JP4723934B2 - Object search device - Google Patents

Description

  The present invention relates to an object search device that is provided in a vehicle and searches a surrounding pedestrian or the like from a captured image.

  Some vehicles are equipped with a vehicle image pickup device that assists the driver's vision by photographing the front of the vehicle during night driving and displaying pedestrians outside the irradiation area of the headlight on the display. is there. In this vehicle imaging device, for example, near-infrared light is irradiated in front of the vehicle, and light reflected by a pedestrian or the like (near-infrared light) is received by the imaging device and output as a video signal of a visible light image. . By displaying this video signal on the display, a pedestrian or the like in the dark is displayed on the display so that the driver can recognize it.

  In such a vehicle imaging device, in order to accurately display the movement of a pedestrian or the like, for example, the video signal of each frame image is output at 10 frames / second or more. The relative movement of the object can be displayed on the display.

  By the way, when the front of the vehicle is photographed using near infrared light, reflected light from various objects in front of the vehicle is received. In order to provide visual assistance to the driver, it is necessary to accurately extract and display objects such as pedestrians, and for this purpose, image processing is performed on the video signal, It is necessary to extract a control that detects the position and size of the object.

  When extracting an object such as a pedestrian from a frame image, if the position and size of the object are known, it is only necessary to determine the presence or absence of the object, but in general the position and size of the object are unknown. It is. That is, the presence or absence of a pedestrian is unknown, the position and size of the object on the frame image differ depending on the relative position of the vehicle (camera) and the pedestrian, and the relative position of the object and the vehicle changes. As a result, the position and size of the object on the frame image also change.

  In order to extract such an object from the frame image, there is a method of setting a search window and scanning the entire frame image with the search window. At this time, it is necessary to scan the entire frame image while changing the size of the search window, and the object search process is performed on each frame image output in time series from the imaging device. Requires an enormous amount of computation processing and also requires a long computation time.

  From here, as a method of reducing the amount of calculation processing when searching for a specific object and shortening the processing time, in addition to the imaging means using near infrared light, an imaging means using visible light is provided, A method of detecting a lane marking of a road from a visible light image and limiting a search area based on the detected lane marking has been proposed (for example, see Patent Document 1).

  In addition, a proposal has been made to limit the search area on the frame image by providing a laser radar instead of the imaging means using visible light and specifying the distance and direction of the object by the laser radar. (For example, refer to Patent Document 2).

In other words, in Patent Document 1 and Patent Document 2, the search area is limited by using an imaging means for visible light and a laser radar, thereby reducing the amount of calculation for search and the processing time. .
JP 2002-362302 A JP 2003-302470 A

  However, in any of the above proposals, in order to limit the search area, an imaging unit using visible light or laser light is required in addition to an imaging unit using near infrared light.

  Further, in the case of using a laser radar, there arises a problem that anything not detected by the laser radar is excluded from the search target.

  The present invention has been made in view of the above facts, and in addition to an imaging means for display, an efficient search process can be performed without separately providing an imaging means for detecting the position, size, etc. of an object. An object of the present invention is to propose an object search apparatus capable of reducing the amount of calculation for search and the processing time by making it possible.

In order to achieve the above object, the present invention provides an imaging means for imaging a target object at a predetermined frame rate, and outputting frame images obtained by imaging in time series, and a plurality of candidates having different preset sizes. A search time interval, which is an interval between the frame images to be searched , is set for each candidate region so that the candidate region with a smaller size becomes longer, and the frame corresponding to the search time interval is set for each candidate region. Search means for searching for an image of the object having a size corresponding to the size of the candidate area with respect to the image .

The object search device according to the present invention includes an imaging unit that images a target at a predetermined frame rate and outputs a frame image obtained by the imaging in time series, and a plurality of candidates having different preset sizes. A search time interval, which is an interval between the frame images to be searched , is set for each candidate region so as to be shorter as the candidate region is larger in size, and the frame corresponding to the search time interval is set for each candidate region. Search means for searching for an image of the object having a size corresponding to the size of the candidate area with respect to the image .

  According to this invention, the object is searched from the frame image output in time series by the imaging means.

  The object imaged by the imaging means is large when the distance to the object is short, but it is small as it is separated. In addition, as the distance increases, the amount of movement per unit time decreases. Further, the candidate area for searching for an object may be smaller as the object is farther away.

