WO2016202012A1

WO2016202012A1 - Traffic information detection method, acquiring method and acquiring apparatus based on traffic monitoring video

Info

Publication number: WO2016202012A1
Application number: PCT/CN2016/075557
Authority: WO
Inventors: 朱斐; 刘全; 伏玉琛
Original assignee: 苏州大学张家港工业技术研究院
Priority date: 2015-06-17
Filing date: 2016-03-03
Publication date: 2016-12-22

Abstract

A traffic information detection method, acquiring method and acquiring apparatus based on a traffic monitoring video. A video processor is provided corresponding to each monitoring device, and each video processor is connected to a remote server through a network module. The video processor reads a video image on the corresponding monitoring device, calculates a grayscale value X of each pixel in each frame of picture according to obtained pictures, compares the grayscale value X with a pre-stored parameter value X ^-1, assesses, by means of a kernel method, a road congestion degree C _i observed from the video processor, and uploads the road congestion degree C _i. The remote server analyzes the real-time road congestion degrees C _i to obtain a road congestion degree value C of a road from an intersection to a next intersection. A data pusher sends information about a congested road to a data receiver, and the data receiver outputs the information about the congested road. The video processor analyzes the real-time image grayscale, assesses the congestion degree by means of the kernel method, and uploads the congestion degree, the data pusher sends the information in real time, and the data receiver broadcasts the information in real time, so that a driver acquires traffic information in real time and timely plans a driving route, thereby alleviating congestion.

Description

Instruction manual

Invention name: road condition detection method, acquisition method and acquisition device based on traffic monitoring video

Technical field

[0001] The present invention relates to a method for identifying road traffic conditions, and more particularly to a road condition acquisition method and device based on traffic monitoring video.

Background technique

[0002] Transportation is the foundation of modern society and the lifeblood of human society. People's social behavior is closely related to transportation. In a city, the number of motor vehicles and non-motor vehicles is large, and the intersections and road sections are complicated. It is very complicated to deal with such a large-scale, dynamic and highly uncertain distributed system. work. In the absence of new traffic roads, improving traffic utilization efficiency through reasonable traffic control, and thus improving traffic efficiency is an effective way to quickly solve urban traffic problems.

[0003] However, traffic congestion and congestion are now becoming more serious. The reason for the traffic problem is that on the one hand, due to more and more vehicles, traffic planning and design lags behind, on the other hand, many traffic signal control systems are relatively backward, and traffic lights are not well regulated according to actual traffic conditions. To improve the efficiency of traffic. The use of computing technology and machine intelligence to help solve traffic problems has become more and more important, and has become a trend.

[0004] In recent years, a large number of traffic monitoring equipments have been put into use, and traffic video data has been continuously transmitted to traffic management departments, and traffic video data has exploded in data. These traffic video data only needs a lot of time and bandwidth to be transmitted to the server. In addition, the analysis data for processing the video data to obtain traffic conditions requires a large amount of time, so that the obtained data has high requirements on the traffic signal system with high requirements. In terms of traffic signal decision-making, it has a long delay and is not suitable for real-time decision control of traffic signals. Therefore, how to make full use of these traffic video data and improve the control of road traffic signals to improve road traffic efficiency is becoming more and more important.

technical problem Problem solution

Technical solution

[0005] An object of the present invention is to provide a road condition detection method, an acquisition method, and an acquisition device based on a traffic monitoring video. The detection method, the acquisition method, and the acquisition device can obtain information about road conditions of each intersection, and Feedback to the receiver, it is convenient for traffic participants and 吋 to make route planning, to achieve "car-way coordination", thereby reducing congestion and easing traffic pressure.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: A road condition detection method based on traffic monitoring video, comprising: setting a video processor corresponding to each monitoring device, each of the video processors passing through a network module Connect to a remote server, where:

[0007] the video processor reads a video image corresponding to the monitoring device, and obtains a gray value X of each pixel in each frame according to the obtained image, and compares it with the pre-stored parameter value X- ¹ . When it is greater than XX - ¹ >0, Bay ljnum i_l= nuni i-l+l , No Bay Unum ;_0= num i—0+1, by collecting num and num;_0 values, forming a real gray value ei= ( _nU m i_l, nuni iJ)), evaluating a road congestion level ci observed on the video processor by a core method, and uploading to the remote server through the network module;

[0008] The remote server collects the congestion degree c of the road segment uploaded by each video processor. The remote server uses the weight of the original data and the received congestion degree ci of each real road segment to obtain an intersection. The road congestion degree value C of the next intersection is expressed as the actual online traffic state of the road section.

