JPWO2014141891A1 - Imaging device - Google Patents

Abstract

To provide a television camera for controlling a lens iris so that a luminance level of an arbitrary hue portion designated by a user automatically becomes a constant level. In an imaging apparatus having a function of detecting a specific hue for each pixel and correcting the hue and a function of controlling an iris of a lens, a menu for setting the imaging apparatus or an arbitrary hue area in the specific hue An image display unit that displays an image, and the image display unit selects an arbitrary hue region in the specific hue, and a lens so that the luminance level of the arbitrary hue region in the specific hue is constant. Control the iris.

Description

  The present invention relates to an imaging apparatus, and more particularly to lens iris control.

An auto iris control system that automatically controls the iris of a lens so that the brightness level is constant is known in order to reduce the burden of camera operation on the cameraman (see Patent Document 1).
Conventionally, detection is performed based on a specific screen area, such as center-weighted metering, which detects the brightness level at the center of the screen, and split metering, which divides the screen into multiple blocks. . In recent years, an auto iris control system has appeared that selects a person or a pet in a subject using a touch panel or the like and performs iris control so that the luminance level of the object is constant.

  However, with these methods, in a television camera system that switches between multiple cameras such as sports broadcasts, the brightness of the color parts that are easy for the viewer to focus on, such as the blue color of the tennis court and the green color of the soccer ground. When it is going to be constant, since brightness changes for every position of a camera by the sunlight condition etc., it is difficult to make brightness constant by the detection method based on the conventional screen area.

  On the other hand, the television camera has a function of detecting a specific hue for each pixel and correcting the specific hue for each pixel, which is called 6-color independent masking or 12-color masking (see Patent Document 2). ).

JP-A-8-181907 JP-A-9-247701

  In order to solve this drawback, the present invention controls the lens iris so that the luminance level of a specific hue portion such as the blue color of the tennis court or the green color of the soccer ground is constant, not the detection based on the conventional screen area. It aims at realization.

In order to solve the above object, the present invention provides means for detecting a specific hue for each pixel (such as 6-color independent masking and 12-color masking) (and correcting a specific hue for each pixel) and a lens iris. And a means for detecting a luminance signal level of a pixel in an arbitrary hue region in the specific hue, and a luminance signal level of the pixel in the detected arbitrary hue region is constant. Means for controlling the iris of the lens to detect the luminance signal level of a pixel in an arbitrary hue region in the specific hue, and the luminance signal level of the pixel in the detected arbitrary hue region is The imaging apparatus is characterized in that the iris of the lens is controlled to be constant.
Further, in the imaging apparatus, the (R, G, B) primary color hue region and (Ye, Cy, Mg) are detected by calculating the color difference (RG, RB, GB) and detecting the color difference signal level. Detecting means for detecting a luminance signal level of a pixel in an arbitrary hue area within a specific hue in the complementary hue area, color difference calculation (of RG, RB, GB) and the color difference signal level Detection and primary color, complementary color, and intermediate color constant selection, any hue within a specific hue in the primary hue area (for R, G, B), complementary hue area (for Ye, Cy, Mg), and intermediate hue area An image pickup apparatus having at least one of detection means and detection means for detecting a luminance signal level of a pixel in a region.
The imaging apparatus further includes an image display unit (such as a view finder or a monitor display) that displays a setting menu of the imaging apparatus or an arbitrary hue region within the specific hue, and the image display unit includes the specific display. Means for selecting an arbitrary hue area in the hue of the image, and displaying the setting hue menu of the imaging apparatus or an arbitrary hue area in the specific hue on the image display unit, and the specific hue It is an imaging device that selects.

  According to the present invention, it is possible to control the lens iris so that the luminance level of a specific hue portion such as blue of a tennis court or green of a soccer ground is constant.

