JP4517987B2 - Image processing method and image processing apparatus - Google Patents

Description

  The present invention relates to an image processing technique for outputting image data after performing predetermined processing, and more particularly, to a technique for processing image data according to the characteristics of each block obtained by dividing the image data into a plurality of blocks.

In recent years, documents have been configured using materials having various characteristics such as color characteristics such as black and white and characters, characters, photographs, and the like.
For example, when an image of such a document is read with a scanner or the like by a copying machine and printed out based on the image data, the image processing necessary for the image data is performed by a system as shown in FIG. Yes.

The example shown in FIG. 7 is an image processing system (IPS) that processes color image data in a color copying machine.
First, pre-correction (ENL) is applied to the input color image data to convert the RGB color space into a Lab color space of lightness and saturation, detecting edges (pictures) in the image, and MTF correction such as moire prevention Apply (digital filter). Then, based on the background level detected from the pre-corrected image data, background noise is removed from the MTF-corrected image data, the image data is converted to the CMYK color space by post-correction, and further, tone correction and screen generation are performed. And output as print data.
In order to print out the image data in this way, various image processes are performed.

In order to realize flexible and real-time processing in a hardware-configured image processing system, an invention has been proposed in which control of a plurality of image processing modules serially connected by a pipeline is switched by object tags (Patent Literature). 1).
In addition, in an image processing system applied to a color copying machine, an invention has been proposed in which different processing is performed in a case where a pixel of interest in image data represents a text pixel and an edge pixel and in other cases (patent) Reference 2).
Further, an invention has been proposed in which image data is divided into a plurality of blocks and image processing is performed on these blocks in parallel (see Patent Document 3).

As an invention for generating a histogram based on image data and performing image processing, an invention has been proposed in which a density histogram for an entire page of an image is generated and whether the entire image is a character or a ground based on its shape characteristics (Patent Document) 4).
Further, an invention has been proposed in which a density histogram of an entire page of an image is generated and background detection is performed from information such as the maximum frequency and the second highest frequency in the density histogram (see Patent Document 5).

As an invention for controlling print processing of image data, an invention has been proposed in which processing time is predicted based on attribute information included in a data file described in a page description language to control the start timing of a printing operation (Patent Document 6). reference.).
Further, the present invention preliminarily compresses image data to obtain a compressed data amount, and based on the data amount, uses the correlation between the compression rate and the expansion processing time to estimate the expansion processing time and control the output. Has been proposed (see Patent Document 7).
Japanese Patent Laid-Open No. 11-167627 JP 2001-247990 A JP-A-5-268479 JP 2000-059614 A JP 2004-066452 A Japanese Patent Laid-Open No. 10-058798 JP-A-10-023271

As illustrated in FIG. 7, various types of image processing are performed in order to print out image data, and there are necessary processing and processing that is not necessary depending on the characteristics of the material in the image data.
In addition, due to a request for improving the image quality of a printed image, a newly required process may occur depending on the characteristics of the material.

  For such a situation, it is conceivable to prepare a plurality of processes (processing paths) that combine necessary processes according to the characteristics of the material in the image data with a hardware configuration. For example, as shown in FIG. 6B, processing paths A and B having a hardware configuration are prepared, and image data is subjected to image processing in parallel through these processing paths A and B, and the characteristics determined by the determination process. Thus, it is conceivable to select (MPX) and output the processing result of one of the processing paths.

However, according to such a method, there is a problem that processing is performed in an unnecessary processing path that is not output as a result, resulting in waste. In addition, since it has a hardware configuration, it is difficult to add a new processing path, and there is a problem in that it cannot respond flexibly to user requests.
In addition, the former problem is particularly serious when the image processing system is configured by software, and the processing load of the processor increases because unnecessary processing paths are executed in parallel, resulting in a significant decrease in processing speed. The problem of doing it was also invited.

