JPH1127685A - Image pickup device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an image with high image quality by applying an arithmetic operation to pluralities of pixel data adjacent in a vertical direction as to each line of image pickup data respectively and generating a luminance signal from the processing result. SOLUTION: Two pixels adjacent in the vertical direction of each line are added as 1st line+2nd line, 2nd line+3rd line, and 3rd line+4th line, in the case of luminance signals and of raw data by one image pattern into data of a form of a received signal respectively. That is, pixels of lines with closer correlation in the vertical direction are added. Thus, the number of lines to produce a contour emphasis signal in the vertical direction on a concerned line and two upper/lower lines and a reference line, in total 4 lines are enough and detailed control is attained. That is, the processing sequence of pixel data in image pickup data in the case of generating the luminance signal and the processing sequence of pixel data in image pickup data in the case of generating color difference signals are different and the luminance signal is generated with the processing with closer correlation.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image pickup apparatus using a solid-state image pickup device such as a CCD, and more particularly to signal processing thereof.

[0002]

2. Description of the Related Art In a conventional image pickup apparatus using a CCD such as a video camcorder, when a digital signal processing circuit (DSP) is used to pick up a frame still image using an image pickup apparatus set for field reading, the following method is required. Such signal processing has been performed.

FIG. 8 shows an image pickup surface of a CCD.
(Cyan), Y (yellow), M (magenta), and G (green) color filters are arranged for each pixel as shown. Image data of one frame captured by the CCD is taken into a memory, and pixel data of the image data is combined as shown in FIG. Note that Wb = C + M,
It is assumed that Gr = Y + G, Gb = C + G, and Wr = Y + M. By combining them as shown in FIG. 9, first, an odd field is created, and then an even field is created.

[0004]

However, when the signal processing method described with reference to FIG. 9 is used, the adjacent pixels in the vertical direction are skipped by one pixel in order to process the odd and even fields separately. There is a problem that the correlation becomes distant, the vertical edge enhancement is biased toward the low frequency region, and false colors increase.

The present invention has been made to solve the above-described problems, and has as its object to provide an imaging apparatus capable of obtaining a high-quality image.

[0006]

According to the present invention, an image pickup means including an image pickup device in which a plurality of color filters are arranged in a predetermined manner for each pixel, and each line of image pickup data obtained from the image pickup means is provided in a vertical direction. And a luminance signal generating means for generating a luminance signal from an output of the arithmetic processing means.

[0007]

FIG. 6 is a block diagram showing a configuration of a digital still camera to which the present invention is applied. In FIG. 6, a subject image (not shown) includes an optical system 1 and an optical LPF.
(Low-pass filter) to form an image on the imaging surface of the CCD 3. The imaging signal as an electric signal photoelectrically converted by the CCD 3 is subjected to noise removal or the like by the CDS circuit 4 and then converted to digital imaging data by the A / D converter 5. One screen of this imaging data is stored in the memory 7 through the memory controller 6.

The raw data in the memory 7 is combined by the memory controller 6 into a signal format conforming to the specifications of the DSP 8 described later, and then input to the DSP 8 for signal processing. The processed data is further compressed by a JPEG circuit 9 and then recorded on a recording medium such as a flash memory 11 or a hard disk 12 via an I / F 10. The system controller 13 controls the memory controller 6 and the CCD driver 14.

FIG. 7 is a block diagram showing the structure of the DSP 8. The raw data for one screen stored in the memory 7 is converted by the memory controller 6 into a signal format conforming to the specifications of the DSP 8, and then input to the DSP 8 in FIG. The input data is a CMAT (color signal processing circuit) 1
5, and a luminance signal processing is performed by a circuit of HLPF (horizontal low-pass filter) 16 or less.

In CMAT15, Wb, Gr, G
R, G, B signals are generated from the b, Wr signals by matrix conversion. Thereafter, each of the R, G, and B signals becomes a color signal through the LPF 17. In the luminance signal processing, first, the HLPF
16 removes the chromatic color aliasing. Next, HAPC
A (horizontal outline emphasis circuit) 18 and a VAPC (vertical outline emphasis circuit) 19 generate horizontal and vertical outline emphasis signals. After each edge enhancement signal is added by the adder 20,
The low-level signal input to the BC (base clip) circuit 21 is clipped to zero. Then, the adder 22
Is added to the signal from the HLPF 16 to form a luminance signal. Then, the luminance signal and the chrominance signal are sent to the JPEG circuit 9 in FIG. 6, where they are compressed and recorded on a recording medium.

Next, a method of converting raw data for one screen into a signal format input to the DSP 8 will be described. FIG.
(A) is a diagram for explaining an array of CCD data for one screen obtained. FIG. 1B illustrates how to create data for generating a luminance signal. In the case of a luminance signal, the addition method in the vertical direction is as follows: the first line + the second line, the second line + the third line, the third line + the fourth line... ing. That is, lines having close correlations in the vertical direction are added.

For this purpose, the reference line for generating the vertical contour enhancement signal by the corresponding line and the upper and lower lines is C.
Only four lines are required on the CD surface, and fine control is possible. On the other hand, in the conventional example, as shown in FIG. 9, if an attempt is made to generate a contour enhancement signal in the vertical direction by the corresponding line and the upper and lower two lines, the reference line becomes a total of six lines on the CCD surface, and It becomes impossible to emphasize a fine contour in a direction.

