Intelligent Monitoring System for Controlled Facilities

Abstract

The intelligent control system according to the present invention may include an image capturing device installed at a predetermined location of a facility to be controlled, which captures a plurality of areas inside the facility to be controlled and generates first image data related to a parking space; an image analysis unit that receives the first image data from the image capturing device and analyzes the first image data to determine the status of the remaining parking space inside the facility to be controlled; and a display device installed at a predetermined location of the facility to be controlled, which receives the analysis result of the image analysis unit and displays location information of the remaining parking space.

Description

Technology Field

[0001] The present invention relates to a control system for a controlled facility using an image capturing device, and more specifically, to an intelligent control system that analyzes images acquired from an image capturing device installed at a pre-set location of the controlled facility and automatically guides a vehicle entering the controlled facility to a suitable parking space. Background Technology

[0002] With the increase in parking demand, the efficient management and operation of large parking lots and similar facilities are emerging as an important challenge.

[0003] Generally, existing parking management systems used a method of detecting whether a vehicle was parked by installing individual sensors in each parking space. This individual sensor method had the drawbacks of high installation and maintenance costs, as well as the inability to provide accurate parking space information in the event of sensor failure or malfunction.

[0004] Furthermore, as dedicated parking areas for various vehicle types—such as electric vehicles, vehicles for the disabled, and vehicles for pregnant women—are expanded, it has become necessary to effectively guide users to parking spaces suitable for each vehicle type. However, existing parking control systems have limitations in that they merely indicate parking availability and fail to provide the functionality to automatically identify vehicle types and guide users to appropriate parking zones.

[0005] Furthermore, conventional parking lot management systems had several issues regarding safety management within the parking lot. For instance, major problems have been pointed out, such as the risk of pedestrian safety accidents in blind spots, the difficulty of rapid detection and response in the event of a fire, and the lack of effective management measures for vehicles violating no-parking zones. In particular, for large parking lots, it was difficult to effectively address these safety issues with a limited number of management personnel.

[0006] Furthermore, existing parking lot guidance systems are limited to simply displaying the number of remaining parking spaces by floor or zone, causing drivers to spend a significant amount of time searching for actual available parking spaces. This increases unnecessary vehicle movement within the parking lot, exacerbating congestion and contributing to energy waste and increased environmental burden.

[0007] Therefore, a method to resolve these various problems is required. Prior art literature

[0008] Korean Patent Publication No. 10-2006-0094276 The problem to be solved

[0009] The present invention is devised to solve the problems of the aforementioned prior art and aims to automatically determine the type of vehicle within a controlled facility by analyzing images acquired through an image capturing device, efficiently guide the vehicle to a parking space suitable for it, and ensure the safety of pedestrians within the controlled facility.

[0010] The problems of the present invention are not limited to those mentioned above, and other unmentioned problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0011] The intelligent control system of the present invention for achieving the above-mentioned purpose may include: a video capturing device installed at a predetermined location of a facility to be controlled, which captures a plurality of areas inside the facility to be controlled and generates first video data related to a parking space; a video analysis unit that receives the first video data from the video capturing device and analyzes the first video data to determine the status of the remaining parking space inside the facility to be controlled; and a display device installed at a predetermined location of the facility to be controlled, which receives the analysis result of the video analysis unit and displays location information of the remaining parking space.

[0012] And the image analysis unit can perform a parking space analysis algorithm comprising the steps of: (a-1) detecting the boundary of a parking space from the first image data; (a-2) determining whether a vehicle exists in an area within the detected boundary; (a-3) registering an area determined not to have a vehicle as a remaining parking space; and (a-4) transmitting location information of the registered remaining parking space to a display device.

[0013] In addition, the video recording device can additionally generate second video data related to a vehicle that has entered the controlled facility.

[0014] At this time, the image analysis unit may perform a vehicle analysis algorithm comprising the steps of: (b-1) determining the type of vehicle by analyzing the characteristics of the vehicle from the second image data; (b-2) selecting a remaining parking space where the vehicle can be parked among the remaining parking spaces according to the determined type of vehicle; and (b-3) transmitting location information of the selected remaining parking space to a display device.

[0015] In addition, the types of vehicles identifiable in step (b-1) may include electric vehicles, vehicles for the disabled, vehicles for pregnant women, light vehicles, and regular vehicles.

[0016] In addition, step (b-1) may include step (b-1-1) of detecting a license plate area of ​​a vehicle from the second image data, step (b-1-2) of detecting a front area of ​​a vehicle from the second image data, step (b-1-3) of detecting the outer boundary of a vehicle from the second image data and calculating the size of the vehicle, step (b-1-4) of determining whether it is an electric vehicle by analyzing the color of the license plate detected in step (b-1-1), step (b-1-5) of determining whether it is a vehicle for a disabled person or a vehicle for a pregnant woman by analyzing the presence of a disabled person symbol or a pregnant woman symbol in the front area detected in step (b-1-2), and step (b-1-6) of determining whether it is a light car by analyzing whether the size of the vehicle calculated in step (b-1-3) is less than or equal to a preset standard size.

[0017] Meanwhile, step (b-2) may include step (b-2-1) of classifying remaining parking spaces into electric vehicle-only spaces, disabled-only spaces, pregnant women-only spaces, light-car-only spaces, and general parking spaces; step (b-2-2) of determining whether there is a dedicated parking space that matches the type of vehicle identified in step (b-1); and step (b-2-3) of selecting an empty remaining parking space among the dedicated parking spaces if a matching dedicated parking space exists.

