JP2843912B2 - Image processing method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an image processing circuit. [Prior Art] In conventional image processing in which a negative film image such as a silver halide negative film is captured by a scanner or the like and a negative / positive reversal process is performed, a predetermined fixed reversal curve is used, or A means for visually determining a reversal curve for each input image in accordance with a long-standing feeling is used. [Problems to be Solved by the Invention] In the above conventional example, there is a problem that it is difficult to obtain good color reproducibility. An object of the present invention is to provide an image processing method capable of obtaining good color reproducibility. [Means for Solving the Problems] The present invention provides a negative image data inputting step of scanning a color film and inputting negative image data composed of a plurality of color components; A histogram creation step of independently creating a histogram; and the plurality of colors being based on a highlight point and a dark point in each of the color components set according to the histogram created independently for each of the plurality of color components. A negative / positive conversion data generating step of generating negative / positive conversion data for each of the components; and performing a conversion process using the negative / positive conversion data for each of the plurality of generated color components, thereby obtaining the negative image data. A conversion step of performing a negative / positive conversion process and a normalization process on each of the color components. is there. [Operation] The present invention provides a negative image data input step of scanning a color film and inputting negative image data composed of a plurality of color components, and a histogram for independently creating a histogram for each of the plurality of color components. A negative / positive operation for each of the plurality of color components based on a highlighting point and a dark point in each of the color components, which are set according to a histogram independently generated for each of the plurality of color components. A negative / positive conversion data generating step of generating conversion data; and performing a conversion process using the negative / positive conversion data for each of the plurality of generated color components. And a conversion stage for performing a negative / positive conversion process and a normalization process, so that good color reproducibility can be obtained. You. Embodiment FIG. 1 is a block diagram showing an embodiment of the present invention. This embodiment includes a CPU 10 for controlling the entire image processing circuit.
And a scanner 20 for reading a negative image such as a silver halide negative film.
A buffer 21 for temporarily holding an input signal from the scanner 20, a ROM 30 for storing a program for realizing the flowcharts shown in FIG. 4 and thereafter, a VRAM 40 for storing a negative / positive inverted image signal, a television monitor 50 and are provided. The CPU 10 and the ROM 30 are an example of a unit that takes a histogram of luminance components for black and white negative image data and a histogram of color components for color negative image data. Is an example of a means for determining a point of approximately 1% from the dark point as a dark point, and determining a point of approximately 99% from a darker cumulative frequency of the histogram as a highlight point,
It is also an example of a means for performing a negative / positive inversion between the determined dark point and highlight point. FIG. 2 is a view showing an example of an image captured from the scanner 20 in the above embodiment. This is a color image, that is, is composed of three planes of R (red) plane, G (green) plane, and B (blue) plane.
It is composed of an image of 80 dots × 8 bits. Hereinafter, only the R plane will be described, but the G plane and the B plane can be described similarly to the R plane. Since the luminance data is 8 bits, it can take a value from 0 to 255. FIG. 3 is a diagram showing an example of a histogram (cumulative frequency distribution) per screen for the luminance data and an example of a negative / positive inversion curve. This histogram shows the points where the cumulative frequency is accumulated from 0 from the darker direction to the brighter direction and the accumulated frequency becomes 1% of the whole as the dark point DP, and the point where the accumulated frequency becomes 99% as the highlight point HP. It is. And the above dark point DP, highlight point
Using HP, the input signal d
Is converted into an inverted output signal, thereby performing negative / positive inversion. When d ≦ DP, inverted output = 255 formula When DP <d <HP, inverted output = 255 × {1− (d−P) / (HP−DP)} formula When HP ≦ d, inverted Output = 0 Equation The inverted output obtained by these equations, that is, the negative / positive inversion curve is shown in FIG. Next, the operation of the above embodiment will be described. FIG. 4 is a flowchart showing the overall operation in the above embodiment. First, a histogram of the R plane is obtained (S1), and its dark point DP and highlight point HP are determined (S1).
2) Negative / positive inversion is performed on the R plane (S3). Similarly, a histogram of the G plane is obtained, DP and HP are determined, and negative / positive inversion is performed (S4, S5, S6). Further, the same operation as the above operation is performed for the B plane (S7, S8, S9). In this embodiment, D is obtained independently for each of the RGB color components.
Performs negative / positive conversion and normalization processing on each color component based on P and HP using formulas and normalizes each color component within a common signal level range, thus achieving good color reproduction. Can be. FIG. 5 is a flowchart showing the operation of taking a histogram in the above embodiment. In this case as well, only the R plane will be described.
The same applies to the plane and the B plane. First, a variable X indicating a dot position in the X direction and a variable Y indicating a dot position in the Y direction are respectively set to “1”, and all variables HIST (0) to HIST (255) indicating respective luminance values are set as variables. It is cleared to "0" (S10). Since the pixels of one screen are 640 dots in the X direction and 480 dots in the Y direction, the total number of dots is 640 × 480, and the value of the buffer 21 holding the data of each pixel, that is, the variable BUFFER Substitute each input signal d for (X, Y) (S11). Further, the dot position for taking the histogram is incremented by 1 (S12), the value of the dot position X is incremented by 1 (S13), and the above operation (S11 to S13) is performed by X
Is repeated until is larger than 640 (S14), that is,
Repeat until one line of the image ends. Then, the dot position in the X direction is set to "0" (S15).