[0009] In the above technical solution, the pre-stored parameter value X- ^{1 is} stored on a remote server, including the gray value G i of the road color and the error value _ei , and the gray color of the road is changed as the daytime and the weather change. One or two of the degree value G i and the error value ε i are sent by the remote server to each of the video processors through the network module

[0010] A further technical solution is that, according to the change of the daytime, each time a small 吋~4 small 吋 calls the data pre-stored in the remote server, the gray value G of the road color on each video processor i and the error value ε i are reassigned.

[0011] Further, the gray value of each pixel in each frame picture is X=0.11xr+0.59xg+0.3xb, where r represents red brightness, g represents green brightness, and b represents blue brightness, when ( XG i) ² - ε; > 0, then num i _1 = num _; _1+1, no Bay Unum _; _0= num _; _0+1 .

[0012] In the above technical solution, the “core method” is used to evaluate the road congestion period observed on the video processor. The way ci" is:

[0013] a remote server corresponding to each video processor i has a data storage space, storing the previous gray value data ε, where j = l, 2 ... ... N;

[0014] b each video processor i calls the e data in the remote server, using formula (1) to compare the similarity with the currently obtained real gray value ei;

(1)

[0016] wherein

&, =, m: im, -I,

[0017] c. Each video processor i further determines the degree of congestion of the road segment that can be observed by the current monitoring according to formula (2)

C i ;

(2)

[0019] d finally through the network module to upload the congestion degree c i of the road segment and the real gray value e i to the remote server

[0020] In the above technical solution, the “remote server uses the weight of the original data” is a set of grayscale value data uploaded by the video processor of the road segment in the remote server storage space by using a method of random gradient descent. Calculated weight

, the congestion degree value of the road section from one intersection to the next intersection

, where / represents a collection of all video processors for that segment.

[0021] In the foregoing technical solution, the network module is an Ethernet cable module or a wireless data transmission network module.

[0022] The present invention further includes another technical solution: a road condition acquisition method based on traffic monitoring video, including setting a video processor corresponding to each monitoring device, each of the video processors being connected to a remote server via a network module a data pusher disposed beside the video processor and a data receiver disposed on the vehicle, wherein:

[0023] the video processor reads a video image corresponding to the monitoring device, and obtains a gray value X of each pixel in each frame according to the obtained image, and compares it with the pre-stored parameter value X- ¹ . Greater than XX - ¹

>0, 贝ljnum i_l= nuni i-l+l , no Bay Unum i_0= num i—0+1, by collecting num i_l and num i_0 values, forming a real gray value ei=( _nU m i_l, nuni iJ)), evaluating, by the core method, the road congestion level ci observed on the video processor, and then uploading to the remote server through the network module;

[0024] the remote server collects the congestion degree c of the road segment uploaded by each video processor. The remote server uses the weight of the original data and the received congestion degree ci of each real road segment to obtain an intersection. The road congestion degree value C of the next intersection is expressed as the actual online traffic state of the road section;

[0025] the data pusher sends a request signal including a current location to the remote server, acquires a real online traffic state signal corresponding to the transmission request signal segment, and then sends a link congestion degree value C to the data receiver;

[0026] The data receiver receives the actual online traffic state information sent by the data pusher through a wireless network, and outputs the data.