[0018] And if the type of vehicle identified in step (b-1) is determined not to be a general vehicle, and in step (b-3-2) it is determined that there is no dedicated parking space or that all of them are occupied, step (b-2) may further include step (b-2-4) of selecting an empty remaining parking space among general parking spaces.

[0019] Additionally, the image analysis unit can perform a pedestrian safety algorithm comprising: a step (c-1) of detecting a pedestrian from first image data; a step (c-2) of calculating the direction of travel of a vehicle from second image data; a step (c-3) of determining whether the direction of travel of a vehicle intersects with the location of a pedestrian; and a step (c-4) of predicting the possibility of a collision between a vehicle and a pedestrian and outputting a warning signal when it is determined that the direction of travel of a vehicle intersects with the location of a pedestrian.

[0020] In addition, the display device is installed at the entrance and internal intersection of the controlled facility and displays the direction of available remaining parking spaces, the number of remaining parking spaces in that direction, and the density, wherein the density can be divided into multiple pre-set levels and displayed in different colors. Effects of the invention

[0021] The intelligent control system of the present invention, designed to solve the aforementioned problem, has the advantage of significantly reducing installation and maintenance costs compared to existing systems that require individual sensors to be installed in each parking space, by capturing multiple areas within a controlled facility in real time through a video recording device and analyzing the results.

[0022] In addition, the present invention has the advantage of minimizing unnecessary vehicle movement within the controlled facility and improving parking efficiency by accurately identifying the status of remaining parking spaces within the controlled facility in real time through an image analysis unit and effectively guiding drivers through this information via a display device, thereby enabling drivers to quickly find available parking spaces.

[0023] Furthermore, unlike conventional individual sensor methods, the present invention determines the condition of the parking space through image analysis, thereby having the advantage of enabling stable operation of the parking control system without being affected by hardware problems such as sensor failure or malfunction.

[0024] 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 in the claims. Brief explanation of the drawing

[0025] FIG. 1 is a schematic diagram showing an intelligent control system according to one embodiment of the present invention. FIG. 2 is a diagram showing each process of a parking space analysis algorithm performed by an image analysis unit in an intelligent control system according to an embodiment of the present invention. FIG. 3 is a diagram illustrating various display methods through a display device in an intelligent control system according to an embodiment of the present invention. FIG. 4 is a diagram showing each process of a vehicle analysis algorithm performed by an image analysis unit in an intelligent control system according to an embodiment of the present invention. FIG. 5 is a diagram showing a detailed process for determining the type of vehicle among the vehicle analysis algorithms performed by the image analysis unit in an intelligent control system according to an embodiment of the present invention. FIG. 6 is a diagram showing a detailed process of selecting remaining parking spaces among vehicle analysis algorithms performed by an image analysis unit in an intelligent control system according to an embodiment of the present invention. FIG. 7 is a diagram showing each process of a pedestrian safety algorithm performed by an image analysis unit in an intelligent control system according to an embodiment of the present invention. FIG. 8 is a diagram showing each process of a fire detection algorithm performed by an image analysis unit in an intelligent control system according to an embodiment of the present invention. Specific details for implementing the invention

[0026] In this specification, where a component (or region, layer, part, etc.) is described as being "on," "connected," or "combined" with another component, it means that it may be directly placed / connected / combined with the other component, or that a third component may be placed between them.

[0027] Identical reference numerals denote identical components. Additionally, in the drawings, the thicknesses, proportions, and dimensions of the components are exaggerated for the effective illustration of the technical content.

[0028] "And / or" includes all one or more combinations that the associated configurations can define.

[0029] Terms such as "first," "second," etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. A singular expression includes a plural expression unless the context clearly indicates otherwise.

[0030] Additionally, terms such as "below," "lower side," "above," and "upper side" are used to describe the relationships between the components depicted in the drawings. These terms are relative concepts and are described based on the directions indicated in the drawings.

[0031] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Additionally, terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and are explicitly defined herein unless interpreted in an ideal or overly formal sense.

[0032] Terms such as "include" or "have" are intended to indicate 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.

[0033] In addition, in the present invention to be described below, the term "controlled facility" is a concept that includes various types of facilities. For example, a controlled facility may refer to spaces requiring safety management for vehicles and pedestrians, such as parking lots, routes for preschool children to and from school, areas with private academies, and commercial districts. More specifically, it may include parking facilities at large shopping malls or supermarkets, outdoor / indoor parking lots within residential complexes, above-ground and underground parking lots, and parking spaces at public institutions or companies. Furthermore, areas where the movement paths of vehicles and pedestrians intersect, such as child protection zones around schools, pedestrian-dense areas in busy commercial districts, and areas around the entrances of complex commercial facilities, may also be considered controlled facilities.

[0034] The intelligent control system of the present invention can be effectively applied to these various types of controlled facilities and can provide customized safety management solutions tailored to the characteristics of each facility.

[0035] Furthermore, in the embodiments of the present invention described below, the facility subject to control is exemplified as a parking lot; however, since this is implemented as one embodiment, it goes without saying that the present invention is not limited by said embodiment.

[0036] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0037] FIG. 1 is a schematic diagram showing an intelligent control system according to one embodiment of the present invention.

[0038] As illustrated in FIG. 1, in this embodiment, the intelligent control system may include an image capturing device (100), an image analysis unit (200), and a display device (300).

[0039] A video recording device (100) can be installed at a pre-set location of a controlled facility to generate first video data related to a parking space.

