detection frame and the second detection feature point are determined based on the second video frame; the second video frame is a previous frame of the first video frame;

and determining a target video frame and a corresponding target preceding vehicle detection frame set based on the first video frame, the first target preceding vehicle detection frame, the continuous video frames and the preceding vehicle detection frame corresponding to each frame.

determining a third feature pattern based on the third detection feature points, and determining a fourth feature pattern based on the fourth detection feature points; the fourth video frame is a frame before the third video frame; the fourth feature pattern is matched with the third feature pattern;

judging whether the third front vehicle detection frame shakes or not according to the similar properties of the graphs on the basis of the third front vehicle detection frame, the fourth front vehicle detection frame, the third characteristic graph and the fourth characteristic graph; wherein the third preceding vehicle detection frame and the third detection feature point are determined based on the fourth video frame; the fourth front vehicle detection frame and the fourth detection feature point are determined based on the fourth video frame;

if the third front vehicle detection frame is determined to shake, determining a third target front vehicle detection frame;

and determining a target video frame and a corresponding target front vehicle detection frame set based on the third video frame, the third target front vehicle detection frame, the continuous video frames and the front vehicle detection frame corresponding to each frame.

determining the distance change between the vehicle and the front vehicle based on the front vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames;

and determining the starting of the vehicle ahead according to the preset starting condition of the vehicle ahead based on the distance change between the vehicle and the vehicle ahead, and sending the reminding information of the starting of the vehicle ahead.

According to the method for reminding the start of the preceding vehicle provided by the invention, the start of the preceding vehicle of the vehicle is determined according to the preset condition of the start of the preceding vehicle based on the distance change between the vehicle and the preceding vehicle, and the information for reminding the start of the preceding vehicle is sent, and the method specifically comprises the following steps:

judging whether the distance change between the vehicle and the front vehicle meets the starting preset condition of the front vehicle or not based on the distance change between the vehicle and the front vehicle; the starting preset conditions of the front vehicle are as follows: the distance between the vehicle and the front vehicle is sequentially positioned in a first preset distance interval, a second preset distance interval and three preset distance intervals;

and if the change of the distance between the vehicle and the front vehicle meets the preset condition of the front vehicle starting, determining the front vehicle starting of the vehicle, and sending the reminding information of the front vehicle starting.

The invention also provides a preceding vehicle starting reminding system, which comprises: the device comprises a collecting unit, a detecting unit and an alarming unit;

the acquisition unit is used for continuously acquiring video frames based on a vehicle-mounted camera of the vehicle when the vehicle is in a parking state; the vehicle-mounted camera is used for acquiring images in front of the vehicle;

the detection unit is used for determining a front vehicle detection frame and at least one detection characteristic point based on each video frame; wherein, the detection characteristic points are positioned in the detection frame of the front vehicle;

and the alarm unit is used for determining the starting of the vehicle ahead and sending the reminding information of the starting of the vehicle ahead based on the detection frame and the detection characteristic point of the vehicle ahead corresponding to each video frame in the continuous video frames.

The embodiment of the invention provides electronic equipment, which comprises a memory and a processor, wherein the processor and the memory finish mutual communication through a bus; the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the steps of the preceding vehicle starting reminding method.

The embodiment of the invention provides a non-transitory computer readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the preceding vehicle start reminding method.

The method and the system for reminding the start of the preceding vehicle, the electronic device and the storage medium provided by the embodiment of the invention jointly determine whether the preceding vehicle of the current vehicle starts or not by combining the outside of the preceding vehicle detection frame and at least one detection characteristic point. Compared with a method for detecting by simply using a video frame extraction detection frame, the method combines a detection mode of detecting the characteristic points, can accurately judge whether the front vehicles are the same target, and reduces the influence of inaccurate extraction of the detection frame or missed detection on the detection result. The vehicle starting detection method has the advantages that whether the front vehicle starts or not is accurately detected, prompt is timely carried out when the front vehicle starts, and passing efficiency is improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

FIG. 1 is a flow chart of a preceding vehicle start reminding method provided by the invention;

FIG. 2 is a schematic flow chart of a preceding vehicle starting reminding method provided by the invention;

FIG. 3 is a schematic diagram of video frame detection according to the present invention;

FIG. 4 is a second schematic diagram of video frame detection according to the present invention;

FIG. 5 is a schematic diagram of a process for determining a starting preset condition of a preceding vehicle according to the present invention;

FIG. 6 is a schematic structural view of a preceding vehicle start reminding system provided by the invention;

fig. 7 is a schematic physical structure diagram of an electronic device provided in the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Fig. 1 is a flowchart of a preceding vehicle start reminding method provided by the present invention, and as shown in fig. 1, the preceding vehicle start reminding method provided by the present invention includes:

step S1, when the vehicle is in a parking state, continuously collecting video frames based on the vehicle-mounted camera of the vehicle; the vehicle-mounted camera is used for acquiring images in front of the vehicle;

step S2, determining a front vehicle detection frame and at least one detection feature point based on each video frame; wherein, the detection characteristic points are positioned in the detection frame of the front vehicle;

and step S3, determining the front vehicle starting of the vehicle based on the front vehicle detection frame and the detection characteristic point corresponding to each video frame in the continuous video frames, and sending the reminding information of the front vehicle starting.