In addition, there is a method of using a histogram for characteristic determination. For example, in order to speed up the determination process, for example, the lightness values of 256 gradations obtained from image data are gathered for every 16 gradations to reduce the data amount. A method of creating and determining a histogram is conceivable.
However, according to this method, although the discrimination processing can be performed at a high speed by suppressing the data amount, the gradation characteristics are uniformly roughened, so that the characteristics of the histogram are lost and correct discrimination cannot be performed. was there.

Further, for example, when image data subjected to image processing is output to another processing device for further processing, the image processing and the output processing may be performed in series by performing the output processing after completion of the image processing. It is more efficient and desirable to process each of them in parallel.
However, output processing is usually faster than image processing, and simply performing parallel processing causes the output processing to overtake image processing, resulting in a situation where image processing must be waited on the way.
Thus, there has been a problem of how to perform output without delay while performing image processing and output processing in parallel.

The present invention has been made in view of the above-described conventional circumstances, and an object thereof is to avoid execution of an unnecessary processing pass in image processing.
Another object of the present invention is to enable image processing to be realized by a software configuration without causing an unnecessary increase in the burden on the processor, and to flexibly respond to user requests.

Another object of the present invention is to improve the discrimination accuracy while improving the discrimination speed when using a histogram in the discrimination of characteristics.
Another object of the present invention is to perform output processing without delaying output in the middle while performing image processing and output processing in parallel when outputting image processed image data to a subsequent apparatus. .

  The present invention relates to an image processing apparatus that outputs image data after performing predetermined processing, an image processing method that can be performed by the image processing apparatus, and a program that is executed by a computer to configure the image processing apparatus as software. Realized.

In the present invention, the block forming means divides the image data into a plurality of blocks, the characteristic determining means determines the characteristics of these blocks, and the image processing means applies a preset process to each block according to the characteristics.
Therefore, it is possible to select and execute a processing path corresponding to the characteristics of the material in the image data in units of blocks.

Further, in the present invention, the blocking unit divides the image data into a plurality of blocks, and the subdividing unit specifies a block to be further subdivided based on a preset condition, and further subdivides the block into a plurality of blocks. Then, the characteristic discriminating unit discriminates the characteristics of the block and the subdivided block, and the image processing unit applies a preset process to the block and the subdivided block according to the characteristic.
Therefore, by further subdividing the blocks, the material characteristic accuracy of the blocks and the subdivided blocks can be improved, and a processing path corresponding to the characteristics can be selected and executed.

Here, characteristics can be determined using a histogram. In particular, block characteristics are determined based on a brightness histogram of a block image, and subdivided block characteristics are further determined based on a saturation histogram of a subdivided block image. It is preferable to discriminate.
That is, the background background block can be determined using the feature that there is one peak in the highlight portion of the lightness histogram, and the other background blocks are further subdivided into black characters using the lightness histogram and saturation histogram, Other characteristics such as color letters, color photographs, black and white photographs can be distinguished.

At this time, a gradation pattern having at least a partially rough gradation to quantize the image data of the block or subdivided block is prepared in advance for each material characteristic, and the material characteristic designated for processing the image data is prepared. It is preferable to create a lightness histogram and a saturation histogram based on pixel values included in a block or a subdivided block using a gradation pattern according to the above.
In other words, by changing the quantization width according to the material characteristics to be preferentially image processed, creating a histogram suitable for determining the material characteristics and using it for the determination makes it easy to determine the corresponding block or subdivided block Thus, appropriate image processing can be performed, and the amount of data in the histogram is reduced, so that discrimination processing can be performed at high speed.

The present invention also has an output means for outputting image data for one frame subjected to image processing, and outputs based on the data amount of the portion subjected to image processing in the image data for one frame. Calculate the time required and calculate the time until the end of processing based on the data amount of the portion of the image data that has not yet undergone image processing using the determined characteristics. Compared with the time, when the time required for output exceeds the time until the end of processing, the output means starts outputting image data for one frame.
That is, when the image processing and the output processing are performed in parallel, the output processing is started at a timing at which the output processing does not overtake the image processing, so that the output can be performed without delay.