[0040] The video recording device (100) can be implemented as various types of video recording devices, such as CCTV cameras and IP cameras, and it is also possible to utilize existing video recording equipment already installed in the controlled facility. Utilizing existing video recording equipment within the controlled facility can have the advantage of reducing additional installation costs.

[0041] And the pre-set locations of the control target facility may refer to locations selected to minimize blind spots within the control target facility. For example, the pre-set locations may be set on the ceiling, pillars, or walls of the control target facility, and multiple video recording devices (100) may be arranged to complement each other to record the entire area of ​​the control target facility.

[0042] In addition, the first image data related to the parking space may include various information such as the boundary line of the parking space, the presence or absence of a vehicle within the parking space, special markings on the parking space (e.g., for disabled persons, for pregnant women), no-parking lines, vehicle movement paths, and pedestrian movement paths.

[0043] In addition, the video recording device (100) can additionally generate second video data related to a vehicle that has entered the control target facility.

[0044] The second image data can be used to determine the type of vehicle by analyzing the characteristics of the vehicle. Here, the types of vehicles may include general vehicles, electric vehicles, vehicles for the disabled, vehicles for pregnant women, and light vehicles.

[0045] As will be described later, the type of vehicle can be identified through the color of the license plate (e.g., in the case of electric vehicles), special markings on the front of the vehicle (e.g., disability symbol, pregnant woman symbol), and the size of the vehicle (e.g., identification of light vehicles).

[0046] In addition, the second video data can be used to calculate the vehicle's direction of travel and speed, which can also be used to predict the possibility of collision with pedestrians in blind spots within the controlled facility.

[0047] The video analysis unit (200) receives first video data from the video recording device (100) and analyzes the first video data to determine the status of the remaining parking space inside the facility under control.

[0048] For example, in this embodiment, the image analysis unit (200) may be equipped with a deep learning-based real-time object recognition algorithm and may simultaneously detect and classify multiple objects within a single image. Additionally, the image analysis unit (200) may simultaneously predict the location and type of objects from image data, thereby enabling real-time analysis of the situation of the facility under control.

[0049] In addition, the video analysis unit (200) can be implemented as various types of computing devices, such as local servers and edge computing devices, and can also utilize server devices already installed in the controlled facility.

[0050] The display device (300) is installed at a pre-set location of the facility to be controlled and can display location information of the remaining parking space by receiving the analysis result of the image analysis unit (200).

[0051] Here, the pre-set location may correspond to the entrance of the controlled facility, the ceiling, pillar, or floor of an internal intersection point. Additionally, the display device (300) can display the number of parking spaces available for each direction, and can divide the density of the controlled facility area into multiple levels and display them in different colors.

[0052] Additionally, the display device (300) may be configured to operate only when a vehicle or pedestrian is detected, thereby reducing power consumption.

[0053] In addition, the display device (300) can be implemented in various forms such as an LED display board or an LCD monitor, and it is also possible to utilize a display device already installed in the controlled facility.

[0054] Meanwhile, in this embodiment, the image analysis unit (200) can perform various algorithms as described below.

[0055] The image analysis unit (200) can perform a parking space analysis algorithm. The parking space analysis algorithm detects the boundaries of the parking space from the first image data and determines whether there is a vehicle in the area within the detected boundaries to register the remaining parking space.

[0056] The image analysis unit (200) can perform a vehicle analysis algorithm. The vehicle analysis algorithm analyzes the characteristics of a vehicle from the second image data to determine the type of vehicle, and selects remaining parking spaces where the vehicle can be parked according to the determined type of vehicle. Here, the types of vehicles may include electric vehicles, vehicles for the disabled, vehicles for pregnant women, light cars, and general vehicles.

[0057] The image analysis unit (200) can perform a pedestrian safety algorithm. The pedestrian safety algorithm detects a pedestrian from the first image data and calculates the direction of travel of the vehicle from the second image data to predict the possibility of a collision between the vehicle and the pedestrian.

[0058] In addition, the video analysis unit (200) can additionally perform algorithms such as monitoring no-parking zones, fire detection, and density analysis of controlled facilities. The analysis results of each algorithm can be displayed in real time through an administrator terminal or a display device (300).

[0059] The above algorithms can be performed independently or in conjunction with each other in the image analysis unit (200), and the specific processing steps of each algorithm will be explained in detail below.

[0060] FIG. 2 is a diagram showing each process of a parking space analysis algorithm performed by an image analysis unit (200) in an intelligent control system according to an embodiment of the present invention.

[0061] Referring to FIG. 2, in this embodiment, the parking space analysis algorithm may include steps (a-1) through (a-4).

[0062] (a-1) Step is to detect the boundaries of the parking space from the first image data.

[0063] In this step, the boundary of the parking space can be implemented as a boundary line, such as a solid line or a dotted line, that distinguishes the parking space. Accordingly, the image analysis unit (200) can detect the location and shape of the boundary line through an edge detection technique and can define the area of ​​each parking space by recognizing the corner points of the parking space. In addition, special symbols or characters marked on the parking space can also be recognized to determine the use of the corresponding parking space.

[0064] (a-2) Step is a step of determining whether a vehicle is present in the area within the detected boundary.

[0065] In this step, the presence of a vehicle can be determined through object recognition technology. The image analysis unit (200) can analyze the amount of pixel change within the parking space or detect feature points of the vehicle. In addition, the presence of a vehicle can be determined more accurately through various image processing techniques, such as shadow area analysis and brightness value distribution analysis within the parking space.

[0066] Step (a-3) is the step of registering areas determined to be vacant as remaining parking spaces.