It should be noted that the execution subject of the method may be a computer device.

Optionally, the vehicle-mounted camera collects video information in real time, and can assist driving and perform traffic accident processing and positioning. In the invention, the vehicle-mounted camera is used for collecting images in front of the vehicle.

When a vehicle is driven, if the vehicle is in traffic jam or red light at an intersection, the vehicle needs to stop for waiting. At this time, it may be determined that the current vehicle is in a parking state according to an on-board GPS module, an Inertial Measurement Unit (IMU), a vehicle bus, or the like. It should be noted that, in the present invention, the determination method for determining whether the vehicle is in the parking state may be set according to actual requirements, and the present invention is not limited to this.

In step S1, video frames are continuously captured based on the onboard camera mounted on the vehicle while the vehicle is currently parked. It can be understood that if there is a vehicle in front of the current vehicle, the video frame includes image information of the preceding vehicle.

In step S2, image recognition is performed on the video frames based on each video frame, and a preceding vehicle detection frame and at least one detection feature point are determined. The preceding vehicle detection frame is an identification frame for determining the position of the preceding vehicle in the video frame, and the detection feature points are located in the preceding vehicle detection frame, it can be understood that the specific number and the corresponding positions of the feature points (for example, a tail light, a tail mark, a certain point on a license plate, and the like) can be set according to the actual situation, which is not limited in the present invention.

It can be understood that, the method for determining the preceding vehicle detection frame and detecting the feature points according to the video frame may be to input the video frame into a trained deep learning detection network to determine the detected preceding vehicle detection video frame. And cutting the image according to the ROI (region of interest) of the detected video frame of the front vehicle obtained by identification, determining a detection feature point in the cut image according to a feature extraction algorithm, and recording the position information of the detection identification frame corresponding to the video frame and the detection feature point in the video frame.

It should be noted that the determination of the front vehicle Detection frame in the present invention can be realized based on Object Detection (Object Detection) algorithms such as yolo (you Only Look one), R-CNN (Region-CNN), Fast R-CNN (Fast Region-CNN), etc. In addition, the determination of the detection Feature points may be implemented based on Feature extraction (Feature extraction) algorithms such as orb ((organized FAST and Rotated BRIEF)), surf (speed Up route features), and SIFT (Scale-innovative Feature transform). In the practical application process of the present invention, the specific algorithm used may be adjusted according to the practical situation, which is not limited in the present invention.

In step S3, the motion state of the preceding vehicle is analyzed based on the preceding vehicle detection frame and the detection feature point corresponding to each of the consecutive video frames, and it is determined whether the preceding vehicle starts. And if the front vehicle starting of the vehicle is determined, sending the front vehicle starting reminding information.

It is to be understood that the method for determining the motion state of the preceding vehicle based on the detected continuous video frames may be selected according to actual conditions (for example, whether the preceding vehicle moves or not and the moving distance is determined according to the comparison between the lower edge position of the preceding vehicle detection frame and the ground relative position).

The preceding vehicle starting reminding information can be voice reminding information, main console display screen display reminding information, signal lamp flickering reminding information and the like, and in addition, the preceding vehicle starting reminding information can also be in other forms, and the invention is not limited to the above.

It should be noted that, because of the limited ability of the neural network or the algorithm to process the images, the frequency of acquiring the continuous video frames collected by the camera may be set according to the actual situation (for example, 5 video frames are selected in time sequence every second with a fixed time difference). The invention is not limited in this regard.

The method for reminding the start of the preceding vehicle provided by the embodiment of the invention determines whether the preceding vehicle of the current vehicle starts or not by combining the outside of the preceding vehicle detection frame and at least one detection characteristic point. Compared with a method for detecting by simply using a video frame extraction detection frame, the method combines a detection mode of detecting the characteristic points, can accurately judge whether the front vehicles are the same target, and reduces the influence of inaccurate extraction of the detection frame or missed detection on the detection result. The vehicle starting detection method has the advantages that whether the front vehicle starts or not is accurately detected, prompt is timely carried out when the front vehicle starts, and passing efficiency is improved.

Further, in an embodiment, according to the preceding vehicle starting reminding method provided by the present invention, based on the video frame, the preceding vehicle detection frame and the at least one detection feature point are determined, which specifically includes:

determining a front vehicle detection frame based on the video frame;

Optionally, based on the video frame, determining a preceding vehicle detection frame and at least one detection feature point specifically includes:

and based on the acquired video frame, detecting the position of the front vehicle in the video frame according to a target detection algorithm, and determining a front vehicle detection frame.

And based on the obtained video frame, determining at least three detection feature points in the front vehicle detection frame according to a feature extraction algorithm, and ensuring that all the detection feature points are not positioned on the same straight line.

It can be understood that on the premise that all the detection feature points are not located on the same straight line, 3 feature points can form a triangle, and the monitoring feature points in the front vehicle detection frame can form a graph.

Since the preceding vehicle starts based on the forward movement of the current vehicle, the proportion of the preceding vehicle in the image shot by the camera of the current vehicle is reduced, the detection frame of the preceding vehicle is reduced, and the graph formed by the detection characteristic points is also reduced. In consideration of the similarity of the images, the reduction ratio of the front vehicle detection frame relative to the complete image frame is the same as the reduction ratio of the images formed by the detection characteristic points, and the corresponding positions of the images formed by the detection characteristic points in the front vehicle detection frame are unchanged.

