JP4594197B2 - Automatic tracking device - Google Patents

Description

  The present invention relates to an automatic tracking device that can track a tracking target in an image by controlling a shooting direction of an imaging device.

  Conventionally, an automatic tracking device for a monitoring system or the like has been proposed. The automatic tracking device is configured to detect a tracking target such as a person from an image by image processing, and to rotate the imaging device by controlling a pan / tilt mechanism so as to face the detected tracking target.

A conventional automatic tracking device designates the position of a tracking target in one frame by an input operation from the outside, registers an image of an area centered on the designated position as a template, and searches for an area correlated with the template A range is set, and a region having the highest correlation with the template is searched as a tracking target from the search range of subsequent frames. This search process is performed for each successive frame. Based on the search result, the pan / tilt / zoom function of the imaging apparatus is controlled to track the tracking target. A conventional automatic tracking device is disclosed in Patent Document 1, for example.
Japanese Patent Laid-Open No. 11-187387 (page 4-5, FIGS. 1 and 2)

  However, in the conventional automatic tracking device, tracking is forcibly attempted even in a situation where tracking by image processing is difficult, and therefore an appropriate tracking image cannot be obtained, and further, the tracking target is out of the angle of view. There was a problem that sometimes.

  For example, an imaging apparatus capable of setting the accumulation mode shown in FIG. 7 is conventionally known. The accumulation mode is a mode for obtaining a bright image in a dark environment such as at night. In the accumulation mode, a plurality of frames of images are accumulated, and one frame of images is obtained. In the example of FIG. 7, one frame is obtained by accumulating three frames. This one frame is displayed three times. Therefore, the imaging apparatus outputs a video signal whose image changes every three frames. The number of stored images is referred to as storage magnification, and in the example of FIG. 7, the storage magnification is 3 times. In the accumulation mode, the moving subject image is slightly blurred and the movement becomes rough. Therefore, feature extraction by image processing becomes difficult, and there is a high possibility of erroneous tracking. As the accumulation magnification increases, the possibility of erroneous tracking increases.

  FIG. 8 shows another example in which the tracking process becomes difficult. In FIG. 8, a pedestrian crossing is photographed and the tracking target is a person. In this example, there are many pedestrians and the pedestrian crossing is congested. In such a crowded situation, there are a large number of similar regions in the search range of the tracking target, and therefore there is a high possibility that erroneous tracking will occur. The more crowded, the greater the chance of false tracking.

  The conventional automatic tracking device has a problem that, as in the above example, tracking is forcibly attempted even in a situation where tracking by video processing is difficult, and an appropriate tracking video cannot be obtained. Furthermore, there is a possibility that the real tracking target may be out of the angle of view as a result of performing an excessive tracking process.

  The present invention has been made to solve the conventional problems, and an object of the present invention is to provide an automatic tracking device capable of improving convenience by performing processing adapted to a situation where tracking is difficult.

  The automatic tracking device of the present invention processes the video signal of the imaging device and controls the shooting direction of the imaging device so as to track the specified tracking target. The automatic tracking device includes an input unit that inputs a video signal of the imaging device, a tracking target designation receiving unit that receives designation of a position of a tracking target in the video signal, and the tracking target reflected at a designated position as the video. A search processing unit for searching within, a search difficulty level parameter acquiring unit for acquiring a search difficulty level parameter indicating a difficulty level of search processing by the search processing unit, and a tracking mode and a non-tracking mode according to the search difficulty level parameter In the tracking mode, the movement position of the imaging device is determined so as to track the tracking target in accordance with the processing result of the search processing unit, and in the non-tracking mode, the tracking target designation reception is performed. Movement for determining the movement position of the imaging device so that the designated position received by the unit is fixedly projected at a predetermined non-tracking mode position in the video A location determination unit, and a movement control unit for controlling the moving mechanism of the imaging device in accordance with the movement position determined by the movement position determination unit.

  With this configuration, the automatic tracking device acquires a search difficulty level parameter indicating the difficulty level of the search processing for the tracking target, and switches between the tracking mode and the non-tracking mode according to the search difficulty level parameter. In the non-tracking mode, the moving mechanism of the imaging device is controlled so that the designated position is fixedly displayed at a predetermined non-tracking mode position (for example, the center of the screen) in the video. Therefore, excessive tracking in situations where tracking is difficult is avoided. In the non-tracking mode, although tracking is not performed, it is possible to avoid the tracking target from being out of the angle of view, and to obtain the tracking target video. In this way, even if a situation where tracking is difficult occurs, a non-tracking mode can be set to obtain an appropriate video, processing suitable for the situation where tracking is difficult can be performed, and convenience can be improved. .

