1. Technical Field
The present invention relates to a digital camera equipped with a solid-state imaging device (CCD, CMOS and the like) having a photoelectric conversion function.
2. Related Art
In recent years, through popularization of digital cameras (digital still cameras, digital video cameras, mobile cameras and the like), digitalization of photographic imagery can be done readily. Particularly, there are increasing opportunities of treating photographic imagery as digital image data on personal computers.
Further, through popularization of application software which can correct or process an image, a user can freely perform image processing on the personal computer. For example, correction processing to brighten an image photographed darkly and processing treatment to make wrinkles inconspicuous by applying LPF to a flesh-color area can be exemplified.
From such background, there is a need for an image processing technique that eliminates the need for correction and processing treatment of imagery during shooting.
In recent years, it has been becoming possible to detect a person's face from an image and the accuracy of the detection is improving as well. As an effect of facial detection, correction of luminance of a facial portion in correcting backlight imagery is considered effective.
In view of this, when making an album or a list of photos of individual faces, there is a need for improved visibility by shooting and performing image processing so as to assure uniform levels of luminance (brightness) and gradation under different photographic conditions.
For example, Japanese Unexamined Patent Application Publication No. 2000-354196 discloses a technique of providing proper exposure of a photographic subject through optimum exposure control. Japanese Unexamined Patent Application Publication No. 2006-18465 discloses a technique of extracting a facial area and changing its level of luminance.
In prior art techniques, proper exposure and diaphragm control are conducted according to photographic conditions. But since there are no uniform standards regarding the level of luminance signal of the photographic subject, the brightness of the photographic subject varies under different photographic conditions. Further, even if the same level of brightness is set with respect to different photographic conditions, assignment of gradation is provided to the entire surface of the image so that the gradations of the photographic subjects differ from each other.
In the digital camera of the present invention, exposure adjustment, diaphragm control, gain control, and gradation correction are performed such that the luminance level of a specified area (flesh-color area, AF area, facial area by facial recognition and the like) is on a uniform level regardless of the photographic conditions. While the gradation of an area whose luminance level is to be changed is maintained, the luminance level of the area is changed (the gamma characteristic is changed), thereby retaining high-definition of the specified area. When an album shooting mode is entered by means of the mode dial, a frame of a face is shown by OSD. By placing a face in that frame and shooting, the size of the face is fixed. Further, the above-mentioned processing is applied to that area.
According to the digital camera of the present invention, since the luminance level of the photographic subject remains constant even if the photographic conditions differ, it becomes easier to make a comparison of photographic subjects (faces, upper halves of bodies, food samples and the like). Further, when carrying out gain control, the gradation in the vicinity of the photographic subject is retained, and hence, the high-definition of the photographic subject is retained, thereby improving recognition of the photographic subject. Still further, the frame is shown by the OSD, and by placing the photographic subject in that frame and shooting it, the shooting can be performed at a constant size regardless of the size of the photographic subject and the photographic conditions. Moreover, it becomes easier to compare photographic subjects.
With reference to the drawings, the embodiments of the present invention will be described as follows.
A construction of a digital camera according to a first embodiment is shown in
The imaging device 2 converts light from a photographic subject by means of a solid-state imaging devices such as a CCD and a CMOS into a video signal RAW1 corresponding to R, G, B and outputs the video signal RAW1. The video signal adjustment portion 3 subjects each of the color components (R, G, B) of the video signal RAW1 from the imaging device 2 individually to gain adjustment, white balance adjustment, offset addition, scratch correction, and the like, and outputs RAW1 as an adjusted video signal RAW2. The image data conversion portion 4 converts the adjusted video signal RAW2 from the video signal adjustment portion 3 into image data YC1. The OSD overlap portion 5 overlaps an OSD (On Screen Display) on the image data YC1 and outputs it as an OSD overlap signal YC2. The display device 6 displays the OSD overlap signal YC2, e.g., a liquid crystal display or an organic EL display. The area detection portion 7 detects an area to control (a facial area, a flesh-color area, and the like) from the image data YC1. The area specifying portion 8 obtains information on the area detected by the area inspection portion 7, and specifies a gain adjustment area and the like in the video signal adjustment portion 3 based on the information. The luminance signal detection portion 9 detects a luminance signal from the adjusted video signal RAW2. The gradation assignment portion 10 gives an instruction to the image data conversion portion 4 on assignment of gradation based on the information detected by the luminance signal detection portion 9. The switch 11 gives instructions on selection of the shooting mode, shooting start, shooting end, ON/OFF of display output, change in gain adjustment value, and the like. The setting holding portion 12 holds settings (ISO sensitivity, diaphragm, exposure time and the like). The control device 13, upon switching of the shooting mode, reads a prior setting of the switched shooting mode from the setting holding portion 12 and controls each of the above-mentioned portions.