According to the characteristic relation of the graph formed by the video frame, the front vehicle detection frame and the detection characteristic points, the shake and the error detection of the detection frame can be effectively identified, the error detection frame is further removed or corrected, and the identification accuracy is improved.

The method for reminding the start of the preceding vehicle provided by the embodiment of the invention determines whether the preceding vehicle of the current vehicle starts or not by combining the outside of the preceding vehicle detection frame and at least three detection characteristic points which are not on the same straight line. Compared with a method for detecting by simply using a video frame to extract a detection frame, the method combines a detection characteristic point to detect, can effectively identify the shake and error detection of the detection frame according to the characteristic relation of a graph formed by the video frame, the front vehicle detection frame and the detection characteristic point, further eliminates or corrects the wrong detection frame, improves the accuracy of identifying whether the front vehicle starts or not, timely reminds when the front vehicle is determined to start, and improves the passing efficiency.

Further, in an embodiment, according to the preceding vehicle start reminding method provided by the present invention, the preceding vehicle start of the vehicle is determined based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames, and the sending of the preceding vehicle start reminding information specifically includes:

Optionally, when the current vehicle is in a stopped state, a situation that the vehicle ahead starts may occur, and when the vehicle ahead is too far away from the current vehicle, the vehicle-mounted camera of the current vehicle may not capture image information of the vehicle ahead, or the vehicle ahead drives away, and there is a situation that a new vehicle is located in front of the current vehicle (for example, the vehicle ahead changes lane and drives away from the current position when the vehicle ahead is in a traffic jam). Therefore, in order to improve the accuracy of recognition, it is necessary to determine whether a preceding vehicle (i.e., a preceding vehicle) in the captured video frame is the same target.

The method comprises the steps of determining the front vehicle starting of a vehicle based on a front vehicle detection frame and detection characteristic points corresponding to each video frame in continuous video frames, and sending reminding information of the front vehicle starting, and specifically comprises the following steps:

considering that the distance of the front vehicle moving is small in the time difference between two continuous frames of video frames, based on the front vehicle detection frame and the detection characteristic point corresponding to each video frame in the continuous video frames, whether the target front vehicle contained in the video is the same vehicle is detected according to the similarity of the images, and the video frames corresponding to all the same target front vehicle and the corresponding front vehicle detection frames are used as the target video frames and the corresponding target front vehicle detection frame sets.

It can be understood that, because the scale of the reduction of the preceding vehicle detection frame relative to the complete image frame is the same as the scale of the reduction of the pattern formed by the detection feature points in consideration of the similarity of the patterns, the corresponding positions of the patterns formed by the detection feature points in the preceding vehicle detection frame are unchanged. And according to the reduction ratio of the detection frame of the front vehicle among different frames, comparing the reduction ratio with the reduction ratio of the graph formed by the detection characteristic points, and according to whether the ratios are the same or whether the ratios are within an error allowable range, determining whether the front vehicles in different frames are the same target.

In addition, since the number of the detection feature points can be set according to actual conditions, when the number of the determined detection feature points is large, whether the vehicle is the target front vehicle or not can be confirmed by detecting the relevance between the post points. When the number of the detected feature points is small, a target tracking algorithm (for example, an SORT (simple on line And real tracking) algorithm, a depsort algorithm And the like) can be used for tracking the detection frame And the detected feature points of the preceding vehicle in combination with the detection frame of the preceding vehicle, And whether the preceding vehicle is the same target or not is determined according to the overlapping rate of the detection frames in different video frames And the matching degree of the feature points. In the practical application process of the present invention, the specific method for determining whether the front vehicle is the same target may be set according to the practical situation, which is not limited in the present invention.

And judging whether the preceding vehicle starts or not based on the target video frame corresponding to the target preceding vehicle and the corresponding target preceding vehicle detection frame set, and if the preceding vehicle starts, sending the preceding vehicle start reminding information.

It should be noted that, in a case where the preceding vehicle is not directly facing the current vehicle and the train starts to have a certain distance from the current vehicle, the acquired video frame may only include the local image information of the preceding vehicle. According to the detection characteristic points, whether the front vehicle is the same as other video frames or not can be still determined, the application range of the front vehicle starting reminding scheme provided by the invention is expanded, and the identification capability is improved.

According to the method for reminding the starting of the preceding vehicle, whether the preceding vehicle of the current vehicle is the same target or not is determined by combining the outside of a preceding vehicle detection frame and at least three detection characteristic points which are not on the same straight line, and whether the preceding vehicle starts or not is judged based on related data of the same target. Compared with a method for detecting by simply using a video frame extraction detection frame, the method combines a detection characteristic point detection mode, can screen data from the same target according to the characteristic relation of a graph formed by the video frame, a front vehicle detection frame and the detection characteristic point, effectively identifies the jitter and the error detection of the detection frame, further rejects or corrects the wrong detection frame, improves the accuracy of identification of whether a front vehicle starts, timely reminds when the front vehicle is determined to start, and improves the passing efficiency.

