System and method for detecting jaywalking

Abstract

The present invention relates to a jaywalking detection system and method, and discloses a technology that prevents accidents caused by jaywalking by identifying a road section where jaywalking occurs through a camera capturing a vehicle road, detecting the occurrence of jaywalking through real-time video analysis of the road section, and providing warning information to the jaywalking pedestrian or providing jaywalking guidance information to a vehicle traveling on the road section through a warning device such as an electronic display installed near the road section or a vehicle terminal.

Description

Technology Field

[0001] The present invention relates to a jaywalking detection system and method, and more specifically, to a technology that prevents accidents caused by jaywalking by identifying a road section where jaywalking occurs through a camera capturing a vehicle road, detecting the occurrence of jaywalking through real-time video analysis of the road section, and providing warning information to the jaywalking pedestrian or providing jaywalking guidance information to a vehicle traveling on the road section through a warning device such as an electronic display installed near the road section or a vehicle terminal. Background Technology

[0003] Traffic accidents caused by jaywalking frequently occur on roads. Unlike other types of traffic accidents, jaywalking accidents are the most dangerous type of accident, with a fatality rate ten times higher than other types of traffic accidents, because the pedestrian receives the full impact of the impact.

[0004] Generally, even if a moving vehicle is visible on the road, its speed appears slow from a distance; however, since the actual speed is significantly faster than what a person visually perceives, a sudden attempt to jaywalk can cause a serious accident.

[0005] Various measures are being proposed to prevent such jaywalking accidents.

[0006] Conventional technologies provide technical specifications for monitoring crosswalks via cameras to detect jaywalking and warning of such occurrences or providing information about the jaywalking situation to the surroundings.

[0007] However, because drivers pay more attention near crosswalks, they can recognize jaywalkers more quickly and respond immediately. On the other hand, on roads without nearby crosswalks, drivers do not anticipate jaywalking, so their response may be delayed after recognizing a pedestrian, potentially leading to a serious accident.

[0008] Therefore, it is necessary to devise measures to detect jaywalking and provide warnings or guidance not only on roads where crosswalks exist nearby, but also on roads where they do not.

[0009] Furthermore, when determining a jaywalking situation based on footage captured by a conventional camera, errors frequently occur in which moving objects in the footage are not accurately identified as people. For example, there is a problem in that errors in determining jaywalking occur when a person-shaped advertisement on a vehicle such as a truck or bus is recognized as a person jaywalking, or when a person riding a motorcycle or scooter is mistaken for a jaywalker.

[0010] Going a step further, in the case of conventional technology, a central server receives video footage from cameras in various locations to perform video analysis regarding jaywalking. This results in a high dependency on network performance, leading to a problem where the accuracy of the analysis is low due to the processing of the footage at low resolution. Prior art literature

[0012] Korean Published Patent No. 10-2022-0074356 The problem to be solved

[0013] The present invention aims to solve the problems of the prior art as described above by detecting jaywalking through captured video of the road section where jaywalking occurs, and by providing warning information and jaywalking guidance information through a warning device and a vehicle terminal, thereby preventing accidents caused by jaywalking.

[0014] In particular, this invention aims to solve the problem where conventional technology detects jaywalking near crosswalks, but jaywalking on roads other than crosswalks leads to more serious accidents.

[0015] Furthermore, we aim to resolve the error of failing to accurately determine whether a moving object in the recorded video is a person.

[0016] Going a step further, this invention aims to solve the problem of low analysis accuracy resulting from processing captured images at low resolution due to high dependence on network performance, which is inherent in conventional technology where a central server receives images from cameras in various locations to perform video analysis regarding jaywalking.

[0017] The objectives of the present invention are not limited to those foregoing, and other objectives and advantages of the present invention not mentioned may be understood from the following description. means of solving the problem

[0019] To achieve the above technical objective, an embodiment of a pedestrian crossing detection system according to the present invention may include: a camera that photographs a vehicle road; a pedestrian crossing detection device positioned around the vehicle road that receives real-time footage from a nearby camera, analyzes the footage to determine pedestrian crossing, generates pedestrian crossing information, and transmits it to a warning device; and a warning device positioned around the road that receives pedestrian crossing information from the nearby pedestrian crossing detection device and provides pedestrian crossing warning information to nearby vehicles and pedestrians.