According to the present invention, it is possible to select a processing path corresponding to the material characteristic in the image data and execute the image processing in units of blocks. Therefore, it is possible to avoid execution of an unnecessary processing path in the image processing. That is, as shown in FIG. 6B, image data is processed in parallel through processing paths A and B, and the processing result of one of the processing paths is selected and output according to the characteristics determined by the determination processing. In the present invention, as shown in FIG. 5A, in the present invention, either processing path A or B is selected and executed according to the characteristic determined by the determination process, and the processing result is obtained by the optimum processing path. Can be output.
In addition, since the present invention selects a necessary processing path and executes the processing, image processing can be realized by a software configuration without causing an unnecessary increase in the burden on the processor, and flexibly responding to user requests. Can respond.

  Further, according to the present invention, according to material characteristics to be preferentially image-processed, a pattern suitable for discrimination of the characteristics is quantized and a histogram with reduced gradation frequency is created and used for discrimination. Blocks and subdivided blocks can be easily discriminated and necessary image processing can be performed, and the amount of data in the histogram is reduced, so that the discrimination speed is improved.

  Further, the present invention predicts the time until the end of the image processing and the time required for the output in parallel with the image processing and the output processing, and starts the output processing at a timing that does not overtake the image processing. The output process can be performed without delay, and the output destination apparatus can receive data without interruption.

Hereinafter, the present invention will be specifically described based on examples.
FIG. 1 shows a main functional configuration of an image processing apparatus 1 according to an example of the present invention. The image processing apparatus 1 is provided as an image processing system of a copying machine.

  The image processing apparatus 1 includes an input unit 2 to which image data of a document obtained by scanner reading or the like is input, a blocking unit 3 for dividing the image data into a plurality of blocks (for example, 6 × 6 mm), and each divided block First characteristic discriminating means 4 for discriminating the characteristics of the image, specifying a block to be further subdivided based on a preset condition (for example, a block having a background characteristic of the image is not subdivided), Subdividing means 5 for subdividing into blocks (for example, 3 × 3 mm), second characteristic discriminating means 6 for discriminating the characteristics of the subdivided blocks, blocks corresponding to the image data according to the characteristics of the blocks and the characteristics of the subdivided blocks And image processing means 7 for performing image processing in a processing path set in advance for the subdivided blocks, and output means for outputting image processed image data as print data 8 is provided.

Each of the functional units 3 to 7 may be realized by a hardware configuration. In this example, the functional units 3 to 7 are configured by executing the program according to the present invention by the computer hardware of the image processing apparatus 1. .
The first characteristic discriminating means 4 and the second characteristic discriminating means 6 may be configured by using the same function means.

A process in which the characteristic image processing means 8 selects a processing path for the corresponding block and subdivided block according to the characteristics determined by the first characteristic determining means 4 and the second characteristic determining means 6 is shown in FIG. Will be described with reference to FIG.
2 exemplifies a case where the input image data 10 is divided into blocks and a processing path for each block is selected for the sake of simplicity, the same applies to the subdivided blocks.

The first characteristic discriminating means 4 (second characteristic discriminating means 6) applies each block 11 (subdivided) to the image data 10 divided into a plurality of blocks in a matrix form by the blocking means 3 (subdividing means 5). The image processing means 7 selects a processing path for performing image processing on a corresponding block (subdivided block) in the image data based on, for example, a preset correspondence table 12. And run.
For example, the block 11 determined as the image characteristic A is subjected to image processing in a processing pass including each processing step such as ENL + TRC + YYY, and the block 11 determined as the image characteristic B is processed through a processing pass including each processing step such as ENL + color conversion + XXX + YYY. The block 11 that has been processed and determined to have the image characteristic A + B is subjected to image processing in a processing pass including each processing step such as ghost correction + YYY.

Next, image characteristic determination processing performed by the first characteristic determination unit 4 and the second characteristic determination unit 6 will be described with reference to FIG.
FIG. 3 shows a histogram of lightness (L) and a histogram of saturation (a, b) according to the material included in the image of the block (subdivided block), where FIG. Color characters, (c) is a color photographic image, (d) is a black and white photographic image, and (e) is a background background.