Further, in an embodiment, according to the preceding vehicle starting reminding method provided by the present invention, based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames, and according to the similarity of the graphs, the target video frame and the corresponding target preceding vehicle detection frame set are determined, which specifically includes:

determining a first feature pattern based on the first detected feature points;

Optionally, fig. 2 is a schematic flow chart of the preceding vehicle starting reminding method provided by the present invention, and as shown in fig. 2, when the preceding vehicle detection frame is confirmed by the video frame, a situation (missing of the target detection algorithm) occurs that the preceding vehicle detection frame is lost, that is, the video frame has corresponding detection feature points, but lacks the preceding vehicle detection frame. At this time, the missing front vehicle detection frame is filled according to the similar property of the graph.

It should be noted that, in the description of the present invention, the prefixes of the first, second, etc. are only used for distinguishing different video frames, detection frames, and feature points, and have no practical meaning, and are not used as limitations of the present invention. Further, it can be understood that the specific number of the detected feature points corresponding to the video frame may be set according to the actual situation, and the number of the feature points is not limited herein.

Based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames, determining a target video frame and a corresponding target preceding vehicle detection frame set according to the similar properties of the graph, specifically comprising:

the video frame with the missing detection condition is determined to be a first video frame (a front vehicle detection frame corresponding to the missing of the first video frame), and first detection feature points of feature points are determined based on the first video frame (the number is not limited, and the detection feature points can be a detection feature point set).

Fig. 3 is a schematic diagram of video frame detection provided by the present invention, fig. 4 is a schematic diagram of video frame detection provided by the present invention, and as shown in fig. 3 and fig. 4, the present invention is explained by taking a shape of a triangle as an example.

A first feature pattern is determined based on the first detected feature points. For example: three detected feature points (such as A, B and C in fig. 3, and a ', B' and C in fig. 4) which are not on the same straight line are selected from the detected feature points corresponding to the first video frame to form a first feature pattern (triangle). In addition, four feature points can be selected to construct a quadrilateral, and the construction rule of the specific feature graph can be set according to the actual situation, which is not limited by the invention.

And after the first video frame is determined to have the missing detection problem, filling the missing detection frame based on the previous video frame of the first video frame. And determining a video frame which is a frame before the first video frame as a second video frame based on the time sequence in the continuous video frames, and determining a second front vehicle detection frame and a second detection feature point which correspond to the second video frame. And determining a second characteristic graph based on the same rule of the generated first characteristic graph, and ensuring that the second characteristic graph is matched with the first characteristic graph.

It can be understood that, when determining the first feature pattern and the second feature pattern, a matching algorithm (e.g., ransac (random Sample consensus) algorithm, flann (fast approximation Nearest Neighbor Search library) matching algorithm, etc.) may be used to match feature points in the two frames of images, and an optimal matching feature point is selected to form the feature pattern, thereby improving the accuracy of feature pattern construction. It should be noted that the method for constructing a feature pattern with the aid of a matching algorithm is only used as a specific example to describe the present invention, and besides, in practical applications, the specific method for constructing a feature pattern may be set according to practical situations, which is not limited by the present invention.

Based on the second preceding vehicle detection frame, the second characteristic graph and the first characteristic graph, according to the similarity of the graphs and the proportion of the line length, the relative relationship between the preceding vehicle detection frame and the characteristic graph and the related line length can be calculated, the preceding vehicle detection frame is filled based on the relative relationship, and the first target preceding vehicle detection frame corresponding to the first video frame is determined.

For example: the scheme of filling the front vehicle detection frame according to the present invention is explained by taking fig. 4 as a first video frame and fig. 3 as a second video frame as an example.

And determining the pixel positions of three characteristic points constructing the first characteristic graph in the first video frame in the video frame, namely A ', B ' and C '. And the corresponding first preceding vehicle detection frame is missed due to the first video frame. At this time, the previous frame is determined to be a second video frame, a second detection frame (b 5 and b7 are rectangles with side lengths) corresponding to the second video frame is taken as a Point of Interest (POI), and pixel positions of three feature points corresponding to the second feature graph in the video frame are determined according to the same rule, wherein the pixel positions are A, B and C respectively.

And matching the characteristic points of the front frame image and the rear frame image, and performing triangular similarity calculation to find out a characteristic point pair with higher similarity. The characteristic point matching A, B and C with A ', B' and C 'can form three characteristic point pairs of A-A', B-B 'and C-C'. The feature point determination with high similarity may be based on featuresWhether the difference value of the side length ratios of the characteristic graphs is larger than a preset threshold value is judged, for example: set up K_minAs a preset threshold value, ensuring that the side length of the formed triangle needs to meet the formula of a1/b1-a2/b2<K_minAnd a2/b2-a3/b3<K_min。

Because the ratio of the distance between the detection characteristic points of the front frame and the rear frame to the length and the width of the whole vehicle is not changed, the position of the detection frame of the vehicle before missed detection can be calculated.

For example: the side lengths a5 and a7 of the first target front vehicle detection frame can be deduced according to the side lengths b5 and b7 of the second front vehicle detection frame, the transverse distance b4 and the longitudinal distance b6 of the second feature pattern in the second front vehicle detection frame, and the transverse distance a4 and the longitudinal distance a6 of the second feature pattern in the second front vehicle detection frame, and the relative positions of the detection frame and the feature pattern can be determined according to the length relationship of the line segment, so that the missed front vehicle detection frame can be accurately filled.