[0020] Preferably, the jaywalking detection device may include: an object determination unit that identifies an object through pattern analysis of a captured image and determines a pedestrian based on a motion pattern according to the object's direction of movement and speed; a jaywalking detection unit that tracks the movement of an object determined to be a pedestrian to determine whether jaywalking has occurred; and a jaywalking information providing unit that generates jaywalking information upon detection of jaywalking and provides it to the warning device.

[0021] Furthermore, the object determination unit can determine the frequency of jaywalking on a vehicle road through previously accumulated footage to set a jaywalking section, and extract an ROI centered on the jaywalking section from real-time footage.

[0022] For example, the object determination unit determines a vehicle based on a motion pattern according to the movement direction and speed of the object, the jaywalking detection unit tracks the movement of the object determined to be a vehicle to calculate the time to reach the object determined to be a pedestrian, determines a warning level based on the time to reach, and the jaywalking information providing unit can generate different jaywalking information for each warning level.

[0023] Preferably, the above-mentioned jaywalking information providing unit can provide the jaywalking information to the terminal device of the vehicle when the warning level is higher than the reference level.

[0024] Furthermore, it may further include a central server that analyzes jaywalking patterns on a vehicle road through video footage accumulated in the past, analyzes the frequency of jaywalking on the vehicle road to set jaywalking sections on the vehicle road, and provides a jaywalking analysis algorithm for the video footage to the jaywalking detection device.

[0025] In addition, an embodiment of the jaywalking monitoring method according to the present invention may include: a road image acquisition step in which a jaywalking monitoring device acquires real-time captured images of a vehicle road; a jaywalking determination step in which the jaywalking monitoring device determines jaywalking through image analysis of the captured images and generates jaywalking information; and a warning information provision step in which the jaywalking monitoring device provides the jaywalking information to a warning device placed around the road, thereby providing jaywalking warning information to nearby vehicles and pedestrians through the warning device.

[0026] Preferably, the jaywalking determination step may include: an object identification step for identifying an object through pattern analysis of a captured video; a pedestrian determination step for determining a pedestrian based on a motion pattern according to the direction and speed of movement of the object; a pedestrian tracking step for determining whether jaywalking has occurred by tracking the movement of an object determined to be a pedestrian; and a jaywalking information generation step for generating jaywalking information when jaywalking is detected.

[0027] Furthermore, the above-mentioned jaywalking determination step can extract an ROI centered on the jaywalking section from the captured video based on a jaywalking analysis algorithm from a central server, and determine jaywalking through video analysis of the extracted ROI.

[0028] Going a step further, the above-mentioned jaywalking information generation step may include: a vehicle determination step that determines a vehicle based on a motion pattern according to the movement direction and speed of the object; and a warning level-specific jaywalking information generation step that tracks the movement of an object determined to be a vehicle to calculate the time to reach an object determined to be a pedestrian, determines a warning level based on the time to reach, and generates different jaywalking information for each warning level.

[0029] Preferably, the warning information providing step may provide the unauthorized crossing information to the terminal device of the vehicle when the warning level is higher than the reference level. Effects of the invention

[0031] According to the present invention, accidents caused by jaywalking can be prevented by detecting jaywalking through captured images of the road section where jaywalking occurs and providing warning information and jaywalking guidance information through a warning device and a vehicle terminal.

[0032] In particular, in the present invention, since a pedestrian crossing monitoring device is positioned near the location of a camera capturing the vehicle road, real-time high-definition video footage can be received from the camera, thereby enabling more accurate video analysis.

[0033] In addition, the present invention enables accurate detection of whether an object is a pedestrian by identifying an object in a captured image and determining the type of object based on the object's motion pattern.

[0034] Furthermore, by providing jaywalking information according to warning levels that consider the possibility of collision between jaywalking pedestrians and nearby vehicles, the warning device displays jaywalking warning information differently according to path levels, allowing jaywalking pedestrians and nearby pedestrians to intuitively sense the degree of danger more immediately. Additionally, since jaywalking guidance information is delivered differently according to warning levels through the vehicle terminal, vehicle drivers can pay closer attention and prevent collisions with jaywalking pedestrians in advance.

