KR101149651B1 - Parking lot real time detection system using image analysis of omnidirectional camera and detection method thereof - Google Patents

Abstract

PURPOSE: An available parking space detection system through multi-faced image analysis of a fisheye camera and a detection method thereof are provided to easily detect an available parking space by distinguishing an occupation state of a parking space. CONSTITUTION: An omni-directional camera(310) is installed on the ceiling of a parking lot. The omni-directional camera omni-directionally records the lower side of the ceiling. The omni-directional camera comprises a fisheye lens. A network video server(300) converts an optical image into image data. The network video server encodes video data. A parking monitoring server(200) is operated in order to distinguish an occupation state of a parking space. A parking space management server(100) detects and transmits an available parking space on a real time basis. A display panel(400) displays a parking state of each level. A lamp(410) displays an occupation state of each parking section.

Description

TECHNICAL FIELD [0001] The present invention relates to a real-time detection system for a parking space using a multi-sided image analysis of a fisheye camera,

The present invention relates to a parking real time real time sensing system and a sensing method thereof, and more particularly, to a fisheye camera capable of more efficiently managing a parking space by analyzing a parking space taken by an omnidirectional camera equipped with a fish- The present invention relates to a real time parking system and a method of detecting the same.

In recent years, parking problems have become a serious social problem due to the increasing trend of vehicles, especially in places where large numbers of people gather, such as large discount stores, department stores, and performance theaters. Therefore, researches on techniques for efficient parking management are being actively carried out.

However, parking management has a lot of management items such as real-time checking of parking area, recording of vehicle information, calculation of parking time, identification of vehicle owner, charge adjustment, illegal parking control, lighting facility management and it is inefficient It costs a lot.

A typical parking lot operation method is that a plurality of parking administrators reside in a parking lot, and when the vehicle enters the parking lot,

The parking start time and end time are checked to collect parking fees for the time at the time of departure. This type of operation not only requires a lot of manpower, but also causes a problem of illegal use of parking errors and parking errors.

In order to solve such a problem, an automatic parking management system using a parking gate bar is being introduced in a large-scale outdoor parking lot or a building-installed parking lot so that a considerable portion of the parking lot operation and management can be automated.

For example, by using a loop detector and a parking card issuer, the time of entry and exit of the vehicle is automatically recorded, and the parking fee is computed and processed when the vehicle leaves the vehicle, thereby minimizing the overall manpower required for the parking lot management. However, in such an automatic parking management system, a parking manpower is required to monitor the settlement of parking fees and illegal departure of the vehicle.

1 is a photograph showing a state in which a parking space is photographed by a plurality of cameras provided on a raceway W of a parking lot among conventional parking lot management systems. As shown in FIG. 1, in a conventional parking surface management system, a raceway W is installed on a parking lot ceiling along both sides of an entrance and exit, and camera modules C1 and C2 each having a plurality of cameras And is constructed in such a manner that it is installed on the raceway W and takes a picture of the parking side across the entrance and exit lanes.

1, a conventional method using a camera having a unidirectional viewing angle and photographing a vehicle on both sides of a parking zone is also used. In addition, a parking vehicle is photographed by each of right and left two- And a parking lot management system for detecting the presence or absence of an entrance and exit of the vehicle or recognizing the vehicle number.

In such a method, the number of cameras required for installation is large and the need to install raceways W on both sides necessitates an increase in cost, and there is a problem in simplifying the system.

In order to solve such a problem, the present invention introduces a high-quality omnidirectional camera capable of taking a panoramic picture, captures a large number of parking faces in one camera, distinguishes tolerances and gaps, and provides a fisheye And to provide a real time parking system capable of parking through multi-sided image analysis of a camera.

It is another object of the present invention to provide a real time parking system capable of distinguishing tolerance and fullness by histogram similarity analysis.

Another object of the present invention is to provide a real time parking system capable of recognizing the number of vehicles entering and leaving the parking lot and facilitating the settlement of charges according to the parking time.

