CN102436739B - Traffic jam discrimination method in expressway toll plaza based on video detection technology - Google Patents

Abstract

The invention relates to the technical field of the detection of the traffic state of a road and particularly discloses a method for distinguishing the traffic jam of a toll plaza of a highway based on a video detection technology, which comprises the following steps: (1) shooting the video of the road of a toll plaza; (2) extracting the picture of the road of the toll plaza from the video; (3) establishing and updating a background model of the picture; (4) extracting a foreground image from the picture; and (5) obtaining the energy value and the absolute value of the variable quantity of the energy value of the foreground image and judging whether the road of the toll plaza is jammed or not. According to the method for distinguishing the traffic jam of the toll plaza of the highway based on the video detection technology, only the parameter of the energy value of the road needs to be obtained, and a jam distinguishing model is constructed, so that the judgment to the jam state of the toll plaza of the highway is finished. The algorithm is simple, the operation expense is little, the real-time property is strong, the problem of distinguishing the traffic jam of the toll plaza of the highway can be accurately and effectively solved, and a jam caution is also output at the time of jam, so that powerful information support is provided for a manager to master a site condition in time and make a management decision so as to reduce the potential safety hazard of pass in traffic.

Description

Traffic jam discrimination method in expressway toll plaza based on video detection technology

technical field

The invention relates to the technical field of road traffic state detection, in particular to a method for judging traffic congestion in expressway toll plazas based on video detection technology.

Background technique

With the rapid development of my country's economic construction, the operating mileage of expressways has increased rapidly, the number of vehicles passing through expressways has increased sharply, and the safety hazards of expressway toll stations have also increased. As a special road section where vehicles gather, expressway toll plazas are particularly prominent in terms of traffic safety. Especially during peak traffic hours, vehicles tend to queue up in toll plazas and cause congestion, which affects the driving behavior of drivers in toll plazas. Their behavior of choosing lanes with short service times leads to intensified traffic conflicts between vehicles, which in turn leads to traffic accidents. . Therefore, it is of great significance to maintain the traffic safety of the toll plaza to monitor the congestion events of the expressway toll plaza and strengthen the on-site supervision of the toll plaza. At the same time, because there are often many random factors in the occurrence of toll plaza congestion events, it is not possible to simply rely on the time period to delineate the peak period of vehicles, but to monitor the congestion status of expressway toll plazas in real time through scientific and technological means.

At present, using the video monitoring system of the toll plaza, the real-time recording of vehicle entry and exit conditions and vehicle types in the toll plaza has been realized. However, the detection of congestion events in toll plazas is still achieved by staff observing video image sequences through manual monitoring, and the automatic detection of congestion events in toll plazas has not been realized. Therefore, how to use video detection technology to automatically detect the congestion state of the toll plaza in real time is of great significance for traffic operation managers to make management decisions in real time and improve the service level of toll plazas.

The existing road traffic congestion event detection method based on video detection technology obtains a large number of traffic state parameters, such as: traffic flow, road occupancy rate, speed, inter-vehicle distance, queue length, etc., and then selects multiple parameters and uses traditional congestion discrimination algorithms to achieve Detection of road traffic congestion events. This method requires the use of image processing technology to calculate multiple parameters, which is complex and expensive, and is not conducive to real-time monitoring of road congestion events. And in the literature published at home and abroad, no detection method for toll plaza congestion event based on video detection technology has been found.

Therefore, there is an urgent need for a method to automatically detect congestion events in toll plazas, to realize timely and reliable detection of toll plaza congestion events, and to provide powerful information support for managers to grasp the site conditions in a timely manner and make management decisions, thereby reducing traffic traffic Security risks.

Contents of the invention

In view of this, the present invention provides a method for judging traffic congestion in expressway toll plazas based on video detection technology with low computing overhead and strong real-time performance.

The object of the present invention is achieved through the following technical solutions: the method for judging traffic jams in expressway toll plazas based on video detection technology comprises the steps:

1) Capture the road video of the toll plaza;

2) Extract the picture of the toll plaza road from the video;

3) Establish and update the background model of the picture;

4) extract the foreground image from the picture;

5) Obtain the energy value of the foreground image and the absolute value of the energy value change, and judge whether the current frame image is a toll plaza congestion image.