The calculation formula is as follows: b7= a7/a6 b6, b5= a5/a4 b4, b8= a8/a6 b6 and b9= a9/a4 b 4. It can be understood that the proportion of the front vehicle detection frame in the front frame and the rear frame can be calculated according to the side length of the feature graph in the front frame and the rear frame and according to the property of the similar triangle. Except for marking line segments in the figure, the length of other line segments can be calculated, and the specific calculation method has the same conception as the method and is not repeated herein.

And determining a first target preceding vehicle detection frame, namely filling the condition of missing detection in the continuous video frame sequence, and determining a target video frame and a corresponding target preceding vehicle detection frame set based on the first video frame, the first target preceding vehicle detection frame, the continuous video frame and a preceding vehicle detection frame corresponding to each frame.

The missing front vehicle detection frame may be filled in by identifying the front vehicle detection frame after each time the front vehicle detection frame is determined, and filling the front vehicle detection frame in real time if the front vehicle detection frame is determined to be missing. In addition, a mode of determining a continuous video frame set containing a target number of video frames once, identifying the detection frames after determining the previous vehicle detection frames corresponding to all the video frames in all the sets, judging whether the detection frames are missing or not, and filling the missing previous vehicle detection frames can be adopted. In the practical application process, the detection missing recognition of the detection frame and the filling time of the detection missing frame can be set according to the practical situation, and the invention is not limited to this.

The method for reminding the start of the preceding vehicle provided by the embodiment of the invention determines whether the preceding vehicle of the current vehicle starts or not by combining the outside of the preceding vehicle detection frame and at least one detection characteristic point. Compared with a method for detecting by simply using a video frame extraction detection frame, the method combines a detection mode of detecting the characteristic points, can accurately judge whether the front vehicles are the same target, and reduces the influence of inaccurate extraction of the detection frame or missed detection on the detection result. Under the condition that the detection omission of the front vehicle detection frame is determined, the missing front vehicle detection frame is filled based on the detection graph formed by the detection characteristic points according to the similarity of the graph, the situation that the corresponding video frame cannot be used or the detection omission occurs due to the detection omission is avoided, the accuracy of whether the front vehicle starts to detect is improved, the prompt is timely carried out when the front vehicle starts to be determined, and the passing efficiency is improved.

Optionally, fig. 2 is a schematic flow chart of the preceding vehicle starting reminding method provided by the present invention, as shown in fig. 2, when the preceding vehicle detection frame is confirmed through the video frame, since the camera may be affected by factors such as light brightness change and decoding noise, even if the preceding vehicle is not started, the microscopic pixels of the video frame have great changes, which causes the problem of shaking of the preceding vehicle detection frame, and causes false detection of the state of the preceding vehicle (for example, the preceding vehicle is not started, but the recognition result is the starting of the preceding vehicle due to the serious shaking of the preceding vehicle detection frame). At this time, the jittered preceding vehicle detection frame is corrected according to the similar property of the graph.

In the description of the present invention, the prefixes of the third, fourth, etc. are merely used to distinguish different video frames, detection frames, and feature points, and are not practical, and are not intended to limit the present invention. Further, it can be understood that the specific number of the detected feature points corresponding to the video frame may be set according to the actual situation, and the number of the feature points is not limited herein.

And detecting whether the front vehicle detection frame of the next video frame is too large in jitter amplitude compared with the front video frame or not based on any two adjacent video frames in the continuous video frames.

And determining the video frame to be detected as a third video frame, and taking the previous frame image of the third video frame as a fourth video frame. And determining a third feature pattern according to the third feature point corresponding to the third video frame, and determining a fourth feature pattern according to the fourth detection feature point corresponding to the fourth video frame.

It should be noted that the method for determining the feature pattern here is the same as the method for determining the feature pattern in the method for filling the missed front vehicle detection frame, and is not described herein again.

And judging whether the third front vehicle detection frame shakes or not according to the similar properties of the graphs based on the third front vehicle detection frame, the fourth front vehicle detection frame, the third characteristic graph and the fourth characteristic graph.

For example: the scheme of the present invention for correcting the shake of the leading car detection frame is explained by taking fig. 4 as a third video frame and fig. 3 as a fourth video frame as an example.

And determining the pixel positions of three feature points constructing a third feature graph in the third video frame in the video frame, wherein the pixel positions are A ', B ' and C ', respectively. The pixel positions of three feature points corresponding to the fourth feature pattern in the video frame are determined by the same rule, wherein the pixel positions are A, B and C respectively.

Thus, it can be determined that the third feature pattern is a triangle A ' B ' C ' and the fourth feature pattern is a triangle ABC. And determining the side length ratio of the two triangles of the third characteristic graph and the fourth characteristic graph according to the similarity of the triangles, wherein the side length ratio of the corresponding third front vehicle detection frame and the fourth front vehicle detection frame can be determined because the ratio of the distance between the front and rear two frames of detection characteristic points to the length and width positions of the whole vehicle is not changed.

It can be understood that, the method for judging whether the third preceding vehicle detection frame shakes may be to calculate a difference between a mean value of the side length ratios of the feature patterns and a mean value of the side length ratios of the preceding vehicle detection frames, compare the difference with a preset error threshold, and if it is determined that the difference is greater than the error threshold, consider that the third preceding vehicle detection frame shakes.