In the histogram of this example, pixel values (brightness values or saturation values) are quantized with a gradation of 256 gradations, for example, the horizontal axis and the number of pixels corresponding to each pixel value on the vertical axis, The distribution of each pixel value for each gradation is shown.
Here, in this example, the lightness (L) histogram indicates that the smaller the pixel value is, the brighter the pixel value is, and the larger the pixel value is, the darker the pixel value is. This indicates that the green color is darker in the negative direction and the red color is darker in the positive direction. The saturation (b) histogram shows that the center of the pixel value is gray with no color, and goes in the negative direction. The blue color is dark, and the yellow color increases toward the positive direction.
Further, instead of representing the number of pixels on the vertical axis as in the present example, for example, the appearance ratio may be represented on the vertical axis and may be used in the form of a histogram.

  Each block (subdivided block) exhibits different histogram characteristics depending on the image material. For example, an image block (a) containing black characters has a lightness histogram with a peak in highlight and high density and a color with one peak in the center. An image block (b) including a color character is a brightness histogram having a highlight and a peak in high density, and a saturation histogram having two peaks at positions shifted from the center and the center, and an image including a color photographic image. The block (c) has a lightness histogram with a gentle and undefined peak and a saturation histogram with a gentle and unsteady peak, and the image block (d) containing a black-and-white photo image has a lightness histogram with a gentle and undefined peak and one peak in the center. Is a saturation histogram, and the image block (e) including the background background is highlighted. One peak the saturation histogram in brightness histogram and the center there is one peak in bets.

Therefore, the first characteristic discriminating means 4 and the second characteristic discriminating means 6 can discriminate the image characteristics of blocks and subdivided blocks based on the histogram characteristics as described above.
Then, the image processing means 7 selects and executes a processing path for performing image processing on the corresponding block and subdivided block as shown in FIG. 2 according to the determined image characteristics.

FIG. 4 shows the details of the processing path of the image processing apparatus according to this example in more detail. For the sake of brevity, explanations regarding blocking and subdivision will be omitted.
In the example shown in FIG. 4, “color photograph”, “color character and surrounding background”, “color background”, “black and white photograph”, “monochrome character and surrounding background”, and “monochrome background” are set as image characteristics. A processing path including different processing steps according to these characteristics is set in advance. That is, a processing path suitable for image characteristics is set in advance.

For example, for the image characteristic “color photograph”, a processing path for sequentially performing the processes of “background removal”, “digital filter”, “following color correction”, “tone correction”, and “screen generation” is set.
In addition, a processing path is set for the image characteristics “color character and surrounding background”, which sequentially performs the processes of “background removal”, “edge extraction”, “digital filter”, “following color correction”, “gradation correction”, and “screen generation”. ing.
In addition, a processing path for sequentially performing the processes of “background removal”, “rear color correction”, “tone correction”, and “screen generation” is set for the image characteristic “color background”.

In addition, a processing path for sequentially performing “background removal”, “digital filter”, “gradation correction”, and “screen generation” processing is set for the image characteristic “monochrome photograph”.
Further, for the image characteristic “monochrome character and peripheral background”, a processing path for sequentially performing the processes of “background removal”, “edge extraction”, “digital filter”, “tone correction”, and “screen generation” is set.
In addition, a processing path for sequentially performing “background removal”, “gradation correction”, and “screen generation” processing is set for the image characteristic “monochrome background”.

  Therefore, the image data whose image characteristics for each block has been determined by the characteristic determination means 4 (6) is subjected to image processing by the image processing means 7 by selecting a processing path set in advance for each block and printing. Output as data.

FIG. 5 shows the details of the block subdivision according to the image processing apparatus according to this example more specifically.
In the example shown in the figure, the block size is set to three types of X × Y, X ′ × Y ′, and X ″ × Y ″ (X> X ′> X ″, Y> Y ′> Y ″). Although subdivision is performed in two stages, whether or not subdivision is adopted and the number of stages in subdivision may be set arbitrarily.