[0027] In the above technical solution, the pre-stored parameter value X- ^{1 is} stored on a remote server, including the gray value G i of the road color and the error value _ei , and the gray color of the road is changed as the daytime and the weather change. One or two of the degree value G i and the error value ε i are sent by the remote server to each of the video processors through the network module

[0028] A further technical solution is that, according to the change of the daytime, each time a small 吋~4 small 间隔 calls the data pre-stored in the remote server, the gray value G of the road color on each video processor i and the error value ε i are reassigned.

[0029] Further, the gray value of each pixel in each frame picture is X=0.11xr+0.59xg+0.3xb, where r represents red brightness, g represents green brightness, and b represents blue brightness, when (XG i) ² - ε; > 0, then num i_l= Num _; _1+1 , no shell IJ num;_0= num;_0+ 1 .

[0030] In the above technical solution, the manner of evaluating the road congestion degree c i observed on the video processor by using the “nuclear method” is:

[0031] a remote server corresponding to each video processor i has a data storage space, storing the previous gray value data ε, where j = l, 2 ... N;

[0032] b each video processor i calls the e data in the remote server, using formula (1) to compare the similarity with the currently obtained real gray value _ei ;

(1)

[0034] wherein

— : — Qing —: Record

[0035] c. Each video processor i further determines the degree of congestion of the road segment that can be observed by the current monitoring according to formula (2)

C i;

[0036]

■ ^— — ^Li ^ ¹ . : .”•^C

■ ■ ^^: " ' - ^Λ ■■:

(2)

[0037] d finally through the network module to upload the congestion degree c i of the road segment and the real gray value e i to the remote server

[0038] In the above technical solution, the “remote server uses the weight of the original data” is a set of grayscale value data uploaded by the video processor of the road segment in the remote server storage space by using a method of random gradient descent. Calculated weight

, where / represents a collection of all video processors for that road segment.

[0039] In the above technical solution, the data receiver includes a receiving module, a voice module or a display module, and the receiving module receives a signal and sends a request signal to the data pusher via the wireless network, and the voice module or the display module The signal acquired by the receiving module is converted into voice information or text information and output on the vehicle.

[0040] The present invention further includes another technical solution: a traffic condition acquisition device based on traffic monitoring video, including a video processor, a network module, a remote server, a data pusher, and a data receiver, wherein:

[0041] a video processor: disposed on a monitoring device of each intersection, configured to read a video image on the monitoring device, and analyze the video image;

[0042] a network module: a wireless network or a wired network, configured to connect the video processor and the remote server to transmit data information;

[0043] a remote server: receiving an analysis result sent by the video processor, and transmitting the result to the data pusher through a network module;

[0044] a data pusher: receiving a request signal sent from the data receiver, and transmitting the request signal to the remote server, acquiring corresponding data, and transmitting the data to the data receiver, where the data pusher is installed on the monitoring device Side

[0045] Data Receiver: The information transmitted by the data pusher is received by a wireless network and output.

[0046] In the above technical solution, the data transmitter is a radio frequency transceiver, and the radio frequency transceiver is connected to the remote server via the network module, and sends a request signal to the remote server, where the remote server passes the network. The module transmits a corresponding signal to the radio frequency transceiver, the radio frequency transceiver transmitting electromagnetic waves to the receiver.

[0047] In the above technical solution, the data receiver includes a receiving module, a voice module or a display module, and the receiving module receives a signal and sends a request signal to the data pusher via the wireless network, and the voice module or the display module The signal acquired by the receiving module is converted into voice information or text information output.

Advantageous effects of the invention

Beneficial effect [0048] Due to the above technical solutions, the present invention has the following advantages over the prior art:

[0049] 1. The invention installs a video processor on the original monitoring device of each road segment, and obtains image data of each frame in real time, and calculates a gray value of each pixel in the image, and the actual gray value is obtained. Compared with the gray value of the road segment in the past, the nuclear method is used to analyze and evaluate the congestion degree of the road segment, which is the actual traffic state of the road segment. Since the analysis of the real data is processed by the video processor, the analysis result Uploaded to a remote server, the amount of communication data between the server and the video processing is small, does not cause transmission pressure on the network, and the data transmission speed is fast, providing good data support for the remote server to obtain the actual road condition, so as to be instant Control the road signal lights to help improve road conditions;