From this point, in the present invention, when searching for a distant object, the size of the candidate area is reduced and the search time interval is increased. Further, as the distance gets closer, the size of the candidate region used for searching for the target object is increased and the search time interval is shortened. The search interval for the time of output frame image time series can be replaced by a frame interval of a frame image, with respect to a large candidate region size, but to shorten the frame interval to be searched, the size of the candidate region The smaller the becomes, the wider the frame interval used for the search.

  Accordingly, it is possible to reduce the amount of calculation for searching for an object and shorten the processing time for each frame image output in time series.

It said searching means of the present invention as described above, in the case of small the candidate region size searches the frame image for each first number of predetermined frame, in the case of large the candidate region size the first The frame image is searched for every second frame number smaller than the predetermined number of frames.

Further, the image pickup means of the present invention is provided in a vehicle provided with a vehicle speed detection means for detecting a vehicle speed and images the front of the vehicle, and the search means detects the faster the vehicle speed detected by the vehicle speed detection means. The frame image is searched by shortening the search time interval for each candidate region .

According to the present invention, the number of frames is set according to the search time interval, and the object is searched from the frame image using the candidate area having a size corresponding to each set number of frames. In this case, for the not small candidate area size, searches for every first predetermined number of frames, for large candidate region size, each first number of the second frame is less than a predetermined number of frames To explore.

Further, when searching for a front object provided in the vehicle, the object relatively moves in a direction approaching the host vehicle. Since the amount of movement of the object at this time affects the vehicle speed, the search time interval for each search region is changed so that the search time interval becomes shorter as the vehicle speed increases according to the vehicle speed. This allows precise search of the vehicle front side objects.

Furthermore, in the present invention, the size of the size of the candidate area to be provided for the search is set in a plurality of stages, and the number of frames that are the search time interval or the frame interval to search for each set search area is set, An object may be searched for every number of frames set for the nuclear search region.

  As described above, according to the present invention, when searching for and extracting an object from a frame image input in time series, different size candidate areas and search times corresponding to the sizes of the candidate areas By setting the interval and performing a search based on the setting, it is possible to obtain an excellent effect that it is possible to reduce the amount of calculation for searching each frame image and shorten the processing time.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a schematic configuration of a vehicle imaging system 10 applied to the present embodiment. The vehicle imaging system 10 includes an imaging device 12, an image processing device 14 that is provided as a search device to which the present invention is applied and that performs predetermined image processing on a video signal output from the imaging device 12, and image processing A display 16 for displaying an image corresponding to the video output output from the device 14 is included.

  The vehicle imaging system 10 is provided in a vehicle, captures the front of the vehicle, and displays it on a display 16 provided in the passenger compartment, thereby providing visual assistance to the driver.

  The imaging device 12 includes a light unit 18 and a camera unit 20. The light unit 18 emits light having a predetermined wavelength toward the front of the vehicle. The camera unit 20 includes an imaging device such as a CCD or CMOS image sensor, and an optical system (both not shown) that forms incident light on the imaging device, and has a predetermined wavelength incident from the front side of the vehicle. Receive the light.

  Thereby, an electrical signal corresponding to the image formed on the image sensor by the optical system can be obtained. Further, the camera unit 20 performs each process such as A / D conversion, correlated double sampling, gain control, and predetermined digital signal processing on the electrical signal output from the image sensor, and thereby corresponds to one frame. A video signal (image data of a frame image) is output.

  In addition, the imaging device 12 is provided with an imaging control unit 22. The imaging control unit 22 outputs a predetermined synchronization signal and exposure control signal. Thereby, in the imaging device 12, the operation | movement of the lamp unit 18 and the camera unit 20 is controlled, and imaging is performed. Further, the imaging control unit 22 performs control so that the video signal is output from the camera unit 20 in time series. A known configuration for outputting a video signal corresponding to a captured image can be applied to the basic configuration of such an imaging apparatus 12.

  The imaging device 12 used in the vehicle imaging system 10 can use near infrared light of infrared light as light of a predetermined wavelength. At this time, the light unit 18 emits near infrared light toward the front of the vehicle, and the camera unit 20 can receive near infrared light by using, for example, a near infrared light transmission filter. It only has to be.

  Thereby, in the imaging device 12, the light irradiated from the lamp unit 18 is reflected by the object in front of the vehicle and is incident on the camera unit 20, whereby the object is imaged. At this time, by using near-infrared light, it is possible to image a region that is difficult for the driver to visually recognize outside the headlight irradiation region during night driving. By displaying the image picked up by 12 on the display means, the driver's visual assistance is made possible.