First, as shown in FIG. 5B, for example, image data including a black and white character description 21, a black and white photo 22, a color photo 23, a color character description 24, and a black and white character description 25 are provided on a black and white background paper 20. When input, the process is executed according to the procedure shown in FIG.
The image data input by the blocking unit 3 is divided into X × Y blocks (step S1), and the characteristic determining unit 4 determines the image characteristics of each block (step S2). As a result, for the block determined as “monochrome or color background”, the characteristics are notified to the image processing means 7 and each is subjected to image processing through a predetermined processing path (step S7).

On the other hand, blocks with characteristics other than “monochrome or color background” are subdivided by the subdividing means 5 into X ′ × Y ′ blocks (step S 3), and the characteristic discriminating means 6 images the subdivided blocks. The characteristics are determined (step S4). As a result, for the subdivided blocks determined as “monochrome or color photographs”, the characteristics are notified to the image processing means 7 and are subjected to image processing by a predetermined processing path (step S7).
On the other hand, blocks with characteristics other than “monochrome or color photographs” are further subdivided into blocks of X ″ × Y ″ by the subdivision means 5 (step S5), and the characteristic discrimination means 6 is divided into subdivision blocks. Is determined (step S6). As a result, for the subdivided blocks determined as “monochrome or color characters”, the characteristics are notified to the image processing means 7 and each is subjected to image processing through a predetermined processing path (step S7).

FIG. 8 shows the characteristic discriminating means 4 (6) of this example in more detail.
The characteristic determination unit 4 (6) includes a histogram table 21 that holds a histogram for each material characteristic used for characteristic determination, a pixel value extraction unit 22 that extracts a pixel value of each pixel from a block or a subdivided block, and an extracted pixel Histogram creation means 23 for creating a histogram based on the value, and matching means 24 for discriminating the characteristics of the block or subdivided block with reference to the histogram table 21 based on the created histogram are provided.

  The image processing apparatus of this example accepts designation of material characteristics that prioritize image quality from the user. For example, “character”, “photo”, “character / photo mixed”, and the like are designated. The material characteristic is not limited to the user designation, and for example, a material characteristic previously determined by another device based on image data may be used.

The histogram creating means 23 prepares a gradation pattern in which the gradation to be quantized is at least partially rough for each material characteristic in advance, and blocks or subdivisions are made using the gradation pattern according to the designated material characteristic. Create a histogram of the block.
In addition, the histogram table 21 applies a gradation pattern to a typical histogram for each material characteristic as shown in FIG. 3 to obtain a reference histogram having the same quantization width as that of a histogram of a block or a subdivided block. It is possible to easily compare with the histogram of the created block or subdivided block.

As will be described later, the gradation pattern is set to perform quantization suitable for the determination of the histogram of the material characteristics, and by determining using the histogram created by applying this, the corresponding block or Discrimination of subdivided blocks becomes easy.
That is, for example, when “character” is designated as the material property to be prioritized, the histogram of the mixed block of characters and photos is a histogram that emphasizes the character feature, and is identified as the “character” property. Since the image processing is performed based on the processing path “”, the image processing suitable for “character” is performed, and the image printed out can be made clear.

Creation of a histogram using the gradation pattern will be described with reference to FIG.
FIG. 9A shows an example of the brightness histogram of “character”. This lightness histogram has a constant quantization width as shown by dividing the horizontal axis by a to y, and the quantization number is 256 gradations (8 bits).

Here, as shown in FIG. 3 (a) or (b), the brightness histogram of “character” shows a relatively large peak in the highlight part (low density part) and one of the high density part. It has the characteristic of showing a relatively small peak at the position.
Therefore, as shown in FIG. 9B, the gradation pattern of the brightness histogram for determining the “character” characteristic has a relatively fine quantization width (a to g: the above) so that the peak shape can be seen in the low density portion. 9 (A), and it is only necessary to know that one peak appears at any position in the high concentration portion, so that a relatively coarse quantization width (z: h in FIG. 9A) y is set to be one quantization width), and the histogram data amount is compressed while retaining the characteristics of the histogram.

