JP2006267924A - Multi-gradation image generation method and multi-gradation image generation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve gradation representation without being affected by movements of an image. <P>SOLUTION: A multi-gradation image generation device includes a frame detection part 1 for detecting consecutive identical frames in an input image VIN and a dither processing part 2 for performing time-base dither processing based on rules independent per frame with respect to a plurality of frames of the input image VIN, which are determined to be identical by the frame detection part 1. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a multi-tone image generation method and a multi-tone image generation apparatus that improve gradation expression of an image by dither processing.

  Conventionally, a dither method has been used to display a multi-tone image in a display device that can display only binary values, for example. As the dither method, there are a systematic dither method that regularly converts the gray level of an input image and a random dither method that randomly converts the level (for example, see Non-Patent Document 1).

Supervised by the Image Processing Standard Textbook Editorial Committee, “Image Processing Standard Textbook”, Foundation for the Promotion of Image Information Education, 1997, p. 146-148

  According to the conventional dither method, the gradation expression can be improved by artificially increasing the number of gradations of the image. However, the conventional dither method has a problem that the resolution of the image is deteriorated instead of improving the gradation expression. Therefore, there is a method of suppressing degradation of resolution by performing time direction dither processing based on independent rules for each frame for a plurality of frames. However, according to this time direction dither, when the image is moved, there is a problem that the gradation is inconsistent before and after the movement, and the gradation expression is impaired.

  The present invention has been made to solve the above-described problem. Even when an input image is a moving image, a multi-tone image generation method and a multi-tone image generation method capable of improving gradation expression while suppressing deterioration in resolution are provided. An object is to provide a gradation image generating apparatus.

The multi-tone image generation method of the present invention includes a frame detection procedure for detecting consecutive identical frames in an input image, and a plurality of frames of the input image determined to be the same in the frame detection procedure for each frame. And a dither processing procedure for performing time direction dither processing based on independent rules.
In the configuration example of the multi-gradation image generation method of the present invention, the input image is obtained by telecine-converting N frames of film images per second into M frames (N <M) video images per second. is there.
In addition, the multi-tone image generation apparatus of the present invention includes a frame detection unit that detects consecutive identical frames in an input image, and a frame for a plurality of frames of the input image that are determined to be the same by the frame detection unit. And a dither processing unit that performs time direction dither processing based on independent rules.

  According to the present invention, by performing time direction dither processing on a plurality of frames of the input image determined to be the same in the frame detection procedure, the gradation expression of the image is improved without being influenced by the motion of the image. can do.

  In the present invention, the gradation expression can be effectively improved by applying it to an input image obtained by converting a film image into a video image.

[First Embodiment]
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a multi-tone image generation apparatus according to the first embodiment of the present invention. The multi-tone image generating apparatus according to the present embodiment detects a frame detection unit 1 that detects consecutive identical frames in the input image VIN, and a plurality of frames of the input image VIN that are determined to be the same by the frame detection unit 1. And a dither processing unit 2 that performs time direction dither processing based on independent rules for each frame.

  The frame detection unit 1 detects consecutive identical frames in the input image VIN, and outputs to the dither processing unit 2 a frame identification signal FID that designates frames determined to be identical. For determining the frame identity, for example, the input image VIN of the previous frame is stored in a frame memory (not shown) in the frame detection unit 1, and the input image VIN of the previous frame and the current input image VIN are stored. Compare.

  Next, the time direction dither processing by the dither processing unit 2 will be described. 2-6 is a figure for demonstrating the time direction dither process by the dither process part 2. FIG. The time direction dither processing is to perform dither processing in units of a plurality of frames so that the number of apparent gradations increases when a plurality of frames are regarded as one frame. The squares in FIGS. 2 to 6 represent pixels, and the numerical values in the squares in FIGS. 3 to 6 represent the luminance of the pixels. That is, in the example of FIG. 3 to FIG. 6, four pixels of horizontal 2 pixels × vertical 2 pixels of A11, A21, A12, and A22 shown in FIG. 2 are described. 3 and 4 show the operation when the number of output bits is smaller than the number of input bits, and FIGS. 5 and 6 show the operation when the number of input bits is the same as the number of output bits.