  The imaging device provided as the imaging means is not limited to this, and the camera unit 20 may receive infrared rays (far infrared rays) emitted from an object. In this case, the lamp unit 18 is unnecessary (omitted). )can do. The imaging device 12 is not limited to infrared light such as near-infrared light and far-infrared light, and may be one that performs imaging with visible light. In this case, a light projecting unit that emits visible light is used. These may be provided as necessary, and will be described below without specifying the light source type.

  As the display 16 for displaying a frame image, any display means for directly or indirectly displaying an image, such as an LCD (liquid crystal display), can be applied. Note that when the imaging device 12 performs imaging using near-infrared light, a monochromatic image is displayed on the display 16. When the camera unit 20 receives visible light, a color image may be displayed on the display 16.

  On the other hand, the image processing apparatus 14 is provided with a specific image extracting unit 24 and an image superimposing unit 26. The specific image extraction unit 24 extracts, for example, a predetermined object such as a pedestrian from the video signal input as a frame image from the imaging device 12.

As shown in FIGS. 2A and 2B, the camera unit 20 of the imaging device 12 is attached, for example, above the center of the front shield window of the vehicle 28 and images the front of the vehicle 28. The imaging range of the camera unit 20 at this time is, for example, as shown in FIG. 2A, the vertical direction is the range of the vertical field angle φ _V , and the horizontal direction is horizontal as shown in FIG. the range of the angle of view φ _H.

  Further, the imaging device 12 outputs a video signal corresponding to the frame image 30 as shown in FIG. 3A by performing imaging with the camera unit 20.

  In the specific image extraction unit 24, a search window 32 having a predetermined size (window size) is set as a search area, and the search window 32 is set in the horizontal direction (horizontal direction, arrow direction in FIG. 3A) of the frame image 30. Scanning is performed by shifting predetermined pixels in the vertical direction (vertical direction), and the presence or absence of an object to be searched is determined in the search window 32. The search for the target object using the search window 32 is performed by, for example, scanning the search window 32 with respect to image data obtained by binarizing the video signal in order in the horizontal direction by a predetermined number of pixels (for example, 2 When scanning is performed while shifting one pixel at a time, and scanning for one column is completed in the horizontal direction, scanning is performed while shifting by a predetermined pixel (for example, two pixels) in the vertical direction. By repeating this, it is possible to apply a known method such as determining whether or not there is an image approximated to the object in the search window 32, and a detailed description thereof will be omitted here.

  In the specific image extraction unit 24, when an object is extracted by scanning using the search window 32, image data including the position and size of the extracted object on the frame image 32 is generated, and the generated image data is converted into an image. Output to the superimposing unit 26.

  The image superimposing unit 26 superimposes the object on the frame image 32 by combining the image data of the object generated by the specific image extracting unit 24 with the video signal of the frame image input from the imaging device 12, and displays 16 is output.

  Thus, in the vehicle imaging system 10, an image clearly showing the object is displayed on the display 16 on the frame image 32 output from the camera unit 20.

By the way, as shown in FIG. 1, in the vehicle imaging system 10 applied to the present embodiment, a search setting unit 34 is provided in the image processing device 14. The search setting unit 34 sequentially counts the frame images 30 (video signals) captured by the camera unit 20 and input from the imaging device 12 in time series, and a search used for searching for an object according to the count value. The window size of the window 32 is set.

  The specific image extraction unit 24 performs the object search process on the frame image selected by the search setting unit 34 using the search window 32 having the window size set by the search setting unit 34. .

  As shown in FIG. 4, when the object to be searched is a pedestrian (hereinafter referred to as object 36), the size of the object 36 changes according to the distance from the vehicle 28. If it is close to 28, the object 36 becomes a large image and becomes smaller on the frame image 30 as it is farther from the vehicle 28.

In order for the specific image extraction unit 24 to accurately detect the object 36, it is necessary to use a search window 32 corresponding to the size (window size) of the object 36 on the frame image 30 .

  Accordingly, in the present embodiment, when the camera unit 20 uses an optical system having a vertical angle of view φV of 12.57 ° and an image sensor having a vertical pixel count of 420 pixels, the window size of the search window 32 is The number of pixels in the vertical direction (the number of vertical pixels) is set in a plurality of stages.