Further, said step 3) specifically includes the following steps:

31) Calculate the initial background by using the average value method for N ₀ pictures of toll plaza roads, N ₀ >20;

32) Setting the threshold, and extracting the background from the initial background through threshold judgment;

33) Use the background extracted in step 32) to establish and update the background model.

Further, in the step 4), according to the background model obtained in the step 3), the foreground image is extracted from the picture by using the difference method.

Further, step 5) specifically includes the following steps:

51) Obtain the energy value Energy(k) of the foreground image of the kth frame of the current frame;

52) Obtain the absolute value ΔEnergy(k) of the energy value difference between the foreground image of the kth frame of the current frame and the energy value of the foreground image of the k-1th frame of the previous frame, and obtain ΔEnergy(k-1), ΔEnergy(k-1) in the same way 2);

53) judge whether Energy (k) is greater than threshold value T1, if yes, then perform the next step; if not, then determine that the kth frame is a non-congested frame;

54) Determine whether the absolute value ΔEnergy(k) of the energy value difference between the energy value of the foreground image of the kth frame of the current frame and the k-1th frame of the previous frame is less than the threshold T2, if yes, then perform the next step; if not, Then it is determined that the kth frame is a non-congested frame;

55) Determine whether the absolute value ΔEnergy(k-1) of the energy value difference between the foreground image of the k-1th frame and the energy value of the foreground image of the k-2th frame of the previous frame is less than the threshold T2, if so, then perform the next step; If not, it is determined that the kth frame is a non-congested frame;

56) Determine whether the absolute value ΔEnergy(k-2) of the energy value difference between the foreground image of the k-2th frame and the energy value of the foreground image of the k-3th frame of the previous frame is less than the threshold T2, and if so, then determine that the kth frame is congestion frame; if not, then determine the kth frame as a non-congestion frame.

Further, T1=0.48, T2=0.076.

Further, the step 2) also includes the step of converting the extracted picture of the toll plaza road from a color picture to a grayscale picture.

Further, the step 4) also includes the step of performing morphological denoising on the extracted foreground image.

Further, in step 5), the method for obtaining the energy value of the foreground image is as follows: define the image sequence as k is the frame number, and N is the total frame number of the video, then the difference image is dif _k (x, y)=fr _k (x, y)-fr _k-1 (x, y); for the kth frame difference image dif _k (x, y), calculate its global threshold level, and then perform binarization processing on the difference image, and force it to be converted into a binary image; the obtained binarized difference image is It has m×n pixels; the energy of the binarized difference image is calculated by the following formula:

$\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; DB</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span>}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>$

Where b(i, j) is the value of the pixel point (i, j), that is, the energy value of the point.

Further, there are following steps after step 5):

6) According to the congestion discrimination result of the kth frame image in step 5), it is judged whether the current detection cycle is a toll plaza congestion cycle, and the specific steps are as follows:

Taking 10 seconds as a detection period, when more than 90% of the image frames in the detection period are congested frames, this period is marked as a congested period, otherwise, it is marked as a non-congested period.

Further, the following steps are also included:

7) According to the congestion discrimination result of the current detection period in step 6), judge whether the current toll plaza road is congested, and output or remove the alarm information accordingly, which specifically includes the following steps:

71) Establish a sliding window voting model, take 10 seconds as a detection period, define the capacity of the sliding window as 6 periods, and determine that the toll plaza is congested when 3 or more periods in the sliding window are congestion periods, otherwise it is non-congestion state.

72) Compare the voting result of the current cycle with the voting result of the previous cycle, if the voting result of the previous cycle is non-congested state, and the voting result of the current cycle is congested state, then output congestion warning; if the voting result of the previous cycle If the result is a congestion state, and the voting result of the current cycle is a non-congestion state, the congestion warning will be released.

The beneficial effect of the present invention is that it can accurately and efficiently solve the problem of judging traffic congestion in expressway toll plazas, and output congestion warnings at the time of congestion, thereby providing powerful information support for managers to grasp the site conditions in time and make management decisions. Thereby reducing traffic safety hazards. The present invention aims at the shortcomings of the traditional road traffic congestion discrimination method that needs to obtain a large number of traffic state parameters, high computational cost, and poor real-time performance. The present invention only needs to obtain road energy value parameters to construct a congestion discrimination model, and then completes the highway toll plaza Judgment of the congestion state, the algorithm is simple, the operation cost is small, and the real-time performance is strong.