  In the automatic tracking device of the present invention, the search difficulty level parameter acquisition unit acquires storage magnification information of an accumulation mode in the imaging device as the search difficulty level parameter, and the operation mode switching unit has a predetermined accumulation magnification. The tracking mode is switched to the non-tracking mode when the threshold magnification is equal to or higher. With this configuration, the accumulation magnification is used as the search difficulty level parameter, a tracking difficult situation can be detected based on the accumulation magnification, and an appropriate image can be obtained even in a tracking difficult situation by setting the non-tracking mode.

  Further, in the automatic tracking device of the present invention, the search difficulty level parameter acquisition unit processes the video signal as the search difficulty level parameter to detect a congestion level of the shooting location, and the operation mode switching unit includes the operation mode switching unit, The tracking mode is switched to the non-tracking mode when the congestion level is equal to or higher than a predetermined threshold. With this configuration, it is possible to use a congestion degree as a search difficulty level parameter, detect a difficult tracking situation based on the congestion degree, set a non-tracking mode, and obtain an appropriate video even in a difficult tracking situation.

  The automatic tracking device of the present invention further includes a zoom control unit that controls a zoom function of the imaging device, and the zoom control unit is configured such that in the non-tracking mode, the zoom magnification becomes a predetermined non-tracking mode magnification. The zoom function is controlled. With this configuration, in the non-tracking mode, the angle of view can be widened by controlling the zoom function so as to capture a wide range including the specified position when the tracking target is specified. As a result, even when tracking is not performed, the tracking target is positioned in the shooting range for a longer time, and the tracking target can be projected for a longer time.

  Another aspect of the present invention is a program for processing a video signal of an imaging apparatus and causing a computer to execute processing for controlling a shooting direction of the imaging apparatus so as to track a specified tracking target. The program includes a tracking target designation receiving step for accepting designation of a tracking target position in the video signal, a search processing step for searching the video for the tracking target reflected in the designated position, and a difficulty level of the search processing. A search difficulty level parameter acquisition step for acquiring a search difficulty level parameter to be represented; an operation mode switching step for switching between a tracking mode and a non-tracking mode according to the search difficulty level parameter; and in the tracking mode, according to a processing result by the search process The movement position of the imaging device is determined so as to track the tracking target. In the non-tracking mode, the specified position received by the tracking target specification receiving unit is fixed to a predetermined non-tracking mode position in the video. A moving position determining step for determining the moving position of the imaging device so as to be reflected in the image, and according to the determined moving position. And a movement controlling step of controlling the moving mechanism of the imaging device causes the computer to execute Te. The above-described advantages can be obtained by this configuration as well.

  The present invention switches between the tracking mode and the non-tracking mode according to the tracking difficulty parameter as described above, and in the non-tracking mode, the specified position is fixedly displayed at a predetermined non-tracking mode position in the video. By controlling the moving mechanism, it is possible to provide an automatic tracking device that has an effect of improving the convenience by performing processing adapted to a situation where tracking is difficult.

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

  An automatic tracking device according to a first embodiment of the present invention is shown in FIG. The automatic tracking device 1 according to the present embodiment processes a video signal generated by the imaging device 3 with a pan / tilt / zoom (hereinafter referred to as PTZ) function and controls the direction of the camera so as to track a specified tracking target. It is a device to do.

  In FIG. 1, an image pickup apparatus 3 with a PTZ function (hereinafter simply referred to as an image pickup apparatus 3) is provided on a turntable having a pan / tilt mechanism, and is rotatable in the pan direction and the tilt direction. Furthermore, the imaging device 3 has a zoom function. Then, the rotation and zoom of the imaging device 3 are controlled by the automatic tracking device 1.

  The automatic tracking device 1 is composed of a computer having an image processing function. The automatic tracking device 1 includes an input unit 5 that inputs a video signal generated by the imaging device 3, a first image memory 7 and a second image memory 9 that store the video signal, and the position of the tracking target in the video signal. A tracking target designation receiving unit 11 that receives designation and a search processing unit 13 that searches for a tracking target in sequentially input frame images are provided.