In order to achieve the above object, the present invention provides a parking sensing surface real-time sensing system using a multi-sided image analysis of a fisheye camera, comprising a fisheye lens mounted on a ceiling of a parking lot for photographing a lower surface of the ceiling, And a controller for converting the circular optical image input through the camera into circular image data, compressing and storing the image data for one screen, encoding the stored optical image data into a signal capable of receiving input from a display device, A network video server for processing the encoded data by any one of an MPEG 4 moving picture decoding technique, a technique for converting decoded image data into a panorama image, and a technique for playing back a panorama image or an image in an enlarged form of a specific portion, If the output video of the video server is more than 4 A parking monitoring server operable to compare the restored image with the divided image information corresponding to the divided areas of the main parking space and the compressed and stored screen data to identify the tolerance and the difference according to the degree of similarity; A parking space management server for receiving a parking space for each parking area and sensing a space available for parking by a predetermined area and transmitting the parking space in real time, a floor display panel for displaying a floor parking state in cooperation with the parking space management server, And a ramp unit interlocked with the management server or the parking monitoring server so as to be positioned at an upper portion of the parking zone so as to indicate tolerance or fullness of each parking zone.

The parking monitoring server extracts a histogram of the image of the parking space on the basis of a tolerance or a full parking position, stores the extracted histogram as basic data, extracts histogram data of the compressed and stored screen data, compares the similarity with the basic data, Wherein the histogram data is compared with the histogram data so that the median of the histogram becomes a reference for the similarity measurement.

The network video server includes a CMOS sensor, which is an image pickup device that converts the circular optical image input through the camera into circular image data and outputs the image data as a digital image, An NTSC / PAL encoder for converting the image data stored in the first DRAM into an analog signal that can be received by the TV, and outputting the converted analog signal; (Contrast), color, etc. of the image in the window by moving the size or position of the window within the image screen, and automatically detects a moving object in the screen as the signal is output, A motion detector for detecting only motion and outputting a signal, and a Linux-based program for controlling each component But comprises a lash memory, it is desirable to handle so that it can be stored directly in the storage device, to compress the compressed image data stored in the first DRAM as MPEG 4 using the network interface using a transmission or yueseubi interface to the PC.

Wherein the tolerance notification display and the full-time notification display each emit different colors, and recognizes the number of the vehicle to be entered into the parking lot and the number of the vehicle to be shipped, and transmits the recognized image number to the parking space management server And the parking space management server records the arrival time and the departure time of the vehicle by the number so that the parking lot management server can automatically collect the parking fee.

The parking monitoring server compares the median value of the histogram with the histogram data so that the median value of the histogram is used as a reference for the similarity measurement, and judges the case where the similarity of the histogram is out of the error for a predetermined time or longer.

Meanwhile, a real-time parking method using a multi-faceted image analysis of a fisheye camera provided with a fish-eye lens mounted on a ceiling of a parking lot and equipped with a fish-eye lens so as to photograph the lower surface of the ceiling lower part including the parking surface in all directions (a) converting a circular optical image input through the camera into circular image data, compressing the image data by one screen data, storing the image data, and encoding the resized image data into a signal capable of being received by a display device b) comparing an image obtained by restoring the output image of step (a) into the divided image information corresponding to at least four divided areas of the parking plane, and comparing the compressed and stored image data to determine a tolerance and a step difference according to the degree of similarity; (C) receiving the tolerance data and the blanking data for each parking zone identified in the step (b) (D) displaying the floor level parking status on the electric signboard of each floor with the data transmitted in the step (c); and (e) On the display means.

The step (b) may include extracting a histogram of a tolerance or a parking position image and storing the extracted histogram as basic data, extracting histogram data of the compressed and stored screen data, comparing the similarity with the basic data, do.