[0050] 2. Since the analysis of the real data in the present invention adopts the nuclear method, environmental factors are considered, and not only the proportion of pixels in the entire picture, but also the color of the road or the change of weather and time. Change the nuclear parameters to ensure that the degree of congestion assessed is not affected by the huge changes in the environment, and the data obtained is more accurate;

[0051] 3. In the analysis of the congestion situation from one intersection to the next intersection, the gradient descent method is used to calculate the weight, and the weight is added to the analysis of the congestion of the road section, which increases the correctness of the entire accounting process and avoids Local optimal solution

[0052] 4. In the present invention, the combination of the data pusher and the data receiver is used to enable the driver to obtain the congestion of the nearby road section through the data receiver in the vehicle, and plan the driving route in advance to achieve reasonable shunting, effective Alleviate traffic pressure.

Brief description of the drawing

DRAWINGS

1 is a schematic diagram of a partial network topology according to Embodiment 1 of the present invention;

2 is a schematic diagram of a partial network topology according to Embodiment 2 of the present invention.

Embodiments of the invention

[0055] The present invention is further described below in conjunction with the accompanying drawings and embodiments:

Embodiment 1 Embodiment 1: Referring to FIG. 1 , a method for detecting an online traffic state based on traffic monitoring video includes setting a video processor corresponding to each monitoring device (TMS320 C66x of Texas Instruments) DSP series TMS320C6670 products), each of the video processors via a network module (Ethernet Or 2G\3G\4G wireless network) Connect to a remote server, where:

[0057] the video processor reads a video image corresponding to the monitoring device, and obtains a gray value X of each pixel in each frame according to the obtained image, and compares it with the pre-stored parameter value X- ¹ . When greater than XX- ¹ > 0, Bay ljnumi_l = nunii - l + l, No Bay Unum i_0 = num i - 0 + 1, by collecting the values of num i_l and num i_0, constitute the real gray value ei = (n _U Mi_l, nuniiJ)), evaluating the road congestion level ci observed on the video processor by the kernel method, and uploading to the remote server through the network module;

[0058] The remote server collects the congestion degree ci of the road segment uploaded by each video processor, and the remote server analyzes the weight of the original data and the received congestion degree ci of each real road segment, and obtains from a junction to The road congestion degree value C of the next intersection is expressed as the actual online traffic state of the road section.

[0059] The pre-stored parameter value X ^{1 is} stored on the remote server, including the gray value G i of the road color and the error value ε i , and changes the gray value G i of the road color with the change of the daytime and the weather. The error value _{ei is} trained by the monitoring device's past weather and illumination to a similar video and stored in a remote server. According to the change of the daytime, the parameter value pre-stored in the remote server is called every interval of 1 to 4 small ,, and the gray value G i and the error value ε i of the road color of the remote server for each video processor. Reassign.

[0060] The specific method is:

[0061] (1) assigning a core e of each video processor on the road segment through Ethernet, where =l, 2, ..., 5 [0062] (2) each pair on the road segment via Ethernet Video processor ί parameter G and

Assignment.

[0063] (3) For the video processor i, it belongs to: . 3⁄4 - 6.j _; 14- §

, taking a frame of the surveillance video, for each pixel, if

then

Otherwise,

. Where r is the red brightness, g is the green brightness, and b is the blue brightness.

[0064] (4) Video Processor According to the formula

Find the similarity of ^ with each one, where

[0065] 5) The video processor is based on the formula

, to monitor the degree of congestion C of the road segment that can be observed. And pass the value of _C; to the server via Ethernet.

[0066] (6) The server passes a set of weights

Calculate the congestion level of a section from one intersection to another

£ =,

. '

, where / represents a collection of all video processors for that road segment,

It is a set of weights obtained by calculating the gray value data of the road segment uploaded by the video processor of the road segment in the remote server storage space by the method of random gradient descent. [0067] If the environment on the road section changes greatly, such as the roadside green belt and the background color of the road section changes, turn to step (1). Otherwise, if the set interval is reached or the weather changes, the steering step ( 2) . Otherwise, go to step (3).