The block diagram which shows one Example of the television camera of this invention Schematic diagram showing R / G / B magnitude relationship and corresponding hue range The block diagram which shows the structure of the hue detection correction | amendment part of one Example of this invention. Explanatory drawing of the hue area | region in the color tone correction | amendment of one Example of this invention The conceptual diagram of the hue area of one Example of this invention Explanatory drawing of calculation principle of primary color component and complementary color component of one embodiment of the present invention Explanatory drawing of the color tone correction process by the 6 color independent color tone correction method of one Example of this invention. Correction characteristic diagram of one embodiment of the present invention The block diagram which shows the structure of the hue detection correction | amendment part of other one Example of this invention. Explanatory drawing of the hue area | region in the color tone correction of other one Example of this invention. Conceptual diagram of a hue area according to another embodiment of the present invention. Explanatory drawing of calculation principle of primary color component and complementary color component of another embodiment of the present invention Correction characteristic diagram of another embodiment of the present invention The schematic diagram which shows the menu screen of one Example of this invention

An embodiment of the present invention will be described below with reference to FIGS.
FIG. 1 is a block diagram showing an embodiment of a television camera of the present invention. Incident light from the subject is imaged by the lens unit 31, and the formed incident light is decomposed into red light, green light, and blue light by the prism unit 32 of the television camera 30, and each is a CCD (Charge Coupled Device) unit. Photoelectric conversion is performed by 33R, 33G, and 33B. The photoelectrically converted R / G / B signal is subjected to correlated double sampling, gain correction, and analog-to-digital conversion by an AFE (analog front end processor) 34, and a video signal processing unit 35 having a hue detection correction function. Various video signal processing such as color correction, contour correction, gamma correction, knee correction, and the like are performed.

After various video signal processing and the like are performed in the digital signal processing unit 5, Y = 0.2126R + 0.7152G + 0.0722BPb = 0.5389 (B−Y) Pr = 0.6350 (R−Y) R / G / B is converted into a luminance signal (Y) and a color difference signal (Pb / Pr). Then, it is converted into a serial video signal by the parallel-serial converter 7 and output to the outside.

  A CPU (Central Processing Unit) 39 controls each part of the television camera 1. In addition, the image display unit 40 of the viewfinder or the monitor display displays a setting menu of the imaging apparatus and an arbitrary hue region in the specific hue.

  Here, the hue detection / correction unit 38 in the video signal processing unit 35 with the hue detection / correction function shown in FIG. 3 of the block diagram showing the configuration of the hue detection / correction unit according to the embodiment of the present invention is the R / G / B. The hue range in which the color of the subject is detected is detected from the magnitude relationship between the signal levels. FIG. 2 shows the hue range corresponding to the magnitude relationship of R / G / B. Although the hue is displayed in 6 divisions here, the hue can be further subdivided by further subdividing the magnitude relationship between the signal levels of R / G / B.

  The CPU 9 passes information on an arbitrary hue range set by the user to the hue detection correction unit 38 in the video signal processing unit 35 with a hue detection correction function, and the hue detection correction unit in the video signal processing unit 35 with a hue detection correction function. 38 passes the area information that matches the user-set hue range to the CPU 9. Based on the area information, the CPU 9 gates the luminance signal from the matrix conversion unit 36 in the video signal processing unit 35 with the hue detection correction function, and the luminance signal level of the area in the arbitrary hue range set by the user is The iris of the lens 31 is controlled so as to be an arbitrary level set by the user. When the set hue range area is out of the subject, the iris control is stopped (the previous state is maintained) until the hue range area comes into the subject again.

In the viewfinder or monitor display 40, the menu screen is superimposed on the subject image, and the user sets the hue range and the luminance signal level while viewing the menu screen. In addition, in order to check whether the hue range set by the user matches the target subject's color, it is superimposed on the 40 subject video in the viewfinder or monitor display and matches the set hue range Markers may be displayed in these areas.
FIG. 14 is a schematic diagram showing a menu screen according to an embodiment of the present invention. In TARGET LEVEL, the luminance signal level that is the convergence target of iris control is set in the range of 0 to 100%. COLOR DEPEND performs conventional screen area detection type iris control when set to OFF, and performs hue-dependent type iris control of the present invention when set to ON. When set to ON, the hue range (R-Mg, Mg-B, B-Cy, Cy-G, G-Ye, Ye-R) and each hue range in PHASE below it are set. Are further subdivided into 1 to 100 hues. When MARKER below is turned ON, a striped marker is superimposed on the video portion that matches the actually selected hue, making it easy to select the target color for auto iris control. I can do it.

  As described above, according to the present invention, by controlling the lens iris so that the luminance signal level in an arbitrary hue range set by the user is constant, the brightness of the tennis court blue, the soccer ground green, etc. Can be made constant automatically.