  The input unit 5 has a function of inputting a video signal of the imaging device 3, performing A / D conversion processing on the video signal, and further separating the digital video signal into a luminance signal and a color difference signal. The first image memory 7 and the second image memory 9 store video signals of the current and several frames before.

  The tracking target designation receiving unit 11 receives the designation position of the tracking target (tracking object) input from the outside. In the process of specifying (instructing) the tracking target, the image in the first image memory 7 is displayed on the display. Then, an input device such as a pointing device is operated to specify the position of a tracking target such as a person. The coordinate data of this designated position is input and accepted by the tracking target designation receiving unit 11.

  The search processing unit 13 includes a feature amount extraction region determination unit 15, a search range determination unit 17, a template registration unit 19, and a correlation value calculation unit 21. The feature amount extraction region determination unit 15 determines a region having a predetermined size around the specified position coordinates of the tracking target received by the tracking target specification reception unit 11 as a feature amount extraction region. The search range determination unit 17 determines a search range for searching for a region having the same feature amount as the feature amount extraction region. The template registration unit 19 registers the feature amount extraction area as a template. The correlation value calculation unit 21 calculates a correlation value between the template and the surrounding area in order to extract an area that matches the template.

  Here, a candidate area is set within the search range, and a correlation value between the candidate area and the template is obtained. The candidate areas are shifted little by little within the search range, and the correlation value of each candidate area is calculated. A candidate area having the largest correlation value is obtained. If the correlation value (maximum value) of the candidate area is greater than or equal to a predetermined value, the candidate area is specified as a tracking target.

  The automatic tracking device 1 further includes a storage mode information acquisition unit 23 that acquires storage magnification information of the storage mode from the imaging device 3 as a camera state information signal, and the operation mode of the automatic tracking device 1 according to the storage magnification information. And an operation mode switching unit 25 for switching between the non-tracking mode and the non-tracking mode.

  The accumulation mode is a function provided in the imaging apparatus 3 and is a mode for obtaining a bright image in a dark environment such as nighttime. As shown in the example of FIG. 7, in the accumulation mode, a plurality of frames of images are accumulated and one frame of images is obtained. In the example of FIG. 7, one frame is obtained by accumulating three frames. This one frame is displayed three times. Therefore, the imaging apparatus outputs a video signal whose image changes every three frames. The accumulated number is called the accumulation magnification. In the example of FIG. 7, the accumulation magnification is 3 times.

  In the accumulation mode, signals of a plurality of frames are accumulated, so that the brightness is increased. On the other hand, in the accumulation mode, the image of the moving subject is blurred, and the movement becomes stepwise and rough. Therefore, feature extraction by image processing becomes difficult, search by the search processing unit 13 becomes difficult, and there is a high possibility of erroneous tracking. As the accumulation magnification increases, the difficulty of searching increases. Therefore, in the present embodiment, the accumulation magnification is used as a search difficulty level parameter that represents the difficulty level of search processing by the search processing unit 13.

  The accumulation mode is automatically set by the imaging device 3. Typically, the brightness level of the image is determined, and the accumulation mode is set when the brightness level falls below a predetermined threshold level. Then, the accumulation magnification is changed according to the luminance level. The accumulation magnification increases as the luminance level decreases.

  The information of the accumulation magnification can be output from the imaging device 3 through one of the terminals for RS232C as one of the camera state information signals. Therefore, the accumulation mode information acquisition unit 23 inputs a camera state information signal from the imaging device 3 through state inquiry processing to the imaging device 3, and thereby acquires an accumulation magnification. In this way, the accumulation mode information acquisition unit 23 functions as a search difficulty level parameter acquisition unit by acquiring the accumulation magnification information.

  The accumulation mode information acquisition unit 23 provides the accumulation magnification information to the operation mode switching unit 25. The operation mode switching unit 25 switches between the tracking mode and the non-tracking mode according to the accumulation magnification that is the search difficulty level parameter. The tracking mode is a mode in which tracking is performed by rotating the imaging device 3, and the non-tracking mode is a mode in which the tracking operation is limited. The operation mode switching unit 25 switches the tracking mode to the non-tracking mode and sets the non-tracking mode when the accumulation magnification is equal to or greater than a predetermined threshold magnification. The operation mode switching unit 25 sets the tracking mode if the accumulation magnification is less than the threshold magnification.