In addition, the side length of the front vehicle detection frame can be calculated according to the proportion obtained by calculating the characteristic graph, and whether the front vehicle detection frame shakes or not is determined based on the difference between the calculated value and the actual value of the side length. Or, calculating the transverse distance proportion and the longitudinal distance proportion of the feature patterns in the two frames of images in the detection frame, and determining whether the detection frame of the front vehicle shakes or not according to the comparison between the transverse distance proportion and the longitudinal distance proportion. In the practical application process of the present invention, the specific method for determining whether the front vehicle detection frame shakes according to the similar property of the image may be set according to the practical situation, which is not limited in the present invention.

In addition, it can be understood that, in the process, on the basis of ensuring that the used feature points are correspondingly matched in the front and rear two frames of video frames, the detection of whether the front vehicle detection frame shakes can be realized only by using the line segment formed by the feature points, and a feature graph does not need to be formed necessarily. The specific application method can be set according to the actual situation, which is not limited in the present invention.

And if the third front vehicle detection frame is determined to shake, correcting the third front vehicle detection frame and determining a third target front vehicle detection frame.

It should be noted that the step of correcting the third preceding vehicle detection frame is to generate a new third preceding vehicle detection frame corresponding to the third video frame based on the fourth video frame, the fourth preceding vehicle detection frame, the fourth detection feature point, the third video frame, and the third detection feature point according to the similar property of the graph. The specific generation steps are the same as the filling method for the missed front vehicle detection frame, and are not described herein again.

And determining a third target preceding vehicle detection frame, namely correcting a jittering preceding vehicle detection frame in the continuous video frame sequence, and determining a target video frame and a corresponding target preceding vehicle detection frame set based on the third video frame, the third target preceding vehicle detection frame, the continuous video frame and a preceding vehicle detection frame corresponding to each frame.

The method for correcting the jittering front vehicle detection frame may be to determine whether the front vehicle detection frame jitters after each determination of the front vehicle detection frame, and correct the front vehicle detection frame in real time if the determination of the jitter of the front vehicle detection frame is made. In addition, a mode of determining a continuous video frame set containing a target number of video frames at a time, identifying the detection frames after determining the front vehicle detection frames corresponding to all the video frames in all the sets, judging whether the front vehicle detection frames are jittered or not, and correcting the jittered front vehicle detection frames can be adopted. In the actual application process, the identification of whether the front vehicle detection frame shakes or not and the timing for correcting the shake detection frame can be set according to the actual situation, and the invention is not limited to this.

The method for reminding the start of the preceding vehicle provided by the embodiment of the invention determines whether the preceding vehicle of the current vehicle starts or not by combining the outside of the preceding vehicle detection frame and at least one detection characteristic point. Compared with a method for detecting by simply using a video frame extraction detection frame, the method combines a detection mode of detecting the characteristic points, can accurately judge whether the front vehicles are the same target, and reduces the influence of inaccurate extraction of the detection frame or missed detection on the detection result. Under the condition that the front vehicle detection frame shakes, the shaken front vehicle detection frame is corrected based on a detection graph and the front vehicle detection frame which are formed by detection feature points of adjacent video frames in the continuous video frame sequence according to the similarity of the graphs, the situation that false detection occurs due to shaking of the front vehicle detection frame is avoided, the accuracy of whether a front vehicle starts to detect is improved, prompt is timely given when the front vehicle starts to be determined, and the passing efficiency is improved.

Optionally, fig. 2 is a schematic flow chart of the preceding vehicle start reminding method provided by the present invention, and as shown in fig. 2, the method for determining the starting of the vehicle in front and sending the preceding vehicle start reminding information based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames specifically includes:

based on a previous vehicle detection frame and a detection feature point corresponding to each video frame in the continuous video frames, the initial positions of the current vehicle and the previous vehicle in a static state are used as references, the distance between the current vehicle and the previous vehicle corresponding to different video frames is determined, and the distance change between the vehicle and the previous vehicle is determined.

It will be appreciated that since the distance between the current vehicle and the preceding vehicle is generally short when traffic is jammed or waiting for a red light, the distance can be directly considered as the distance between the two vehicles. Further, it can be considered that the moving distance of the preceding vehicle is considered, and since the current vehicle is in a stopped state, the relative distance between the current vehicle and the preceding vehicle is ignored initially. For example: the distance is 0 when the front vehicle does not move, and the distance is 1m when the front vehicle drives forwards by 1 m.

And judging whether the front vehicle of the current vehicle starts or not according to the preset front vehicle starting condition set by the realization based on the distance change between the vehicle and the front vehicle.

And if the previous vehicle starting of the current vehicle is determined, sending the previous vehicle starting reminding information.

It should be noted that the preset condition for starting the preceding vehicle may be that the distance of the preceding vehicle exceeds a preset threshold, the speed of the preceding vehicle exceeds a preset threshold, the distance between the current vehicle and the preceding vehicle continuously increases within a preset time range, and the like, and the specific preset condition for starting the preceding vehicle may be set according to an actual situation, which is not limited in the present invention.