Hereinafter, another embodiment of the present invention will be described with reference to FIGS.
In FIG. 3, first, the arithmetic / comparators 1, 2, and 3 calculate the color difference signals RG, RB, and GB from the input video signals R, G, and B, and compare the magnitudes thereof. Are supplied to the hue region determination circuit 4 and the primary color component amount and complementary color component amount calculation circuit 5.
Therefore, the hue area determination circuit 4 first determines the hue area as shown in FIG. 5 based on the calculation results of the calculation / comparators 1, 2, and 3. FIG. 5 is a conceptual diagram of this hue region, in which a straight line from the center point in each color direction is used as a reference line, and is divided into six hue regions.
Further, the primary color component amount and complementary color component amount calculation circuit 4 compares the levels of the signals R, G, and B to determine the maximum level, the intermediate level, and the minimum level as shown in FIG. Then, in the process of this comparison and determination, the level difference between the maximum level and the intermediate level is obtained and used as the primary color component amount, and further the level difference between the intermediate level and the minimum level is obtained and used as the complementary color component amount. Here, the maximum level color corresponds to the primary color, and the minimum level component corresponds to the white component. Then, the complementary color can be determined from the information of the maximum level color and the minimum level color. As a result, as shown in FIG. 4, the primary color component and the complementary color component can be determined.

In the example of FIG. 6, since the maximum level is R and the intermediate level is G, the primary color component is R, and the complementary color component is Ye (yellow), which is the hue between R and G. The primary color component amount is RG, the complementary color component amount is GB, and the minimum level B amount is the white component amount. Therefore, in the case of FIG. 8, the result shown second from the bottom in FIG. 4 is obtained.
The determination result of the hue region by the determination circuit 4 is supplied to the constant selection circuit 6, and a specific gain constant is selected according to the determination result and is supplied to the multipliers 7 and 8, so that the calculation circuit 5 calculates it. Correction is performed by multiplying the primary color component amount and the complementary color component amount. For this reason, in the constant selection circuit 6, specific gain constants corresponding to the respective hue regions from the region 1 to the region 6 are set in advance.
The primary color component amount and the complementary color component amount thus multiplied by the gain constant by the multipliers 7 and 8 are sent to the data selection / addition circuit 11 for selecting addition / subtraction and connection selection for the video signals R, G, B, respectively. In the other hand, they are respectively supplied via complements (-1 times multipliers) 9 and 10 respectively. Then, after the addition destination is selected by the data selection / addition circuit 11, it is supplied to each adder 12, 13, 14 and added to the video signals R, G, B. Therefore, the above processing is shown in the flowchart in FIG.

Therefore, now, when correcting the tone of the signal R, for example, in the case of correction in the saturation direction, the primary color component amount RG is multiplied by a specific constant Kr and then added to the video signal R. At this time, if the ratio based on the constant Kr is in the range of −1 to 1 times, the level difference between the intermediate level and the minimum level (complementary color component amount) and the minimum level amount (white component amount) are also obtained by this correction. It does not change.
When correcting the saturation of the signal Ye, the complementary color component amount GB is multiplied by a specific constant Ky and then added to R and G, respectively. Also at this time, if the ratio according to the constant Ky is in the range of −1 to 1 times, the level difference between the maximum level and the intermediate level (primary color component amount) and the minimum level amount (white component amount) are also obtained by this correction. Does not change.

  Therefore, in this case, if the constants Kr and Ky are operated, the saturation direction of the primary color R and the complementary color Ye can be independently corrected while maintaining the white balance. In the above-described six-color independent tone correction method, the chromaticity direction can be corrected independently, and even when the input video signal is in a different hue, independent correction is possible as well. Is omitted.

  Hereinafter, a color tone detection and correction apparatus according to another embodiment of the present invention will be described in detail with reference to the illustrated embodiments. First, FIG. 9 shows an embodiment of the present invention, in which 15 is an intermediate hue setting circuit, 16 is a primary color / complementary color area determination circuit, 17 is an α / β, β / α calculation circuit, 18 is a constant selection circuit, 20 is a multiplier, 21 is a data selection addition / subtraction circuit, and the others are the same as in the prior art shown in FIG.