  The automatic tracking device 1 further includes a movement position determination unit 27, a rotary table zoom control unit 29, a display processing unit 31, and an output unit 33.

  The movement position determination unit 27 determines the movement position of the imaging device 3 in the pan direction and the tilt direction. The movement position varies depending on the operation mode. When the tracking mode is set, the movement position is determined so as to track the tracking target according to the processing result by the search processing unit 13. At this time, even if the tracking target moves, the camera movement position is determined so as to continue to appear at a predetermined position on the screen (for example, the center of the screen).

  On the other hand, when the non-tracking mode is set, the designated position coordinates received for designating the tracking target by the tracking target designation receiving unit 11 are fixedly displayed at a predetermined non-tracking mode position in the video. The moving position is determined. In the present embodiment, the predetermined non-tracking mode position is the center of the screen (view angle center). That is, the movement position is determined so that the designated position coordinates when the tracking target is designated are projected to the center of the view angle regardless of the movement of the tracking target after the tracking target is designated.

  The movement position determination unit 27 also determines the control amount of the zoom function. In the tracking mode, the zoom magnification is determined so that a predetermined range including the tracking target can be photographed. This zoom magnification may be a predetermined value. The zoom magnification may be changed according to the size of the tracking target. In the non-tracking mode, the zoom magnification is determined to be a predetermined non-tracking mode magnification. The non-tracking mode magnification is set smaller than the zoom magnification in the tracking mode, so that a wider range of images can be obtained in the non-tracking mode than in the tracking mode.

  For example, it is assumed that the automatic tracking device 1 is provided in the monitoring system and the monitoring place is a large intersection. At this time, the zoom magnification in the tracking mode is set large in order to close up the person to be tracked. The zoom magnification in the non-tracking mode is set to be small so that the entire intersection is reflected.

  The turntable zoom control unit 29 functions as a movement control unit and a zoom control unit of the imaging device 3. The turntable zoom control unit 29 controls the PTZ function by sending a turntable control signal and a zoom control signal to the imaging device 3, that is, controls the pan / tilt mechanism of the turntable and controls the zoom function of the camera. The PTZ function of the imaging device 3 is controlled according to the control amount determined by the movement position determination unit 27. Therefore, in the tracking mode, the rotary base zoom control unit 29 rotates the imaging device 3 so as to track the tracking target in accordance with the determination of the movement position determination unit 27, and further performs shooting at the specified zoom magnification. Further, in the non-tracking mode, the turntable zoom control unit 29 rotates the imaging device 3 so that the specified position received by the tracking target specification receiving unit 11 is at the center of the screen according to the determination of the movement position determining unit 27. Then, the imaging device 3 is fixed and photographing is performed at a predetermined non-tracking mode magnification.

  The display processing unit 31 performs a process of adding a frame and a mark to the tracking target in the video. The frame may be the same size as the template. The mark is placed at the center of the frame. The frame and mark addition processing is performed when the tracking mode is set.

  The output unit 33 outputs the video signal to the display unit 35. The display unit 35 is a display. The output unit 33 converts the video signal composed of the luminance signal and the color difference signal into a form suitable for the display unit 35. At this time, D / A conversion is performed and an analog video signal is output.

  FIG. 2 shows the operation of the operation mode switching unit 25. The process of FIG. 2 is repeatedly performed. The operation mode switching unit 25 acquires the accumulation magnification from the accumulation mode information acquisition unit 23 (S11), and determines whether or not the accumulation magnification is equal to or greater than the threshold magnification (S13).

  If the accumulation magnification is greater than or equal to the threshold magnification, the operation mode switching unit 25 determines whether or not the current mode is the tracking mode (trackable mode) (S15). If step S15 is Yes, the operation mode switching unit 25 sets the current operation mode to the non-tracking mode (tracking impossible mode) (S17). That is, the operation mode is switched from the tracking mode to the non-tracking mode. Then, the operation mode switching unit 25 sets a timer (S19). The timer is set to prevent frequent switching (hunting) of the mode. Until the timer value becomes 0, the non-tracking mode is maintained even if the accumulation magnification decreases.

  If the current mode is not the tracking enable mode in step S15, it is determined whether or not the timer value is 0 (S21). If the timer value is 0, a timer is set (S23). Here, even if the timer value becomes 0, the accumulation magnification is still more than the threshold magnification, so the timer is set again.

