JP2012008021A - Monitoring device, method and program, and monitoring system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To monitor birds approaching that may cause a bird strike and give warning.SOLUTION: A radar device (10) periodically outputs radar image data obtained by scanning a periphery. A background creation function (32a) of a monitoring program that operates on a computer (30) creates background image data from the radar image data of multiple numbers of scenes, and a background removal function (32c) removes the background image data from subsequent radar image data. A binarization function (32d) binarizes echo images, and a center determination function (32e) determines a center of each echo image. A rain/snow removal function (32f) removes rain and snow. A tracking function (32g) tracks a bird from the result of rain/snow removal of the multiple scenes. A warning function (32h) outputs warning when a bird enters a prescribed area.

Description

  The present invention relates to a monitoring apparatus, method and program for monitoring a moving body such as a bird, and a monitoring system.

  Bird collisions have become a major problem at airfields and wind power generation facilities. For this reason, a monitoring system for monitoring approaching birds is required.

  For example, Patent Document 1 describes that birds are observed by using an airport monitoring radar, a weather observation radar, an aircraft tracking (tracking) radar, a marine monitoring radar, and the like. The sound generated by the birds is collected by the sound collecting means while night-visioning by the visual means, and the night vision direction of the night vision means 30 and the sound collection direction of the sound collection means are adjusted in accordance with the movement of the birds, and the adjustment position A bird observing apparatus and a bird observing method for easily and accurately acquiring the position of a bird by holding

  Further, in Patent Document 2, a millimeter wave radar is turned upward, and whether a target body is a helicopter or a bird is identified based on distance information acquired by the millimeter wave radar and intensity information of a received signal. It is described. A small object such as a bird or a small flying object such as a helicopter or a small airplane is distinguished based on the reception intensity of the reflected wave.

JP 2005-189103 A Japanese Patent Laid-Open No. 2005-233763

  In a conventional monitoring system using a radar, an observer looks at a radar measurement screen and visually observes what seems to be a bird and its approach, and cannot accurately detect the degree of approach of an approaching bird, Therefore, it is difficult to warn at an appropriate timing for approaching birds. Also, with the conventional technology, it has been difficult to collectively monitor a wide area.

  In addition to birds, there is a desire to monitor the approach of grouped insects, and there are fields where it is necessary to monitor the approach of flying objects such as radio-controlled airplanes and small airplanes.

  An object of the present invention is to provide a monitoring device, method and program, and a monitoring system capable of efficiently and efficiently monitoring a moving object such as birds approaching a predetermined area.

  A monitoring apparatus according to the present invention is radar image data obtained by scanning a predetermined angle range, and monitors a moving body according to radar image data of a plurality of scenes sequentially supplied over time. A background generation unit that generates background image data by combining the radar image data of a predetermined number of scenes, and subtracts the background image data from each radar image data supplied after the background image data is generated. Background removing means for removing, binarizing means for binarizing the radar image data from which the background has been removed by the background removing means, and generating binarized echo image data including an echo image of the moving object, and the binarization Center determining means for determining an echo center indicating the center of the echo image of each moving body included in the echo image data, and an echo determined by the center determining means According to the rain / snow removal means for removing the echo center indicating either rain or snow in the mind and the rain / snow removal result by the rain / snow removal means for a predetermined number of radar image data of three or more continuous in time. And a tracking means for tracking the mobile body, and a warning means for outputting a warning when the mobile body enters a predetermined area as a result of tracking by the tracking means.

  The monitoring method according to the present invention is radar image data obtained by scanning a predetermined angle range, and allows a computer to monitor a moving body in accordance with radar image data of a plurality of scenes sequentially supplied over time. A background generation step in which the computer generates background image data by combining the radar image data of a predetermined number of scenes, and the computer generates the background image data from each radar image data supplied after the background image data is generated. A background removal step for removing the background by subtracting the background image data, and the computer binarizes the radar image data from which the background has been removed in the background removal step, and includes a binarized echo image including an echo image of the moving object A binarization step for generating data and the computer are included in the binarized echo image data. A center determining step for determining an echo center indicating the center of the echo image of each mobile object, and the computer removing an echo center indicating either rain or snow among the echo centers determined by the center determining step. A rain and snow removal step, a tracking step in which the computer tracks the mobile body according to a result of rain and snow removal by the rain and snow removal step with respect to a predetermined number of radar image data of three or more continuous in time, The computer includes a warning step of outputting a warning when the moving body enters a predetermined area as a result of tracking by the tracking step.

