JP2011160063A - Image processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor which, if a deficient pixel is included in an image taken by an imaging means when displaying an image taken by a plurality of imaging means installed on a mobile object, can provide a vehicle driver with a beautiful image by interpolating a pixel of a corresponding location of an image taken by anther imaging means to the position of a missing pixel. <P>SOLUTION: In the case that a pixel of a defective pattern stored in a defective pixel pattern storing means 9 exists in an image taken by an imaging means 1a, an image recognizing means 5 recognizes that the pixel of a defective pattern exists. An image synthesizing means 6 determines whether a mapping table 8 has a pixel corresponding to a defective pixel and the pixel imaged by another imaging means. The image synthesizing means 6 interpolates the pixel having a defective pattern with a pixel imaged by the another imaging means 1b with reference to the mapping table 8. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing apparatus in which a plurality of imaging means are mounted on a vehicle body, a plurality of images around the photographed vehicle body are combined, and the combined image is provided to a driver or the like.

  When displaying an image captured by a camera as a conventional image processing device, the original pixel information is lost due to a pixel defect that occurs in the image of the imaging result, or a pixel that occurs due to smear, halation, or the like. Are interpolated and displayed (for example, see Patent Document 1).

JP 2006-157192 A

  In a conventional image processing apparatus, as an interpolation method for a missing pixel, an address corresponding to a pixel with missing information is searched from an image captured in the past, and there is image data in which the pixel of the part is not missing. In this case, the pixel data of the corresponding part is cited from the past captured image, and the data is interpolated to the missing pixel. However, when the object to be imaged changes from moment to moment, for example, an imaging means mounted on a moving body such as a vehicle cannot use an image captured in the past, and thus the conventional method cannot be applied. There was a problem.

  The present invention has been made in order to solve the conventional problem, and when displaying an image captured by a plurality of imaging means installed on a vehicle body, a missing pixel is included in an image captured by a certain imaging means. The image processing apparatus can provide a beautiful image to the driver of the vehicle by interpolating the pixel of the corresponding portion of the image captured by the other imaging means to the position of the missing pixel. Objective.

  The present invention is configured to include an image synthesis unit that interpolates a defective pixel region captured by the first imaging unit with a pixel captured by the second imaging unit using a mapping table.

  In the present invention, the pixel recognizing unit recognizes whether or not the image captured by the first imaging unit includes a pixel having a defective pattern stored in the defective pixel pattern storage unit, and the image synthesizing unit is configured based on the mapping table. By searching for the pixel imaged by the second imaging means corresponding to the defective pixel and interpolating the pixel, a beautiful image is provided to the vehicle driver even if smear or halation occurs in the display image for a certain period of time. can do.

1 is a block diagram of an image processing apparatus according to an embodiment of the present invention. The figure explaining the position which attaches the image pick-up means which is the principal part to a vehicle body The figure explaining the example of an image image | photographed with the imaging means which is the principal part The figure which shows the operation explanation flow of the same device The figure which shows the mapping table which shows an overlapping area | region in the same apparatus.

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

  An image processing apparatus of this embodiment is shown in FIG. In FIG. 1, the image processing apparatus includes four imaging units 1 a to 1 d for imaging at least the periphery of the vehicle body, and information signals captured by the imaging unit 1 are input to the ECU 2. The CPU 4 is connected to the ECU 2, receives image information from the imaging unit 1 via the ECU 2, has an image recognition unit 5 that recognizes defective pixels of the image, and a pixel synthesis unit 6 that combines the interpolation pixels, and interpolates the defective pixels. To do. The display unit 3 displays a processing result by the CPU 4, and the storage unit 7 includes a mapping table 8 and a defective pixel pattern storage unit 9 as a storage unit for data necessary when the CPU 4 performs image processing.

  The imaging unit 1 includes a plurality of color or monochrome imaging means having a solid-state imaging device such as a CCD or CMOS device. As the image pickup means 1a to 1d, for example, a device capable of photographing a wide range using a high viewing angle lens such as a fisheye lens is used.

  FIG. 2 is an example in which four imaging means 1a to 1d are installed on the vehicle body so that the entire periphery of the vehicle body can be confirmed by the captured image. The imaging means 1a is installed near the bumper in front of the vehicle body, the imaging means 1b and 1c are installed on the side mirrors on the side of the vehicle body, and the imaging means 1d is installed on the rear side of the vehicle body. Images actually photographed by the respective imaging means 1a to 1d are images as indicated by 31, 32, 33, and 34 in FIG.