The step (a) includes the steps of: (a-1) converting a circular optical image input through the fish-eye lens into circular image data and inputting the image data in YCbCr format; (a-2) (A-3) encoding the edited image using an MPEG codec, (a-4) checking a recording mode to determine whether or not to store the image data, and (a-5) If it is determined in the step (a-4) that the recording mode is always the mode, it is determined whether both the network and the USB hard disk are connected, and only the video is transferred to the PC when all the disks are connected and the recording flag is transmitted together with the data a moving picture decoding technique, a technique of converting decoded image data into a panoramic image, and a technique of processing the converted image in real time to apply the panorama image or (A-7) if it is determined to be the motion-picture recording mode in the step (a-4), when a motion of the object is detected in the set area, the motion- (A-8) enabling a reserved recording flag at a predetermined time when it is determined in the schedule recording mode in the step (a-4) to enable network recording or network recording And a step of judging whether or not recording is to be performed in transmission or in a hard disk drive of the USB hub.

As described above, according to the parking real time real-time detection system using the multi-plane image analysis of the fisheye camera of the present invention, a high-quality omnidirectional camera capable of omni-directional photographing can be installed and the tolerance or full- It is possible to construct a management system.

In addition, according to the present invention, the parking-enabled real-time detection system using the multi-plane image analysis of a fisheye camera can quickly and accurately identify the tolerance and the vacant space through a small number of cameras and histogram analysis.

According to the parking real time real-time detection system using the multi-sided image analysis of the fisheye camera of the present invention, by recognizing the number of the vehicle entering and leaving the parking lot, not only can the fare adjustment according to parking time be made easier, Security and security functions can also be performed.

FIG. 1 is a photograph showing a state in which a parking lot is photographed by a plurality of cameras installed on a raceway of a parking lot among conventional parking lot management systems,
2 is a configuration diagram showing the configuration of an embodiment of the present invention,
3 is a detailed block diagram of an NVR in an embodiment of the present invention,
4 is a view showing the image data moving path of the present invention
FIG. 5 is a flowchart for explaining the operation of the real time parking system capable of detecting a parking available surface through multi-faceted image analysis of a fisheye camera of the present invention.
And,
6 is a detailed flowchart for explaining the omnidirectional image processing method of FIG.

It is to be understood that the words or words used in the present specification and claims are not to be construed in a conventional or dictionary sense and that the inventor can properly define the concept of a term in order to describe its invention in the best possible way And should be construed in light of the meanings and concepts consistent with the technical idea of the present invention.

Throughout the specification, when an element is referred to as "comprising ", it means that it can include other elements as well, without excluding other elements unless specifically stated otherwise. It should be noted that the terms such as " part, "" module, " .

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 2 is a block diagram illustrating a main configuration according to an exemplary embodiment of the present invention. As shown in FIG. 2, a real-time parking system capable of detecting a parking available through a multi-faceted image analysis of a fisheye camera of the present invention includes an omnidirectional camera A network video server (NVR) 300 for restoring an image taken by the omnidirectional camera 310, a video network server 300 for comparing the output image of the NVR 300 with a basic image, A monitoring server 200, a parking space management server 100 for providing a parking space in real time, a floor lighting board 400 for displaying a floor parking status, a lamp unit 410 for displaying a parking zone and an impossible zone as a lamp, As shown in FIG.

The omnidirectional camera 310 is composed of a camera mounted on a ceiling of a parking lot and equipped with a fisheye lens so that the lower surface of the ceiling including the parking surface can be photographed in all directions.

The omnidirectional camera 310 is installed on the ceiling of the parking lot and photographs all over the ceiling including the parking surface and the entrance / exit lane. In reality, the omnidirectional camera 310 It is preferable to determine the installation place in consideration of the surface. In addition, input / output lane image information to be used for crime prevention as well as the parking plane can be obtained through the omnidirectional camera 310. [

The omnidirectional camera 310 is installed on the ceiling of the entrance / exit lane, but may be fixed to the ceiling according to the facilities inside the car park, and may be installed lower than the ceiling. Also, the omnidirectional camera 310 It is preferable to use an intelligent camera to which an IP (Internet Protocol) address is attached to the NVR 300. In this case, a network (e.g., LAN, Local Area Network) do.

The network video server (NVR) 300 converts the circular optical image inputted through the omnidirectional camera 310 into circular image data, compresses the image data by one screen data, stores the image data, resizes the stored image data, A technique for converting the encoded data into MPEG4 video decoding technology, a technique for converting decoded video data into a panorama image, and a technique for reproducing a panorama image or an image in which a specific portion is enlarged And is configured to process a single image.