  The monitoring program according to the present invention is radar image data obtained by scanning a predetermined angle range, and causes a computer to monitor a moving body according to radar image data of a plurality of scenes sequentially supplied over time. A background generation function for causing the computer to synthesize the radar image data of a predetermined number of scenes to generate background image data, and the background image data from each radar image data supplied after the generation of the background image data. A background removal function for removing the background by subtraction, and a binarization function for binarizing the radar image data from which the background has been removed by the background removal function and generating binary echo image data including an echo image of the moving object And determining an echo center indicating the center of the echo image of each mobile object included in the binary echo image data. A center determination function, a rain / snow removal function that removes an echo center indicating either rain or snow from the echo center determined by the center determination function, and a predetermined number of three or more radar images continuous in time A tracking function for tracking the mobile object according to the rain / snow removal result for the data, and a warning function for outputting a warning when the mobile object enters a predetermined area as a result of tracking by the tracking function; It is characterized by realizing.

  The monitoring system according to the present invention repeatedly scans a predetermined angle range, sequentially generates and outputs radar image data of the predetermined angle range, and according to the radar image data of a plurality of scenes from the radar device, A monitoring system comprising a monitoring device that monitors a moving body existing in a predetermined angle range, wherein the monitoring device synthesizes a predetermined number of scenes of the radar image data to generate background image data; and A background removing unit that removes the background by subtracting the background image data from each radar image data supplied after the generation of the background image data, and binarizing the radar image data from which the background is removed by the background removing unit, Binarization means for generating binarized echo image data including an echo image of the moving body, and each of the binarized echo image data included in the binarized echo image data Center determining means for determining an echo center indicating the center of an echo image of a moving object; and rain / snow removing means for removing an echo center indicating either rain or snow among echo centers determined by the center determining means; According to the rain / snow removal result by the rain / snow removal means for a predetermined number of radar image data of three or more continuous in time, the tracking means for tracking the mobile body, and the tracking result by the tracking means Warning means for outputting a warning when entering a predetermined area is provided.

  According to the present invention, it is possible to effectively track a moving body such as a bird from radar image data, and to accurately detect entry into a predetermined area.

1 shows a schematic block diagram of an embodiment of the present invention. The operation | movement flowchart of the main routine of the monitoring program which operate | moves with the computer 30 is shown. 3 shows an operation flowchart of bird recognition / tracking processing in the bird measurement processing of FIG. Fig. 4 shows a detailed flowchart of the preprocessing of Fig. 3. Fig. 4 shows a detailed flowchart of this processing in Fig. 3. The detailed flowchart of the bird tracking by a tracking function is shown. It is explanatory drawing of the procedure which tracks and connects an echo center point. It is explanatory drawing of another procedure which tracks and connects an echo center point. It is explanatory drawing of another procedure which tracks and connects an echo center point.

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

  FIG. 1 shows a schematic block diagram of an embodiment of the present invention. The radar apparatus 10 emits a radio wave pulse beam in the horizontal direction while rotating the antenna unit 12 at a constant speed (for example, one rotation in 2.5 seconds) over 360 degrees in a horizontal plane by the motor 14, and reflects the reflected wave. To detect. In this embodiment, in order to reduce ground surface reflection (ground clutter), the slot antenna that radiates radio waves is slightly upward.

  The horizontal width of the beam radiated from the antenna unit 12 determines the azimuth resolution, and the vertical width determines the altitude resolution. The pulse width determines the distance resolution. Basically, the narrower the beam, the higher the resolution. The azimuth resolution is the minimum azimuth angle at which two targets that are equidistant but have slightly different azimuth angles can be distinguished. The distance resolution is the shortest distance between two targets that can distinguish two targets located at different distances in the same direction. The narrower the pulse width, the higher the distance resolution. The altitude resolution is a value obtained by multiplying the sine value of the vertical beam angle by the distance.