[0067] In this stage, the area registered as a remaining parking space can be stored in a database and managed in real time. The image analysis unit (200) can store information such as the location, size, and use of the registered remaining parking space, and can periodically update the overall status of the remaining space of the controlled facility.

[0068] (a-4) Step is to transmit location information of the registered remaining parking space to the display device (300).

[0069] In this step, the location information may include coordinate values, zone information, or direction information within the controlled facility. The image analysis unit (200) can calculate the number of remaining parking spaces for each direction based on the transmitted location information and can calculate and transmit the density of the controlled facility together. Through this, the driver can efficiently find available parking spaces.

[0070] In this way, the parking space analysis algorithm in the present embodiment detects the boundaries of the parking space from the first image data, determines whether a vehicle exists in the area within the detected boundaries, registers the area where no vehicle exists as a remaining parking space, and then transmits the location information of the registered remaining parking space to the display device (300), thereby assisting in the efficient parking of vehicles within the controlled facility.

[0071] FIG. 3 is a diagram illustrating various display methods through a display device (300) in an intelligent control system according to one embodiment of the present invention.

[0072] As described above, the display device (300) can be installed at the entrance and internal intersection of the facility under control and can provide various display methods as follows.

[0073] Referring to FIG. 3, the display device (300) can indicate the direction of the remaining available parking space. In this embodiment, the direction indication can be implemented in the form of an arrow and can indicate various directions such as left, right, front, and rear.

[0074] Additionally, the display device (300) can display the number of remaining parking spaces in the corresponding direction. In this embodiment, the display of the number of remaining parking spaces can be implemented as a number and can be displayed separately for each direction.

[0075] Additionally, the display device (300) can display the density of each direction. In this embodiment, the density can be divided into a plurality of preset levels and displayed in different colors. For example, it can be displayed in red when full, yellow when crowded, and green when there is space.

[0076] At this time, the display device (300) can display the direction indicator and the number of remaining parking spaces in combination with colors according to density. In this embodiment, the arrow and number may have different colors depending on the density of the corresponding direction. For example, if the parking area in a specific direction is full, the arrow and number in that direction may be displayed in red; if it is crowded, they may be displayed in yellow; and if there is space, they may be displayed in green. Through this, the driver can intuitively understand the direction, the number of remaining spaces, and the density.

[0077] In addition, the display device (300) may display guidance on dedicated zones for each type of vehicle. For example, electric vehicle charging zones, zones for the disabled, zones for pregnant women, zones for light cars, etc., may be distinguished and displayed, and intuitive guidance can be provided by using unique icons or symbols for each zone.

[0078] In addition, the display device (300) may also display information such as the zone name or floor number of the facility under control. For example, the zone name may be expressed in alphabets or numbers, and may include additional information indicating the characteristics of each zone.

[0079] In this way, the display device (300) in the present embodiment can provide information about available parking spaces to the driver intuitively and efficiently through various display methods.

[0080] FIG. 4 is a diagram showing each process of a vehicle analysis algorithm performed by an image analysis unit (200) in an intelligent control system according to one embodiment of the present invention.

[0081] Referring to FIG. 4, in this embodiment, the vehicle analysis algorithm may include steps (b-1) through (b-3).

[0082] Step (b-1) is a step of determining the type of vehicle by analyzing the characteristics of the vehicle from the second image data.

[0083] At this stage, the types of vehicles may include electric vehicles, vehicles for the disabled, vehicles for pregnant women, light vehicles, and general vehicles. The image analysis unit (200) can determine the type of vehicle by analyzing various features such as the color of the vehicle's license plate, special markings on the front of the vehicle, and the size of the vehicle.

[0084] Step (b-2) is a step of selecting a remaining parking space where the vehicle can be parked among the remaining parking spaces, based on the type of vehicle identified.

[0085] In this step, the image analysis unit (200) can classify the remaining parking spaces into electric vehicle-only spaces, disabled-only spaces, pregnant women-only spaces, light vehicle-only spaces, and general parking spaces. It can also determine whether there is a dedicated parking space that matches the identified type of vehicle.

[0086] (b-3) Step is to transmit location information of the selected remaining parking spaces to a display device (300).

[0087] At this stage, the image analysis unit (200) can transmit direction information to the space, along with location information of the remaining parking space where the vehicle can be parked, to the display device (300).

[0088] In this way, the vehicle analysis algorithm in the present embodiment can assist in the efficient operation of dedicated parking areas for each type of vehicle by automatically determining the type of vehicle and selecting and guiding a parking space suitable for the vehicle.

[0089] FIG. 5 is a diagram showing a detailed process for determining the type of vehicle among the vehicle analysis algorithms performed by the image analysis unit (200) in an intelligent control system according to one embodiment of the present invention.

[0090] Referring to FIG. 5, the process of determining the type of vehicle in this embodiment, i.e., the aforementioned step (b-1), may include steps (b-1-1) through (b-1-6) in detail.

[0091] Step (b-1-1) is a step of detecting the license plate area of ​​a vehicle from the second image data.

[0092] In this step, the image analysis unit (200) can detect the area where the license plate is located on the front of the vehicle. The detection of the license plate area can be achieved by analyzing complex features such as the rectangular shape, border characteristics, and character arrangement pattern of the license plate.

[0093] In addition, the image analysis unit (200) can normalize license plates taken from various angles and convert them into a consistent form, and can process them to enable accurate color analysis by correcting the influence of illumination or shadows in the license plate area.

[0094] Step (b-1-2) is a step of detecting the front area of ​​the vehicle from the second image data.