  First, the operation when the number of output bits is smaller than the number of input bits will be described. Here, as an example, a case where a 2-bit input image VIN is converted into a 1-bit output image VOUT will be described. When the dither processing unit 2 recognizes from the frame identification signal FID from the frame detection unit 1 that three consecutive frames in the input image VIN are the same, the dither processing unit 2 performs the processing shown in FIG. FIGS. 3A, 3 </ b> B, and 3 </ b> C show three frames detected by the frame detection unit 1.

  The conversion rule for the first frame shown in FIG. 3A is to multiply the luminance value of each pixel by 1/2. Multiplying a pixel with luminance “2” by 1/2 gives “1”, while multiplying a pixel with luminance “1” by 1/2 gives “0.5”. However, since the luminance value of “1” or “0” exists in the 1-bit output image VOUT, when the multiplication result is “0.5”, the output luminance is “0”. As a result, the first frame converted by the dither processing unit 2 becomes as shown in FIG.

  The conversion rule for the second frame shown in FIG. 3B is to multiply the luminance value by 1/2 and subtract 1 from the multiplication result. Multiplying a pixel with luminance “2” by 1/2 and further subtracting 1 results in “0”, while multiplying a pixel with luminance “1” by 1/2 and subtracting 1 further results in “ −0.5 ”. However, when the multiplication result is “−0.5”, the output luminance is “0” for the same reason as described above. As a result, the second frame converted by the dither processing unit 2 becomes as shown in FIG.

  The conversion rule for the third frame shown in FIG. 3C is to multiply the luminance value by 1/2 and add 1 to this multiplication result. Multiplying a pixel of luminance “2” by 1/2 and adding 1 further gives “1”, while multiplying a pixel of luminance “1” by 1/2 and adding 1 further adds “1”. 1.5 ". However, when the multiplication result is “1.5”, the output luminance is “1” for the same reason as described above. As a result, the third frame converted by the dither processing unit 2 becomes as shown in FIG.

  FIG. 3G shows an image when the three frames in FIGS. 3D to 3F are regarded as one frame. Since the luminance value when the upper left pixel (position A11 in FIG. 2) of the first frame to the third frame is regarded as one pixel is obtained as an average value of the three frames, it apparently becomes “2/3”. The luminance value when the lower left pixel (position A12) of the first to third frames is regarded as one pixel is also apparently “2/3”. In addition, the luminance value when the pixel at the upper right (position A21) of the first frame to the third frame is regarded as one pixel is apparently “1/3”, and the lower right (A22) of the first to third frames. The luminance value when the pixel at the position is regarded as one pixel is apparently “1/3”. Although a 2-bit gradation can be expressed by a 1-bit image, when the time direction dither processing is performed by the dither processing unit 2 on three consecutive frames, apparently “0”, “1/3”, “ Four luminance values of “2/3” and “1” can be expressed, and four gradations can be expressed.

  On the other hand, when the dither processing unit 2 recognizes that two consecutive frames in the input image VIN are the same by the frame identification signal FID from the frame detection unit 1, it performs the processing shown in FIG. FIGS. 4A and 4B show two frames detected by the frame detection unit 1.

  The conversion rule for the first frame shown in FIG. 4A is to multiply the luminance value of each pixel by 1/2. As a result, the first frame converted by the dither processing unit 2 becomes as shown in FIG. The conversion rule for the second frame shown in FIG. 4B is set so that the rule is different for pixels adjacent in the horizontal and vertical directions, and the rule is the same for pixels adjacent in the diagonal direction. For example, for the pixels at positions A11 and A22, the luminance value is multiplied by 1/2, and 1 is added to the multiplication result. For the pixels at positions A21 and A12, the luminance value is 1/2. Multiplication is performed, and 1 is subtracted from the multiplication result. As a result, the second frame converted by the dither processing unit 2 becomes as shown in FIG.

  FIG. 4E shows an image when the two frames in FIGS. 4C and 4D are regarded as one frame. The luminance value when the upper left pixel of the first frame and the second frame is regarded as one pixel is apparently “1”, and the luminance value when the lower left pixel of the first frame and the second frame is regarded as one pixel. Is apparently “1/2”. When the upper right pixel of the first frame and the second frame is regarded as one pixel, the luminance value is apparently “0”, and the lower right pixel of the first frame and the second frame is regarded as one pixel. The apparent luminance value is “½”. When the dither processing unit 2 performs time direction dither processing on two consecutive frames, apparently three luminance values of “0”, “1/2”, and “1” can be expressed. Can be expressed.