  The rotary encoder 16 detects the rotational angle position of the antenna unit 12 and outputs a pulse signal at a predetermined angular interval. The rotary encoder 16 is also adjusted in advance so as to output a trigger signal when the antenna unit 12 faces a reference direction (for example, north).

  The imaging circuit 18 generates gray scale image data (radar image data) indicating the intensity distribution of the reflected wave received by the antenna unit 12 in accordance with the pulse signal and trigger signal from the rotary encoder 16. The generated radar image data includes circular image data within a fixed distance range centered on the antenna unit 12, or image data based on a video signal obtained by scanning such circular image data in the horizontal direction and the vertical direction. Become. Each pixel of the former image data is specified by a polar coordinate system in the distance direction and the azimuth direction, and the latter image data is specified by an orthogonal coordinate system in the horizontal direction and the vertical direction. In the case of a radar image, the farther away from the radar apparatus 10, the more widely the geographical range represented by one pixel becomes wide since the radiation beam diffuses in the horizontal and vertical directions.

  Each image data generated by the imaging circuit 18 for each rotation of the antenna unit 12 has a fixed azimuth (reference direction) in which uneven rotation of the antenna unit 12 due to wind or the like is eliminated by a trigger signal from the rotary encoder 16. It will be suitable for. The imaging circuit 18 also outputs the generated radar image data to the computer 30 in synchronization with the trigger signal. Radar image data obtained by the antenna unit 12 rotating once from the reference direction is set as image data of one scene.

  The computer 30 functions as a monitoring device that uses the radar image data from the radar device 10 to monitor whether or not a bird to be monitored enters a predetermined management area. Similar to the general configuration of the computer, the computer 30 includes a CPU 32, a keyboard 34 and a mouse 36 for inputting various instructions, a memory 38 serving as a calculation work area of the CPU 32, and a display device 40 for displaying various images and texts. It comprises. The CPU 32 realizes various functions 32a to 32h by a monitoring program described later.

  As a means for storing radar image data observed by the radar apparatus 10 and image data obtained by processing the radar image data for each scene for observation, the computer 30 includes a data memory 42 capable of storing data for four scenes. The data memory 42 stores scene memories 42-1 and 42, each storing image data of one scene obtained by processing radar image data at a series of different times tc, tc-T, tc-2T, and tc-3T. -2, 42-3, 42-4. tc is the current time, and T is the time required for the antenna unit 12 to make one rotation. Actually, the storage area of the memory 38 is used as the data memory 42. However, in order to facilitate understanding of the present embodiment, the data memory 42 (scene memories 42-1, 42-) is provided as a means different from the memory 38. 2, 42-3, 42-4).

  A hard disk device (HDD) 50 stores a monitoring program that operates on the CPU 32, observation data, and results. In addition, the computer 30 includes a data capturing device 52 that captures radar image data observed by the radar device 10. The data capturing device 52 captures radar image data from the imaging circuit 18 in accordance with a trigger signal from the rotary encoder 16 of the radar device 10 and stores it in the memory 38.

  2 to 6 show operation flowcharts of the monitoring program operating on the computer 30. FIG. FIG. 2 shows an operation flowchart of the main routine.

  When the monitoring program is started, the operator inputs a display scale on the screen of the display device 40 (S1), and the management area setting function 32a is operated by the operator through the keyboard 34 and the mouse 36 to display one or more monitoring areas. And a mask area is set (S2, S3). The monitoring area is an area for monitoring the approach of birds, and the mask area is an area that does not require such monitoring, and a plurality of them can be set. The monitoring area and the mask area can be set as a circle, an ellipse, or a rectangle.

  The operator designates alarm conditions (number of wings, rate of increase, etc.), designates a rain / snow removal threshold, etc., and then instructs the execution of the measurement. The monitoring program starts the measurement according to this instruction from the operator. (S4).