By applying this gradation pattern to FIG. 9 (A), as shown in FIG. 9 (C), as shown in FIG. A lightness histogram is created, and similarly, the “character” characteristic is easily discriminated by comparing with the reference histogram stored in the histogram table 21 created in advance by applying the gradation pattern. be able to.
In addition, since the histogram of FIG. 8C has eight quantization numbers a to g and z, that is, 3 bits, the amount of data is much larger than that of the 8-bit histogram shown in FIG. There are few, and the burden of a discrimination | determination process is reduced and it can discriminate | determine at high speed.

  Further, for example, the brightness histogram of the “photograph” characteristic has a characteristic that the peak is gentle and insufficient as shown in FIG. 3C or 3D. In other words, it is not necessary to know the presence or absence of peaks or the peak shape to determine the “photograph” characteristics, and it is only necessary to know that the lightness values are appropriately dispersed. It is sufficient to use a gradation pattern in which the entire value is divided into eight equal quantization widths). By applying this gradation pattern, a simplified lightness histogram in which the lightness distribution degree is easy to understand is created, and the “photograph” characteristic can be easily determined.

As described above, the gradation pattern is set to leave the shape as a fine quantization width for the part that requires the shape of the histogram for discrimination, and to collect the coarse quantization width for the part that does not need the shape. That's fine.
That is, the gradation pattern is applied to a typical histogram for each material characteristic as shown in FIG. 3 to create a reference histogram and hold it in the histogram table 21 in advance. Then, a histogram is created by applying a gradation pattern corresponding to the material characteristic designated to give priority to image quality for the block or subdivided block. Since these histograms are all simplified by leaving a characteristic portion, it is easy to compare the histogram of the block or subdivided block and the reference histogram, so that the characteristics can be easily discriminated.

  In this example, the histogram table 21 holds a typical reference histogram for each material characteristic, and the matching means 24 compares the histogram of the block or the subdivided block with the reference histogram to determine the characteristic. However, for example, by holding a conditional expression indicating the characteristics of the histogram for each material characteristic in the histogram table 21 and determining the characteristic depending on whether or not the histogram of the block or the subdivided block satisfies a condition, or the like, You may make it discriminate | determine by.

  FIG. 10 shows another example of the image processing apparatus 1 according to the present invention. In addition to the functional units 2 to 8 included in the image processing apparatus illustrated in FIG. Processing end estimation means 31 for calculating the time required for processing the unprocessed portion and the time required for outputting the processed portion, and image data for one frame by the output means 8 when the time required for output exceeds the time required for processing The image data output by the output unit 8 is, for example, print processing by the image forming apparatus, image enlargement / reduction by another image processing apparatus, or another image. Processing such as synthesis is performed.

Each of the functional units 3 to 7, 31, and 32 may be realized by a hardware configuration. In this example, the computer hardware of the image processing apparatus 1 executes the program according to the present invention. It is configured.
The computer hardware of this example includes a plurality of processors, the blocking means 3, the subdividing means 5 and the characteristic discriminating means 4 (6) as one processor, the image processing means 7 as another processor, and the processing. The end estimation means 31 and the output control means 32 are further distributed to other processors and processed in parallel to speed up the processing. For example, the processing is speeded up by a plurality of computer hardware equipped with one processor. The image processing apparatus 1 may be configured by distributing each process.

The image processing apparatus of this example divides input image data for one frame into blocks (subdivided blocks), and performs characteristic determination by the characteristic determination means 4 (6) and image processing by the image processing means 7 respectively. Are applied in turn to the blocks (subdivided blocks).
Here, at a certain point in the process, the remaining processing time required for performing the remaining characteristic determination and image processing is T2, and output by the output means 8 of the data portion subjected to the characteristic determination and image processing is required. The output time is T1.