  The operation of the image processing apparatus configured as described above will be described below with reference to the flowchart of FIG.

  First, images obtained from the plurality of imaging means 1a to 1d for photographing the periphery of the vehicle body are input to the CPU 4 via the ECU 2 (step S1). The image recognition unit 5 of the CPU 4 compares the images captured by the imaging units 1 a to 1 d with the defective pixel pattern storage unit 9 of the storage unit 7. The defective pixel pattern storage means 9 stores a shape pattern of pixels that cause overexposure due to the influence of a strong light source such as smear or halation. The image recognition unit 5 compares the captured image with the defective pixel pattern storage unit 9, calculates a defective pixel region, and recognizes the defective pixel region in the captured image (step S2). The phenomenon that the pixel recognizing unit 5 can recognize the shape pattern of the defective pixel is not only due to the influence of a strong light source such as smear or halation, but the imaged pixel does not change even after a certain time. It is also possible to apply to an image having a deposit on -1d or a deposit covering a cover that covers the imaging means 1a-1d.

  When the pixel recognizing unit 5 recognizes a defective pixel causing smear, halation, and the like, and when the recognition continues for a certain time (step S3), the image composition unit 6 refers to the mapping table 8, and the smear, halation is detected. It is determined whether or not the pixel causing the above is not picked up by other image pickup means. FIG. 5 is a diagram illustrating a method of acquiring pixel coordinates by referring to a mapping table and referring to the other captured image capturing the same portion from one captured image. The mapping table 8 is a table in which, in the captured image 51 of the image capturing unit 1a and the captured image 52 of the image capturing unit 1b, the coordinates capturing the same part are associated one-to-one. When it is desired to know whether the pixel corresponding to the coordinate 53 of the captured image on the captured image 51 exists in the captured image 52, the pixel composition unit 6 refers to the pixel coordinate list of the mapping table 8 (step S4). . The image synthesizing unit 6 obtains the coordinates 54 of the captured image 52 corresponding to the coordinates 53 of the captured image 51 from the mapping table 8 (step S5). When displaying the composite image on the display unit 3, the image composition means 6 preferentially uses pixels without smear or halation and performs interpolation on the pixels that have smear or halation (step S6). This is performed by real-time processing while the moving body is moving.

  If such an image processing apparatus according to the above-described embodiment is mounted on a vehicle body, the pixel recognizing unit determines whether the image captured by a certain imaging unit includes a defective pattern pixel stored in the defective pixel pattern storage unit. The image composition means finds a pixel imaged by another imaging means corresponding to the defective pixel from the mapping table of the storage unit, and interpolates the pixel so that smear, halation, etc. are generated in the display image for a certain period of time. Even if it occurs, a beautiful image can be provided to the driver of the vehicle.

  As described above, the image processing apparatus according to the present invention recognizes whether or not the image captured by the pixel recognition unit using the first imaging unit includes a pixel having a defective pattern stored in the defective pixel pattern storage unit. The image synthesizing unit finds a pixel imaged by the second imaging unit corresponding to the defective pixel from the mapping table of the storage unit, and interpolates the pixel, so that smear, halation, etc. occur in the display image for a certain time. However, it is useful as an image processing apparatus that can provide a beautiful image to the driver of the vehicle.

DESCRIPTION OF SYMBOLS 1 Imaging part 5 Image recognition means 6 Pixel composition means 8 Mapping table 9 Defective pixel pattern storage means

Claims (2)

  Two or more imaging means for photographing the periphery of the vehicle body, a defective pixel pattern storage means for storing a defective pixel pattern, an image captured by the first imaging means in the imaging means, and the defective pixel pattern Corresponding to the imaging range of the image captured by the first imaging means, and the second imaging means among the imaging means. A mapping table indicating an imaging range captured by the image capturing unit, and an image synthesis unit that interpolates a defective pixel area captured by the first imaging unit using a pixel captured by the second imaging unit using the mapping table. An image processing apparatus. It consists of a vehicle body and an image processing device installed on this vehicle body.
The image processing apparatus includes two or more imaging units that capture the periphery of the vehicle body, a defective pixel pattern storage unit that stores a defective pixel pattern, and the first imaging unit among the imaging units. A pixel recognition unit that recognizes a defective pixel region by comparing an image and a defective pixel pattern of the defective pixel pattern, and corresponds to an imaging range of an image captured by the first imaging unit, The mapping table showing the imaging range imaged by the second imaging unit, and the pixel imaged by the second imaging unit using the mapping table, the defective pixel area imaged by the first imaging unit. A vehicle comprising image composition means for interpolation.