  If the accumulation magnification is not greater than or equal to the threshold magnification in step S13, the operation mode switching unit 25 determines whether or not the tracking mode is currently set (S23). If the tracking mode is not set, the operation mode switching unit 25 determines whether the timer value is 0 (S25). If the timer value is 0, the operation mode switching unit 25 sets the tracking mode (S27).

  As described above, the operation mode switching unit 25 compares the accumulation magnification with the threshold magnification, sets the non-tracking mode if the accumulation magnification is equal to or higher than the threshold magnification, and sets the non-tracking magnification if the accumulation magnification is less than the threshold magnification. Set the tracking mode. Mode switching is performed using a timer so that hunting does not occur.

  Next, the overall operation of the automatic tracking device 1 of the present embodiment will be described. First, the operation when the accumulation magnification is smaller than the threshold magnification and the tracking mode is set will be described. The video signal of the imaging device 3 is input by the input unit 5 and stored in the first image memory 7 and the second image memory 9. The first image memory 7 and the second image memory 9 store video signals of the current and several frames before. The video of the first image memory 7 is displayed, the position of the tracking target (person or the like) is specified by the user, and the coordinate data of the specified position is received by the tracking target designation receiving unit 11.

  Then, the feature amount extraction region determination unit 15 determines the region around the specified position coordinates of the tracking target as the feature amount extraction region, the search range determination unit 17 determines the search range of the tracking target, and the template registration unit 19 performs the feature. Register the quantity extraction area as a template.

  The correlation value calculation unit 21 performs a process of calculating a correlation value between the image of the second memory 9 and the template within the search range, and detects a region having the highest correlation value within the search range (matching region). Here, the regions are shifted little by little, the calculation of the correlation value is repeated, and the matching region having the highest correlation value is specified. In this way, the image in the second image memory 9 is compared with the template, and the matching area having the maximum correlation value is specified as the tracking target. A plurality of consecutive frames are provided from the second image memory 9. Then, a tracking target is detected from each of the plurality of frames by the same process. Thus, even if the tracking target moves in the video, the tracking target is suitably captured by the search processing unit 13.

  The movement position determination unit 27 determines the movement position of the imaging device 3 so as to track the tracking target according to the processing result by the search processing unit 13. Further, the movement position determination unit 27 determines the zoom magnification so that a predetermined range including the tracking target can be photographed. Then, the turntable zoom control unit 29 controls the PTZ function of the imaging device 3 according to the control amount determined by the movement position determination unit 27. Therefore, the imaging device 3 rotates so as to track the tracking target, and further performs imaging at the designated zoom magnification.

  The display processing unit 31 adds a frame and a mark to the tracking target of the video signal. The video signal to which the frame and the mark are added is output to the display unit 35 by the output unit 33 and displayed.

  In this way, the imaging device 3 rotates so as to track the tracking target in accordance with the control of the automatic tracking device 1. The display unit 35 can continuously display the tracking target together with the frame and the mark.

  Next, the operation when the non-tracking mode is set will be described. As described above, when the accumulation magnification is equal to or higher than the threshold magnification, the non-tracking mode is set by the operation mode switching unit 25. When the non-tracking mode is set, the moving position determination unit 27 moves so that the designated position coordinates received by the tracking target designation receiving unit 11 are fixedly displayed at a predetermined non-tracking mode position in the video. Determine the position. In the present embodiment, the predetermined non-tracking mode position is the center of the angle of view. Regardless of the movement of the tracking target after specifying the tracking target, the movement position is determined so that the position coordinates when the tracking target is specified are reflected in the center of the angle of view. The zoom magnification is determined to be a predetermined non-tracking mode magnification.

  The turntable zoom control unit 29 controls the imaging device 3 according to the control amount determined by the movement position determination unit 27. Therefore, the imaging device 3 rotates so that the designated position is at the center of the screen, and performs imaging at a predetermined non-tracking mode magnification. The video signal is output from the output unit 33 and displayed on the display unit 35.