[0035] The effects of the present invention are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art from the description below. Brief explanation of the drawing

[0037] FIG. 1 illustrates a configuration diagram of an embodiment of a pedestrian crossing monitoring system according to the present invention. FIGS. 2 and 3 illustrate configuration diagrams of an embodiment of a pedestrian crossing detection device of a pedestrian crossing monitoring system according to the present invention. FIGS. 4 and 5 illustrate configuration diagrams of an embodiment of a warning device for a pedestrian crossing monitoring system according to the present invention. FIG. 6 illustrates a flowchart of an embodiment of a method for monitoring unauthorized crossing according to the present invention. FIGS. 7 to 9 illustrate an example of identifying and tracking an object through the present invention. FIG. 10 illustrates a flowchart of another embodiment of the jaywalking monitoring method according to the present invention. FIG. 11 illustrates an example of providing jaywalking information through the present invention. Specific details for implementing the invention

[0038] In order to explain the present invention, the operational advantages of the present invention, and the objectives achieved by the implementation of the present invention, preferred embodiments of the present invention are illustrated below and examined with reference thereto.

[0039] First, the terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention; singular expressions may include plural expressions unless the context clearly indicates otherwise. Furthermore, in this application, terms such as "comprising" or "having" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0040] In describing the present invention, if it is determined that a detailed description of related known components or functions may obscure the essence of the invention, such detailed description is omitted.

[0042] The present invention presents a technology that prevents accidents caused by jaywalking by identifying road sections where jaywalking occurs through a camera capturing a vehicle road, detecting the occurrence of jaywalking through real-time video analysis of the road section, and providing warning information to jaywalking pedestrians or jaywalking guidance information to vehicles traveling on the road section through warning devices such as electronic display boards installed near the road section and vehicle terminals.

[0044] FIG. 1 illustrates a configuration diagram of an embodiment of a pedestrian crossing monitoring system according to the present invention.

[0045] The jaywalking monitoring system according to the present invention may include a jaywalking monitoring device (100), a camera (200), a warning device (300), a central server (500), etc.

[0046] The jaywalking monitoring device (100) can be placed around a vehicle road. The jaywalking monitoring device (100) receives real-time footage from a camera (200) that photographs the vehicle road, analyzes the footage to determine jaywalking on the vehicle road, generates jaywalking information, and provides it to a nearby warning device (300) and a vehicle terminal of a nearby vehicle (400).

[0047] The camera (200) includes a plurality of CCTVs (200a, 200b, ..., 200n) distributed and installed adjacent to the vehicle road, and can provide real-time footage of the vehicle road and its surroundings to the unauthorized crossing monitoring device (100).

[0048] The warning device (300) is positioned around the vehicle road and receives jaywalking information from a nearby jaywalking detection device (100), and can provide jaywalking warning information so that vehicles (400) and pedestrians traveling nearby can detect the danger of jaywalking.

[0049] The central server (500) can be installed in a separate, isolated central control facility and can receive and analyze various captured images from the illegal crossing monitoring devices (100) distributed across multiple regions.

[0050] The central server (500) can analyze the frequency of jaywalking on the vehicle road in the area by analyzing the captured video of the area and, based on this, set the jaywalking section on the vehicle road. The central server (500) can also provide information about the jaywalking section on the vehicle road to a jaywalking monitoring device (100) placed near the jaywalking section on the vehicle road.

[0051] The central server (500) can perform artificial intelligence learning for jaywalking detection using previously accumulated captured images as learning data and provide a jaywalking analysis algorithm based on the learning results to each corresponding jaywalking monitoring device (100).

[0052] That is, in the present invention, an ignition monitoring system is constructed using an edge computing method so that each ignition monitoring device (100) performs analysis of real-time captured video to determine ignition, and the central server (500) provides an ignition analysis algorithm through artificial intelligence learning on cumulatively acquired captured video, thereby increasing the efficiency of the ignition analysis of the ignition monitoring device (100).

