JP2008015770A - Image recognition device and image recognition method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accelerate processing speed while reducing the amount of templates in recognition of a pedestrian, and furthermore enhance recognition accuracy. <P>SOLUTION: An image of a pedestrian who exists at a location of predetermined distance (for example, a most distant end for recognition of a pedestrian) from a vehicle is used as a template. Frames of determination object domains A10 and A20 imported from an image shot by an onboard camera are reduced according to the size of the template and reduced as images A11 and A12 respectively and compared with a template. If they are in agreement, the images A11 and A12 are determined as the pedestrian. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image recognition apparatus and method for recognizing a specific object from an input image captured by a camera, and more particularly to an image recognition apparatus and image recognition method for recognizing a pedestrian using an in-vehicle camera.

  It is particularly important to avoid collisions with pedestrians while the vehicle is running. Therefore, in recent years, a technique for recognizing pedestrians around the own vehicle using image recognition and radar detection has been devised (see, for example, Patent Document 1).

  When recognizing a pedestrian by image recognition, the size of the image of the pedestrian that is photographed varies depending on the distance from the vehicle to the pedestrian, but a template that is different for each size of the pedestrian image is prepared. Since the amount increases, for example, as disclosed in Patent Document 2, it has been devised to enlarge and reduce the target image and match it with the template.

JP 2002-362302 A JP 2003-196655 A

  However, if the size ratio between the target image and the template is unknown, it will search for the optimal enlargement / reduction ratio while sequentially changing the magnification for the image, and the processing load will increase even if the template amount is reduced. However, there is a problem that processing time required for pedestrian recognition increases.

  On the other hand, when recognizing a pedestrian, since the difference in clothing or the like affects the recognition accuracy, it is necessary to apply an appropriate filter that absorbs such individual differences.

  That is, in the pedestrian recognition by the conventional image processing, there is a problem that it is impossible to achieve both reduction of the template amount, increase in processing speed, and improvement of recognition accuracy.

  The present invention has been made in order to solve the above-described problems in the prior art and to solve the problems. An image recognition apparatus and an image in which the processing speed is increased while the amount of templates is reduced, and the recognition accuracy is further improved. An object is to provide a recognition method.

  In order to solve the above-described problems and achieve the object, the image recognition apparatus and the image recognition method according to the present invention cut out the determination target region from the input image captured by the in-vehicle camera, reduce the image in the determination target region, The reduced image and the reference image are compared to recognize the presence of a specific object.

  According to the present invention, the image recognition apparatus and the image recognition method reduce the storage medium capacity necessary for holding the template, increase the type of template if there is enough storage medium capacity, and perform the reduction process for individual differences among pedestrians. Therefore, it is possible to obtain an image recognition apparatus and an image recognition method that can increase the processing speed while reducing the amount of template and further improve the recognition accuracy.

  Exemplary embodiments of an image recognition apparatus and an image recognition method according to the present invention will be explained below in detail with reference to the accompanying drawings.

  First, an image recognition method according to the present invention will be described with reference to FIG. In the present invention, an image of a pedestrian existing at a predetermined distance from the vehicle is used as a template. Here, the predetermined distance is preferably the farthest distance that requires pedestrian recognition, that is, the maximum distance (detection distance) at which a pedestrian should be detected from the vehicle, for example, about 50 m. Therefore, the own vehicle recognizes a pedestrian existing within 50 m from the own vehicle and gives a warning.

  Then, when the determination target areas A10 and A20 are cut out from the image captured by the in-vehicle camera as shown in FIG. 1, they are reduced according to the size of the template to be reduced images A11 and A12, respectively, and the reduced images A11 and A12 are used as templates. It is determined that the person is a pedestrian if they match.

  Here, the reduction ratio is determined based on the position of the extracted determination target area in the input image, specifically, the height of the lower end of the determination target area in the input image.

  When the pedestrian image is included in the input image, the position of the pedestrian's foot corresponds to the distance from the own vehicle, and the closer the distance from the own vehicle to the pedestrian, The size at will increase. Therefore, if the distance from the host vehicle is estimated assuming that the lower end of the determination target region is the foot of the pedestrian, it is necessary to match the size of the template, that is, the size of the pedestrian at a predetermined distance (for example, 50 m) from the host vehicle. Can be obtained.

  By calculating the reduction ratio in this way, it is not necessary to search for the optimum enlargement ratio / reduction ratio while sequentially changing the magnification, and the processing speed can be greatly reduced and the processing speed can be increased.

  In addition, as the distance from the host vehicle to the pedestrian increases, the difference in clothes and the like becomes difficult to see, and the ratio of the information on the contour portion in the pedestrian image increases. Therefore, by using a long-distance pedestrian image as a template and reducing and comparing the determination target areas, it is possible to provide an effect as a filter that removes the individual difference of pedestrians.