[0095] In this step, the image analysis unit (200) can detect an area including the front bumper, grille, headlights, etc. of the vehicle. The detection of the front area can be performed by recognizing the characteristic morphological pattern of the vehicle, and the contrast ratio of the image can be corrected so as not to be affected by changes in illumination or reflected light of the surrounding environment.

[0096] In addition, the image analysis unit (200) can divide the front area into several detailed areas and independently analyze whether there is a special mark for each area.

[0097] Step (b-1-3) is a step of detecting the outer boundary of the vehicle from the second image data and calculating the size of the vehicle.

[0098] In this step, the image analysis unit (200) can detect the overall outline of the vehicle and calculate the size information of the vehicle. The detection of the outer boundary of the vehicle can be performed through an edge detection algorithm, and perspective correction can be performed so that accurate size measurement is possible even in images taken from various angles, such as the front, side, and rear of the vehicle. In addition, the image analysis unit (200) can estimate the actual physical size of the vehicle based on the detected outer boundary, and for this purpose, camera calibration information and depth information can be utilized.

[0099] Step (b-1-4) is a step of determining whether it is an electric vehicle by analyzing the color of the detected license plate.

[0100] In this step, the image analysis unit (200) can determine whether the color of the license plate is blue, which represents an electric vehicle. Color analysis can be performed in various color spaces such as RGB, HSV, etc., and a color correction algorithm can be applied to correct for changes in lighting conditions or the influence of shadows.

[0101] In addition, the image analysis unit (200) can analyze the color distribution of the license plate as a histogram to verify the uniformity of the color, thereby minimizing false detection.

[0102] Step (b-1-5) is a step of determining whether a vehicle is a disabled vehicle or a pregnant woman vehicle by analyzing the presence of a disabled person symbol or a pregnant woman symbol in the detected front area.

[0103] In this step, the image analysis unit (200) can determine the type of vehicle by analyzing the shape and color of the special marking attached to the front. The detection of the special marking may utilize techniques such as template matching and pattern recognition, and a scale-invariant feature point detection technique may be applied to enable accurate recognition even when the size or position of the marking varies.

[0104] In addition, the image analysis unit (200) can additionally check security elements such as detailed patterns or holograms of the mark to determine whether the special mark is authentic.

[0105] Step (b-1-6) is a step for determining whether a vehicle is a light car by analyzing whether the calculated vehicle size is smaller than or equal to a pre-set standard size.

[0106] In this step, the image analysis unit (200) can determine whether the vehicle is a light car by comparing the vehicle size information with a preset light car standard. The standard size for determining whether the vehicle is a light car can be set by reflecting the specifications of a light car according to regulations, and may include an allowable range considering measurement error.

[0107] In addition, the video analysis unit (200) can improve the accuracy of the judgment by comprehensively analyzing various dimensions such as the length, width, and height of the vehicle.

[0108] As such, the vehicle type identification process in this embodiment can automatically distinguish electric vehicles, vehicles for the disabled, vehicles for pregnant women, light cars, etc. by comprehensively analyzing various characteristics of the vehicle.

[0109] FIG. 6 is a diagram showing a detailed process of selecting a remaining parking space among a vehicle analysis algorithm performed by an image analysis unit (200) in an intelligent control system according to an embodiment of the present invention.

[0110] Referring to FIG. 6, in this embodiment, the process of selecting remaining parking spaces, i.e., the aforementioned step (b-2), may include steps (b-2-1) through (b-2-4).

[0111] Step (b-2-1) is the step of classifying remaining parking spaces into electric vehicle-only spaces, spaces for the disabled, spaces for pregnant women, spaces for light vehicles, and general parking spaces.

[0112] In this step, the image analysis unit (200) can analyze road markings, guide signs, etc. displayed in each parking space within the controlled facility to distinguish the use of the parking space.

[0113] In addition, the video analysis unit (200) can identify the distribution of dedicated spaces by zone of the controlled facility and manage the location information of each dedicated space by creating a database. At this time, the classification of dedicated spaces can be set so that it can be flexibly changed according to the control facility management policy.

[0114] Step (b-2-2) is a step for determining whether there is a dedicated parking space that matches the type of vehicle identified in Step (b-1).

[0115] At this stage, the video analysis unit (200) can determine whether there is a dedicated parking space designated for the same purpose as the type of vehicle. At this time, the existence of a dedicated parking space can be determined by considering the ratio of dedicated parking spaces to the total number of parking spaces in the area, the current occupancy status of each dedicated parking space, etc.

[0116] In addition, the video analysis unit (200) can analyze the distribution of dedicated parking spaces for each floor or zone of the controlled facility and recommend the optimal parking location.

[0117] Step (b-2-3) is the step of selecting the remaining empty parking spaces among the dedicated parking spaces if a matching dedicated parking space exists.

[0118] In this step, the image analysis unit (200) can identify a currently empty space among the dedicated parking spaces where the vehicle can be parked. At this time, the selection of the parking space can be made by considering various conditions such as the distance from the vehicle's current location, the difficulty of parking, and the congestion of the surrounding space.

[0119] In addition, the video analysis unit (200) may also calculate and provide the optimal path to the selected parking space.

[0120] Step (b-2-4) is the step of selecting an empty remaining parking space among general parking spaces when there is no dedicated parking space or all are occupied.

[0121] In this step, if there is no dedicated parking space for the vehicle or all spaces are in use, the image analysis unit (200) can select a suitable space from among general parking spaces. At this time, the selection of general parking spaces may be made by considering the size of the vehicle, parking convenience, etc., and if necessary, a different type of dedicated space may be temporarily allocated.