The NVR will be described in more detail with reference to the detailed block diagram of the NVR of FIG.

The circular optical image entered through the fisheye lens 310 is converted into circular image data through the CMOS sensor 320 which is an image pickup element and is transmitted to the digital main processor 330. The fisheye lens 310 uses a wide field of view (180 degrees or more), but uses a wide-angle convex lens in view of the characteristics of the system. In addition, it is necessary to increase the number of pixels of the CMOS sensor in order to take advantage of the characteristics of a fisheye lens capable of shooting a wide range and to obtain a clearer image after software processing.

The circular image data photographed in digital form through the CMOS sensor 320 is encoded into a moving picture using the MPEG-4 codec in the main processor 330. The Moving Picture Experts Group 4 codec (MPEC4) has higher compression efficiency than JPEG (Joint Photographic Expert Group), which is widely used in surveillance cameras. Using a high-compression codec reduces the resources required for data processing and transmission and reduces the amount of data that is ultimately stored.

In addition, the CMOS sensor 320 is configured to convert the circular optical image received through the fisheye lens 310 into circular image data using the BT656 interface and output it in the YCbCr 422 format.

A memory unit 370 including a flash memory 372 and a first DRAM (Dynamic Random Access Memory) 374, a second DRAM 376 and a third DRAM 378 is provided with a software processing, display and display controller The main processor 330 may be configured to recover an image by using a special technique (Map Projection) for recovering a photographed distorted image of the camera unit, so that a rectangular panoramic image or a perspective image And the like are processed into images that are easier to see.

It is also desirable to apply Day & Night, White Balance, backlight correction and self noise reduction functions to the CMOS sensor and software in order to improve the quality of a clear image and a dark region or night image .

The main processor 330 also allows the image processing software to pan, pan, tilt, zoom in or zoom out, It is more preferable to be able to select functions such as enlarging and displaying an image of a specific part.

As described above, the above-mentioned image conversion processing for converting a circular optical image into a human-friendly image such as a rectangular panorama image or a perspective image requires a lot of coordinate transformation for converting polar coordinates into rectangular coordinates using a trigonometric function, It is not a specific part of the present invention, and thus a detailed description thereof will be omitted.

The encoded image data is transmitted to a PC through a TCP / IP (Transmission Control Protocol / Internet Protocol) protocol through an Ethernet interface 340. Since a large amount of data is transmitted to obtain a clear image, a large amount of data is transmitted through Ethernet. Therefore, it is desirable to construct a dedicated network in which a camera and a PC are connected with little network delay.

Further, the image photographed in digital form through the CMOS sensor 320 is converted from an NTSC / PAL encoder 360 into an analog signal of the NTSC / PAL type using a general signal processing technique and output separately, It is possible to directly check the status of the image.

On the other hand, if a USB hard disk such as an external storage device is additionally configured, the encoded video data can be directly stored in the USB hard disk 410 through the USB interface 350 without any additional equipment. USB high speed driver creation, USB device management technology, and file system technology are required, but they are general and therefore not described. It is desirable that this function be developed for an environment that requires low cost, high efficiency surveillance imaging and storage.

The present invention uses motion detection technology, which is one of image coding techniques, to store an image if the motion of the specific parking surface is confirmed without storing the image data when there is no motion on the parking surface . With this feature, you can drastically reduce the amount of data stored when there is no movement for most of the time and reduce the time required to analyze the entire image.

For this, the motion detector 390 detects a change in the contrast (contrast), color, etc. of the in-window image by allowing the size or position of the window to be moved within the image screen to be searched, Or detects only motion within the window and outputs a signal.

That is, the main processor 330 determines whether or not to store the screen data in the first DRAM according to the presence or absence of the detection signal of the motion detector 390, and outputs the data (1600x1200) received from the CMOS sensor 320 (1200x1200) and stores it in the first DRAM.