  At this time, as shown in FIG. 3B, a search window 32a having a vertical pixel count of 32 pixels is set as the minimum size, and the window size (vertical pixel count) is increased by 20% by 32%. Fifteen search windows 32 from the search window 32b to 385 pixels 32o are set. That is, in the specific image extraction unit 24, 32 pixels (search window 32a), 38 pixels (search window 32b), 45 pixels, 54 pixels, 64 pixels, 76 pixels, 91 pixels, 109 pixels (search window 32h), 130 pixels. The object 36 can be extracted from the frame image 30 using each of the search window 32o of 156 pixels, 187 pixels, 224 pixels, 268 pixels, 321 pixels (search window 32n) and 385 pixels. It has become. Note that the vertical angle of view φ, the pixels, the window size and the number of search windows 32 are examples, and are not limited to the above numerical values.

  As described above, the elevation angle φ (see FIG. 4), which is the vertical angle of view when the object 36 is imaged by the camera unit 20, increases as the distance between the vehicle 28 and the object 36 is short. The longer the distance of the object 36, the smaller.

  At this time, on the image sensor of the camera unit 20, the number of pixels to be imaged changes according to the size of the object 36. The smaller the object 36 appears on the display 16, the smaller the number of vertical pixels.

  Here, Table 1 shows the distance (m) from the object 36 to the number of vertical pixels set to the window size when the camera unit 20 images the object 36 having a height H of 1.6 m. The elevation angle φ (°), which is the vertical field angle, is shown.

  When set in this way, the distance of the object 36 searched for in the search window 32a is 108 m, the elevation angle φ at that time is 0.85 °, and the object searched for in the search window 32o. The distance of 36 is 9 m.

  Further, for example, the distance difference Δ between when the object 36 is extracted with the search window 32a (vertical pixel number 32 pixels) and when the object 36 is extracted with the search window 32b (vertical pixel number 38 pixels) that is one size larger. Is 17 m. Table 1 shows the distance difference when the size of the search window 32 is changed by one step as the distance difference Δ (m).

  On the other hand, if the number of frame images 30 output per unit time (the number of frames) is the same, the change in the distance between the traveling vehicle 28 and the object 36 is mainly caused by the traveling speed of the vehicle 28 being large. to cause.

  The imaging device 12 outputs the frame images 30 in time series. At this time, if the output rate of the frame images 30 is 30 frames / sec, the frame interval is 33 msec.

  Table 2 shows the travel distance between frames (33 msec) for each speed of the vehicle 28.

  As shown in Table 2, for example, when the speed is 40 (km / h), the vehicle 28 moves 0.37037 (m) while the frame image 30 for one frame is obtained, and moves to the object 36. approach.

  As shown in Table 1, when the object 36 moves from the position extracted by the search window 32a to the position extracted by the search window 32b, the vehicle 28 traveled 17 m because the distance difference Δ is 17 m. It will be. At this time, if the vehicle 28 is traveling at a speed of 40 (km / h), the frame image 30 for about 46 frames is output from the imaging device 12 and input to the image processing device 14 from Table 2. The

  Further, the object 36 moves from the position (distance 54 m) extracted by the search window 32 of 64 pixels, which is one step of the search window 32, to the position (distance 45 m) detected by the search window 32 of 76 pixels. In this case, if the vehicle 28 is traveling at a speed of 40 km / h, the frame image 30 for about 24 frames is input to the image processing device 14. become.

  From here, the search setting unit 34 of the image processing apparatus 14 sets a search time interval for each of the search windows 32 according to the vehicle speed. At this time, the search window 32 sets a search time interval with reference to about ½ of the time until one step changes, and based on this search time interval, the frame interval used for the search, that is, every frame is searched. Set the number of frames to perform.

  Table 3 shows an example of setting the number of frames at this time.

  The search setting unit 34 stores the settings shown in Table 3 as, for example, a map of vehicle speed and search window size.

  As shown in FIG. 1, a vehicle speed sensor 38 is connected to the search setting unit 34. For example, the vehicle speed sensor 38 outputs a signal corresponding to the number of rotations of the wheels of the vehicle 28 (not shown in FIG. 1), and the search setting unit 34 uses the output of the vehicle speed sensor 38 to determine the travel speed ( Vehicle speed).