[0053] In particular, since the existing centralized method of determining jaywalking requires a central server to receive video data from cameras located at a distance, there was a problem where the accuracy of video analysis was degraded due to the high dependency on the network, which prevented the transmission of data in high-quality video.

[0054] In the present invention, by applying an edge computing method, the jaywalking monitoring device (100) positioned near each jaywalking occurrence point directly analyzes the captured video, thereby receiving high-quality data and performing video analysis, which has the advantage of increasing accuracy.

[0056] We will examine each component of the jaywalking monitoring system according to the present invention in more detail through an example.

[0057] FIGS. 2 and 3 illustrate configuration diagrams of an embodiment of a pedestrian crossing detection device of a pedestrian crossing monitoring system according to the present invention.

[0058] The jaywalking detection device (100) may include a jaywalking determination unit (110), a jaywalking information providing unit (130), a communication unit (150), etc.

[0059] The jaywalking detection unit (110) can analyze whether there is jaywalking by receiving real-time footage of the vehicle road from a nearby camera (200).

[0060] For example, the jaywalking determination unit (110) may include an object determination unit (111), a jaywalking detection unit (115), etc.

[0061] The object determination unit (111) can identify an object present in the video through video analysis of the captured video. The object determination unit (111) can identify the motion pattern by comprehensively analyzing the movement direction and speed of the identified object and determine the type of object based on this. For example, the object determination unit (111) can distinguish whether the identified object in the captured video is a person or a vehicle.

[0062] The jaywalking detection unit (115) can track the movement of a specified object. For example, the jaywalking detection unit (115) can track the movement of an object determined to be a pedestrian, and can determine whether the pedestrian is jaywalking by distinguishing between a pedestrian path and a vehicle road.

[0063] Additionally, the unauthorized crossing detection unit (115) can track the movement of an object determined to be a vehicle, calculate the time to reach an object determined to be a pedestrian, and determine a warning level based on the time to reach.

[0064] Furthermore, the jaywalking determination unit (110) can receive a jaywalking analysis algorithm from the central server (600) and apply it to determine jaywalking in the captured video. In addition, the jaywalking determination unit (110) can receive jaywalking section information from the central server (600) and extract and analyze an ROI centered on the jaywalking section in the real-time captured video.

[0065] The jaywalking information providing unit (130) can generate jaywalking information based on the jaywalking judgment result of the jaywalking judgment unit (110) and provide it to a warning device (300) placed nearby or to a vehicle terminal of a vehicle (400) driving nearby.

[0066] Preferably, when the jaywalking judgment unit (110) determines jaywalking as a warning level, the jaywalking information providing unit (130) can generate and provide different jaywalking information for each path level.

[0067] Furthermore, the jaywalking information providing unit (130) may not provide separate jaywalking information to the vehicle terminals of nearby vehicles when the warning level is lower than the reference level, but may provide jaywalking information to the vehicle terminals of nearby vehicles when the warning level is higher than the reference level.

[0068] For example, the jaywalking information provider (130) can provide jaywalking information by applying the MQTT (Message Queuing Telemetry Transport) messaging protocol.

[0069] The communication unit (150) can provide communication with the central server (500), nearby camera (200), warning device (300), and vehicle terminal of nearby driving vehicle (400). Here, the communication unit (150) can apply wired or wireless communication methods depending on the situation, and can also apply various wireless communication methods considering the form of transmitted and received data.

[0071] FIGS. 4 and 5 illustrate configuration diagrams of an embodiment of a warning device for a pedestrian crossing monitoring system according to the present invention.

[0072] The warning device (300) may be placed near the unauthorized crossing section and may include a warning information providing unit (310), a control unit (330), a communication unit (350), etc.

[0073] The warning information providing unit (310) receives jaywalking information from a nearby jaywalking detection device (100) and can provide jaywalking warning information to warn nearby vehicles and pedestrians of jaywalking.

[0074] For example, the warning information providing unit (310) may include a display (311) that visually displays and notifies jaywalking warning information and a speaker (315) that emits the jaywalking warning information as an audible signal.