[0122] Meanwhile, in step (b-2-1) of the present embodiment, after dividing the remaining parking spaces into dedicated spaces and general parking spaces, the image analysis unit (200) can select an empty remaining parking space among the general parking spaces without going through steps (b-2-2) and (b-2-3) if the vehicle identified in step (b-1) is a general vehicle.

[0123] In addition, during the processing process, the image analysis unit (200) can select an optimal parking space by considering the location of the general parking space, accessibility, and the degree of congestion in the surrounding space. Furthermore, the image analysis unit (200) can calculate the shortest path to the selected general parking space and provide guidance through the display device (300).

[0124] As such, the remaining parking space selection process in this embodiment can enable efficient parking guidance by systematically selecting a suitable parking space according to the type of vehicle.

[0125] FIG. 7 is a diagram showing each process of a pedestrian safety algorithm performed by an image analysis unit (200) in an intelligent control system according to one embodiment of the present invention.

[0126] Referring to FIG. 7, in this embodiment, the pedestrian safety algorithm may include steps (c-1) through (c-4).

[0127] (c-1) Step is a step of detecting pedestrians from the first image data.

[0128] In this step, the video analysis unit (200) can detect pedestrians within the controlled facility in real time. Pedestrian detection can be performed through a deep learning-based object recognition algorithm, and information such as the location, direction of movement, and speed of the pedestrian can be calculated together.

[0129] In addition, the video analysis unit (200) can utilize a model trained under various environmental conditions to accurately detect pedestrians even in blind spots or low-light areas within the controlled facility.

[0130] (c-2) Step is the step of calculating the vehicle's direction of travel from the second image data.

[0131] In this step, the video analysis unit (200) can predict the direction of travel by analyzing the vehicle's movement trajectory. The calculation of the vehicle's direction of travel can be performed by tracking changes in the vehicle's position in consecutive frames, and the vehicle's current speed and acceleration information can also be calculated.

[0132] In addition, the video analysis unit (200) can also utilize additional information such as the vehicle's turn signal status or steering angle.

[0133] (c-3) Step is a step for determining whether the vehicle's direction of travel intersects with the pedestrian's location.

[0134] In this step, the video analysis unit (200) can analyze whether there is a possibility that the expected path of the vehicle and the location or path of the pedestrian may intersect. At this time, the determination of whether there is an intersection can be made by comprehensively considering the current location, speed of movement, and direction of travel of the vehicle and the pedestrian.

[0135] In addition, the video analysis unit (200) can calculate the risk level by considering the structural characteristics of the facility under control or factors limiting visibility.

[0136] (c-4) is a step of outputting a warning signal by predicting the possibility of a collision between the vehicle and the pedestrian when it is determined that the vehicle's direction of travel intersects with the pedestrian's location.

[0137] In this step, the video analysis unit (200) can calculate the expected time and location of the collision between the vehicle and the pedestrian and generate an appropriate warning signal. The warning signal may be output through a display device (300) or a separate warning light, alarm sound, etc., and the intensity or form of the warning may vary depending on the risk of collision.

[0138] In addition, the video analysis unit (200) can transmit the situation in real time to the administrator terminal to enable immediate response.

[0139] As such, the pedestrian safety algorithm in this embodiment can detect and prevent potential collision risks in advance by analyzing the movements of vehicles and pedestrians within the controlled facility in real time.

[0140] FIG. 8 is a diagram showing each process of a fire detection algorithm performed by an image analysis unit in an intelligent control system according to an embodiment of the present invention.

[0141] Referring to FIG. 8, in this embodiment, the fire detection algorithm may include steps (d-1) through (d-3).

[0142] (d-1) Step is a step of detecting fire characteristics from the first image data.

[0143] In this stage, the image analysis unit (200) can detect fire flames or smoke that may occur within the controlled facility in real time. At this time, the detection of fire characteristics can be achieved by analyzing changes in the color of the flames, the spread pattern of the smoke, and rapid changes in brightness values.

[0144] (d-2) Step is the step of specifying the location of the detected fire features.

[0145] At this stage, the image analysis unit (200) can identify the exact location where the fire occurred. Location identification can be achieved by comprehensively analyzing images obtained from multiple image capturing devices (100), and the fire occurrence point can be mapped on the structural diagram of the facility under control.

[0146] Step (d-3) is the step of generating a warning signal for the occurrence of a fire.

[0147] In this step, the video analysis unit (200) can generate an appropriate warning signal according to the location and scale of the fire. The warning signal can be output through a display device (300) or a separate alarm system, and the fire occurrence situation can be transmitted in real time to an administrator terminal.

[0148] In this way, the fire detection algorithm in this embodiment can minimize damage caused by fire by enabling rapid detection and response to the occurrence of a fire within a controlled facility.

[0149] Meanwhile, the intelligent control system of the present invention may further include identification and control functions for no-parking zones within the controlled facility.

[0150] To this end, the image analysis unit (200) can automatically identify no-parking zones by comparing and analyzing the drawing information of the controlled facility with real-time video. These zones may include emergency exits, fire hydrants, ramps for the disabled, pedestrian walkways, etc.

[0151] The video analysis unit (200) performs continuous monitoring of the no-parking zone and can immediately detect when a vehicle stops or parks in the zone. When a vehicle is detected in the no-parking zone, the system can deliver a warning message to the driver of the vehicle. This can be done through a display device (300) or an in-vehicle communication system.

[0152] In addition, real-time notifications can be simultaneously sent to administrators to enable a rapid response. The system records and analyzes instances of no-parking zone violations and can suggest additional safety measures or improved guidance for areas where issues frequently occur.