Next, FIGS. 2A and 2B illustrate how to create data for generating a color signal. Since the order of sending color signals to the DSP 8 for color generation is determined, the method is basically the same as that of the conventional example. That is, signals are input into odd (a) and even (b) fields, and color signals are generated in the following order. odd
In the (a) field, the 1st line + 2nd line 3rd line + 4th line 5th line + 6th line ... In the even (b) field, the 2nd line + 3rd line 4th line + 5th line 6th line + 7th line ..., in the odd field, the data of the next even line is added to each odd line, and in the even field, the data of the next odd line is added to the even line.

That is, since the processing order of each pixel data in the image data in the case of generating a luminance signal is different from the processing order of each pixel data in the image data in the case of generating a color signal, the processing order of the luminance signal is different. It can be generated by processing having a close correlation.

Next, as another embodiment, a case where a CCD 3 having a color filter array different from that of the above embodiment is used will be described. FIG. 3 shows the color filter array, which can generate and process color signals as in the array of the above-described embodiment even when thinning in the vertical direction is performed.

FIG. 4 illustrates how to generate a luminance signal in the filter array. The luminance signal is generated by adding two pixels in the vertical direction as in the above embodiment. Of course, the combination is not the interlaced combination (the correlation in the vertical direction is far) as in the conventional example, but the combination having the close correlation in the vertical direction as in the above embodiment. Therefore, the contour in the vertical direction can be clearly reproduced.

FIG. 5 illustrates how to generate a color signal. The color signals need to be input to the DSP 8 in a predetermined combination, and simply adding in the vertical direction does not result in a predetermined signal. Therefore, interpolation is performed in the horizontal direction so that a predetermined combination is obtained. For example, the Wr signals of the second horizontal pixel and the vertical 2 + 3 pixel are generated by adding the adjacent M of the second row and the adjacent Y of the third row in the horizontal direction. That is, Wr = (M12 + M32 + Y13 + Y33) / 2 is generated. Similarly, the third horizontal pixel, the vertical 2 + 3
The Gb signal of the pixel is generated by horizontally adding the adjacent G in the second row and the adjacent C in the third row. That is, Gb = (G22 + G42 + C23 + C43) / 2

As described above, signals of an appropriate combination are input to the DSP 8 by interpolation processing to generate color signals. By performing the interpolation processing, a slight reduction in the band in the horizontal direction can be expected. However, in practice, band limitation is usually performed at the stage when a color signal is formed, and this does not cause a serious problem. On the other hand, S / N may be improved by addition.

In this manner, by generating a luminance signal and a color signal for the entire screen in a time-division manner and then combining them, so-called R, G, and B plane image data is obtained. Needless to say, when it is better to separate luminance and color as in JPEG compression, the compression processing may be performed without synthesis.

In each of the above-described embodiments, the signal of each line is created by adding two pixels in the vertical direction to generate a signal of each line. May be created, and these processes may be selectable.

[0021]

As described above, according to the present invention, a high-quality image can be generated. In particular, a high-quality image can be obtained when a frame image is captured by an imaging device having an existing digital signal processing circuit for field reading.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a color filter array of a CCD and generation of a luminance signal according to an embodiment of the present invention.

FIG. 2 is a configuration diagram illustrating generation of a color signal.

FIG. 3 is a configuration diagram showing a color filter array of a CCD according to another embodiment of the present invention.

FIG. 4 is a configuration diagram illustrating generation of a luminance signal.

FIG. 5 is a configuration diagram illustrating generation of a color signal.

FIG. 6 is a block diagram of an imaging device according to the present invention.

FIG. 7 is a block diagram of a digital signal processing circuit.

FIG. 8 is a configuration diagram showing an arrangement of color filters of a conventional CCD.

FIG. 9 is a configuration diagram showing conventional signal processing.

[Explanation of symbols]

 Reference Signs List 3 CCD 6 Memory controller 7 Memory 8 Digital signal processing circuit (DSP) 13 System controller 15 Color signal processing circuit 16 Horizontal low-pass filter 18 Horizontal outline enhancement circuit 19 Vertical outline enhancement circuit

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Hideaki Yakimaki, 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc. (72) Ryoji Kubo 3-30-2, Shimomaruko, Ota-ku, Tokyo Canon (72) Inventor Takao Sasakura 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc. (72) Inventor Nobuhiro Takeda 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc. ( 72) Inventor Masato Ikeda 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc.

Claims (5)

[Claims] 1. An image pickup device including an image pickup device in which a plurality of color filters are arranged in a predetermined manner for each pixel, and two pixel data vertically adjacent to each line of image data obtained from the image pickup device. An image pickup apparatus, comprising: an arithmetic processing unit for performing arithmetic processing by means of an arithmetic processing unit; and a luminance signal generating unit for generating a luminance signal from an output of the arithmetic processing unit. 2. The method according to claim 1, wherein the arithmetic processing is an addition processing, a subtraction processing,
The imaging apparatus according to claim 1, wherein the imaging apparatus is any one of an averaging process and a rearrangement process. 3. The imaging apparatus according to claim 1, wherein said luminance signal generating means generates horizontal and vertical contour emphasizing signals from an output of said arithmetic processing means and adds the signals to the output. 4. An image pickup apparatus according to claim 1, further comprising a color signal generating means for generating a color signal by processing each pixel data of said image pickup data in a predetermined order. 5. The color signal generation means generates a color signal of an odd field by adding a next even line to an odd line of the imaging data, and adds a next odd line to an even line of the imaging data. The imaging apparatus according to claim 4, wherein the color signal of the even field is generated by performing the operation.