  In this way, in the non-tracking mode, the direction of the imaging device 3 is fixed so that the designated position is located at the center of the screen. Then, the fixed image of the imaging device 3 is displayed on the display unit 35. Therefore, in a situation where tracking is difficult, an excessive tracking operation is suppressed. However, since the imaging device 3 faces the original designated position, the tracking target can be shot. In addition, since the non-tracking zoom magnification is set to be small and a wide range is photographed, the tracking target can be photographed over as long a time as possible.

  For example, in the above example, the automatic tracking device 1 is provided in the monitoring system, and the monitoring location is a large intersection. At this time, in the tracking mode, the zoom magnification is set large, and the person to be tracked is closed up. On the other hand, in the non-tracking mode, the zoom magnification is set small, the entire intersection is projected, and the person to be tracked can be projected until the intersection is crossed.

  In the non-tracking mode, the frame and mark addition processing by the display processing unit 29 is suppressed.

  Moreover, the search process of the search process part 13 may be suppressed and may be performed continuously. In the latter case, the tracking rotation control is suppressed.

  As a modification of the present embodiment, the operation mode may be switched according to whether the accumulation mode is on or off. When the accumulation mode is turned on, the non-tracking mode is set. In this modification, the mode is switched when the accumulation magnification becomes larger than 1. This modification also corresponds to mode switching when the accumulation magnification becomes equal to or greater than a predetermined threshold value, and also corresponds to mode switching according to the search difficulty level parameter.

  The automatic tracking device 1 according to the first embodiment of the present invention has been described above. In the present embodiment, the automatic tracking device 1 acquires a search difficulty level parameter indicating the difficulty level of the tracking target search process, and switches between the tracking mode and the non-tracking mode according to the search difficulty level parameter. In the non-tracking mode, the moving mechanism of the imaging device is controlled so that the designated position is fixedly displayed at a predetermined non-tracking mode position in the video. Therefore, excessive tracking in situations where tracking is difficult is avoided. In the non-tracking mode, although tracking is not performed, it is possible to avoid the tracking target from being out of the angle of view, and to obtain the tracking target video. In this way, even if a situation where tracking is difficult occurs, a non-tracking mode can be set to obtain an appropriate video, processing suitable for the situation where tracking is difficult can be performed, and convenience can be improved. .

  In addition, according to the present embodiment, the accumulation magnification is used as a search difficulty parameter, a tracking difficult situation can be detected based on the accumulation magnification, a non-tracking mode is set, and an appropriate video can be obtained even in a difficult tracking situation. Can be obtained.

  Further, according to the present embodiment, in the non-tracking mode, the zoom function is controlled so that the zoom magnification becomes the predetermined non-tracking mode magnification. With this configuration, in the non-tracking mode, the angle of view can be widened by controlling the zoom function so as to capture a wide range including the specified position when the tracking target is specified. As a result, even when tracking is not performed, the tracking target is positioned in the shooting range for a longer time, and the tracking target can be projected for a longer time.

  Next, FIG. 3 shows an automatic tracking device according to the second embodiment. In the first embodiment, the accumulation magnification in the accumulation mode is used as a search difficulty level parameter. In the second embodiment, the congestion level of the shooting location is detected by video processing, and the congestion level is used as a search difficulty level parameter. Hereinafter, the second embodiment will be described focusing on the differences from the first embodiment.

  As shown in FIG. 3, in the second embodiment, the automatic tracking device 101 includes a congestion degree detection unit 41. The congestion level detection unit 41 processes the video signal and detects the congestion level of the shooting location. The degree of congestion is supplied to the operation mode switching unit 25.

  FIG. 4 shows a configuration of the congestion degree detection unit 41. The congestion degree detection unit 41 includes a difference calculation unit 43, a binarization unit 45, and a motion region ratio calculation unit 47.

  The congestion degree detection unit 41 acquires two consecutive frames of images from the first image memory 7 and the second image memory 9. The difference calculation unit 43 calculates the difference between the two frames and generates a difference image. The binarization unit 45 binarizes the difference image. The value of each pixel (difference between the same pixels in two frames) is compared with a threshold value. If the difference pixel value is equal to or greater than the threshold value, the pixel value is converted to 1, otherwise the pixel value is converted to 0. A portion having a pixel value of 1 represents a portion where the subject has moved. The motion area ratio calculation unit 47 calculates the ratio of the area where the pixel value is 1. The value of this ratio is the degree of congestion.

  In the above example, the degree of congestion is the proportion of moving parts in the image. If the percentage of moving parts is large, there are many moving objects, and it is considered that congestion has occurred. Therefore, the degree of congestion obtained as described above can appropriately represent the degree of congestion.