Since the gradation of the image data YC1 is generally less than the gradation of the video signal RAW1, as shown in
At this point, if a target mean luminance is set for the facial area (specified area), then, as shown in
Further, if the gradation assignment curve is no longer smooth due to the gradation assignment, gain adjustment and the like, in order to prevent effects at a bend point from appearing significantly in images, the bend point may be approximated to a non-linear curve as shown in
Furthermore, when adding an offset, it is presumed that an addition of an offset only to the specified area (facial area) may result in unnaturalness since continuity to peripheral image areas is lost. The gradation assignment curve of
Further, a luminance signal threshold value may be set so that the luminance signal detection portion 9 conducts a detection operation only to a portion where out of the luminance signals in the specified area (facial area and the like), the luminance signal does not surpass the luminance signal threshold value. An image concept diagram of this instance is shown in
Still further, it may also be acceptable to overlap the specified area on the OSD in the form of a frame (OSD frame) indicating the specified area as shown in, for example,
At this point, if the video signal RAW1 which the imaging device 2 outputs is in complimentary colors (Ye, Cy, Mg, and G), then as shown in
A construction of a digital camera according to a second embodiment is shown in
Normally, when performing AF control, there are several AF areas (focus areas) to focus on. In this case, seven square frames (AF frames) in
A construction of a digital camera according to a third embodiment is shown in
Even in the case where exposure time change and gain adjustment cannot be performed in the imaging device 2 with respect to an image whose axis of time is shifted, such as correcting an image at an instant the shutter is pressed, since the memory access portion 31 stores the video signal RAW1 in and takes it out of the memory device 30, by loading the video signal RAW1 once on the memory device 30, the image data YC1 can be generated as often as desired with respect to the same video signal RAW1. Consequently, it is possible to approach the target image data closer.
It should be noted that the present invention is not confined to the above-mentioned embodiments but may be implemented as follows.
(1) In the above-mentioned embodiment, the formulas Ye=R+G, Cy=B+G, Mg=R+B, and G, are used for conversion from a complimentary color to a primary color. But it is not limited to this. For conversion, coefficients a to f may be used such as Ye=aR+bG, Cy=cB+dG, Mg=eR+fB, and G.
(2) In the above-mentioned embodiment, overlap data OSD1 overlaps the image data YC1, but a value of the image data YC1 itself may be adjusted.
(3) In the above-mentioned embodiment, area recognition is made from the image data YC1 in facial recognition and flesh-color recognition. However, area recognition may be performed from the video signals RAW1 and RAW2.
(4) In the above-mentioned embodiment, offset adjustment and the like are performed with respect to the target mean luminance. However, as shown in
(5) In the above-mentioned embodiments, there is described an example of providing the AF control portion 20 which performs AF control. In lieu of this, as shown in
(6) In the above-mentioned embodiments, the mean luminance of the luminance signal is adjusted through offsetting by the video signal adjustment portion 3. However, it is acceptable to adjust the mean luminance by adjusting exposure time and gain in the imaging device 2.
(7) In the above-mentioned embodiments, luminance signal detection is performed with respect to the adjusted video signal RAW2. However, as shown in
(8) In the above-mentioned embodiment, correction is made linearly in changing gradation assignment in
According to the digital camera of the present invention, the luminance level of a face (specified area) is made constant regardless of photographic conditions, thus facilitating to compare photographic subjects (e.g., faces and upper body halves). Further, when performing gain control, maintaining gradation in the vicinity of a face contributes to keeping the high-definition of the face and improving the recognition of the face. Furthermore, by showing a frame of the face through OSD, placing the face in the frame, and shooting it, it is possible to take a picture of a face in a preset size regardless of the size of a person's face or photographic conditions. Moreover, it is made easier to compare photographic subjects. From the foregoing, when making an album or a list using facial photos, shooting and performing image processing so as to obtain uniform levels of luminance (brightness) and gradation makes it possible to improve visibility.
Number | Date | Country | Kind |
---|---|---|---|
2006-299372 | Nov 2006 | JP | national |
This non-provisional application claims priority based on Patent Application No. 2006-299372 filed in Japan on Nov. 2, 2006, the entire contents of which are hereby incorporated by reference.