  FIG. 3 shows an operation flowchart of the bird recognition / tracking process in the measurement process of step S4 of FIG. The computer 30 takes in the radar image data from the radar apparatus 10 for the number of scenes designated by the operator, combines them using the background generation function 32b to form a background image, and stores the image data of the background image in the memory 38. (S11). For example, the background generation function 32b calculates the minimum luminance, the maximum luminance, and the average luminance for each pixel from the radar image data of all scenes. If the difference between the minimum brightness and the average brightness (α) is greater than or equal to the difference between the average brightness and the maximum brightness (β), the maximum brightness is set as the background brightness. In other cases, the average brightness + α is set as the background brightness. And Even for radar images that are easily affected by temperature, humidity, floating dust, and the like, the background image can be determined with high accuracy by determining the background luminance in units of pixels in this manner. However, radar image data of all scenes may be simply added and averaged. The background image consists only of a still life, and as will be described later, by subtracting the background image data from the radar image data including birds, it is possible to extract each echo of birds, rain, snow, and other moving objects.

  Next, online measurement preprocessing is executed (S12). In this embodiment, the images of the immediately preceding three scenes are used to track the birds. Therefore, as preprocessing, radar image data of these scenes is captured, and processing for tracking is performed, and the scene memory is sequentially recorded. Stored in 42-1, 42-2, and 42-3.

  Details of the preprocessing (S12) will be described with reference to FIG. FIG. 4 shows a detailed flowchart of the preprocessing (S12).

  1 is substituted into the loop variable n (S21). One scene of radar image data is taken from the radar apparatus 10 and temporarily stored in the memory 38 (S22). When the radar image data stored in the memory 38 is incomplete such that frames are dropped (S23), the radar image data of one scene is again taken from the radar apparatus 10 and temporarily stored in the memory 38 (S22). When the data capturing device 52 sequentially captures the radar image data from the radar device 10 and stores it in the memory 38, steps S22 and S23 confirm whether the radar image data has been stored in the memory 38. Change to processing.

  When the radar image data stored in the memory 38 is not dropped (S23), the background removal function 32c subtracts the background image data acquired in step S11 from the radar image data to remove the background image (S24). ). The radar image data from which the background image has been removed becomes echo image data mainly including mobile echoes including bird echoes.

  The binarization function 32d binarizes the echo image data from which the background image has been removed with a predetermined threshold, and removes noise if necessary (S25). In the case of measurement by the radar apparatus 10, as described above, the farther away moving body has a smaller reflected wave intensity from the moving body. Therefore, in general, the binarization threshold is smaller as the distance is longer. It is desirable to do. However, the purpose of the present embodiment is to detect an approaching bird, and a constant binarization threshold is adopted regardless of the distance so as not to miss such a bird.

  In this embodiment, each pixel of the binarized echo image data obtained by the binarization function 32d has a binary value “1” for a moving object including a bird, and a binary value “0” for the rest. It becomes. Reflecting the size of the moving body, the pixel values of a plurality of adjacent pixels are “1”.

  After binarization, the binarization function 32d generates an echo that covers a size sufficiently smaller than the target bird size and an echo that covers a size sufficiently larger than the target bird size. Remove as. Since an object farther away from the radar apparatus 10 is expressed in a larger size, the noise determination threshold is also larger as the distance is longer. When the image data is expressed in polar coordinates, the noise determination threshold expressed by the number of pixels is smaller for both a small noise and a large noise as the distance increases. On the other hand, when expressed in an orthogonal coordinate system, the noise determination threshold expressed by the number of pixels may be constant regardless of the distance, whether for small noise or for large noise.

  The center determination function 32e determines the center of each echo of the binarized echo image data from which noise has been removed by the noise removal function 32d (S26). Specifically, adjacent pixels having a pixel value of “1” express an echo of one moving body, and a pixel that is the center of gravity of the group of pixels is specified as the echo center. For simplicity, the center in the horizontal direction and the vertical direction may be adopted.