  FIG. 11 conceptually shows a state in the middle of processing for image data for one frame. Immediately after the image data is input, all the image data is unprocessed. Processing is performed on the unprocessed portion, and the processed portion is gradually processed, and the processed portion is output by the output unit 8. That is, the remaining processing time required for characteristic determination and image processing of the unprocessed portion is T2, and the output time required for outputting the processed portion is T1.

  The processing end estimation unit 31 obtains the data amount d1 of the processed portion in the image data for one frame according to the progress of the processing by the image processing unit 7 in order to calculate the output time T1 and the remaining processing time T2. Further, based on the determination result by the characteristic determination means 4 (6), the data amount d2b of the “background” characteristic portion, the data amount d2t of the “character” characteristic portion, and the “photograph” characteristic portion of the unprocessed portion The data amount d2p and the data amount d2e of the other part are acquired.

  Further, the processing end estimation means 31 uses the “background” characteristic as the output time coefficient t1 required for outputting the unit data amount and the processing time coefficient for each characteristic required for the image processing of the unit data amount as shown in FIG. The processing time coefficient t2b, the processing time coefficient t2t of the “character” characteristic, the processing time coefficient t2p of the “photograph” characteristic, and the processing time coefficient t2e of the other part are set in advance according to the processing capability of the image processing apparatus 1. The output time T1 = d1 × t1 and the remaining processing time T2 = d2b × t2b + d2t × t2b + d2p × t2p + d2e × t2e.

  Then, the output control means 32 compares the output time T1 with the remaining processing time T2, and when the output time T1 ≧ residual processing time T2, the output means 8 starts outputting image data for one frame. .

The timing for starting output of image data will be described with reference to FIG.
When the image processing starts (41), the output time T1 and the remaining processing time T2 are calculated at fixed timings indicated by an ellipse 45 in the figure, and it is determined whether or not the output start condition, that is, T1 ≧ T2 is satisfied. When the condition is satisfied, output of image data for one frame is started (42).
When the output is started at this timing, the output process ends (44) at the same time as or after the end of the image process (43). That is, an unprocessed portion remains at the time when output is started, but when the output time T1 has elapsed, the remaining processing time T2 has already elapsed and all the image data for one frame has been processed. The image data can be output without any delay, and the output destination apparatus can receive the image data without interruption.

FIG. 14 shows a processing flow of output control.
In response to the start of image processing by the image processing means 7, the processing end estimation means 31 calculates the output time T1 and the remaining processing time T2 (step S21). And it is determined whether the output control means 32 satisfy | fills the conditions of T1> = T2 (step S22). If the condition is not satisfied, the system waits for a certain time (step S23), recalculates T1 and T2 (step S21), and re-determines whether the condition is satisfied (step S22). If the condition of T1 ≧ T2 is satisfied, the image data can be output without delay even if the output is started, so the output control means 32 starts the output of the image data for one frame by the output means 8 (step S24).

1 is a configuration diagram of an image processing apparatus according to an embodiment of the present invention. It is a figure explaining the process selection according to the block characteristic which concerns on one Example of this invention. It is a figure explaining the discrimination process of the block characteristic which concerns on one Example of this invention. It is a figure explaining the image processing according to the characteristic which concerns on one Example of this invention. It is a figure explaining the subdivision process of the block which concerns on one Example of this invention. It is a conceptual diagram explaining the difference between this invention and the past. It is a figure explaining an example of an image processing system. It is a figure explaining the characteristic discrimination means based on one Example of this invention. It is a figure explaining the creation process of the histogram which concerns on one Example of this invention. It is a figure explaining the image processing apparatus which concerns on one Example of this invention. It is a figure explaining the processing condition of the image data which concerns on one Example of this invention. It is a figure which shows the processing time coefficient used for calculation of the remaining processing time which concerns on one Example of this invention. It is a figure explaining the start timing of the output process which concerns on one Example of this invention. It is a figure explaining the processing flow of the output control which concerns on one Example of this invention.