The method for reminding the start of the preceding vehicle provided by the embodiment of the invention determines whether the preceding vehicle of the current vehicle starts or not by combining the outside of the preceding vehicle detection frame and at least one detection characteristic point. Compared with a method for detecting by simply using a video frame extraction detection frame, the method combines a detection mode of detecting the characteristic points, can accurately judge whether the front vehicles are the same target, and reduces the influence of inaccurate extraction of the detection frame or missed detection on the detection result. The method comprises the steps of determining the change of the distance between a current vehicle and a preceding vehicle based on the detection of continuous video frames, realizing the accurate detection of whether the preceding vehicle starts or not according to a preset starting condition of the preceding vehicle, timely reminding when the preceding vehicle starts, and improving the passing efficiency.

Further, in an embodiment, according to the method for prompting vehicle start ahead provided by the present invention, based on a distance change between a vehicle and a vehicle ahead, determining vehicle start ahead of the vehicle according to a preset condition for vehicle start ahead, and sending vehicle start ahead prompting information, specifically including:

Optionally, based on the distance change between the vehicle and the preceding vehicle, determining the starting of the preceding vehicle of the vehicle according to a preset starting condition of the preceding vehicle, and sending a reminding message of the starting of the preceding vehicle, the method specifically includes:

fig. 5 is a schematic diagram of a process for determining a starting preset condition of a preceding vehicle, and as shown in fig. 5, it is determined whether a change in the distance between the vehicle and the preceding vehicle meets the starting preset condition of the preceding vehicle based on a change in the distance between the vehicle and the preceding vehicle.

The preset starting condition of the front vehicle divides the starting of the front vehicle into three states, namely an initial stage, a preparation reminding stage and a reminding stage. Each stage corresponds to a preset distance interval. For example: the initial stage corresponds to a first preset distance interval of 0-2m, the preparation reminding stage corresponds to a first preset distance interval of 2-3m, and the reminding stage corresponds to a third preset distance interval of more than 3 m. And determining that the starting preset condition of the front vehicle is met only if the distance between the current vehicle and the front vehicle is ensured to be sequentially located in a first preset distance interval, a second preset distance interval and a third preset distance interval.

It should be noted that, the specific value ranges corresponding to the different intervals may be set according to actual situations, and the present invention is not limited to this.

It can be understood that, when the detection of whether the front vehicle starts or not is carried out at present, a preset threshold value is generally determined, and when the distance is greater than the preset threshold value, the front vehicle starts to be reminded. However, due to the fact that the detection frame shakes, for example, the preset threshold is set to be 3m, and if the distance between the current vehicle and the vehicle in front is determined to exceed 3m, the vehicle in front is determined to start. Since the detection frame shakes around 3m all the time, the situation that the distance obtained by recognition exceeds 3m may occur, and false reminding is caused. The recognition accuracy is seriously degraded and the user experience is poor.

According to the invention, different distance intervals are set, so that the continuous starting action of the front vehicle is identified when the vehicle starts, the influence of accidental errors such as shaking on the identification result is avoided, and the situation of mistaken identification is avoided.

The method for reminding the start of the preceding vehicle provided by the embodiment of the invention determines whether the preceding vehicle of the current vehicle starts or not by combining the outside of the preceding vehicle detection frame and at least one detection characteristic point. Compared with a method for detecting by simply using a video frame extraction detection frame, the method combines a detection mode of detecting the characteristic points, can accurately judge whether the front vehicles are the same target, and reduces the influence of inaccurate extraction of the detection frame or missed detection on the detection result. The method comprises the steps of determining the change of the distance between a current vehicle and a preceding vehicle based on the detection of continuous video frames, determining the starting of the preceding vehicle only when the distance change sequence of the current vehicle and the preceding vehicle is sequentially located in three preset distance intervals according to preset starting preset conditions of the preceding vehicle, and timely reminding when the starting of the preceding vehicle is determined, so that the influence of accidental errors on identification results is effectively avoided, the occurrence of false identification conditions is avoided, and the identification accuracy and the user experience are improved.

Fig. 6 is a schematic structural view of a preceding vehicle start reminding system provided by the present invention, and as shown in fig. 6, the present invention further provides a preceding vehicle start reminding system, which includes: the system comprises a collecting unit 610, a detecting unit 620 and an alarming unit 630;

the acquisition unit 610 is used for continuously acquiring video frames based on a vehicle-mounted camera of a vehicle when the vehicle is in a parking state; the vehicle-mounted camera is used for acquiring images in front of the vehicle;

a detection unit 620, configured to determine a preceding vehicle detection frame and at least one detection feature point based on each video frame; wherein, the detection characteristic points are positioned in the detection frame of the front vehicle;

and the alarm unit 630 is configured to determine the starting of the vehicle ahead based on the detection frame and the detection feature point corresponding to each of the consecutive video frames, and send the starting prompt information of the vehicle ahead.

The vehicle-mounted camera collects video information in real time, and can assist driving and carry out traffic accident processing and positioning. In the invention, the vehicle-mounted camera is used for collecting images in front of the vehicle.

And the acquisition unit 610 is used for continuously acquiring video frames based on a vehicle-mounted camera loaded on the vehicle when the current vehicle is in a parking state. It can be understood that if there is a vehicle in front of the current vehicle, the video frame includes image information of the preceding vehicle.

And the detection unit 620 is configured to perform image recognition on the video frames based on each video frame, and determine a preceding vehicle detection frame and at least one detection feature point. The preceding vehicle detection frame is an identification frame for determining the position of the preceding vehicle in the video frame, and the detection feature points are located in the preceding vehicle detection frame, it can be understood that the specific number and the corresponding positions of the feature points (for example, a tail light, a tail mark, a certain point on a license plate, and the like) can be set according to the actual situation, which is not limited in the present invention.