When the network is connected to the Ethernet interface 340 which is a network interface and the USB interface is connected to the USB interface 350 at the same time, the main processor 330 does not store data in the USB interface through the USB interface 350, 340, and controls data to be stored in the USB interface only when the network interface 340 is not connected to the network.

The video data moving path of the present invention will be described in detail with reference to FIG.

The CMOS sensor 320 outputs the image data to a CCD (charge coupled device) intermediate frequency (CCD) block 322 in the YCbCr 422 format using the BT656 interface. In the SIDEFA 322 block, the image data is received in the YCbCr 422 format with the BT656 interface, and the image data is stored in the first DRAM 374 in YCbCr 422 format.

The image data stored in the first diram 374 is too large to output directly to the TV. Therefore, the image is converted into a small image such as 720x480 pixels through the resize engine 324, and is stored again in the second dyram 376. However, it is a matter of course that the size of the pixel can be varied when a precise image is desired as in the cloud distribution measurement of the weather station.

The image data stored in the second diram 376 is converted into an analog signal through the NTSC / PAL encoder 360 and output to the display device such as a TV. The NTSC / PAL encoder 360 may use an external chip, or may use a block in a chip if the video processor is a dedicated video processor.

Meanwhile, the video data stored in the first diram 374 is compressed by the software MPEG 4 using a video encoder 380 and stored in the third diram 378. The compressed image data stored in the third dirram 378 can be transferred to the PC using the network interface 340 or directly stored in the USB hard disk using the USB interface 350. [

Preferably, the main processor 330 transmits data through the network interface 340 without storing data in the USB bus when the network is connected to the network interface 340 and the USB interface is connected to the USB interface 350 at the same time, Controls that data is stored in the USB drive only when it is not connected.

At this time, the data stored in the PC may be circular image data such as the image seen from a fish-eye lens. Therefore, the data stored in the PC can be expressed in various forms according to the necessity in the future by applying the MPEG 4 moving picture decoding technique, the technique of converting the decoded image data into the panorama image, and the technique of processing the converted image in real time. It is possible to store only the panoramic image or the image of the enlarged form of the specific part separately.

In addition, by using the reserved recording function to record video data only in a specific time period, it is possible to reduce the amount of data to be stored and to reduce image analysis time.

The NVR 300 may be configured to be integrated with various network video encoder cameras and network monitoring devices of other companies, and may support pan / tilt / zoom and digital input / output. In addition, can do.

In addition, an account can be unlimitedly registered for user login and user protection, and an alarm function can be automatically performed when a predetermined event occurs.

Since these additional functions are general camera control and image processing functions, a detailed description will be omitted.

The parking monitoring server 200 compares an image obtained by restoring an output image of the NVR 300 into divided image information corresponding to four or more divided areas of the parking plane and the compressed and stored image data, And is operated to identify the fullness.

To this end, the parking monitoring server 200 extracts the histogram of the tolerance or full-length parking surface image as basic data, extracts the histogram data of the compressed and stored screen data, compares the similarity with the basic data, The central value of the histogram is set to be a reference for the similarity measurement, and the histogram data is compared to identify the tolerance or the fullness state.

More specifically, it will be described with reference to the following table.

Table 1 shows histogram data extracted from the basic image of the tolerance state and stored as basic data. Table 2 shows the histogram data of the full state extracted, and the similarity of the two histogram data has data out of the error range. The degree of similarity of the histogram data is numerically calculated to recognize that the vehicle is parked.

That is, the histogram data are compared so that the median value (arrow portion) of the two histograms becomes the reference for the similarity measurement, and the case where the similarity of the histogram is out of the error for a predetermined time or more is judged as a full state.

In Table 1, which is the histogram of the tolerance state, the median is located between 'x3' and 'x4', but the median is located in 'x4' (sky blue) Able to know.

Since the error range may vary depending on the place or the surrounding environment, it is preferable to set the error range by viewing the histogram of the tolerance and the tolerance state inputted when the parking surface is installed.

 The median value of the histogram is used as the reference for the similarity measurement because the median value of the histogram is shifted according to the change in the surrounding brightness and the characteristic of the graph is changed within the error range so that the median value of the histogram is compared with the image data .