[0075] The control unit (330) can control the warning information providing unit (310) to selectively provide unauthorized crossing warning information according to unauthorized crossing information transmitted from the unauthorized crossing detection device (100).

[0076] For example, the control unit (330) can display the level of danger by controlling various elements such as the brightness and color change of the display (311) according to the warning level. In addition, the control unit (330) can display the level of danger by controlling the volume and type of the sound emitted through the speaker (315) differently according to the warning level.

[0077] The communication unit (350) can provide communication with the nearby unauthorized crossing detection device (100). The communication unit (150) can apply wired or wireless communication methods depending on the situation, and can also apply various wireless communication methods considering the form of transmitted and received data.

[0079] In addition, the present invention presents a method for monitoring jaywalking implemented through the jaywalking monitoring system described above. Below, the method for monitoring jaywalking according to the present invention will be explained through examples.

[0080] FIG. 6 illustrates a flowchart of an embodiment of a method for monitoring unauthorized crossing according to the present invention.

[0081] The pedestrian crossing monitoring device (100) can acquire real-time footage of the nearby road from the camera (200) (S110).

[0082] The jaywalking monitoring device (100) can determine jaywalking through image analysis of the acquired captured image. More specifically, the jaywalking monitoring device (100) can recognize and identify (S120) an object present in the captured image through pattern analysis of the captured image.

[0083] The jaywalking monitoring device (100) can analyze the movement pattern according to the movement direction and speed of a specified object and determine the type of object based on this (S130). For example, the jaywalking monitoring device (100) can analyze the movement pattern according to the movement direction and speed of an object to distinguish whether the object is a pedestrian or a vehicle.

[0084] The jaywalking monitoring device (100) tracks the identified object (S140), and, for example, can determine whether jaywalking is occurring by tracking the movement of an object determined to be a pedestrian.

[0085] Refer to FIGS. 7 to 9 to examine the process in which a pedestrian crossing monitoring device (100) determines the type of object and tracks its movement to determine whether it is a pedestrian crossing.

[0086] As shown in FIG. 7 above, the jaywalking monitoring device (100) can identify objects (611, 513, 621, 623, 631, 641) in real-time captured images (610 to 640). And as shown in FIG. 8 above, the jaywalking monitoring device (100) can determine whether the object is a person by identifying the direction and speed of movement of the object (611, 621, 631, 641) and analyzing the movement path. For example, in the case of FIG. 9 above, by identifying the speed and direction of movement of the object, it can determine whether the object is a pedestrian (651), a person riding a quiverboard (653), an advertisement (656, 657) displayed on a vehicle (655), etc.

[0087] And when the jaywalking monitoring device (100) determines jaywalking, it generates jaywalking information and provides the jaywalking information to a warning device (300) placed around the road, thereby providing jaywalking warning information to nearby vehicles and pedestrians through the warning device (300).

[0088] As such, in the present invention, by placing a pedestrian crossing monitoring device near the placement location of a camera that photographs the vehicle road, real-time high-definition video footage can be received from the camera, thereby enabling more accurate video analysis.

[0089] In addition, the present invention enables accurate detection of whether an object is a pedestrian by identifying an object in a captured image and determining the type of object based on the object's motion pattern.

[0091] Furthermore, FIG. 10 illustrates a flowchart of another embodiment of the jaywalking monitoring method according to the present invention.

[0092] The central server (500) can accumulate and analyze footage captured by the camera (200) in the area to identify and set (S210) a section on the vehicle road in the area where jaywalking frequently occurs. The central server (500) can also provide information on the jaywalking section setting to a jaywalking monitoring device (100) placed near the jaywalking section.

[0093] In addition, the central server (500) can perform artificial intelligence learning for jaywalking detection using previously accumulated captured images as learning data and provide a jaywalking analysis algorithm based on the learning results to the jaywalking monitoring device (100).

[0094] The jaywalking monitoring device (100) can extract an ROI (S220) centered on the jaywalking section from the captured video of a nearby camera (200) and analyze the captured video by applying a jaywalking analysis algorithm provided by a central server (500).