The intermediate hue setting circuit 15 functions to enable setting of an intermediate color that is newly set as a reference color. For example, a flesh color (hue F) that is an intermediate color between R and Ye is preset. The primary color / complementary color area determination circuit 16 determines the hues of the input video signals R, G, and B based on the data from the hue area determination circuit 4 and the hue F given from the intermediate color hue setting circuit 15 to obtain a predetermined color. It serves to generate the control signal S.
The α / β, β / α calculating circuit 17 functions to calculate predetermined constants α / β, β / α based on the data supplied from the intermediate hue setting circuit 15. These constants α / β and β / α will be described later. The constant selection circuit 18 functions to select and output one of constants α / β and β / α according to the control signal S.

  Multipliers 19 and 20 multiply the primary color component and the complementary color component output from the primary color component amount and complementary color component amount calculation circuit 5 by one of the constants α / β and β / α selected by the constant selection circuit 18. To work. The data selection addition / subtraction circuit 21 selects data according to the determination result by the hue area determination circuit 4 and the control signal S, and performs predetermined addition / subtraction. Details of the operation of this circuit will be described later.

  Next, the operation of this embodiment will be described. 10 and 11 are diagrams showing saturation (color saturation) and chromaticity (hue) for explaining the operation principle of the present invention. In these diagrams, the directions away from the origin O are saturation and saturation. The direction perpendicular to (the direction in which the circle is drawn) represents the chromaticity.

  Here, the present invention can be applied to any intermediate color correction, but it is considered that the present invention is often applied particularly to skin color correction. In this embodiment, therefore, correction of skin color will be mainly described below as an example. Then, since the hue of this skin color is located in the region between R and Ye, that is, the region 6, in these FIGS. 10 and 11, only the region 6 from R (red) to Ye (yellow) is shown. Is represented by a point F.

  Therefore, this point is set as the auxiliary reference color F as shown in the figure, and the data is set in the intermediate hue setting circuit 15 as described above.

  Thus, the region 6 is divided into two auxiliary regions, that is, the region (1) and the region (2) by an axis passing from the center point O to the auxiliary reference color F point, that is, the auxiliary reference line. . Next, the hue of the input video signal is divided by the primary color / complementary color area determination circuit 16 into an area (1) between R and F and an area (2) between F and Ye as shown in FIG. Judgment. Then, first, in this case, since the operation is performed when the determination result of the hue region determination circuit 4 is the region 6, the primary color component amount and the complementary color component amount output from the calculation circuit 5 are respectively It is as follows.

  Primary color component amount = RG = Rc complementary color component amount = GB = Yc Next, the primary color / complementary color region indicates whether the hue of the input video signal is in these regions (1) or (2). Identification is made by the determination of the determination circuit 16, and correction is performed separately as shown below.

  <Correction Processing in Region (1)> At this time, the output from each circuit is as follows. First, the constant selection circuit 18 selects the constant β / α and outputs the constant β / α to the multipliers 19 and 20. Next, the data selection addition / subtraction circuit 21 outputs a signal [Rc−Yc × (β / α)], a signal (−Yc), and a signal [Yc × (β / α)]. Further, the constant selection circuit 6 selects constants Kr and Kf and outputs these constants Kr and Kf to the multipliers 7 and 8.

  As a result, the data selection / addition circuit 11 first outputs a signal [Rc−Yc × (β / α)] × Kr + Kf × [Yc × (β / α)] to the adder 12 to generate a signal R Then, the signal [(−Yc) × Kf] is output to the adder 14 and added to the signal B.

  Therefore, in FIG. 10, now, if the point A is the coordinates of the input video signal and is represented by a vector A, this vector A is represented by the synthesis of the R component vector R1 and the skin color component vector F1.

  A = R1 + F1 Next, assuming that Kr is a gain constant dedicated to adjusting the saturation direction of R and Kf is a gain constant dedicated to adjusting the saturation direction of skin color, when performing color correction in the saturation direction of R, | R1 | × Kr may be added in the saturation direction of R, that is, added to R. When color correction is performed in the saturation direction of the skin color, | F1 | × Kf may be added in the skin color saturation direction. .