And the alarm unit 630 is configured to analyze a motion state of the preceding vehicle based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the consecutive video frames, and determine whether the preceding vehicle starts. And if the front vehicle starting of the vehicle is determined, sending the front vehicle starting reminding information.

The preceding vehicle starting reminding system provided by the embodiment of the invention determines whether the preceding vehicle of the current vehicle starts or not through combining the outside of the preceding vehicle detection frame and at least one detection characteristic point. Compared with a method for detecting by simply using a video frame extraction detection frame, the method combines a detection mode of detecting the characteristic points, can accurately judge whether the front vehicles are the same target, and reduces the influence of inaccurate extraction of the detection frame or missed detection on the detection result. The vehicle starting detection method has the advantages that whether the front vehicle starts or not is accurately detected, prompt is timely carried out when the front vehicle starts, and passing efficiency is improved.

Fig. 7 is a schematic entity structure diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 7, the electronic device may include: a processor (processor) 711, a communication interface (communication interface) 712, a memory (memory) 713 and a communication bus (bus) 714, wherein the processor 711, the communication interface 712 and the memory 713 complete communication with each other through the communication bus 714. The processor 711 may call the logic instructions in the memory 713 to execute the preceding vehicle start reminding method, which includes: continuously acquiring video frames based on a vehicle-mounted camera of a vehicle when the vehicle is in a parking state; the vehicle-mounted camera is used for acquiring images in front of the vehicle; determining a front vehicle detection frame and at least one detection feature point based on each video frame; wherein, the detection characteristic points are positioned in the detection frame of the front vehicle; and determining the front vehicle starting of the vehicle based on the front vehicle detection frame and the detection characteristic points corresponding to each video frame in the continuous video frames, and sending the reminding information of the front vehicle starting.

In addition, the logic instructions in the memory 713 may be implemented in the form of software functional units and stored in a computer readable storage medium when sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like.

In another aspect, an embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions, and when the program instructions are executed by a computer, the computer is capable of executing the preceding vehicle start reminding method provided by the foregoing method embodiments, including: continuously acquiring video frames based on a vehicle-mounted camera of a vehicle when the vehicle is in a parking state; the vehicle-mounted camera is used for acquiring images in front of the vehicle; determining a front vehicle detection frame and at least one detection feature point based on each video frame; wherein, the detection characteristic points are positioned in the detection frame of the front vehicle; and determining the front vehicle starting of the vehicle based on the front vehicle detection frame and the detection characteristic points corresponding to each video frame in the continuous video frames, and sending the reminding information of the front vehicle starting.

In another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented to perform the method for reminding a vehicle to start in front provided by the foregoing embodiments when executed by a processor, and the method includes: continuously acquiring video frames based on a vehicle-mounted camera of a vehicle when the vehicle is in a parking state; the vehicle-mounted camera is used for acquiring images in front of the vehicle; determining a front vehicle detection frame and at least one detection feature point based on each video frame; wherein, the detection characteristic points are positioned in the detection frame of the front vehicle; and determining the front vehicle starting of the vehicle based on the front vehicle detection frame and the detection characteristic points corresponding to each video frame in the continuous video frames, and sending the reminding information of the front vehicle starting.

The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods of the various embodiments or some parts of the embodiments.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A preceding vehicle starting reminding method is characterized by comprising the following steps:

2. The preceding vehicle starting reminding method according to claim 1, wherein the preceding vehicle detection frame and the at least one detection feature point are determined based on a video frame, and the method specifically comprises the following steps:

determining a front vehicle detection frame based on the video frame;

3. The preceding vehicle starting reminding method according to claim 2, wherein the preceding vehicle starting of the vehicle is determined based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames, and the preceding vehicle starting reminding information is sent, and the method specifically comprises the following steps:

4. The preceding vehicle starting reminding method according to claim 3, wherein the determining of the target video frame and the corresponding target preceding vehicle detection frame set based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames according to the similarity of the graphs specifically comprises:

determining a first feature pattern based on the first detected feature points;

5. The preceding vehicle starting reminding method according to claim 3, wherein the determining of the target video frame and the corresponding target preceding vehicle detection frame set based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames according to the similarity of the graphs specifically comprises:

6. The preceding vehicle starting reminding method according to any one of claims 1 to 5, wherein the preceding vehicle starting of the vehicle is determined based on the preceding vehicle detection frame and the detection feature point corresponding to each video frame in the continuous video frames, and the preceding vehicle starting reminding information is sent, and specifically comprises:

7. The preceding vehicle starting reminding method according to claim 6, wherein the preceding vehicle starting of the vehicle is determined according to a preset preceding vehicle starting condition based on the distance change between the vehicle and the preceding vehicle, and the preceding vehicle starting reminding information is sent, and specifically comprises:

8. The utility model provides a preceding car start warning system which characterized in that includes: the device comprises a collecting unit, a detecting unit and an alarming unit;

9. An electronic device, comprising a memory and a processor, wherein the processor and the memory communicate with each other via a bus; the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the preceding vehicle start reminding method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the preceding vehicle start reminding method according to any one of claims 1 to 7.