  The rain / snow removal function 32f removes rain and snow echoes (S27). In an area where there is rainfall or snowfall, it appears as an echo even after the processing of steps S24 to S26, but since rain or snowfall is expressed in a state where small echoes are densely packed with a spatial density of a certain level or more, In this embodiment, when a certain number or more of echoes having a certain size or more exist within a certain inspection radius, they are determined to be raining or snowing. More specifically, paying attention to each echo center determined by the center determination function 32e in order, whether there is an echo center of another echo of a certain size or more within a predetermined inspection radius from the focused echo center Find out. The inspection radius is determined from the actual rainfall or snow observation situation. When there are a certain number or more of other echo centers within the inspection radius, deletion marks are added to all the echo centers of interest and other echo centers found within the inspection radius. When this process is completed for all the echo centers in the echo image, the echo center to which the deletion mark is added and the echo are deleted from the binary echo image data.

  The binary echo image data obtained after finishing the processes of steps S24 to S27 indicates the location of a moving object having a size to be monitored, and many of the moving objects indicate birds to be monitored.

  The monitoring program operating on the CPU 32 stores the binary echo image data and echo center data obtained after finishing the processes of steps S24 to S27 in the scene memory 42-4 when n = 1 and when n = 2. Is stored in the scene memory 42-3, and in the case of n-3, it is stored in the scene memory 42-2, respectively (S28).

  n is incremented (S30), and step S22 and subsequent steps are repeated until n exceeds 3 (S31). When n exceeds 3 (S31), necessary observation data for 3 scenes is obtained, so the preprocessing (S12) is terminated.

  In the flow shown in FIG. 4, in step S28, the scene memory as the data storage destination is switched in accordance with the loop variable n. At this time, the data processed by the dense data deletion function 32f is stored in the scene memory regardless of n. 42-1 before storing the next radar image data, that is, before proceeding from step S30 to step S22, the data in the scene memories 42-1 to 42-3 are respectively stored in the scene memories 42-2 to 42-. 4 may be transferred.

  Subsequent to the pre-processing (S12), this processing for tracking the bird is executed (S13). FIG. 5 shows a detailed flowchart of this process (S13).

  The radar image data of one scene is taken from the radar apparatus 10 (S41) and temporarily stored in the memory 38 (S42). When the radar image data stored in the memory 38 is incomplete such that frames are dropped (S42), the radar image data of one scene is again taken from the radar device 10 and temporarily stored in the memory 38 (S41). When the radar image data stored in the memory 38 has not lost frames (S42), as in the case of the preprocessing, background removal (S43), binarization and noise removal (S44), center determination (S45), and Rain / snow removal (S46) is executed. Then, the binary echo image data and echo center data processed by the rain / snow removal function 32f are stored in the scene memory 42-1 (S47).

  At this time, the binary echo image data and echo center data indicating the position of the bird at four consecutive times (or scenes) are stored in the scene memories 42-1 to 42-4. The bird monitoring program tracking function 32g tracks the birds using the echo center data of these four scenes stored in the scene memories 42-1 to 42-4 (S48). The tracking function 32g registers, as a bird, an echo center of each scene that is recognized as a bird in the tracking process, and registers a connection relationship between scenes that have been found. This registration data is stored in the memory 38 (or HDD 50). Details of the tracking function 32g will be described later.

  After the tracking is completed, the CPU 32 transfers the binary echo image data and the echo center data stored in the scene memories 42-1 to 42-3 to the scene memories 42-1 to 42-4 by forward transfer (S49). That is, the data stored in the scene memory 42-3 is transferred to the scene memory 42-4, the data stored in the scene memory 42-2 is transferred to the scene memory 42-3, and stored in the scene memory 42-1. Data to be transferred to the scene memory 42-2. At this time, when the data stored in the scene memory 42-4 is used later, the data in the scene memory 42-4 is transferred to the HDD 50 before the data in the scene memory 42-3 is transferred to the scene memory 42-4. save.

  As a result of tracking, when a bird enters the monitoring area (S50), the warning function 32h generates a warning message (S51).

  The CPU 32 updates the display image of the display device 40 with the radar image that is the latest observation result or the image that has been processed in any of steps S43 to S46 (S52). When the warning function 32h generates a warning message, the warning message is superimposed on the screen of the display device 40.