  In the example of FIG. 8, a pedestrian crossing is photographed. As shown in the figure, when congestion occurs, there are many similar regions within the search range of the tracking target, which makes feature extraction difficult, makes search by the search processing unit 13 difficult, and causes erroneous tracking. Probability is high. The greater the degree of congestion, the greater the difficulty of searching. Therefore, in the present embodiment, the degree of congestion is used as a search difficulty level parameter indicating the difficulty level of search processing by the search processing unit 13.

  Returning to FIG. 3, the operation mode switching unit 25 acquires the congestion level from the congestion level detection unit 41. Then, the operation mode switching unit 25 switches between the tracking mode and the non-tracking mode according to the congestion level that is the search difficulty level parameter. The operation mode switching unit 25 switches the tracking mode to the non-tracking mode and sets the non-tracking mode when the congestion level is equal to or greater than the predetermined congestion level. The operation mode switching unit 25 sets the tracking mode if the congestion level is less than the threshold congestion level.

  The configuration of the automatic tracking device 101 according to the present embodiment has been described above. Other configurations of the automatic tracking device 101 are substantially the same as those in the first embodiment, and a description thereof will be omitted.

  Next, the operation of the automatic tracking device 101 will be described. FIG. 5 shows the operation of the operation mode switching unit 25. The process of FIG. 5 is performed repeatedly. The process of FIG. 5 is the same as the process of FIG. 2 except that the congestion level is acquired and compared with the threshold congestion level.

  The operation mode switching unit 25 acquires the congestion level from the congestion level detection unit 41 (S41), and determines whether or not the congestion level is greater than or equal to the threshold congestion level (S43). If the congestion level is equal to or greater than the threshold congestion level, the operation mode switching unit 25 determines whether or not the current mode is the tracking mode (trackable mode) (S45). If step S45 is Yes, the operation mode switching unit 25 sets the current operation mode to the non-tracking mode (tracking impossible mode) (S47). That is, the operation mode is switched from the tracking mode to the non-tracking mode. Then, the operation mode switching unit 25 sets a timer (S49). The timer is set to prevent frequent switching (hunting) of the mode. Until the timer value becomes 0, the non-tracking mode is maintained even if the congestion level decreases.

  If the current mode is not the tracking enable mode in step S45, it is determined whether or not the timer value is 0 (S51). If the timer value is 0, a timer is set (S53). Here, even if the timer value becomes 0, the congestion level is still more than the threshold congestion level, so the timer is set again.

  If the congestion level is not equal to or greater than the threshold congestion level in step S43, the operation mode switching unit 25 determines whether or not the tracking mode is currently set (S55). If the tracking mode is not set, the operation mode switching unit 25 determines whether or not the timer value is 0 (S57). If the timer value is 0, the operation mode switching unit 25 sets the tracking mode (S59).

  As described above, the operation mode switching unit 25 compares the congestion level with the threshold congestion level, sets the non-tracking mode if the congestion level is equal to or higher than the threshold congestion level, and sets the congestion level to the threshold congestion level. If less than, set the tracking mode. And mode switching is performed using a timer so that hunting does not occur.

  The operation of the operation mode switching unit 25 that uses the degree of congestion in the second embodiment has been described above. The overall operation of the automatic tracking device 101 will be described. When the degree of congestion is small, the tracking mode is set. And according to the search result of the search process part 13, the imaging device 3 rotates so that a tracking object may be tracked. The display unit 35 continuously displays the tracking target. When the degree of congestion increases, the non-tracking mode is set. The direction of the imaging device 3 is controlled by the automatic tracking device 1 so that the specified position received by the tracking target specification receiving unit 11 comes to the center of the screen. And the imaging device 3 is fixed and the display part 35 continues image | photographing the same range. As described above, the operation of the automatic tracking device 101 is the same as that of the first embodiment except that the operation mode switching unit 25 uses the degree of congestion.

  The automatic tracking device 101 according to the second embodiment of the present invention has been described above. Also in the present embodiment, the automatic tracking device 101 is configured to switch between the tracking mode and the non-tracking mode according to the search difficulty parameter. The same advantages as those of the first embodiment can be obtained. That is, it is possible to perform processing adapted to a situation where tracking is difficult, and convenience can be improved. Even in the non-tracking mode in which no tracking is performed, the tracking target can be projected over a longer time.