  Therefore, the calculation method of these amounts | R1 | and | F1 | and the addition method in the skin color saturation direction will be described. For this purpose, all corrections can be expressed as corrections to the R, G, and B components. It ’s fine. Therefore, R component basic vector is R, skin color component basic vector is F, Ye component basic vector is Y, B component basic vector is B, and F = α × Y + β × R = α × (−B) + β × R. And

  Next, the coordinate vector A of the input video signal is expressed by combining the R component and the Ye component. Here, A = Y × Yc + R × Rc and Rc and Yc can be easily obtained as described in the conventional color tone correction method. In this case, R> G> B, and as is apparent from FIG. 9, Rc = RG and Yc = GB.

  Then, A = Y × Yc + R × Rc = (1 / α) × (F−β × R) × Yc + R × Rc = F × Yc / α + R × (Rc−β × Yc / α), and thus | R1 | = Rc- [beta] * Yc / [alpha] | F1 | = Yc / [alpha].

  Therefore, if F × Yc / α is expressed by vector R and vector B, F × Yc / α = (α × (−B) + β × R) × Yc / α = B × (−Yc) + R × (β × Yc / α).

  Therefore, the above results are summarized as follows. That is, to perform color correction in the saturation direction of R, | R1 | Kr = (Rc−β × Yc / α) × Kr may be added to R. Next, in order to perform color correction in the skin color saturation direction, | F1 | × Kf may be added in the skin color saturation direction. This means that −Yc × Kf is added to B and (β Equal to adding (Yc / α) × Kf to R.

  Here, if the angle between the R vector and the skin color vector is θ, α × Sin (60 ° −θ) = β × Sin (θ), and therefore β / α = Sin (60 ° −θ). / Sin (θ).

  Therefore, when θ = 20 °, β / α = 1.794, but when this is set to ≈2.0, the correction at this time is (Rc− 2 × Yc) × Kr may be added to R, and for color correction in the skin color saturation direction, −Yc × Kf may be added to B and 2 × Yc × Kf may be added to R. Then, by changing β / α, the skin color reference axis can be adjusted.

  The above is the description of the correction in the saturation direction, but the description is omitted because the same concept can be applied to the correction in the chromaticity direction.

  <Correction Processing in Region (2)> At this time, the output from each circuit is as follows. First, the constant selection circuit 18 selects a constant α / β and outputs the constant α / β to the multipliers 19 and 20. Next, the data selection addition / subtraction circuit 21 outputs a signal [Yc−Rc × (α / β)], a signal (Rc), and a signal [−Rc × (α / β)]. Further, the constant selection circuit 6 selects constants Ky and Kf and outputs these constants Ky and Kf to the multipliers 7 and 8.

  As a result, the data selection / addition circuit 11 first outputs the signal [Rc × Kf] to the adder 12 and adds it to the signal R, and then the signal − [Yc−Rc × (α / β )] × Ky−Kf × [Rc × (α / β)] is output to the adder 14 and added to the signal B.

  Therefore, in FIG. 3, when the point C is the coordinates of the input video signal and this is represented by a vector C, this vector C is represented by the synthesis of the Ye component vector Y1 and the skin color component vector F2.

  C = Y1 + F2 Next, let Yy be a gain constant dedicated to adjusting the saturation direction of Ye, and Kf be a gain constant dedicated to adjusting the saturation direction of skin color. For color correction in the saturation direction of Ye, | Y1 | × Ky may be subtracted from B. For color correction in the skin color saturation direction, | F2 | × Kf may be added in the skin color saturation direction.

  Next, the calculation method of | Y1 | and | F2 | and the addition method in the skin color saturation direction are the same as those in the above-described region (1), and thus are as follows. C = Y * Yc + R * Rc = Y * Yc + (1 / [beta]) * (F- [alpha] * Y) * Rc = F * Rc / [beta] + Y * (Yc- [alpha] * Rc / [beta]), thus | Y1 | = Yc −α × Rc / β | F2 | = Rc / β.

  Here, when F × Rc / β is expressed by a vector R and a vector B, F × Rc / β = (α × (−B) + β × R) × Rc / β = −B × (α × Rc / β) + R × Rc.

  Therefore, the above results are summarized as follows. That is, when color correction in the saturation direction of Ye is performed, | Y1 | × Ky = (Yc−α × Rc / β) × Ky may be subtracted from B. Next, when performing color correction in the skin color saturation direction, | F2 | × Kf is added in the skin color saturation direction. This means that (−α × Rc / β) × Kf is set to B. It is equivalent to adding Rc × Kf to R.