[0095] As previously explained, the jaywalking monitoring device (100) can identify an object in the captured video and determine its type to determine whether the object is a pedestrian or a vehicle (S230).

[0096] And the jaywalking monitoring device (100) can track the movement (S240) of each object determined to be a pedestrian and an object determined to be a vehicle.

[0097] The jaywalking monitoring device (100) can determine whether there is a collision with the jaywalking pedestrian by tracking the movement path of an object determined to be a vehicle in the vicinity when the jaywalking pedestrian is determined to be jaywalking based on the movement tracking of an object determined to be a pedestrian. For example, the jaywalking monitoring device (100) can determine the distance traveled based on the movement path and speed of the object determined to be a pedestrian and determine the distance traveled based on the movement path and speed of the object determined to be a vehicle in the vicinity to predict the possibility of a collision between the pedestrian and the vehicle.

[0098] The jaywalking monitoring device (100) can calculate the time it takes for an object determined to be a vehicle to reach an object determined to be a pedestrian (S250) and determine a warning level based on the time it takes (S260).

[0099] For example, the jaywalking monitoring device (100) can probabilistically determine the possibility of a collision between a pedestrian and a vehicle, and a warning level can be determined according to a set probabilistic value.

[0100] The pedestrian crossing monitoring device (100) can generate different pedestrian crossing information for each path level (S270) and provide the pedestrian crossing information to a nearby warning device (300) and a vehicle terminal of a nearby moving vehicle (400).

[0101] For example, the jaywalking monitoring device (100) may provide jaywalking information only to a nearby warning device (300) when the warning level is lower than the reference level, and may provide it to a nearby warning device (300) and a vehicle terminal of a nearby moving vehicle (400) when the warning level is higher than the reference level.

[0102] When jaywalking information is received by the jaywalking monitoring device (100), the warning device (300) and the vehicle terminal can display the jaywalking information as visual or auditory information.

[0103] FIG. 11 illustrates an example of providing jaywalking information through the present invention.

[0104] When a pedestrian (710) crossing illegally is detected on a vehicle road and the pedestrian (710) crossing illegally is determined to have a high probability of collision with a nearby moving vehicle (730), the illegal crossing monitoring device (100) can provide corresponding illegal crossing information to the warning device (300) and the vehicle terminal of the vehicle (730).

[0105] The warning device (300) can warn pedestrians jaywalking and nearby pedestrians of jaywalking by changing the color of the screen emitted to the display in a flashy way according to the warning level or by emitting a loud, unpleasant sound.

[0106] In particular, since the warning information provided varies depending on the warning level, pedestrians jaywalking and nearby pedestrians can more immediately sense the degree of danger.

[0107] In addition, by providing different jaywalking guidance information according to the warning level through the vehicle terminal of the vehicle (730), the vehicle driver can pay more attention and prevent a collision with a jaywalking pedestrian in advance.

[0109] The present invention can prevent accidents caused by jaywalking by detecting jaywalking through captured video of a road section where jaywalking occurs and providing warning information and jaywalking guidance information through a warning device and a vehicle terminal.

[0111] The foregoing description is merely an illustrative explanation of the technical concept of the present invention, and those skilled in the art to which the present invention pertains will be able to make various modifications and variations within the scope of the essential characteristics of the present invention. Accordingly, the embodiments described in the present invention are intended to explain, not limit, the technical concept of the present invention, and the technical concept of the present invention is not limited by these embodiments. The scope of protection of the present invention shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of the present invention. Explanation of the symbols

[0113] 100 : Pedestrian crossing detection device, 110 : Jaywalking judgment unit, 111 : Object determination unit, 115 : Pedestrian crossing detection unit, 130 : Jaywalking Information Provision Department, 150, 350 : Communications Department, 200 : Camera, 300 : Warning device, 310 : Warning Information Provider, 311 : Display, 315 speaker, 330 : Control unit.