  FIG. 6 shows a detailed flowchart of bird tracking (S48) by the tracking function 32g. In order to facilitate understanding, four consecutive scenes are expressed as t0, t1, t2, and t3 in the order of time, and echoes included in the binary echo image data after the rain / snow removal processing in the scenes t0 to t3. The echo center is expressed as p0 to p3. That is, the echo centers of the echoes included in the binary echo image data stored in the scene memories 42-1 to 42-4 are p3, p2, p1, and p0, respectively, and the echo center p0 is the scene t0 at time tc-3T. The echo center p1 is included in the scene t1 at time tc-2T, the echo center p2 is included in the scene t2 at time tc-3T, and the echo center p3 is included in the scene t3 at time tc.

  In FIG. 6, first, an arbitrary echo center p0 of the scene t0 is designated (S61), and it is checked whether or not the point has been registered as a bird (S62).

  When the echo center p0 is not registered as a bird (S62), the echo center p0 and the unregistered echo centers p1 to p3 of the scenes t1 to t3 are searched for a continuous combination, and the echo centers p1 to p3 are determined. Register as a bird, and register the connection (S63).

  The search method in step S63 will be described with reference to FIG. With respect to the echo center p0, the area between the inner radius R0a and the outer radius R0b is set as the initial inspection range, and the echo center of the scene t2 that falls within the initial inspection range is searched for as a candidate for p1. A movement error range of radius Re is set around a point pa obtained by extending a straight line connecting the echo center p0 to the p1 candidate by the same distance as the distance between the p0 and p1 candidates, and the scene t2 within the movement error range is set within the movement error range. The echo center is searched for as a p2 candidate. A movement error range of radius Re is set around a point pb obtained by extending a straight line connecting the p1 candidate to the p2 candidate by the same distance as the distance between the p1 candidate and the p2 candidate. The echo center is searched for as a p3 candidate. Each echo center p0, p1 candidate, p2 candidate, p3 candidate with the smallest error in the combination of {p0, p1 candidate, p2 candidate, p3 candidate} is registered as a bird, and its connection relationship is registered. At that time, the average value of the distance between pa and the p2 candidate and the distance between pb and the p3 candidate is evaluated as an error. The operator appropriately selects the radii R0a, R0b, and Re in the observation target and the observation area, and sets them in advance. If no p1 candidate, p2 candidate, or p3 candidate is found, the echo center p0 is not registered as a bird, and the search ends.

  After step S63, it is checked whether or not there is a next unprocessed p0 in the scene t0 (S68). If there is (S68), the next echo center p0 is designated (S69), and the process returns to step S62. If there is no unprocessed echo center p0 (S68), the tracking is terminated.

  When the echo center p0 is registered as a bird (S62), it is checked whether the echo center p1 of the scene t1 connected to the echo center p0 is registered (S64). If the echo center p1 connected to the scene t1 is not registered (S64), the echo center p0 registered as a bird has disappeared in the scene t1. In this case (S64), the processing shifts to the next echo center p0 (S68).

  When the echo center p1 of the scene t1 connected to the echo center p0 has been registered (S64), it is checked whether the echo center p2 of the scene t2 connected to the point p1 has been registered as a bird (S65). When the echo center p2 has been registered (S65), the echo center p3 connectable to these echo centers p0, p1, and p2 is searched from the scene t3 (S66). FIG. 8 is an explanatory diagram of the search process in step S66. A movement error range of radius Re is set around a virtual point p obtained by extending a straight line connecting p1 to p2 by the same distance as the distance between p1 and p2, and the echo center of scene t3 is set within the movement error range. Search and register as the bird tracking the echo center p3 closest to the virtual point p. If the echo center of the scene t3 cannot be found within the movement error range, it is treated as having disappeared in the scene t3.