  Therefore, if the angle θ between the R vector and the skin color vector is set to 20 ° as in the above-described region (1), α / β = 0.5321 is obtained. When correcting color in the saturation direction of Ye, (Yc−0.5Rc) × Ky may be subtracted from B, and in performing color correction in the saturation direction of skin color, −0.5 ×. Rc × Kf may be added to B and Rc × Kf may be added to R.

  The above is the description about the correction in the saturation direction, but the description is omitted because the same concept can be applied to the correction in the chromaticity direction.

  FIG. 4 shows the characteristics obtained by the correction described in each section of the region (1) and the region (2). The characteristics shown in FIG. 4 are obtained by superimposing gain characteristics of color correction in the saturation direction of R, color correction in the saturation direction of Ye, and color correction in the saturation direction of the skin color. In addition, if the skin color saturation direction gain constant Kf is controlled, color correction in the skin color saturation direction can be performed regardless of the R saturation direction gain constant Kr and the Ye saturation direction gain constant Ky. I understand.

  Therefore, according to this embodiment, the influence on R and Ye can be suppressed to a minimum, and the effective tone correction can be performed on the skin color, so that a sense of incongruity can be ensured when the television camera is switched. It can be lost.

  Next, FIG. 5 shows correction characteristics according to another embodiment of the present invention. In this embodiment, a correction function having a gain characteristic with the skin color axis F as the center is generated and extracted. . This is added to the conventional function in the embodiment shown in FIG. 5. According to this method, correction can be performed in such a way as to compensate for an area that cannot be corrected by the conventional method.

  In the above embodiment, the hue range is limited to R and Ye. However, it goes without saying that the present invention can be applied to any hue, and the type and number of reference colors are also applicable. Needless to say, it can be set arbitrarily.

  According to the present invention, it is possible to control the lens iris so that the brightness level of a specific hue portion such as the blue color of the tennis court or the green color of the soccer ground is constant, and to realize a sports broadcast broadcasting camera. Can do.

1, 2, 3: arithmetic / comparator, 4: hue area determination circuit, 5: primary color component amount and complementary color component amount calculation circuit, 6: constant selection circuit, 7, 8: multipliers 9, 10: complement (- 11: data selection adder circuits 12, 13, 14: adder, 15: intermediate hue setting circuit, 16: primary color / complementary color area determination circuit, 17: α / β, α / β calculation circuit, 18 : Constant selection circuit, 19, 20: multiplier, 21: data selection addition / subtraction circuit 30: television camera, 31: lens, 32: prism, 33R, 33G, 33B: CCD (charge coupled device), 34R, 34G, 34B : AFE (analog front end processor), 35: Video signal processing unit with hue detection and correction function, 37: Parallel-serial conversion unit, 39: CPU, 40: Viewfinder,

Claims (3)

In an imaging apparatus having means for detecting a specific hue for each pixel and means for controlling the iris of the lens,
Means for detecting a luminance signal level of a pixel in an arbitrary hue region in the specific hue, and means for controlling an iris of the lens so that the detected luminance signal level of the pixel in the arbitrary hue region is constant; Have
Detecting a luminance signal level of a pixel in an arbitrary hue region in the specific hue, and controlling an iris of the lens so that the detected luminance signal level of the pixel in the arbitrary hue region is constant; An imaging device. The imaging device according to claim 1.
Identification of primary hue areas (R, G, B) and complementary hue areas (Ye, Cy, Mg) by color difference calculation (R-G, RB, GB) and detection of the color difference signal level Detecting means for detecting a luminance signal level of a pixel in an arbitrary hue region in the hue of
By the color difference calculation (of RG, RB, GB), the detection of the color difference signal level, and the constant selection of primary color, complementary color, and intermediate color, the primary color hue region (of R, G, B) and (Ye, Cy An image pickup apparatus comprising: at least one of detection means for detecting a luminance signal level of a pixel in an arbitrary hue area within a specific hue of a complementary hue area and an intermediate hue area of Mg). The imaging apparatus according to claim 1 or 2,
An image display unit for displaying a menu for setting the imaging device and an arbitrary hue region in the specific hue;
Means for selecting an arbitrary hue region of the specific hue in the image display unit;
An imaging apparatus for selecting the specific hue by displaying a menu for setting the imaging apparatus or an arbitrary hue area in the specific hue on the image display unit.

Images