After incrementing the dot position in the direction by 1 (S16), S
The operations from 11 to S13 are repeated, and when the above processing for 480 lines is completed (S17), the process returns. Thereby, frequency distributions HIST (0) to HIST (255) can be obtained. FIG. 6 is a flowchart showing the operation of determining DP and HP in the above embodiment. That is, the frequency distributions HIST (0) to HIST of the luminance data d
(255) is accumulated, and this accumulated same number is 1% of the above frequency distribution.
It is a flowchart which shows operation | movement which makes the brightness value of the point which becomes (= 640 * 480 * 0.01 = 3072) DP and makes the brightness value which becomes 99% HP. First, the frequency Σ and the dot position X in the X direction are initialized to 0 (S20), the frequency of luminance is accumulated (S21), and the accumulated value に な る becomes 3072 while increasing X one by one (S23). (S22), and the dot position X in the X direction when the accumulated value 307 reaches 3072 is determined as the dark point DP (S24). On the other hand, when calculating the highlight point HP, the cumulative value Σ
Is cleared to 0, and the X position in the X direction is set to 255 (S
twenty five). In this flowchart, the speed is increased by accumulating -1 at a time from the luminance value 255. That is, while the X position is reduced one by one (S27), the frequency Σ is accumulated (S26), the X position when the accumulated value becomes 3072 or more is set as the highlight point HP (S29), and the routine returns. FIG. 7 shows that in the above embodiment,
Is a flowchart showing an operation for performing negative / positive inversion according to the formula. First, the X position in the X direction and the Y position in the Y direction are both “1”.
(S30), and the input signal d of each dot position is input from the buffer 21 (S31). If the input signal d is equal to or less than the dark point DP, the inverted output of the input signal d at that time is set to 255. (S33), if the input signal d is larger than DP (S32) and equal to or less than HP (S34), the inverted output is set to 0 (S35). If the input signal d is greater than DP and less than HP, the result value of the above equation is inverted with respect to the input signal d (S36). The above operations (S31 to S36) are performed until the X position becomes 640 (S37, S38), that is, negative / positive inversion processing for one line is performed. Then, the X position is set to "0" and the Y position is incremented by "1" (S39), and the above operation is repeated until the Y position becomes 480 (S40). In the above embodiment, the number of pixels in the X direction and the Y direction are 640 and 480, respectively, and the luminance value is 8 bits. However, other values may be adopted as the number of pixels and the number of gradations. Also,
Although a histogram is obtained by scanning over the entire screen, for example, a histogram may be obtained for image data thinned out every ten points. In this case, a similar result is obtained, and the speed of the negative / positive inversion processing is increased. Further, although the values of DP and HP are set to 1% and 99% of the total frequency, there is no significant difference between DP and HP even if a range of about ± 0.5% is provided. Similar results can be obtained by performing processing based on not only the color data R, G, and B but also a luminance signal, a color difference signal, and the like. Further, although the negative / positive inversion curve obtained by the equation is a straight line, when the input signal is gamma-applied, the negative / positive inversion curve may be determined logarithmically. Conversely, when displaying on a CRT or the like, a gamma curve may be applied, and various functions can be adopted. Further, a video printer may be used instead of the television monitor 50. [Effects of the Invention] According to the present invention, a histogram creation step of independently creating a histogram for each of a plurality of color components, and the above-described histograms independently set for each of the plurality of color components are set. Since there is a negative / positive conversion data generation step of generating negative / positive conversion data for each of the plurality of color components based on a highlight point and a dark point in each color component, good color reproducibility can be obtained. It has the effect of being able to. For example, when the color balance of the input image is broken, according to the present invention, the color balance between the respective color components can be changed satisfactorily.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing one embodiment of the present invention. FIG. 2 is a diagram showing an example of an image read by a scanner in the above embodiment. FIG. 3 is a diagram showing a histogram and a negative / positive inversion curve in the above embodiment. FIG. 4 is a flowchart showing the overall operation in the above embodiment. FIG. 5 is a flowchart showing the operation of creating a histogram in the above embodiment. FIG. 6 is a flowchart for determining DP and HP in the above embodiment. FIG. 7 is a flowchart for performing the negative / positive inversion operation in the above embodiment. 10 ... CPU, 20 ... Scanner, 30 ... ROM, 50 ... TV monitor.

Claims (1)

(57) [Claims] A negative image data inputting step of inputting negative image data composed of a plurality of color components by scanning a color film; a histogram creating step of independently creating a histogram for each of the plurality of color components; Negative / positive conversion data for each of the plurality of color components is generated based on the highlight point and the dark point in each of the color components, which are respectively set according to the histogram independently created for each color component. Performing a conversion process using the negative-positive conversion data for each of the plurality of generated color components, thereby performing a negative / positive conversion process and a normalization process for each of the color components of the negative image data. A conversion processing step of performing a conversion process.