  Next, an operation when the number of input bits is the same as the number of output bits will be described. Here, as an example, a case where a 1-bit output image VOUT is generated from a 1-bit input image VIN will be described. When the dither processing unit 2 recognizes from the frame identification signal FID from the frame detection unit 1 that three consecutive frames in the input image VIN are the same, the dither processing unit 2 performs the processing shown in FIG. FIGS. 5A, 5 </ b> B, and 5 </ b> C show three frames detected by the frame detection unit 1.

  The rule for the first frame shown in FIG. 5A is to output the luminance value of each pixel as it is. As a result, the first frame output from the dither processing unit 2 is as shown in FIG. The conversion rule for the second frame shown in FIG. 5B is to set all luminance values to “0”. Thus, the second frame converted by the dither processing unit 2 becomes as shown in FIG. The conversion rule for the third frame shown in FIG. 5C is to set all luminance values to “1”. Thus, the third frame converted by the dither processing unit 2 becomes as shown in FIG.

  FIG. 5G shows an image when the three frames in FIGS. 5D to 5F are regarded as one frame. The luminance value when the upper left pixel of the first to third frames is regarded as one pixel is apparently “2/3”, and the lower left pixel of the first to third frames is regarded as one pixel. The luminance value is apparently “2/3”. Further, when the upper right pixel of the first frame to the third frame is regarded as one pixel, the luminance value is apparently “1/3”, and the lower right pixel of the first frame to the third frame is regarded as one pixel. In this case, the luminance value is apparently “1/3”. Therefore, four gradations can be expressed as in the case of FIG.

  On the other hand, when the dither processing unit 2 recognizes that two consecutive frames in the input image VIN are the same by the frame identification signal FID from the frame detection unit 1, the dither processing unit 2 performs the processing shown in FIG. FIGS. 6A and 6B show two frames detected by the frame detector 1.

  The rule for the first frame shown in FIG. 6A is to output the luminance value of each pixel as it is. As a result, the first frame output from the dither processing unit 2 is as shown in FIG. The conversion rule for the second frame shown in FIG. 6B is to convert “1” to “0” and “0” to “1”. As a result, the second frame converted by the dither processing unit 2 becomes as shown in FIG.

  FIG. 6E shows an image when the two frames in FIGS. 6C and 6D are regarded as one frame. The luminance value when the upper left pixel of the first frame and the second frame is regarded as one pixel is apparently “½”, and the lower left pixel of the first frame and the second frame is regarded as one pixel. The luminance value also appears to be “½”. In addition, when the upper right pixel of the first frame and the second frame is regarded as one pixel, the luminance value is apparently “½”, and the lower right pixel of the first frame and the second frame is regarded as one pixel. In this case, the luminance value is apparently “½”. Therefore, three gradations can be expressed as in the case of FIG.

  As described above, in the present embodiment, it is possible to improve the gradation expression of an image while suppressing deterioration in resolution by time direction dither processing. Further, since the time direction dither processing is performed in units of the same frame, even if the input image VIN is a moving image, the gradation is not inconsistent before and after the movement, and the gradation expression is impaired. Disappear.

  Note that the frame detection unit 1 and the dither processing unit 2 of the multi-tone image generation apparatus of the present embodiment may be realized by hardware, or a computer including a CPU, a memory, and an interface, and these hardware You may implement | achieve by software with the program which controls resources. When implemented by software, the CPU executes the processing described in the present embodiment in accordance with a program stored in the memory. Further, the time direction dither processing of the present invention is not limited to the method described with reference to FIGS. 2 to 6, and other calculation methods, dither matrices, and the like may be used. Further, in the specific example described in the present embodiment, for ease of explanation, a case where a 2-bit input image is converted into a 1-bit output image and a 1-bit output image is generated from the 1-bit input image. However, in reality, it is assumed that dithering is performed with more tones.