Other advantages, objects and features of the present invention will be set forth in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the investigation and research below, or can be obtained from It is taught in the practice of the present invention. The objects and other advantages of the invention will be realized and attained by the following description and claims.

Description of drawings

Fig. 1 has shown the software processing flow schematic diagram of the highway toll plaza traffic jam discrimination method based on video detection technology;

Fig. 2 has shown the schematic flow chart of step 5), 6), 7), i.e. flow chart of judging traffic jam;

Fig. 3 shows the timing diagram of the absolute value of the energy value variation;

Figure 4 shows a timing diagram of energy values;

Figure 5 shows the actual congestion status output.

Detailed ways

Preferred embodiments of the present invention will be described in detail below. It should be understood that the preferred embodiments are only for illustrating the present invention, but not for limiting the protection scope of the present invention.

Referring to Fig. 1, the method for judging traffic jams in expressway toll plazas based on video detection technology includes the following steps:

1) Capture the road video of the toll plaza;

2) Extract the picture of the toll plaza road from the video, and convert it from a color picture to a gray scale picture;

3) Establish and update the background model of the picture; specifically include the following steps:

31) By using the mean value method to obtain the initial background for the pictures of N ₀ toll plaza roads;

32) Setting the threshold, and extracting the background from the initial background through threshold judgment;

33) Use the background extracted in step 32) to establish and update the background model.

4) extracting the foreground image from the picture, utilizing the difference method to extract the foreground image from the picture, and performing the step of morphological method denoising to the extracted foreground image;

5) Obtain the energy value of the foreground image and the absolute value of the energy value change, and judge whether the current frame image is a toll plaza congestion image. Specifically include the following steps:

k为帧号，N为视频的总帧数，则差分图像为dif_k(x，y)＝fr_k(x，y)-fr_k-1(x，y)；对于第k帧差分图像dif_k(x，y)，计算其全局阈值level，再对该差分图像进行二值化处理，强制转换为二值图像；得到的二值化后的差分图像为其有m×n个像素点；通过下式计算该二值化后的差分图像的能量：51) Obtain the energy value Energy(k) of the foreground image of the kth frame of the current frame. The method of obtaining the energy value Energy(k) of the foreground image is as follows: define the image sequence as

k is the frame number, and N is the total frame number of the video, then the difference image is dif _k (x, y)=fr _k (x, y)-fr _k-1 (x, y); for the kth frame difference image dif _k (x, y), calculate its global threshold level, and then perform binarization processing on the difference image, and force it to be converted into a binary image; the obtained binarized difference image is It has m×n pixels; the energy of the binarized difference image is calculated by the following formula:

$\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; DB</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span>}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>$

Where b(i, j) is the value of the pixel point (i, j), that is, the energy value of the point.

52) Obtain the absolute value ΔEnergy(k) of the energy value difference between the foreground image of the kth frame of the current frame and the energy value of the foreground image of the k-1th frame of the previous frame, and obtain ΔEnergy(k-1), ΔEnergy(k-1) in the same way 2);

53) judge whether Energy (k) is greater than threshold value T1, if yes, then perform the next step; if not, then determine that the kth frame is a non-congested frame;

54) Determine whether the absolute value ΔEnergy(k) of the energy value difference between the energy value of the foreground image of the kth frame of the current frame and the k-1th frame of the previous frame is less than the threshold T2, if yes, then perform the next step; if not, Then it is determined that the kth frame is a non-congested frame;

55) Determine whether the absolute value ΔEnergy(k-1) of the energy value difference between the foreground image of the k-1th frame and the energy value of the foreground image of the k-2th frame of the previous frame is less than the threshold T2, if so, then perform the next step; If not, it is determined that the kth frame is a non-congested frame;

56) Determine whether the absolute value ΔEnergy(k-2) of the energy value difference between the foreground image of the k-2th frame and the energy value of the foreground image of the k-3th frame of the previous frame is less than the threshold T2, and if so, then determine that the kth frame is congestion frame; if not, then determine the kth frame as a non-congestion frame.