[0068] Embodiment 2: Referring to FIG. 2, a road condition acquisition device based on traffic monitoring video includes a video processor, a network module, a remote server, a data pusher, and a data receiver, wherein:

[0069] (1) Video processor: disposed on the monitoring device of each intersection, used to read the video image on the monitoring device, and analyze the video image, and the video processor can select the TMS320C66x DSP system of Texas Instruments. IJTMS320C6670 products, the chip operating temperature is between -40 degrees and 100 degrees, can meet the requirements of outdoor work;

[0070] (2) Network module: a wireless network or a wired network, configured to connect the video processor and the remote server, and transmit data information, and the wired network is Ethernet. If the wireless network is selected, 2G\3G\4G may be selected. One of the networks;

[0071] (3) a remote server: receiving an analysis result sent by the video processor, and transmitting the result to the data pusher through a network module;

[0072] (4) a data pusher: receiving a request signal sent from the data receiver, and transmitting the request signal to the remote server, acquiring corresponding data, and transmitting the data to the data receiver, where the data pusher is installed Side of the monitoring equipment; TI's CC2520 chip can be selected, it is a radio frequency transceiver using ZigBee protocol, the communication range is about 70 meters to meet our communication needs;

[0073] (5) Data receiver: receiving information sent by the data pusher through a wireless network, and outputting, which can be output by voice or text, generally used in voice transmission for more suitable driving process, including receiving module CC2520 The chip and the XF-S4240 voice module of Keda Xunfei, the CC2520 chip receives the data from the data transmitter and transmits it to the XF-S4220, and broadcasts the voice to the driver.

[0074] The specific analysis method is:

[0075] The video processor reads the corresponding video image on the monitor device, in accordance with the obtained image is obtained for each frame picture gradation value X, X ¹ and compared with the stored parameter value is greater than when Χ-Χ -^Ο, 贝 ljn _U m i_l= num i_l+l, otherwise ^11^_0= ^11^_0+1, by num _; _1 and ^^1 i_0 values are collected to form a real gray value _ei = ( _nU m i_l, nuni iJ)), the degree of congestion _{C c} observed on the video processor is evaluated by the kernel method, and then uploaded to the remote server by the network module; [0076] The remote server collects the congestion degree c of the road segment uploaded by each video processor. The remote server uses the weight of the original data and the received congestion degree ci of each real road segment to obtain an intersection. The road congestion degree value C of the next intersection is expressed as the actual online traffic state of the road section.

[0077] The pre-stored parameter value X 1 is stored on the remote server, including the gray value G i of the road color and the error value ε i , and changes the gray value G i of the road color with the change of the day and the weather. The error value _{ei is} trained by the monitoring device's past weather and illumination to a similar video and stored in a remote server. According to the change of the daytime, the parameter value pre-stored in the remote server is called every interval of 1 to 4 small ,, and the gray value G i and the error value ε i of the road color of the remote server for each video processor. Reassign.

[0078] Implementation steps:

[0079] (1) assigning a core e of each video processor on the road segment through Ethernet, where = l, 2, ..., 5 [0080] (2) each pair on the road segment through the Ethernet Video processor ί parameter G and

Assignment.

[0081] (3) For the video processor i, um., 0 — 0. mm; 1 -'-

, taking a frame of the surveillance video, for each pixel, if

then

Otherwise,

[0082] (4) Video processor according to formula

, similarity with each one, where

[0083] (5) The video processor according to the formula ύ ; TM . , : ^ . ί Ί .

, to monitor the degree of congestion of the road segment that can be observed _c; And pass the value of _c; to the server via Ethernet.

[0084] (6) The server passes a set of weights

Calculate the congestion level of a section from one intersection to another

Where / represents the set of all video processors of the road segment, Weng is a set of weights obtained by calculating the gray value data of the road segment uploaded by the video processor of the road segment in the remote server storage space by a method of random gradient descent .

