JP3270620B2 - Image input method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image input method for capturing a two-dimensional image of a three-dimensional object.

[0002]

2. Description of the Related Art Conventionally, when capturing a three-dimensional object as a two-dimensional image, an area sensor (or a line sensor) such as a CCD is used to record two-dimensional information input from an optical system. Further, even if a function for obtaining position information of a three-dimensional object is provided at the same time, the two-dimensional information is not corrected using the information.

[0003]

In order to capture a three-dimensional object as a two-dimensional image, the three-dimensional object must be captured as a two-dimensional image on a sensor, for example, through an optical system. However, the optical system is not a parallel projection but a perspective projection, and does not form an image at a single point except for the focal position. Therefore, for example, when a plurality of sensors are arranged at the time of inputting a color image (for example, three sensors of red, green, and blue are provided as sensors), the deviation from the focus is recorded as the deviation on the image sensor pixel, As a result, when they are integrated, a color shift occurs at a position other than the in-focus position. This is remarkable at the time of inputting a high-definition image, and there is a drawback that a part of the two-dimensional image cannot be viewed even if the two-dimensional image is partially enlarged at a position other than an in-focus position at the time of imaging.

SUMMARY OF THE INVENTION The present invention has been made to solve such a problem, and an object of the present invention is to enable high-definition image input and to correct a color shift in an unfocused portion. It is an object of the present invention to provide an image input method.

[0005]

[0006]

[0007]

According to the present invention, when a three-dimensional object is imaged as a two-dimensional color image, first, a distance image of the three-dimensional object to be imaged and red, Green and blue color images,
Next, the position of the distance image is corrected based on the physical arrangement of the measuring unit and the lens of the optical system, and the distance image is arranged at the same position as the color image. Calculate the shift in each pixel of red, green, and blue according to the value of the distance image for the pixel, and extract the red, green, and blue pixels from the red, green, and blue color images according to the calculated shifts. An image input method for synthesizing and inputting color images is used as means.

[0008]

According to the present invention, in a two-dimensional image obtained by imaging a three-dimensional object with an optical and imaging system, the difference in position from the optical and imaging systems due to the three-dimensional shape is determined by the position information of the three-dimensional object to be imaged. The displacement of the imaged pixels caused by the above is corrected by correcting the characteristics of the optical and the image pickup device arrangement, thereby enabling high-definition image input. For example, in a color image input, an image without color shift can be input. And

[0009]

Embodiments of the present invention will be described below in detail with reference to the drawings.

In this embodiment, it is assumed that a simple lens is used as an optical system, and a plurality of linear sensors (each of which is red, green, and blue) are used as an image pickup system. Is assumed to be a range sensor using infrared laser light. However, the present invention does not operate only in the case of the above-described configuration, and other means may be used to obtain the position information of the three-dimensional object, and other means may be used as the optical system and the imaging system. It does not matter.

FIG. 1 shows an image pickup system, an optical system, and a three-dimensional object to be picked up in this embodiment.

Here, since a system using a linear sensor is used as an example, the optical and imaging system 109 inputs a two-dimensional image of the object 101 while moving in the moving direction 110.

Here, as shown in FIG. 1A, using a red sensor 102, a green sensor 103, and a blue sensor 104,
A color image is input. At this time, the three images that arrive at these three sensors through the lens 105 are the red imaging position 108 and the green imaging position 10 on the imaging target 101, respectively.
7. The information of the blue imaging position 106, and each sensor captures the information of each of them.

As shown in FIG. 1B, for example, if the position of the object 101 to be imaged is different, the imaging position of each object is also different.

Therefore, the red sensor 102, the green sensor 103, and the blue sensor 104 depend on the shape of the object 101 to be imaged.
Will change, and this will be a color shift when a color image of a three-dimensional object is input.

Therefore, by providing the range sensor 111 in the optical and image pickup system 109, the distance 112 between the lens 105 and the image pickup object 101 is measured.

This is repeated until the optical and imaging system 109 moves in the movement direction 110 and captures all the objects to be imaged 101.

FIG. 2 shows a distance image 113 input by the range sensor 111 and a red image 11 input by the red sensor 102.
4, a green image 115 input by the green sensor 103 and a blue image 116 input by the blue sensor 114. here,
As shown in FIG. 1B, the distance image 113 is subjected to position correction based on the physical positional relationship between the range sensor 111 and the lens 105, and thereby the central axis of the lens 105 (optical axis 100).
An image capturing the object to be shot 101 can be obtained. However, the red image 114, the green image 115, and the blue image 116 are slightly shifted from each other depending on the physical positions of the red sensor 102, the green sensor 103, and the blue sensor 104.