6) According to the congestion discrimination result of the kth frame image in step 5), it is judged whether the current detection cycle is a toll plaza congestion cycle, and the specific steps are as follows:

Taking 10 seconds as a detection period, when more than 90% of the image frames in the detection period are congested frames, this period is marked as a congested period, otherwise, it is marked as a non-congested period.

Further, the following steps are also included:

7) According to the congestion discrimination result of the current detection period in step 6), judge whether the current toll plaza road is congested, and output or remove the alarm information accordingly, which specifically includes the following steps:

71) Establish a sliding window voting model, take 10 seconds as a detection period, define the capacity of the sliding window as 6 periods, and determine that the toll plaza is congested when 3 or more periods in the sliding window are congestion periods, otherwise it is non-congestion state.

72) Compare the voting result of the current cycle with the voting result of the previous cycle, if the voting result of the previous cycle is non-congested state, and the voting result of the current cycle is congested state, then output congestion warning; if the voting result of the previous cycle If the result is a congestion state, and the voting result of the current cycle is a non-congestion state, the congestion warning will be released.

T1 and T2 are obtained through experiments. In this embodiment, T1=0.48 and T2=0.076.

Refer to FIG. 2 for the flow diagram of the above steps 5), 6), and 7) (ie, the flow chart of traffic congestion discrimination).

In Figure 2, when the four conditions of Energy(k)>T ₁ , ΔEnergy(k)<T ₂ , ΔEnergy(k-1)<T ₂ , ΔEnergy(k-2)<T ₂ are met simultaneously, it means The energy value of the kth frame is relatively high, while the absolute values of energy value changes of the previous three adjacent frames k-1, k-2, and k-3 are relatively small. Then the kth frame is a congested frame, and the corresponding energy value of this frame is marked in red.

Further, it is judged whether the current detection period is a toll plaza congestion period according to the discrimination result of the frame image congestion situation in the period. The specific rules are: take 10 seconds as a detection cycle, when no less than 9/10 image frames in the detection cycle are congested frames, mark this cycle as a congested cycle, otherwise, mark it as a non-congested cycle.

However, to determine whether the current toll plaza road is congested, it is necessary to combine the sliding window voting model to determine whether the current toll plaza road is congested, and whether to output or release the alarm information accordingly.

In this model, 10 seconds is used as a detection cycle, and the capacity of the sliding window is defined as 6 cycles. When there are 3 or more cycles in the sliding window as congestion cycles, a congestion warning will be output, and the corresponding energy from this moment will be marked. The value line segment is red; on the contrary, when there are 4 or more periods in the sliding window that are non-congested periods, the warning will be released, and the corresponding energy value at this moment will be marked in blue.

Figure 3 shows that when there is a congestion event, the change of the energy value is very slow and very small, and the absolute value of the change of the energy value between 64 frames and 160 frames has been maintained at a small value, and traffic jams occurred during this period . By analyzing the relationship between the energy value and the absolute value of the energy value change, the discriminant rule is finally obtained: while the energy value is at a high value, when the absolute value of the energy value change remains at a small value for a period of time, the toll plaza Congestion occurs.

Finally, it is noted that the above embodiments are only used to illustrate the technical solutions of the present invention without limitation. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be carried out Modifications or equivalent replacements, without departing from the spirit and scope of the technical solution, should be included in the scope of the claims of the present invention.

Claims (9)