[0153] In other words, through these functions, the present invention can improve safety within the controlled facility and enable efficient parking management.

[0154] In addition, the intelligent control system of the present invention may additionally include a safety function for the entrance area of ​​the facility under control.

[0155] To this end, at least one of the video recording devices (100) may be installed around the entrance of a facility under control to record the area. At this time, the video recording device (100) may be implemented as a low-profile CCTV or set to a specific angle of view to protect portrait rights. Such a low-profile CCTV is installed at a low position from the ground so as to acquire necessary video information without directly recording the face of a pedestrian. The specific angle of view setting may be adjusted to focus on capturing pedestrian walkways and vehicle movement paths around the entrance while excluding unnecessary areas.

[0156] The video analysis unit (200) can analyze video data received from the video recording device (100) to monitor pedestrian passageways around the entrance, particularly sidewalk blocks or short crosswalk sections.

[0157] The video analysis unit (200) can detect and track pedestrians, electric scooter or bicycle users, and vehicles exiting the parking lot by utilizing a deep learning-based object recognition algorithm. Additionally, regarding vehicles exiting the parking lot, it can predict the possibility of collision with pedestrians or other means of transportation by comprehensively analyzing the type, size, speed, and direction of travel of the vehicle.

[0158] The display device (300) can output a warning signal when a dangerous situation is predicted based on the analysis results of the image analysis unit (200). This warning signal can be implemented in various forms.

[0159] For pedestrians, voice warnings via speakers installed near entrances or visual warnings via LED floor indicators can be provided. For drivers, warning messages can be delivered in conjunction with in-vehicle displays or voice guidance systems. Additionally, large LED display panels installed near entrances can visualize current hazardous situations to alert all users. In this case, the intensity and frequency of the warning signals can be adjusted in stages according to the severity of the hazardous situation.

[0160] This system may also include data collection and analysis functions. By continuously collecting and analyzing data such as pedestrian traffic volume by time of day and day of the week, vehicle entry and exit frequency, and frequency of hazardous situations, long-term safety improvement measures can be derived. Through this, countermeasures such as changes to entrance designs, installation of additional safety facilities, or development of user education programs can be established.

[0161] Through such a configuration, the present invention can effectively ensure pedestrian safety around the entrances of controlled facilities. In particular, it can significantly improve the safety of pedestrians with low attention spans or users of fast-moving scooters and bicycles, and can also contribute to accident prevention by allowing vehicle drivers to be aware of the presence of pedestrians and users of other means of transportation in advance.

[0162] The algorithms of the present embodiment have been described in detail above. Below, various technologies that may be additionally applied to the present embodiment will be described in detail.

[0163] First, in this embodiment, the image analysis unit (200) can additionally perform a multi-layered security authentication analysis to determine whether a special mark attached to a vehicle (10) is authentic.

[0164] To this end, the image analysis unit (200) can recognize an identification means (e.g., QR code) included in a disabled person mark or a pregnant woman mark. To this end, the identification means may include unique identification information of the issuing agency, validity period information, and issuance serial number, and each information may be stored in an encrypted form.

[0165] And the image analysis unit (200) can verify the authenticity of the information of the recognized identification means by comparing it with an online database in real time, and can restore the information through an error correction function even if the identification means is damaged or partially destroyed.

[0166] Additionally, the image analysis unit (200) can detect dynamic changes in the holographic pattern included in the special mark. In this case, the special mark may refer to an official certification mark attached to the front of the vehicle (10) to identify the type of vehicle.

[0167] For example, special markings may include markings for persons with disabilities and pregnant women, and each marking may be issued by an accredited institution, such as the government or a local government, in accordance with relevant laws and regulations.

[0168] In addition, the disability symbol may use a white wheelchair shape on a blue background as its basic design, and the pregnant woman symbol may use a silhouette of a pregnant woman on a pink background as its basic design. These special symbols may include the following security elements to prevent forgery.

[0169] For example, the hologram pattern can be designed to display a specific pattern change depending on the angle, and this pattern change follows a predefined sequence. The image analysis unit (200) can track the change in the corresponding pattern in consecutive image frames to verify whether the sequence matches, and can verify the depth and three-dimensionality of the hologram through interference pattern analysis of the pattern.

[0170] In addition, the image analysis unit (200) can detect a UV-reactive mark of a special mark using ultraviolet light. To this end, the UV-reactive mark can be implemented as a special pattern that is identifiable only under ultraviolet light, and this pattern can be composed of multiple layers and designed so that each layer reacts to ultraviolet light of a different wavelength.

[0171] And the image analysis unit (200) can more accurately determine whether the special mark has been tampered with by sequentially irradiating ultraviolet light of different wavelengths and comprehensively analyzing the reaction pattern of each layer.

[0172] Through such multi-layered security authentication analysis, the reliability of vehicle type identification can be improved in this embodiment, and fraudulent use can be effectively prevented.

[0173] Additionally, in this embodiment, the image analysis unit (200) can perform intelligent parking space allocation analysis to optimize parking efficiency.

[0174] Specifically, the image analysis unit (200) can perform optimal parking space matching according to the size of the vehicle (10). At this time, the parking spaces can be classified into large, medium, and small sizes according to their size, and the actual size information of each space can be stored in a database.

[0175] And the video analysis unit (200) can calculate the length, width, and height of the vehicle (10) and select the optimal space by comparing it with the size of each parking space. In addition, by considering the clearance space between the vehicle (10) and the parking space, it can prioritize recommending a space where safe parking is possible.

