JPS62159581A - Image pickup device - Google Patents

Abstract

PURPOSE:To perform interpolating with high accuracy by finely displacing an object image and a photodetecting picture element relatively. CONSTITUTION:Potentials V1 and V3 are changed over and an oscillating voltage of an amplitude V4 is impressed to the respective potentials V1, V3 and in this manner, by applying the fine oscillation, a high resolution is further realized. Namely, if a picture element output in the position of the potential V1 and a picture element output in the position of the V3 are simply operated, the interpolation can be performed. However, this method is only applied to the object changing linearly from the position of the V1 to the position of the V3, and if the fine oscillation is applied, the interpolation can be performed with high accuracy.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a solid-stead high-resolution imaging device for increasing the resolution of a finite aperture imaging device. [Prior Art] Conventionally, Like a bimorph COD, a piezo element is used to vibrate a solid-state image sensor at a pitch of half a pixel, increasing the resolution to 2.
Mechanisms such as a doubling method, or a doubling method in which two rows of pixels are superimposed to receive an image, as disclosed in Japanese Patent Application Laid-Open No. 1982, are known.

[Problem to be solved by the invention]

However, either method only doubles the resolution. If the piezo vibration cycle is limited or a polarizing plate is used, the image receiving sensitivity of the image sensor will deteriorate. Furthermore, it has many mechanical parts, is expensive, and requires precision in positioning and assembly. Furthermore, increasing the resolution of a solid-state imaging device not only involves technical difficulties, but also involves deterioration of characteristics and a significant increase in cost.

It is an object of the present invention to eliminate such conventional drawbacks and at the same time provide an imaging device that can measure high resolution simply by causing minute vibrations by deflecting a light beam.

[Means for solving problems]

The present invention includes a plurality of light-receiving pixels arranged at a predetermined pitch for sampling a subject image, a control means for vibrating the plurality of pixels and the subject image relatively at a pitch smaller than the pitch, and This is a calculation means that calculates the output and interpolates the signal of the light receiving pixel.

[For production]

With this configuration, it is possible to interpolate pixel signals with high precision by just making a minute vibration without making a large displacement, and by calculating the vibration output at this time.

〔Example〕

1 to 3 show examples of the present invention, and 21 is a light deflection element (
KDP (KH2PO4), ADP with electro-optic effect
(NH4H2POa).

In a crystalline material such as LiNbO3, the crystal axis is arranged along the Z-axis direction. ) and 2 are made of the same material as 1, with the crystal axis direction opposite to 1 with respect to the Z axis, and are cut in the diagonal direction of the X, Y plane and bonded to 1. The shape of 1.2 joined is, for example, if the thickness in the Y direction is 10 m + s,
X, Z direction husband//30111 (7) Rectangular plate, etc. 3 and 4 are bidirectional 'electrodes that switch between + and -;
Reference numeral 5 denotes an imaging lens, and 6 an imaging element, on the surface of which a plurality of light-receiving pixels are arranged at regular pitches.

Each light-receiving pixel performs photoelectric conversion to sample the subject image, 7 is a control circuit for supplying drive signals to the optical deflection elements 1 and 2, and 8 is an arithmetic operation for processing the output of the image sensor 6. It is a processing circuit.

When the potential between the electrodes 3 and 4 is changed as shown in Figure 2, the incident light is deflected in the X direction centering on the Z axis and enters the image sensor 6, so that it is incident on a predetermined pixel in the image sensor. The sample position of the subject changes from A to B. Here, if the amount of displacement from A to B is selected as the station of the pitch of each pixel, 1 potential ■ 1 and V
By simply switching the voltage between 3 and 3, the same effect as doubling the number of pixels can be obtained.

In this embodiment, in addition to such voltage changes, the aim is to achieve even higher resolution by adding modulation as shown in FIG.

That is, in addition to switching the potentials of ■ and v3, this vl
, v3, an oscillating voltage of amplitude V4 is applied at each potential.

This voltage amplitude v4 is smaller than (Vl - V3),
This is because it is extremely difficult to make the change in v4 more than (VI V3) from the viewpoint of responsiveness.

In this embodiment, higher resolution is achieved by adding minute vibrations in this manner. That is, interpolation is possible by simply calculating the pixel output at the position of potential VI and the pixel output at the position of potential V3. However, this method can only be applied to objects where the subject changes linearly from the position of vl to the position of V3, whereas adding minute vibrations as in this example allows for the following higher precision Capturing m1 becomes possible.

For example, assuming that point C, an intermediate point between points A and B, is on an n-dimensional function that includes points A and B, the zero point can be approximated by this n-dimensional function.

For example, if n=2, the signal level at point 0 y (c)
can be interpolated using the following equation.

y (c) = a d 2 + b d + e
----- (1) In this way, by simply causing minute vibrations at points A and B, more accurate n-th approximation is possible. Moreover, the width of this minute vibration may be smaller than the pitch between points A and B.

The arithmetic processing circuit 8 performs the calculations of equations (1) to (9) above to obtain a signal corresponding to the signals from the four light-receiving pixels from each light-receiving pixel.

Specifically, from the control circuit 7, v=v1+v4srnφ
While the signal is being applied, the accumulated signal from the image sensor is read out, A/D converted, and then processed. n tf f
;1 Tsubaki + Next, with V=V3 +V4 sinφ applied between 'II2 poles 3 and 4, the accumulated signal from the image sensor is read out, A/D converted, and stored in another area of the memory. .

Next, while reading the signals of each pixel stored in each area, calculations of equations (+) to (9) are performed on each other. That is, by performing the above calculation between point A and point B of the same light-receiving pixel, point A of the predetermined pixel and point B of the light-receiving pixel adjacent to the point A of the predetermined pixel, and point A of the pixel adjacent to the point B of the predetermined pixel, 1. Four times as many pixel signals can be obtained using one pixel signal.

In the above explanation, the resolution is increased only in one direction, but it is also possible to perform such image shifting and micro-vibration in two orthogonal directions.By doing so, the detection rate will be increased by 8 times. It becomes possible to perform freezing.

Also, in the example, the image was moved relative to the light-receiving pixel, but
This need only be relative, and it goes without saying that the light-receiving pixel side may be moved using a bimorph or the like. or,
The means of movement for this purpose is not limited to bimorphs or electro-optical elements.

Additionally, as an additional effect of the present invention, even if there is low frequency noise in the light receiving pixel, it is canceled out by the above-mentioned minute vibration! -19, which has the effect of preventing moiré.

〔effect〕

Compared to conventional simple signal interpolation, by making small relative displacements between the subject image and the light-receiving pixels.

Interpolation can be performed with higher precision.

[Brief explanation of drawings]

FIG. 1 is a side view of the configuration of an imaging device according to the present invention, FIG. 2 is a partial signal waveform diagram, and FIG. 3 is a detailed signal waveform diagram. 1.2----1 deflection element 6--m-imaging element 8-111 arithmetic processing circuit

Claims (1)

[Scope of Claims] A plurality of light-receiving pixels arranged at a predetermined pitch for sampling a subject image, a control means for relatively vibrating the plurality of pixels and the subject image in a width smaller than the pitch, and the light-receiving device. An imaging device comprising a calculation means that calculates the output of the pixel and interpolates the signal of the light-receiving pixel.