  If the echo center p2 registered as a bird connected to the echo center p0 does not exist in the scene t2 (S65), it is assumed that it has disappeared in the scene t2, so a virtual point p is set on the scene t3, and the virtual point p The echo center p3 of the scene t3 is searched within a certain distance range from (S67). FIG. 9 is an explanatory diagram of the search process in step S67. A moving error range having a radius 2Re is set around a virtual point p obtained by extending a straight line connecting p0 to p1 by a distance twice the distance between p0 and p1, and the echo of the scene t3 is set within the moving error range. The center is searched, and the point p3 closest to the point p is registered as a bird being tracked. The midpoint of the straight line connecting the echo center p3 and the echo center p1 is registered as the echo center p2 in the scene t2. If the echo center of the scene t3 cannot be found within the movement error range, it is treated as having disappeared in the scene t2.

  After steps S66 and S67, it is checked whether or not there is a next unprocessed echo center p0 in the scene t0 (S68). If there is (S68), the next echo center p0 is designated (S69). Return to S62. If there is no unprocessed echo center p0 (S68), the tracking is terminated.

  When this process (S13) ends, as post-processing, the data stored in the scene memories 42-1 to 42-4 is saved in the HDD 50, and the scene memories 42-1 to 42-4 are cleared (S14).

  In the above embodiment, the radar apparatus 10 autonomously rotates the antenna unit 12 to measure (photograph) the surroundings. However, since the processing capability of the computer 30 is sufficiently faster than the rotation speed of the antenna unit 12, the radar device 10 Even if the apparatus 10 autonomously rotates the antenna unit 12, the computer 30 can sufficiently follow. Conversely, of course, the radar apparatus 10 may measure (photograph) the surroundings by rotating the antenna unit 12 once in accordance with an instruction from the computer 30.

  The radar apparatus 10 may be of a type that repeatedly scans a certain monitoring angle range. When the flight direction of the monitoring target is limited, it is operated as such.

  It is obvious that the computer 30 may be a single stand-alone computer or a system configuration in which a plurality of computers cooperate with each other.

  Although the invention has been described with reference to specific illustrative embodiments, various modifications and alterations may be made to the above-described embodiments without departing from the scope of the invention as defined in the claims. This is obvious to an engineer in the field to which the present invention belongs, and such changes and modifications are also included in the technical scope of the present invention.

10: Radar device 12: Antenna unit 14: Motor 16: Imaging circuit 18: Rotary encoder 30: Computer 32: CPU
32a: area setting function 32b: background removal function 32c: binarization function 32d: noise removal function 32e: center determination function 32f: rain / snow removal function 32g: tracking function 32h: warning function 34: keyboard 36: mouse 38: memory 40 : Display device 42: Data memory 42-1, 42-2, 42-3, 42-4: Scene memory 50: Hard disk device (HDD)

Claims (12)