  In addition, according to the present embodiment, it is possible to detect a situation where tracking is difficult based on the degree of congestion using the degree of congestion as a search difficulty parameter, and set a non-tracking mode so that an appropriate video can be obtained even in a situation where tracking is difficult. Can be obtained.

  FIG. 6 shows an automatic tracking device according to the third embodiment. In the third embodiment, the automatic tracking device 201 includes both the accumulation mode information acquisition unit 23 and the congestion level detection unit 41. The accumulation mode information acquisition unit 23 supplies the accumulation magnification to the operation mode switching unit 25. The congestion level detection unit 41 supplies the congestion level information to the operation mode switching unit 25.

  The operation mode switching unit 25 switches the operation mode based on the accumulation magnification and the degree of congestion. The non-tracking mode is set when the accumulation magnification is greater than or equal to the threshold magnification or when the congestion level is greater than or equal to the threshold congestion level.

  Other configurations and operations of the automatic tracking device 201 are the same as those described in the first and second embodiments, and a description thereof is omitted here.

  The automatic tracking device 201 according to the third embodiment has been described above. Also in this embodiment, the same advantages as those in the first and second embodiments can be obtained. Furthermore, in the present embodiment, the non-tracking mode is set when either the accumulation magnification or the congestion degree increases, so that it is possible to more reliably cope with a situation where tracking is difficult. That is, even when it becomes dark and tracking becomes difficult, or when the degree of congestion increases and tracking becomes difficult, the non-tracking mode can be set to obtain an appropriate video.

  The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above-described embodiments, and it goes without saying that those skilled in the art can modify the above-described embodiments within the scope of the present invention.

  As described above, the automatic tracking device according to the present invention has an effect of improving convenience by performing processing adapted to a situation where tracking is difficult, and is useful as a monitoring system or the like.

The block diagram of the automatic tracking apparatus in the 1st Embodiment of this invention The flowchart which shows operation | movement of the automatic tracking apparatus in the 1st Embodiment of this invention. The block diagram of the automatic tracking apparatus in the 2nd Embodiment of this invention Block diagram showing the configuration of the congestion degree detection unit The flowchart which shows operation | movement of the automatic tracking apparatus in the 2nd Embodiment of this invention. Block diagram of the automatic tracking device in the third embodiment of the present invention Diagram showing accumulation mode Diagram showing the tracking process in a crowded state

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Automatic tracking apparatus 3 Imaging apparatus 5 Input part 7 1st image memory 9 2nd image memory 11 Tracking object designation | designated reception part 13 Search processing part 15 Feature-value extraction area | region determination part 17 Search range determination part 19 Template registration part 21 Correlation value calculation Unit 23 accumulation mode information acquisition unit 25 operation mode switching unit 27 moving position determination unit 29 rotary table zoom control unit 31 display processing unit 33 output unit 35 display unit 41 congestion degree detection unit

Claims (1)

An automatic tracking device that processes a video signal of an imaging device and controls a shooting direction of the imaging device so as to track a specified tracking target,
An input unit for inputting a video signal of the imaging device;
A tracking target designation receiving unit for receiving designation of a position of a tracking target in the video signal;
A search processing unit that searches the video for the tracking target reflected at a specified position;
A search difficulty parameter acquisition unit for acquiring a search difficulty parameter representing the difficulty of the search processing by the search processing unit;
An operation mode switching unit that switches between a tracking mode and a non-tracking mode according to the search difficulty parameter;
In the tracking mode, the movement position of the imaging device is determined so as to track the tracking target according to the processing result by the search processing unit, and in the non-tracking mode, the specified position received by the tracking target designation receiving unit. A moving position determining unit that determines a moving position of the imaging device so that the image is fixedly projected at a predetermined non-tracking mode position in the video;
A movement control unit that controls a movement mechanism of the imaging device according to the movement position determined by the movement position determination unit;
Equipped with a,
The search difficulty parameter acquisition unit acquires accumulation magnification information of an accumulation mode in the imaging device as the search difficulty parameter, and the operation mode switching unit is configured to perform the operation when the accumulation magnification is equal to or greater than a predetermined threshold magnification. An automatic tracking device that switches a tracking mode to the non-tracking mode .

Images