Claims

Claim 1 A camera that photographs a vehicle road; a pedestrian crossing detection device positioned near the vehicle road that receives real-time video footage from a nearby camera, analyzes the video footage to determine jaywalking, generates jaywalking information, and transmits it to a warning device; and a warning device positioned near the vehicle road that receives jaywalking information from the nearby pedestrian crossing detection device and provides jaywalking warning information to nearby vehicles and pedestrians, wherein the pedestrian crossing detection device comprises: an object determination unit that identifies an object through pattern analysis of the video footage, determines the type of the object based on the motion pattern according to the object's direction of movement and speed to distinguish whether the object is a person or a vehicle, and extracts a first object corresponding to a pedestrian and a second object corresponding to a vehicle; and a pedestrian crossing detection unit that tracks the movement of the first object to determine whether jaywalking has occurred. A jaywalking detection system comprising: a jaywalking information providing unit that generates jaywalking information upon detection of jaywalking and provides it to the warning device; wherein the jaywalking detection unit tracks the movement of the second object to calculate the time to reach the first object and determines a warning level based on the time to reach; wherein the jaywalking information providing unit comprises: a display positioned around the vehicle road to visually display and notify jaywalking warning information; and a speaker positioned around the vehicle road to emit the jaywalking warning information as an acoustic signal to provide auditory notification; wherein different jaywalking information is generated according to the determined warning level, wherein the display is controlled so that the brightness and color of the output screen are output differently according to the warning level, and the speaker is controlled so that the volume and type of the emitted sound are output differently according to the warning level. Claim 2 delete Claim 3 delete Claim 4 A pedestrian crossing detection system according to claim 1, wherein the pedestrian crossing information providing unit provides the pedestrian crossing information to the terminal device of the vehicle when the warning level is higher than the reference level. Claim 5 A pedestrian crossing detection system according to claim 1, further comprising a central server that analyzes pedestrian crossing patterns on a vehicle road through cumulatively acquired footage, analyzes the frequency of pedestrian crossing occurrences on the vehicle road to set a pedestrian crossing section on the vehicle road, and provides a pedestrian crossing analysis algorithm for the footage to the pedestrian crossing detection device. Claim 6 A pedestrian crossing detection system according to claim 1, wherein the object determination unit receives information on the pedestrian crossing section from the central server and extracts an ROI centered on the pedestrian crossing section from a real-time captured video. Claim 7 A pedestrian crossing detection device comprises: a road image acquisition step for acquiring real-time captured video of a vehicle road; a pedestrian crossing judgment step for the pedestrian crossing monitoring device for determining pedestrian crossing through image analysis of the captured video and generating pedestrian crossing information; and a warning information provision step for the pedestrian crossing monitoring device for providing the pedestrian crossing information to a warning device placed around the road, thereby providing pedestrian crossing warning information to nearby vehicles and pedestrians through the warning device, wherein the pedestrian crossing judgment step comprises: an object identification step for identifying an object through pattern analysis of the captured video; an object type determination step for determining the type of the object based on a motion pattern according to the direction and speed of movement of the object to distinguish whether the object is a person or a vehicle, and extracting a first object corresponding to a pedestrian and a second object corresponding to a vehicle; and an object tracking step for tracking the movement of the first object to determine whether pedestrian crossing has occurred. A method for detecting jaywalking, comprising: a step of generating jaywalking information upon detection of jaywalking, wherein the object tracking step tracks the movement of the second object to calculate the time to reach the first object and determines a warning level based on the time to reach; wherein the jaywalking information generating step comprises: a step of generating different jaywalking information for each warning level; a step of controlling a display placed around the vehicle road to visually display and notify jaywalking warning information based on the jaywalking information, such that the brightness and color of the output screen are output differently according to the warning level; and a step of controlling a speaker placed around the vehicle road to audibly notify jaywalking warning information as an acoustic signal, such that the volume and type of the emitted sound are output differently according to the warning level. Claim 8 delete Claim 9 A method for detecting jaywalking according to claim 7, wherein the jaywalking determination step is characterized by extracting an ROI centered on the jaywalking section from a captured video based on a jaywalking analysis algorithm provided by a central server, and determining jaywalking through video analysis of the extracted ROI. Claim 10 delete Claim 11 A method for detecting jaywalking according to claim 7, wherein the warning information providing step is characterized by providing the jaywalking information to a terminal device of the vehicle when the warning level is higher than a reference level.

Images