Radar image data obtained by scanning a predetermined angle range, and a monitoring device that monitors a moving body according to radar image data of a plurality of scenes sequentially supplied over time,
Background generation means for generating a background image data by combining the radar image data of a predetermined number of scenes;
Background removal means for removing the background by subtracting the background image data from each radar image data supplied after generation of the background image data;
Binarizing means for binarizing radar image data from which the background has been removed by the background removing means, and generating binarized echo image data including an echo image of the moving body;
Center determining means for determining an echo center indicating a center of an echo image of each moving body included in the binarized echo image data;
Rain / snow removing means for removing an echo center indicating either rain or snow among echo centers determined by the center determining means;
A tracking means for tracking the moving body according to a result of removing rain / snow by the rain / snow removing means for a predetermined number of radar image data of 3 or more continuous in time;
A monitoring apparatus comprising: warning means for outputting a warning when the moving body enters a predetermined area as a result of tracking by the tracking means.   The monitoring apparatus according to claim 1, wherein the binarization unit removes, as noise, an echo image that does not correspond to the echo image of the moving body in the binarized image data.   The rain / snow removing means determines that the echo centers constituting a density of a certain density or more from the echo centers determined by the center determining means are determined as echo centers indicating either rain or snow, and removes them. The monitoring device according to claim 1 or 2, characterized in that Radar image data obtained by scanning a predetermined angle range, and a monitoring method for causing a computer to monitor a moving body according to radar image data of a plurality of scenes sequentially supplied over time,
A background generation step in which the computer generates background image data by combining the radar image data of a predetermined number of scenes;
A background removal step in which the computer removes the background by subtracting the background image data from each radar image data supplied after generation of the background image data;
A binarization step in which the computer binarizes the radar image data from which the background has been removed in the background removal step, and generates binarized echo image data including an echo image of the moving body;
A center determining step in which the computer determines an echo center indicating a center of an echo image of each mobile body included in the binarized echo image data;
A rain / snow removal step in which the computer removes an echo center indicating either rain or snow among the echo centers determined by the center determination step;
A tracking step in which the computer tracks the mobile body according to the rain / snow removal result of the rain / snow removal step with respect to a predetermined number of three or more radar image data continuous in time;
A monitoring method, comprising: a warning step for outputting a warning when the computer enters a predetermined area as a result of tracking by the tracking step.   5. The monitoring method according to claim 4, wherein the binarizing step includes a step in which the computer removes, as noise, an echo image that does not correspond to an echo image of the moving body in the binarized image data. .   In the rain / snow removal step, the computer determines that the echo centers constituting a density of a certain density or more among echo centers determined in the center determination step are echo centers indicating either rain or snow. The monitoring method according to claim 4, wherein the monitoring method is a removing step. Radar image data obtained by scanning a predetermined angle range, which is a monitoring program that causes a computer to monitor a moving body according to radar image data of a plurality of scenes that are sequentially supplied over time.
A background generation function for generating background image data by combining the radar image data of a predetermined number of scenes;
A background removal function for removing the background by subtracting the background image data from each radar image data supplied after the generation of the background image data;
A binarization function that binarizes radar image data from which the background has been removed by the background removal function and generates binary echo image data including an echo image of a moving object;
A center determining function for determining an echo center indicating a center of an echo image of each moving body included in the binarized echo image data;
A rain / snow removal function for removing an echo center indicating either rain or snow among echo centers determined by the center determination function;
A tracking function for tracking the moving object according to the rain / snow removal result of the rain / snow removal function for a predetermined number of radar image data of 3 or more continuous in time;
A monitoring program for realizing a warning function for outputting a warning when the moving body enters a predetermined area as a result of tracking by the tracking function.   The monitoring program according to claim 7, wherein the binarization function includes a function of causing the computer to remove, as noise, an echo image that does not correspond to an echo image of the moving body in the binarized image data. .   The rain / snow removal function determines, in the computer, that the echo centers that constitute a density higher than a certain value among echo centers determined by the center determination function are echo centers indicating either rain or snow. The monitoring program according to claim 7 or 8, wherein the monitoring program has a function of removing. A radar device that repeatedly scans a predetermined angle range, and sequentially generates and outputs radar image data of the predetermined angle range;
In accordance with the radar image data of a plurality of scenes from the radar device, a monitoring system comprising a monitoring device that monitors a moving body existing in the predetermined angle range,
The monitoring device
Background generation means for generating a background image data by combining the radar image data of a predetermined number of scenes;
Background removal means for removing the background by subtracting the background image data from each radar image data supplied after generation of the background image data;
Binarizing means for binarizing radar image data from which the background has been removed by the background removing means, and generating binarized echo image data including an echo image of the moving body;
Center determining means for determining an echo center indicating a center of an echo image of each moving body included in the binarized echo image data;
Rain / snow removing means for removing an echo center indicating either rain or snow among echo centers determined by the center determining means;
A tracking means for tracking the moving body according to a result of removing rain / snow by the rain / snow removing means for a predetermined number of radar image data of 3 or more continuous in time;
A monitoring system comprising: warning means for outputting a warning when the moving body enters a predetermined area as a result of tracking by the tracking means.   The monitoring system according to claim 10, wherein the binarizing unit removes, as noise, an echo image that does not correspond to an echo image of the moving body in the binarized image data.   The rain / snow removing means determines that the echo centers constituting a density of a certain density or more from the echo centers determined by the center determining means are determined as echo centers indicating either rain or snow, and removes them. The monitoring system according to claim 10 or 11, characterized in that

Images