[0085] (7) The data pusher requests the remote server for the congestion degree of the surrounding road section through the network module, and generally can set the search range to be 500-1000 meters, and the road segment information with the congestion degree C reaches a certain degree is transmitted to the automobile through the electromagnetic wave. The data receiver, the inter-language module in the data receiver broadcasts the congestion information voice.

[0086] If the environment on the road section changes greatly, such as the roadside green belt and the background color of the road section change, turn Go to step (1), otherwise, if the set interval is reached or the weather changes, go to step (2). Otherwise, go to step (3).

Claims

Claim

A road condition detection method based on traffic monitoring video, comprising: configuring a video processor corresponding to each monitoring device, wherein each of the video processors is connected to a remote server via a network module, wherein:

The video processor reads a video image corresponding to the monitoring device, and obtains a gray value X of each pixel in each frame according to the obtained image, and compares it with the pre-stored parameter value X 1 when XX >0 , 贝 ljnum i_l= num i_l+l, otherwise num i_0= num i_0+l, by collecting the values of num i_l and num i_0, forming a real gray value e _; =( num _; _1, num _; _0), by the kernel The method evaluates the congestion degree ci of the road segment observed on the video processor, and then uploads to the remote server through the network module; the remote server collects the congestion degree of the road segment uploaded by each video processor c" is utilized by the remote server The weight of the original data is analyzed with the received congestion degree C i of each actual road section, and the congestion degree value C of the road section from one intersection to the next intersection is obtained, that is, the actual online traffic state of the road section is represented.

The road condition detecting method according to claim 1, wherein: said pre-stored parameter value X- ^{1 is} stored on a remote server, including a gray value G i of the road color and an error value _ei , along with the daytime and the weather The change, one or both of the gray value G i , the error value ^ of the change of the road color, is transmitted by the remote server to each of the video processors through the network module. The road condition detecting method according to claim 2, characterized in that: according to the change of the daytime, each time a small 吋~4 small 吋 calls data pre-stored in the remote server, for the road on each video processor The gray value G i of the color and the error value ε i are re-assigned.

The road condition detecting method according to claim 2 or 3, wherein: the gray value of each pixel in each frame picture is X=0.11xr+0.59xg+0.3xb, where r represents red brightness and g represents green brightness. , b represents the blue luminance, when (XG i) ² - ε _; > 0, then num i_l = num i _1+1, no shell IJnum _; _0 = num _; _0+1.

The road condition detecting method according to claim 1, wherein: the method of evaluating a road congestion degree c i observed on the video processor by a "core method" is:

a. Each video processor i in the remote server has a data storage space, which is stored in the past. Gray value data ε , where j=l 2...N;

b. Each video processor i calls the e data in the remote server, and uses formula (1) to compare the similarity with the currently obtained real gray value e i ;

(1)

among them

~(mim, ■■■!::—— 幽; 卿

c. Each video processor i further determines the congestion degree _{C i} of the road segment that can be observed by the current monitoring according to formula (2) _; c : rtxc

3⁄4.: (2)

d. Finally, the congestion degree c i of the road segment and the real gray value e i are uploaded to the remote server through the network module.

[Claim 6] The road condition detecting method according to claim 1, wherein: the "remote server uses the weight of the original data" is a set of methods for dropping by a random gradient, and the original space in the remote server is stored. The weight obtained by calculating the gray value data of the road segment uploaded by the video processor of the road section

, where / represents a collection of all video processors for that segment.

The road condition detecting method according to claim 1, wherein: the network module is an Ethernet cable module or a wireless data transmission network module.