The determination of the predetermined time or longer is intended to improve an instantaneous image error in the vehicle sensing area. When considering the error in the motion in the vehicle-to-vehicle sensing area other than the vehicle, So as to reduce the error by detecting the vehicle only when there is a difference.

The parking space management server 100 is configured to transmit the tolerance and the hourly space data of each parking zone transmitted from the parking monitoring server 200 and to detect whether the parking space is available for each predetermined area and transmit the parking availability in real time.

Specifically, the data transmitted from the parking space management server 100 is transmitted to the floor electric sign board 400 indicating the floor parking status to display the parking status for each parking lot, and the lamp unit 410 To display the tolerance or the completion status of each parking zone.

The layered electric sign board 400 and the ramp unit 410, which display the present state of parking by the floor, have a general configuration, and thus a detailed description thereof will be omitted.

In addition, since the parking space management server 100 can register and set the parking space for each parking drawing and register the camera corresponding to each parking space, the parking space can be grasped more easily. Therefore, It is possible to display the number and the quantity of the tolerance surface and the quantity of the full surface on the layer electric sign board 400.

It is of course possible to display an image of the corresponding NVR 300 on a display device (not shown) when a floor image or a parking surface image to be monitored or to be confirmed is required.

Meanwhile, for remote control, a connection address can be allocated to each parking plane management server to connect to the Internet, and a connection address can be allocated to each NVR 300 to prevent unauthorized viewing of images, and a user ID and a password can be set .

It is preferable that the tolerance notification display and the full-time notification display displayed on the lamp means 410 emit different colors in correspondence with each other so that the user can conveniently check the parking space.

To this end, the lamp unit 410 is mounted with a light emitting diode and illuminates a color light according to the parking surface situation, so that the vehicle state indicator can be displayed in different colors, and the mounting position is set around the omnidirectional camera 310 But it is installed for each parking surface so that the user can easily recognize the tolerance zone by looking at the color.

The video device 310a recognizes the number of the vehicle to be parked in the parking lot, the time of arrival of the vehicle, the number of the vehicle to be shipped, and the shipment time, and transmits the number to the parking space management server 200. The parking space management server 100 then The corresponding data is transmitted so as to collect the parking time and parking time of the vehicle.

Similarly, the parking lot management server 100 stores the number of the goods receipt vehicle transmitted from the video device 310a and the corresponding goods receipt time, and when the number of the dispatch vehicle is transmitted, the goods receipt time and the dispatch time of the vehicle are analyzed Data about the parking charge may also be transmitted.

Hereinafter, the operation of the real time parking detection system through the multi-plane image analysis of a fisheye camera using the above-described apparatus will be described with reference to the drawings.

FIG. 6 is a detailed flowchart for explaining an omni-directional image processing method of FIG. 5; FIG. 6 is a flowchart illustrating an operation of a real time parking system capable of detecting a multi-faceted image of a fisheye camera of the present invention.

Referring to FIG. 5, a parking space image is input to the NVR 300 through a camera 310 (S500).

In step S600, the circular optical image input in step S500 is converted into circular image data, compressed by one screen data, stored and the stored image data is resized and encoded into a signal that can be received by the display device.

An image processing method of the NVR 300 will be described in detail with reference to FIG.

The circular optical image received through the fisheye lens 310 is converted into circular image data using the BT656 interface by the CMOS sensor 320 and input in the YCbCr 422 format (S601).

The main processor finds an area 1200x1200 in which the data (1600x1200) input from the CMOS sensor 320 is actually formed, edits the image (S602), and encodes the edited image using the MPEG4 codec (S604).

In step S606, a recording mode is checked to determine whether to store the image data. If it is determined in step S606 that the recording mode is always the recording mode, it is determined whether a network is connected (S620) It is judged whether or not they are all connected and the recording flag is transmitted together with the data only when all are connected (S621). When the transmission is completed (S622), the process ends.

The PC then applies the MPEG 4 moving picture decoding technique, the technique of converting the decoded image data into the panorama image, and the technique of processing the converted image in real time, so that the panorama image or the enlarged image of the specific part Can be stored separately.