FIG. 3 is a diagram in which a situation in the characteristics of a certain optical and photographing system 109 is modeled. Measurement distance 11
7 can be obtained from the distance image 113 as a three-dimensional position in the image. At the position distant from the optical and imaging system 109 by the focal length 118, the red image 114, the green image 115, and the blue image 116 capture the same position, but at other distances, they differ by a shift 119. This shift is caused by the lens 10 of the optical and imaging system 109.
5, the characteristic 120 can be considered as a slope. Therefore, if the measurement distance 117 is L, the focal length 118 is F, and the characteristic 120 is the inclination f, the displacement d at a certain position on the imaging target 101 can be obtained as d = f * (FL). .

Therefore, by combining the pixels of the red image 114, the green image 115, and the blue image 116 according to the shift d in accordance with the value of the distance image 113, a two-dimensional color image without color shift is obtained. be able to.

FIG. 4 shows a processing flowchart of this embodiment. Optical properties, red, green, and blue, respectively, f _r, f _g, f
_It is known as _b . First, a distance image and a color image of an imaging target are acquired by an optical system and an imaging system (step S1). Next, the position of the range image is corrected based on the physical arrangement of the range sensor and the optical lens, and the range image is arranged at the same position as the color image (step S2). Next, for all the pixels according to the resolution of the color image (or the distance image), (1) The shift in each of the red, green, and blue pixels is calculated according to the value of the distance image (step S3).

[0022] _{d r = f r * (F} -L) d g = f g * (F-L) d b = f b * (F-L) (2) each of the deviation d _r, d _g, d _b According to the red,
Green, respectively d _r for each blue, d _g, synthesized pull out the pixels of position shifted d _b (step S4). Finally, the combined color image is output (step S5). This is the end.

FIG. 5 shows an example of correction according to this embodiment.
FIG. 5A is a schematic view of a color image 121 before correction in which the red image 114, the green image 115, and the blue image 116 are integrated before correction based on the position of the three-dimensional object, and FIG.
The two-dimensional color image 122 after correction and a part of the corrected color image 122 (brightness of an eyeball) are shown. In the color image 121 before the correction, a color shift occurs, the definition is reduced, and the glow of the eyeball is not seen. On the other hand, in the corrected color image 122 according to the present embodiment, the color misregistration is corrected, and the glow of the eyeball is clearly visible.

[0024]

As is apparent from the above description, according to the present invention, the three-dimensional object is corrected by correcting the characteristics of the optical system and the image pickup device arrangement using the position information of the three-dimensional object to be imaged. for correcting the two-dimensional image by correcting the deviation of the color pixels that occur in different positions, other than the focal position
However, a high-definition color image without color shift can be input.

[0025]

[0026]

[Brief description of the drawings]

FIGS. 1A and 1B are diagrams showing an imaging system, an optical system, and a three-dimensional object for explaining an embodiment of the present invention;

FIGS. 2A, 2B, 2C, and 2D are diagrams showing input images of a range sensor in the embodiment.

FIG. 3 is a characteristic model diagram of an optical system and an imaging system in the embodiment.

FIG. 4 is a flowchart showing a processing example in the embodiment.

FIGS. 5A and 5B are diagrams showing an example of correction according to the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 101 ... Imaging target 102 ... Red sensor 103 ... Green sensor 104 ... Blue sensor 105 ... Lens 106 ... Blue imaging position 107 ... Green imaging position 108 ... Red imaging position 109 ... Optical and imaging system 110 ... Moving direction 111 ... Range sensor 112 ... Distance 113 ... Distance image 114 ... Red image 115 ... Green image 116 ... Blue image 117 ... Measurement distance 118 ... Focal length 119 ... Displacement 120 ... Characteristics 121 ... Color image before correction 122 ... Color image after correction

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-2-23790 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06T 1/00 400 G06T 3/00 200

Claims (1)

(57) [Claims] When capturing a three-dimensional object as a two-dimensional color image, first, a distance image of the three-dimensional object to be captured and red, green,
Obtain a blue color image, then perform the position correction of the distance image from the physical arrangement of the measuring means and the lens of the optical system, and arrange it at the same position as the color image,
According to the resolution of the color image or the distance image, the shift of each pixel of red, green, and blue is calculated according to the value of the distance image for all pixels, and the red, green, and blue color images are calculated according to the calculated shifts. An image input method comprising: extracting red, green, and blue pixels from a color image; synthesizing a color image;