  In addition, since the pedestrian image at a long distance is small in size, the storage medium capacity necessary for holding the template can be reduced, and if the storage medium capacity is sufficient, the type of template can be increased to improve recognition accuracy. it can.

  Next, with reference to FIG. 2, a schematic configuration of the in-vehicle image recognition apparatus 10 which is an embodiment of the present invention will be described. As shown in the figure, the image recognition device 10 mounted on the vehicle is connected to a navigation device 30, a camera 31, a radar 33, and a pre-crash ECU 40.

  The navigation device 30 is an in-vehicle device that sets and guides a travel route using the position of the host vehicle specified by communicating with a GPS (Global Positioning System) artificial satellite and map data 30a stored in advance. In addition, the navigation device 30 provides the image recognition device 10 with location information of the host vehicle, surrounding map information, a planned travel route, and the like.

  The camera 31 captures the vicinity of the host vehicle and inputs the captured result to the image recognition apparatus 10. The radar 33 detects an object around the host vehicle and measures a distance to the object and inputs the detected object to the image recognition apparatus 10.

  The pre-crash ECU 40 receives the control of the image recognition device 10 when the image recognition device 10 predicts a collision of the host vehicle, controls the operation of the vehicle by the brake 41 and the engine control device (EFI) 42, and displays the display 43 and the speaker. 44 is an electronic control device that executes notification by 44.

  The display 43 is an output means for notifying the user, that is, the vehicle occupant by display, and the speaker 44 is an output means for notifying by voice. The display 43 and the speaker 44 can be shared by various in-vehicle devices such as the navigation device 30 and an in-vehicle audio device (not shown) in addition to performing output under the control of the pre-crash ECU 40.

  The image recognition apparatus 10 includes a preprocessing unit 11, a vehicle recognition unit 16, a white line recognition unit 17, a pedestrian recognition unit 18, a template storage unit 19, and a collision determination unit 20 therein. Here, the vehicle recognition unit 16, the white line recognition unit 17, the pedestrian recognition unit 18, the template storage unit 19, and the collision determination unit 20 are, for example, a single microcomputer 10a (an arithmetic processing unit including a combination of a CPU, a ROM, and a RAM). It is preferable to realize the above.

  The preprocessing unit 11 performs processing such as filtering, edge detection, and contour extraction on the image captured by the camera 31, and then outputs the image to the vehicle recognition unit 16, the white line recognition unit 17, and the pedestrian recognition unit 18.

  The vehicle recognition unit 16 performs pattern matching on the image output from the preprocessing unit 11 to recognize the vehicle, and outputs the recognition result to the collision determination unit 20. Further, the white line recognition unit 17 recognizes a white line by performing pattern matching or the like on the image output from the preprocessing unit 11, and outputs the recognition result to the collision determination unit 20.

  The pedestrian recognition unit 18 is a processing unit that recognizes a pedestrian image from the image (input image) output by the preprocessing unit 11, and includes a cutout unit 18a, a reduction processing unit 18b, and a recognition processing unit 18c.

  The cutout unit 18a performs a process of cutting out the determination target region from the input image. The reduction processing unit 18b calculates a reduction ratio from the height in the input image of the extracted determination target region, and creates a reduced image. The recognition processing unit 18c performs processing for recognizing the presence of a pedestrian by comparing the reduced image with the template stored in the template storage unit 19.

  In the template storage unit 19, a template used for pedestrian recognition, that is, an image of a pedestrian existing at a predetermined distance from the vehicle is registered in advance. In addition, the template memory | storage part 19 can also memorize | store the template for other recognitions, such as vehicle recognition, in addition to the template for pedestrian recognition. Actually, the shape of the pedestrian at the detection distance of 50 m is imaged with a camera, and this is cut out and stored. There may be a plurality of types of pedestrian recognition templates.

  The collision determination unit 20 uses the recognition results obtained by the vehicle recognition unit 16, the white line recognition unit 17, the pedestrian recognition unit 18, the detection result obtained by the radar 33, and the position information output from the navigation device 30, and automatically detects the collision with the pedestrian and other vehicles. Determine the risk of collision with the vehicle.

  Specifically, the collision determination unit 20 determines the probability of a collision with a pedestrian or another vehicle, the collision time, the distance to the collision position, the collision angle, and the like, and the display 43 based on the determination result. Information display instruction, warning sound output instruction to the speaker 44, brake control instruction, EFI control instruction, and the like are output to the pre-crash ECU 40.

  Next, the processing operation of the entire image recognition apparatus 10 shown in FIG. 2 will be described with reference to FIG. The processing flow shown in the figure is started when a power switch (which may be linked to an ignition switch or the like) is turned on and the camera 31 captures an image, and for each processing of an image frame (for example, every several msec). This process is repeatedly executed.