If it is determined in step 606 that the motion is detected in the motion detection mode, the motion detector 390 may enable the motion recording flag to determine whether recording is to be performed during network transmission or hard disk storage S610).

If it is determined in step S606 that the schedule recording mode is set, the reserved recording flag is enabled at a predetermined time so that network transmission or recording in the USB hard storage can be determined (S614).

If the network is not connected and the USB hard disk is not connected in step S620 and step S630, the image data is initialized in step S650. If it is determined in step S640 that the recording mode is not always the recording mode, If the recording flag is not set, the process proceeds to step S650. If the recording flag is set, the data is controlled to be stored in the USB hard disk (S662).

The image output in step S600 is obtained by comparing the image reconstructed into the divided image information corresponding to the four or more divided areas of the parking space in the parking lot monitoring server 200 and the compressed and stored image data, (S510).

Of course, in step S510, the histogram analysis described above distinguishes the tolerance and the runaway state.

When the parking space management server 100 receives the parking space tolerance and the cumulative data for each parking zone identified in step S510, the parking space management server 100 senses the parking space for each predetermined area and transmits the parking space to the floor lighting board 400 and the lamp unit 410 in real time S520), the floor electric sign board 400 displays floor-level parking status, or the lamp unit 410 displays a tolerance or a full-floor status of each parking zone (S530).

As described above, the present invention extracts characteristics of spatial data when a car is parked and when it is parked from a video input from a fisheye lens camera installed on the ceiling of a parking lot, The parking space management server and the NVR are notified of the space, and the parking space can be notified in real time.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art.

100: parking lot management server 200: parking lot monitoring server
300: NVR 310: camera
400: electric display board for each floor 410: lamp means

Claims (1)

An omnidirectional camera installed on a ceiling of a parking lot and equipped with a fish-eye lens so as to photograph the lower surface of the ceiling lower part including the parking surface in all directions;
A controller for converting the circular optical image inputted through the camera into circular image data, compressing the circular image data by one screen data, storing the image data, resizing the stored image data into a signal capable of being input from the display device, 4 video decoding technique, a technique of converting decoded video data into a panoramic video, and a technique of playing back a panoramic video or a video of a specific portion in an enlarged form;
The video image restored by dividing the output image of the network video server into the divided image information corresponding to the four or more divided areas of the parking surface, and the compressed and stored screen data to identify the tolerance and the difference according to the degree of similarity, Monitoring server;
A parking space management server receiving the parking space clearance data and the parking space data transmitted from the parking monitoring server and detecting a parking space for each predetermined area and transmitting the parking space available in real time;
A floor light display panel interlocked with the parking space management server to display the floor parking status;
A lamp unit configured to be positioned at an upper portion of each parking zone in association with the parking surface management server or the parking monitoring server to display a tolerance or a vacancy state of each parking zone;
/ RTI >
The parking monitoring server
Extracting a histogram of a tolerance or full-length parking plane image as basic data, extracting histogram data of the compressed and stored screen data, comparing the similarity with the basic data to identify a tolerance or a fullness state,
The parking monitoring server
Comparing the median value of the histogram with the histogram data so as to be a reference of the degree of similarity measurement,
The network video server
A CMOS sensor that is an imaging device that converts a circular optical image inputted through the camera into circular image data and outputs the circular image data as digital data;
A CCD IF block for storing output image data of the CMOS sensor in a first DRAM by one screen data;
An NTSC / PAL encoder for converting image data stored in the first DRAM into an analog signal that can be input to a TV, and outputting the converted analog signal;
The size or position of the window can be moved within the image screen to be searched to detect a change in the contrast (contrast), color, etc. of the image in the window, and a moving object in the screen is automatically detected A motion detector for detecting only motion within the window and outputting a signal; And
A flash memory storing a Linux-based program for controlling each component;
And compresses the compressed image data stored in the first DRAM into MPEG-4 and transmits the compressed image data to the PC via the network interface or directly stores the compressed image data in the storage device using the USB interface. Real - time Parking System with Multi - sided Image Analysis.

Images