[Second Embodiment]
Next, a second embodiment of the present invention will be described. According to the first embodiment, even if the input image is a moving image, the gradation expression can be improved while suppressing deterioration in resolution. To obtain such an effect, the input image There must be consecutive identical frames. Therefore, in the present embodiment, a specific application example of the multi-tone image generation apparatus according to the first embodiment will be described. FIG. 7 is a block diagram showing the configuration of the signal conversion apparatus according to the second embodiment of the present invention. The same components as those in FIG. 1 are denoted by the same reference numerals.

  The signal conversion apparatus according to the present embodiment is an apparatus that converts an interlaced input image VIN into a progressive output image VOUT, and includes an I / P conversion unit 3 and a dither processing unit 2. The I / P conversion unit 3 includes a frame detection unit 1 and a conversion unit 4 that converts an interlace signal into a progressive signal. An input image VIN obtained by telecine-converting 24 frames of film video per second into an interlaced video video having a vertical synchronization frequency of 60 Hz is input to the I / P converter 3. The I / P converter 3 converts the input image VIN into a progressive output image VP in accordance with a display device (not shown).

  FIG. 8 is a diagram illustrating a state of interlace-progressive conversion by the I / P conversion unit 3. In FIG. 8, VI1 to VI5 are frames of an interlaced input image VIN having a vertical synchronization frequency of 60 Hz, and VM1 to VM5 are frames of a progressive image VMID having a vertical synchronization frequency of 24 Hz corresponding to 24 frames of film video per second. , VO1 to VO13 are frames of an output image VP having a vertical synchronization frequency of 60 Hz. The black circle f1 represents an odd field, and the white circle f2 represents an even field. In the example of FIG. 8, in order to make the processing easy to understand, the input image VIN is converted into the image VMID and then converted into the output image VP. However, the image VMID is not actually generated, The input image VIN is directly converted into the output image VP.

  As is apparent from FIG. 8, when an input image VIN obtained by converting a film image into a video image is input to the signal conversion device, repetition of 3 frames and repetition of 2 frames appear alternately in the output image VP. For example, the frames VO1, VO2, and VO3 correspond to the frame VM1 (one frame of the film image) of the image VMID, and the frames VO4 and VO5 correspond to the frame VM2 (the next frame of the film image). Therefore, since the output image VP output from the conversion unit 4 has the same continuous frame, the dither processing unit 2 performs time direction dither processing on the output image VP in units of 3 frames or 2 frames. Then, the effects described in the first embodiment can be obtained. As described above, the present invention is effective when applied to, for example, an input image obtained by converting a film image into a video image.

  The present invention can be applied to a technique for improving the gradation expression of an image by dither processing.

1 is a block diagram illustrating a configuration of a multi-tone image generation apparatus according to a first embodiment of the present invention. It is a figure for demonstrating the time direction dither process by the dither process part in the 1st Embodiment of this invention. It is a figure for demonstrating the time direction dither process by the dither process part in the 1st Embodiment of this invention. It is a figure for demonstrating the time direction dither process by the dither process part in the 1st Embodiment of this invention. It is a figure for demonstrating the time direction dither process by the dither process part in the 1st Embodiment of this invention. It is a figure for demonstrating the time direction dither process by the dither process part in the 1st Embodiment of this invention. It is a block diagram which shows the structure of the signal converter device used as the 2nd Embodiment of this invention. It is a figure which shows the mode of the interlace-progressive conversion by the I / P conversion part in the 2nd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Frame detection part, 2 ... Dither processing part, 3 ... I / P conversion part, 4 ... Conversion part, VIN ... Input image, VOUT ... Output image, FID ... Frame identification signal.

Claims (3)

A frame detection procedure for detecting consecutive identical frames in the input image;
A dithering procedure for performing a time direction dithering process based on an independent rule for each of a plurality of frames of the input image determined to be the same in the frame detection procedure. Method. The multi-tone image generation method according to claim 1,
The multi-tone image generation method according to claim 1, wherein the input image is obtained by telecine-converting N frames of film images per second into video images of M frames (N <M) per second. A frame detector for detecting consecutive identical frames in the input image;
A multi-tone image generation comprising: a dither processing unit that performs time-direction dither processing based on an independent rule for each of a plurality of frames of the input image determined to be the same by the frame detection unit apparatus.

Images