A method for acquiring a road condition based on a traffic monitoring video, comprising: configuring a video processor corresponding to each monitoring device, wherein each of the video processors is connected to a remote server via a network module, and is disposed in the video processor. a side data pusher, and a data receiver disposed on the vehicle, where:

The video processor reads a video image corresponding to the monitoring device, and obtains a gray value X of each pixel in each frame according to the obtained image, and compares it with the pre-stored parameter value X 1 when greater than XX > 0, 贝 ljnum i_l= num i_l+l, otherwise num i_0= num i_0+l, by collecting the values of num i_l and num i_0, forming a real gray value e; = ( num; _1, num; _0), by The core method evaluates the congestion degree _{C i} observed on the video processor, and then uploads to the remote server through the network module;

The remote server collects the congestion degree c of the road segment uploaded by each video processor. The remote server uses the weight of the original data and the received congestion degree ci of each real road segment to obtain an intersection from one intersection to the next. The congestion degree value C of the road section is expressed as the actual online traffic state of the road section;

The data pusher sends a request signal including current location information to the remote server, acquires a real online traffic state signal corresponding to the transmission request signal segment, and then sends a link congestion degree value C to the data receiver;

The data receiver receives the actual online traffic state information sent by the data pusher through the wireless network, and outputs the data.

The road condition acquisition method according to claim 8, wherein: the pre-stored parameter value X- ^{1 is} stored on a remote server, and includes a gray value G i of the road color and an error value _ei , along with the daytime and the weather. The change, one or both of the gray value G i , the error value ^ of the change of the road color, is transmitted by the remote server to each of the video processors through the network module. The road condition acquisition method according to claim 9, wherein: according to the change of the day, each time a small 吋~4 small 吋 calls the data pre-stored in the remote server, for each The gray value G i of the road color on the video processor and the error value ε i are re-assigned.

[Claim 11] The road condition acquisition method according to claim 8 or 9, wherein: the gray value of each pixel in each frame picture is X=0.11xr+0.59xg+0.3xb, where r represents red brightness , g means green brightness, b means blue brightness, when (XG i) ² - ε _; >0, then num i_l= num i _1+1, no shell IJnum _; _0= num _; _0+1.

[Claim 12] The road condition acquisition method according to claim 8, wherein: the method for evaluating a road congestion degree c i observed on the video processor by a "core method" is:

a. Each video processor i in the remote server has a data storage space, and stores the previous gray value data ε, where j=l, 2...N;

(1)

among them

c. Each video processor i further determines the congestion degree c _i of the road segment that can be observed by the current monitoring according to formula (2) _;

(2)

[Claim 13] The road condition acquisition method according to claim 8, wherein: the "remote server The weight of the original data is a set of weights obtained by calculating the gray value data of the road segment uploaded by the video processor of the road segment in the remote server storage space by the method of random gradient descent.

[Claim 14] The method for acquiring a road condition according to claim 8, wherein: the data receiver comprises a receiving module, a voice module or a display module, and the receiving module receives a signal and sends a request signal to the wireless network. The data transmitter, the voice module or the display module converts the signal acquired by the receiving module into voice information or text information and outputs the information on the vehicle.

[Claim 15] A traffic condition acquisition device based on traffic monitoring video, comprising: a video processor, a network module, a remote server, a data pusher, and a data receiver, wherein: a video processor: disposed at each intersection On the monitoring device, for reading a video image on the monitoring device, and analyzing the video image;

a network module: a wireless network or a wired network, configured to connect the video processor and the remote server to transmit data information;

a remote server: receiving an analysis result sent by the video processor, and transmitting the result to the data pusher through a network module;

a data pusher: receiving a request signal sent from the data receiver, and sending the request signal to the remote server, acquiring corresponding data, and sending the data to the data receiver, where the data pusher is installed beside the monitoring device;

Data Receiver: Receives information transmitted by the data pusher over a wireless network and outputs it.

[Claim 16] The road condition obtaining apparatus according to claim 15, wherein: the data pusher is a radio frequency transceiver, and the radio frequency transceiver is connected to the remote server via the network module to a remote server. Sending a request signal, the remote server sends the request through the network module Sending a corresponding signal to the radio frequency transceiver, the radio frequency transceiver transmitting electromagnetic waves to the receiver.

[Claim 17] The road condition obtaining apparatus according to claim 15 or 16, wherein: the data receiver comprises a receiving module, a voice module or a display module, and the receiving module receives a signal and sends a request signal via a wireless network. To the data pusher, the voice module or the display module converts the signal acquired by the receiving module into voice information or text information output.