  First, the image recognition apparatus 10 performs processing such as filtering, edge detection, and contour extraction by the preprocessing unit 11 on an image captured by the camera 31 (step S101). Next, the white line recognition process (step S102) by the white line recognition part 16 and the vehicle recognition process (step S103) by the vehicle recognition part 16 are performed.

  Thereafter, the pedestrian recognition unit 18 performs pedestrian recognition (step S104), the collision determination unit 20 performs collision determination (step S105), and outputs the determination result to the pre-crash ECU 40 (step S106). finish.

  Here, the concrete processing content of the pedestrian recognition part 18 shown as step S104 is shown in FIG. As shown in the figure, the pedestrian recognition unit 18 first cuts out the determination target region by the cutout unit 18a (step S201). As the determination target area, a unique image remaining after the background image difference is set as a candidate area, and an area surrounding the specific image is set as a determination target.

  Next, the reduction ratio is calculated from the height (?) In the input image at the lower end of the determination target area cut out by the reduction processing unit 18b (step S202), and the reduced image is reduced by reducing the image in the determination target area. Create (step S203).

  Thereafter, the recognition processing unit 18c compares the reduced image with the template stored in the template storage unit 19 to recognize the presence of a pedestrian (step S204), and the process is terminated. In recognition, the determination is made based on whether or not the degree of coincidence (%) of the shape exceeds a determination threshold.

  Next, the processing operation of the pre-crash ECU 40 will be described with reference to the flowchart of FIG. The processing flow shown in the figure is repeatedly executed during the operation of the pre-crash ECU 40.

  First, the pre-crash ECU 40 acquires a collision determination result from the image recognition device 10 as collision determination information (step S301). When the risk of collision is high (step S302, Yes), the display 43 and the speaker 44 are used to notify the driver (step S303), and the brake 41 and the EFI 42 are controlled to control the traveling state of the host vehicle. Is controlled (step S304), and the process is terminated.

  As described above, the image recognition apparatus 10 according to the present embodiment uses a pedestrian image existing at a predetermined distance from the vehicle as a template, and sets the determination target region from the image captured by the in-vehicle camera to the template size. At the same time, reduce the comparison. Therefore, a filter that reduces the storage medium capacity required to hold the template, increases the type of template if the storage medium capacity is sufficient, improves the recognition accuracy, and reduces the individual differences of pedestrians in the reduction process To improve recognition accuracy.

  Also, the magnification is sequentially changed by calculating the reduction ratio necessary to match the size of the template by estimating the distance from the host vehicle by regarding the lower end of the cut out judgment target area as the foot of the pedestrian. However, it is not necessary to search for the optimum enlargement / reduction ratio, and the processing load can be greatly reduced and the processing speed can be increased.

  In this embodiment, the case where a pedestrian is recognized as a specific object has been described as an example. However, the present invention is not limited to this, and may be applied to other objects such as a falling object on a road. can do.

  As described above, the image recognition device and the image recognition method according to the present invention are useful for image recognition in a vehicle, and are particularly suitable for pedestrian recognition.

It is explanatory drawing explaining the image recognition method in this invention. 1 is a schematic configuration diagram illustrating a schematic configuration of an image recognition apparatus according to an embodiment of the present invention. 4 is a flowchart for explaining a processing operation of the image recognition apparatus 10. It is a flowchart explaining the specific process operation | movement of a pedestrian recognition process. 4 is a flowchart illustrating a processing operation of a pre-crash ECU 40.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image recognition apparatus 10a Microcomputer 16 Vehicle recognition part 17 White line recognition part 18 Pedestrian recognition part 18a Cutout part 18b Reduction process part 18c Recognition process part 19 Template memory | storage part 20 Collision determination part 30 Navigation apparatus 30a Map data 31 Camera 33 Radar 40 Pre Crash ECU
41 Brake 42 EFI
43 Display 44 Speaker

Claims (3)

Reduction means for reducing an image to be a determination target area in an input image captured by the camera;
Recognizing means for recognizing the presence of a specific object by comparing the image reduced by the reducing means with a reference image;
And the size of the reference image corresponds to the size when the specific object is present at the maximum detection distance of the recognition means.   The image recognition apparatus according to claim 1, wherein the reduction unit determines a reduction rate based on a position of the determination target region in the input image. A reduction process for reducing an image to be a determination target area in an input image captured by a camera;
A recognition step of recognizing the presence of a specific object by comparing the image reduced by the reduction means with a reference image;
And the size of the reference image corresponds to the size when the specific object is present at the maximum detection distance of the recognition means.

Images