[0176] In addition, the video analysis unit (200) can dynamically allocate parking spaces by predicting the scheduled parking and exit times. To this end, it can calculate the estimated parking time for each vehicle by learning the location characteristics (commercial district, residential district, etc.) of the controlled facility and parking pattern data by time of day. Based on these predictions, the video analysis unit (200) can optimize the turnover rate of the controlled facility by placing short-term and long-term parked vehicles in different zones.

[0177] And the video analysis unit (200) can calculate the parking difficulty considering the turning radius of the vehicle. To this end, the parking difficulty can be classified into high, medium, and low by comprehensively analyzing the entry angle of the parking space, the distance from pillars or walls, and the distance from adjacent vehicles. The video analysis unit (200) can select a space where the vehicle can be safely parked by considering the size and turning radius of the vehicle (10), and can provide parking guide information according to the difficulty through the display device (300).

[0178] Through such parking efficiency optimization analysis, this embodiment can improve the space utilization of the controlled facility and increase parking convenience for drivers.

[0179] Preferred embodiments according to the present invention have been described above, and it is obvious to those skilled in the art that, in addition to the embodiments described above, the present invention may be embodied in other specific forms without departing from the spirit or scope thereof. Therefore, the embodiments described above should be regarded as illustrative rather than restrictive, and accordingly, the present invention is not limited to the description above but may be modified within the scope of the appended claims and their equivalents. Explanation of the symbols

[0180] 100: Video recording device 200: Video Analysis Department 300: Display device

Claims

Claim 1 An intelligent control system comprising: an image capturing device installed at a pre-set location of a facility subject to control, which captures multiple areas inside the facility subject to control to generate first image data related to a parking space; an image analysis unit that receives the first image data from the image capturing device and analyzes the first image data to determine the status of the remaining parking space inside the facility subject to control; and a display device installed at a pre-set location of the facility subject to control, which receives the analysis result of the image analysis unit and displays location information of the remaining parking space. Claim 2 An intelligent control system according to claim 1, wherein the image analysis unit performs a parking space analysis algorithm comprising: (a-1) a step of detecting the boundary of a parking space from the first image data; (a-2) a step of determining whether a vehicle exists in an area within the detected boundary; (a-3) a step of registering an area determined not to have a vehicle as a remaining parking space; and (a-4) a step of transmitting location information of the registered remaining parking space to the display device. Claim 3 In paragraph 1, the intelligent control system wherein the image capturing device additionally generates second image data related to a vehicle that has entered the control target facility. Claim 4 An intelligent control system that performs a vehicle analysis algorithm, wherein the image analysis unit in paragraph 3 includes: a step (b-1) of determining the type of vehicle by analyzing the characteristics of the vehicle from the second image data; a step (b-2) of selecting a remaining parking space among the remaining parking spaces where the vehicle can be parked according to the determined type of vehicle; and a step (b-3) of transmitting location information of the selected remaining parking space to the display device. Claim 5 In paragraph 4, the types of vehicles identifiable in step (b-1) include electric vehicles, vehicles for the disabled, vehicles for pregnant women, light vehicles, and general vehicles, in an intelligent control system. Claim 6 In claim 5, the above step (b-1) comprises: a step (b-1-1) of detecting a license plate area of ​​a vehicle from the second image data; a step (b-1-2) of detecting a front area of ​​a vehicle from the second image data; a step (b-1-3) of detecting the outer boundary of a vehicle from the second image data and calculating the size of the vehicle; a step (b-1-4) of determining whether it is an electric vehicle by analyzing the color of the license plate detected in step (b-1-1); a step (b-1-5) of determining whether it is a vehicle for a disabled person or a vehicle for a pregnant woman by analyzing the presence of a disabled person symbol or a pregnant woman symbol in the front area detected in step (b-1-2); and a step (b-1-6) of determining whether it is a light car by analyzing whether the size of the vehicle calculated in step (b-1-3) is less than or equal to a preset standard size. Claim 7 In claim 5, the above step (b-2) comprises: a step (b-2-1) of classifying the remaining parking spaces into electric vehicle-only spaces, disabled-only spaces, pregnant women-only spaces, light-car-only spaces, and general parking spaces; a step (b-2-2) of determining whether there exists a dedicated parking space that matches the type of vehicle determined in step (b-1); and a step (b-2-3) of selecting an empty remaining parking space among the dedicated parking spaces if a matched dedicated parking space exists. Claim 8 An intelligent control system according to claim 7, wherein in the case where the type of vehicle identified in step (b-1) is determined not to be a general vehicle, and in step (b-3-2) it is determined that there is no dedicated parking space or that all are occupied, step (b-2) further includes step (b-2-4) of selecting an empty remaining parking space among general parking spaces. Claim 9 An intelligent control system according to claim 3, wherein the image analysis unit performs a pedestrian safety algorithm comprising: (c-1) a step of detecting a pedestrian from the first image data; (c-2) a step of calculating the vehicle's direction of travel from the second image data; (c-3) a step of determining whether the vehicle's direction of travel intersects with the pedestrian's location; and (c-4) a step of predicting the possibility of a collision between the vehicle and the pedestrian and outputting a warning signal when it is determined that the vehicle's direction of travel intersects with the pedestrian's location. Claim 10 An intelligent control system according to claim 1, wherein the display device is installed at the entrance and internal intersection of a facility subject to control and displays the direction of available remaining parking spaces, the number of remaining parking spaces in the said direction, and the density, wherein the density is divided into a plurality of preset levels and displayed in different colors.

Images