  Thereby, the search setting unit 34 sets a search time interval (or the number of frames) for each search window 32 according to the vehicle speed, for example, sets the search window 32 used for the search in the frame number, or searches A frame number to be searched is set for each window 32.

  The search setting unit 34 counts the order (numbers) of the frame images 30 input from the imaging device 12, and uses the search window 32 used for searching for the object 36 for each frame image 30 based on the order that is the count value. The specific image extraction unit 24 extracts the object 36 based on this setting.

  As a result, the image processing apparatus 14 periodically searches for the object 36 for each number of frames set in the search window 32.

  In the vehicular imaging system 10 configured as described above, a camera is controlled while controlling the light projection of the light unit 18 by turning on a power switch (not shown) in a state where an ignition switch (not shown) of the vehicle is turned on. The front of the vehicle 28 is imaged by the unit 20 and displayed on the display 16 while outputting the captured frame images in time series.

  At this time, in the image processing unit 14 provided in the vehicle imaging system 10, when the object 36 is extracted from the frame image 32 input from the imaging device 12 using the search window 32, an image of the extracted object 36 is obtained. Is superimposed on the frame image 30 and displayed on the display 16.

  Thereby, in the vehicle imaging system 10, the driver of the vehicle 28 can accurately visually recognize a pedestrian or the like in an area outside the visible range from the image displayed on the display 16.

  In the vehicle imaging system 10, a search setting unit 34 is provided in the image processing device 14, and the search setting unit 34 searches for an object 36 for each frame image 30 input from the imaging device 12. The window size of the search window 32 used when performing the search is set together with whether or not to perform.

  The specific image extraction unit 24 that extracts the object 36 from the frame image 30 performs a search process using the search window 32 set by the search setting unit 34 on the frame image 30 set by the search setting unit 34. Thus, the object 36 is extracted.

  Here, the processing in the search setting unit 34 will be described with reference to FIG. This flowchart is executed when a power switch (not shown) of the vehicular imaging system 10 is turned on, and ends when the operation of the vehicular imaging system 10 is stopped, for example, when the power switch is turned off.

  In this flowchart, when the speed (traveling speed) of the vehicle 28 detected by the vehicle speed sensor 38 is read in step 100, in step 102, it is confirmed whether or not the speed has changed from the previous time. In the present embodiment, the vehicle speed is set to four levels of 40 (km / h), 60 (km / h), 80 (km / h), and 100 (km / h). The application range is set with the speed of the reference as the reference speed or the center speed. At this time, for example, when the vehicle speed is 40 (km / h), the level is 0 to 50 (km / h), 60 (km / h), 80 (km / h), and 100 (km / h). -70 (km / h), 70-90 (km / h), and 90 (km / h) or more, etc., to determine whether or not the vehicle speed has changed from whether or not the previous application range was deviated. In addition, the present invention is not limited to this, and any determination method according to the vehicle speed at which the search window 32 is set can be applied.

  Here, if the vehicle speed has changed, an affirmative determination is made in step 102 and the routine proceeds to step 104, where a search time interval (frame interval) is set from the search time interval for each search window 32 corresponding to the detected vehicle speed. . At this time, the search time interval is set to be short for the search window 32 having a large window size, and the search time interval is set to be longer as the window size is reduced.

  For one search window 32, the search time interval is set shorter as the vehicle speed becomes faster (higher). When it is determined that there is substantially no change in the vehicle speed, the previous setting is used as it is.

  Furthermore, when performing a search, it is preferable that the search window 32 to be used is set to be distributed. For example, when the vehicle speed is 40 (km / h), the search window 32 of 224 to 385 pixels is every other frame. At this time, for example, the search window 32 of 224 pixels and 385 pixels is set to an odd number, and the search window of 268 pixels and 321 pixels is set to an even number. Thereby, the search process can be distributed.

  On the other hand, in the vehicle imaging system 10, a video signal (hereinafter referred to as a frame image 30) that becomes a frame image 30 is input from the imaging device 12 at 30 frames / sec. It is confirmed whether or not the image 30 has been input. When the frame image 30 has been input, an affirmative determination is made at step 106 and the routine proceeds to step 108.

  In this step 108, the frame number is counted. This frame number is assigned to the frame image 30 input from the imaging device 12 in the order of input. Further, this frame number may be a serial number from when the device is turned on until it is turned off, or when a change in vehicle speed is detected, at a predetermined time interval, or at a predetermined time It may be reset when the value is reached and recounted.