1. the method for judging traffic congestion in expressway toll plaza based on video detection technology, is characterized in that: comprise the steps: 1) Capture the road video of the toll plaza; 2) Extract pictures of toll plaza roads from the video; 3) Establish and update the background model of the picture; 4) Extract the foreground image from the picture; 5) Obtain the energy value of the foreground image and the absolute value of the energy value change, and judge whether the current frame image is a congested image; Among them, step 5) specifically includes the following steps: 51) Obtain the energy value Energy(k) of the foreground image of the kth frame of the current frame; 52) Obtain the absolute value ΔEnergy(k) of the energy value difference between the energy value of the foreground image in the kth frame of the current frame and the foreground image in the k-1th frame of the previous frame, and obtain ΔEnergy(k-1), ΔEnergy(k- 2); 53) Determine whether Energy(k) is greater than the threshold T1, if yes, execute the next step; if not, determine that the kth frame is a non-congested frame; 54) Determine whether the absolute value ΔEnergy(k) of the energy value difference between the energy value of the foreground image in frame k of the current frame and the foreground image in frame k-1 of the previous frame is less than the threshold T2, if yes, execute the next step; if not, Then it is determined that the kth frame is a non-congested frame; 55) Determine whether the absolute value ΔEnergy(k-1) of the energy value difference between the foreground image of the k-1th frame and the energy value of the k-2th frame of the previous frame is less than the threshold T2, and if so, execute the next step; If not, it is determined that the kth frame is a non-congested frame; 56) Determine whether the absolute value ΔEnergy(k-2) of the energy value difference between the foreground image of frame k-2 and the energy value of the foreground image of frame k-3 in the previous frame is less than the threshold T2, and if so, determine that frame k is congestion frame; if not, then determine the kth frame as a non-congestion frame. 2. The method for judging traffic congestion in expressway toll plazas based on video detection technology according to claim 1, characterized in that: said step 3) specifically includes the following steps: 31) Calculate the initial background by using the mean value method for N ₀ pictures of toll plaza roads, N ₀ >20; 32) Set the threshold, and extract the background from the initial background through threshold judgment; 33) Build and update the background model with the background extracted in step 32). 3. The method for judging traffic congestion in expressway toll plazas based on video detection technology according to claim 2, characterized in that: in step 4), the background model obtained in step 3) is extracted from the picture using the difference method foreground image. 4. The method for judging traffic congestion in expressway toll plazas based on video detection technology according to claim 1, characterized in that: in step 5), the method for obtaining the energy value of the foreground image is as follows: define the image sequence as

k is the frame number, N is the total number of frames of the video, then the differential image is dif _k (x,y)=fr _k (x,y)-fr _k-1 (x,y); for the kth frame differential image dif _k (x, y), calculate its global threshold level, and then perform binarization on the difference image, and convert it into a binary image; the obtained binarized difference image is

It has m×n pixels; the energy of the binarized difference image is calculated by the following formula: $\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; DB</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span>}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}}\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>$ Where b(i, j) is the value of the pixel point (i, j), which is the energy value of the point. 5. the method for judging traffic congestion in expressway toll plazas based on video detection technology according to claim 1, characterized in that: T1=0.48, T2=0.076. 6. The method for judging traffic congestion in expressway toll plazas based on video detection technology according to any one of claims 1 to 5, characterized in that: said step 2) also includes extracting the pictures of toll plaza roads from Steps to convert a color image to a grayscale image. 7. The method for judging traffic jams in expressway toll plazas based on video detection technology according to claim 6, characterized in that: in said step 4), it also includes the step of performing morphological denoising on the extracted foreground image . 8. The method for judging traffic congestion in expressway toll plazas based on video detection technology according to any one of claims 1 to 5, characterized in that: after step 5), there are the following steps: 6) According to the congestion judgment result of the kth frame image in step 5), judge whether the current detection cycle is a toll plaza congestion cycle, the specific steps are as follows: Taking 10 seconds as a detection period, when more than 90% of the image frames in the detection period are congested frames, this period is marked as a congested period, otherwise, it is marked as a non-congested period. 9. the method for judging traffic jams in expressway toll plazas based on video detection technology according to claim 8, is characterized in that: also comprise the steps: 7) According to the congestion judgment result of the current detection cycle in step 6), it is judged whether the road in the current toll plaza is congested, and correspondingly output or cancel the alarm information, which specifically includes the following steps: 71) Establish a sliding window voting model, take 10 seconds as a detection cycle, define the capacity of the sliding window as 6 cycles, and determine that the toll plaza is congested when there are 3 or more cycles in the sliding window as congestion cycles, otherwise it is non-congested state. 72) Compare the voting results of the current cycle with the voting results of the previous cycle. If the voting result of the previous cycle is non-congested and the voting result of the current cycle is congested, a congestion warning will be output; if the voting result of the previous cycle If the result is a congestion state, and the voting result of the current cycle is a non-congestion state, the congestion warning will be released.

Images