Explanation of symbols

1: image processing device, 3: blocking means,
4: characteristic discrimination means, 5: subdivision means,
6: characteristic discrimination means, 7: image processing means,
10: image data, 11: block,
12: Correspondence table between image characteristics and processing paths,
21: Histogram table, 22: Image extraction means,
23: Histogram creation means, 24: Matching means,
31: Processing end estimation means 32: Output control means,

Claims (8)

In an image processing apparatus that outputs image data after performing predetermined processing,
Blocking means for dividing the image data into a plurality of blocks,
Characteristic discriminating means for discriminating the characteristics of the block;
Image processing means for performing a preset process on the block according to characteristics ,
The characteristic discrimination means includes
A gradation pattern in which the gradation level for quantizing the image data of the block is at least partially rough is prepared in advance for each material characteristic that can be specified by the user according to the priority image quality. Using a gradation pattern corresponding to the material characteristic specified by the user among the patterns, creating a histogram based on the pixel values included in the block, and determining the characteristic of the block using the created histogram;
An image processing apparatus. The image processing apparatus according to claim 1.
As a material property that can be specified by the user, it has at least a material property indicating a character,
The gradation pattern corresponding to characters is set to a coarser quantization width in the high density part than in the low density part.
An image processing apparatus. The image processing apparatus according to claim 1.
As a material property that can be specified by the user, it has at least a material property indicating characters and a material property indicating photographs.
The gradation pattern corresponding to characters is set to a coarser quantization width in the high density part than in the low density part,
The gradation pattern corresponding to the photo is set to a quantization width that is coarser than the quantization width of the low density portion in the gradation pattern corresponding to the character and is the same width as the whole.
An image processing apparatus. The image processing apparatus according to any one of claims 1 to 3,
Based on the determination result of the characteristics of each block, it comprises a subdivision means for specifying a block to be further subdivided and further subdividing the block into a plurality of blocks,
The characteristic discriminating unit creates a histogram based on pixel values included in the subdivided block using a gradation pattern corresponding to a material characteristic designated by a user from among gradation patterns prepared in advance. Using the histogram to determine the characteristics of the subdivided blocks,
The image processing means performs a preset process according to characteristics on the block and the subdivided block.
An image processing apparatus. The image processing apparatus according to claim 4 .
The characteristic determining means determines the characteristics of the block based on a brightness histogram of a block image, and further determines the characteristics of the subdivided block based on a saturation histogram of the subdivided block image. apparatus. The image processing apparatus according to any one of claims 1 to 5,
Output means for outputting image data for one frame subjected to the processing;
Based on the data amount of the processing performed portion of the image data of one frame, and calculates the time required for output of the data portion, the data amount of the processing is not yet subjected portion in the image data and Processing end estimation means for calculating a time until processing for the data portion is ended based on the determined characteristics ;
By comparing the time until the processing ends and the time required for the output, when the time required for the output exceeds the time until the processing end, the output control means for starting the output of the image data of one frame by the output means And an image processing apparatus. In the method of dividing the image data into a plurality of blocks, determining the characteristics of each block, performing a process set in advance according to the characteristics of each block, and outputting it,
A gradation pattern in which the gradation level for quantizing the image data of the block is at least partially rough is prepared in advance for each material characteristic that can be specified by the user according to the priority image quality. Using a gradation pattern corresponding to the material characteristic specified by the user among the patterns, creating a histogram based on the pixel values included in the block, and determining the characteristic of the block using the created histogram;
An image processing method. A program that realizes a processing function in a computer for performing predetermined processing and outputting image data,
A function for dividing image data into a plurality of blocks;
A function to determine the characteristics of each block;
A program that configures a computer with a function for performing processing set in advance for each block according to characteristics ,
A gradation pattern in which the gradation level for quantizing the image data of the block is at least partially rough is prepared in advance for each material characteristic that can be specified by the user according to the priority image quality. Using a gradation pattern corresponding to the material characteristic specified by the user among the patterns, creating a histogram based on the pixel values included in the block, and determining the characteristic of the block using the created histogram;
A computer program characterized by the above.

Images