  As described above, the search setting unit 34 sets the search window 32 for each order of the frame images 30 based on the vehicle speed and the window size of the search window 32, or the frame for executing the search for each search window 32. In step 110, the window size (search window 32) to be applied to the search is set from the count number of the frame image 30.

  Thereafter, in step 112, the specific image extraction unit 24 is instructed to execute a search process for the object 36 using the set search window 32 for the input frame image 30.

  In response to this instruction, the specific image extraction unit 24 performs a search process using the search window 32 having the window size set for the frame image 30. Thereby, when the object 36 is extracted, the image data of the image position and size is output to the image superimposing unit 26.

  Thereby, the frame image 30 in which the object 36 is clear is displayed on the display 16.

  At this time, an object 36 in a close distance is searched using a large window size search window 32 having a large number of vertical pixels, and a distant object 36 is searched using a small window size search window 32 having a small number of vertical pixels. For the search window 32 having a small window size, the search time interval is made longer than that for the search window 32 having a large window size.

  Thereby, since the object 36 is searched for all the frame images 30 without using all the search windows 32, it is possible to reduce the amount of calculation for the search and the processing time. .

  Further, when the vehicle speed is detected and the vehicle speed is high, the search time interval of the search window 32 of each window size is shortened and the search for the target object 36 is performed in a shorter cycle than when the vehicle speed is low. The image which can grasp | ascertain the motion of 36 and the target object 36 accurately can be displayed on the display 16, and the appropriateness at the time of performing a driver | operator's visual assistance is not impaired.

  In addition, this Embodiment demonstrated above does not limit the structure of this invention. For example, in the present embodiment, the image processing apparatus 14 and the display 16 are provided in the vehicle imaging system 10 as the imaging apparatus, but the present invention is based on a frame image that is captured and output in time series. The present invention can be applied to an image processing apparatus having an arbitrary configuration for extracting an object.

  In the present embodiment, the vehicle imaging system 10 provided on the vehicle 28 has been described. However, the present invention is not limited to this, and the captured video signal is displayed on a display unit such as the display 16. The present invention can be applied to an imaging system having an arbitrary configuration that executes image processing.

It is a schematic block diagram of the imaging system for vehicles applied to this Embodiment. (A) is the schematic which shows the imaging range in the imaging device seen from the vehicle side, (B) is the schematic which shows the imaging range in the imaging device seen from the vehicle upper side. (A) is a schematic diagram illustrating a search on a frame image using a search window, and (B) is a schematic diagram illustrating a search window used for the search. It is the schematic seen from the vehicle side which shows the imaging of a target object. It is a flowchart which shows an example of the setting of the frame to search, and the search window used for a search.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Imaging system for vehicles 12 Imaging device (imaging means)
14 Image processing device 16 Display 18 Light unit (imaging means)
20 Camera unit (imaging means)
22 Imaging control unit (imaging means)
24 Specific image extraction unit (search means)
26 Image superposition unit 28 Vehicle 30 Frame image 32 Search window (search area)
34 Search setting section (search means)
36 Target

Claims (4)

Imaging means for imaging an object at a predetermined frame rate and outputting frame images obtained by imaging in time series;
A plurality of candidate areas having different sizes set in advance, and a search time interval as an interval between the frame images to be searched is set for each candidate area so that the candidate area with a smaller size becomes longer. Search means for searching for an image of the object having a size corresponding to a size of a candidate area with respect to the frame image corresponding to the search time interval ;
An object search device including: Imaging means for imaging an object at a predetermined frame rate and outputting frame images obtained by imaging in time series;
A plurality of candidate areas having different preset sizes are used, and a search time interval as an interval between the frame images to be searched is set for each candidate area so as to be shorter as the candidate area is larger in size. Search means for searching for an image of the object having a size corresponding to a size of a candidate area with respect to the frame image corresponding to the search time interval ;
An object search device including: The search means, in the case of small the candidate region size searches the frame image for each first number of predetermined frame, the smaller than the number of said first predetermined frame in the case of large the candidate region size The object search device according to claim 1, wherein the frame image is searched for every two frames. The imaging means is provided in a vehicle provided with vehicle speed detection means for detecting a vehicle speed, and images the front of the vehicle,
4. The search unit according to claim 1, wherein the search unit searches the frame image by shortening the search time interval for each candidate region as the vehicle speed detected by the vehicle speed detection unit increases. 5. Object search device.

Images