JP2004194050A - Image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of scaling processing and color space conversion processing in response to a demand for image quality and a processing speed. <P>SOLUTION: A controller 5 controls a scaling processor 3 and a color conversion processor 4 to perform processing at a shorter processing time sequence between a case of performing the color space conversion processing by the color conversion processor 4 posterior to the scaling processing by the scaling processor 3, and a case of performing the scaling processing by the scaling processor 3 posterior to the color space conversion processing by the color conversion processor 4. Further, when a plurality of scaling methods are provided in the scaling processor 3, the controller 5 controls to select a scaling method for the scaling processing. By this, the scaling processing and the color space conversion processing can be performed in a shortened processing time to the maximum extent, while using a scaling method for obtaining the image quality designated from a designation input section 6. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a technology for processing digitized images, and for example, to a technology used in various systems that handle digital images, such as a display device or a printer having an image display function. In particular, the present invention relates to an image processing technique including an image scaling process and a color space conversion process.
[0002]
[Prior art]
2. Description of the Related Art In recent years, various digital image obtaining means such as a digital image capturing function attached to a digital camera or a mobile phone and a homepage on the Internet have appeared. Furthermore, with the spread of high-quality printers at home and the emergence of multifunctional color printing systems installed in print shops and convenience stores, users can use these environments to display, edit, and print digital images. Opportunities for processing are increasing dramatically.
[0003]
In such an environment, unlike before, images having various color spaces and image sizes / resolutions are distributed. In order to display / print these images, various types of image processing such as enlargement / reduction, rotation, color space conversion, filtering, and synthesis are required. Above all, enlargement / reduction and color space conversion are essential functions for general user requests, such as displaying / printing an image taken by a digital camera and printing a homepage on the Internet. In particular, as described above, as the digital image obtaining means has diversified, the enlargement function has a small image size of about 100,000 pixels to 1,000,000 pixels, and a display resolution of less than 100,000 pixels. Because low-resolution images are often used, it has become important to print these low-resolution images well.
[0004]
Assuming that such a low-resolution image is printed by a high-resolution printer or the like, if the enlargement processing is performed using the most commonly used nearest neighbor method, the pixels are simply enlarged because the pixels are simply enlarged. In the portion, the boundaries between blocks having the same pixel value are noticeable and rugged. Further, in the edge portion, since the linear edge becomes stair-like and is visually recognized as a jaggy, the image quality is extremely poor.
[0005]
Therefore, as means for enlarging these low-resolution images with better image quality than the nearest neighbor method, there is a method of performing higher-order interpolation (for example, linear interpolation, bicubic interpolation, cubic convolution method, etc.). In addition, there is a method of estimating missing information by repeatedly performing interpolation in a frequency space as described in Patent Document 1, for example. Further, as described in Patent Document 2, for example, there is a method of generating a suitable enlarged image using fractals. Furthermore, as described in Japanese Patent Application No. 2002-76896, a method of obtaining a good edge by performing angle estimation using multi-value pattern matching, or as described in Japanese Patent Application No. 2002-79388. There is also a method of obtaining good image quality by searching for a reduced similar block from the vicinity and performing replacement. As described above, various methods have been proposed as enlargement processing methods.
[0006]
As described above, various methods aimed at enlarging low-resolution images with high image quality have high image quality, but have the disadvantage that they take much longer to process than simple methods such as the nearest neighbor method. is there. As a result, the performance of the entire display / print processing is significantly reduced, and as a result, the productivity of the user is reduced.
[0007]
As long as high image quality is desired, it is unavoidable that the processing becomes complicated and the speed is reduced, but there is a device to suppress the degree of the reduction. Of course, each of the above-described high image quality enlarging methods has a different speed with respect to image quality. For example, the methods described in Japanese Patent Application Nos. 2002-76896 and 2002-79388 minimize the speed reduction while maintaining high image quality. It is designed to keep it to a minimum. Apart from the improvement of the enlargement method itself, there is also an invention in which the processing speed is increased by changing the application method of the enlargement method. As an example, there is a method described in Patent Document 3 or Patent Document 4.
[0008]
The techniques described in Patent Literatures 3 and 4 utilize the fact that the human visual detection ability differs for each color component in a color image. For example, if the input image is in the RGB color space, a high-quality and low-speed enlargement method is used for the most visually sensitive G component, and a medium-quality and medium-speed By using a technique and using a low-quality and high-speed technique for the B component with the lowest sensitivity, the aim is to improve the processing speed of the entire enlargement processing without visually lowering the image quality very much.
[0009]
On the other hand, in consideration of acquiring image data from various sources and outputting image data to various output destinations as described above, a conversion process is also required for the color space of the image data. For example, an RGB color space or a YCbCr color space is used as input image data when compression is performed according to the JPEG-DCT method. As a color space received by an output destination, for example, many display devices such as a display use an RGB color space, and a printer often uses a CMYK color space. Thus, a conversion process of the color space from various color spaces to various color spaces is required.
[0010]
At this time, considering the number of color components in the color space before conversion and the color space of the conversion destination, the number of color components is 3 in the RGB color space and the YCbCr color space, but the number of color components in the CMYK color space is It becomes 4. Considering the case where image processing is performed in a later stage, it is more likely that processing with a smaller number of color components can be performed at a higher speed. Conversely, if a process that increases the number of pixels (for example, the above-described enlargement process) is performed before the color space conversion process, the time required for the color space conversion process increases.
[0011]
As described above, when the data amount of the image data changes in each image processing, the processing time in each processing changes. Therefore, high-speed processing is possible by shortening the processing time only by changing the processing order. However, in the conventional image processing apparatus, the color space of image data to be input and output may be specified, and the order of processing is fixed. Therefore, even when various image data are subjected to enlargement / reduction processing and color space conversion processing at an arbitrary magnification, it has not been possible to increase the speed.
[0012]
In addition, for example, when the compression is performed by the above-described JPEG-DCT method, the resolution may be reduced depending on the color component, but the processing is performed as it is in the processing that can be performed with the reduced resolution. The processing time can be shortened more than performing the processing after returning the resolution. However, conventionally, such a relationship has not been considered. Furthermore, since the image quality reproducibility of the output device is not taken into consideration, there is a disadvantage that high-quality enlargement is uselessly performed.
[0013]
[Patent Document 1]
JP-A-3-204268
[Patent Document 2]
Japanese Patent No. 2968582
[Patent Document 3]
JP-A-2000-151989
[Patent Document 4]
JP 2000-182022 A
[0014]
[Problems to be solved by the invention]
The present invention has been made in view of the above-described circumstances, and has as its object to provide an image processing apparatus capable of performing high-speed scaling processing and color space conversion processing in accordance with image data. Still another object of the present invention is to provide an image processing apparatus capable of performing image processing in accordance with a request for image quality or processing speed, for example, an instruction from an operator or image output capability of an output device.
[0015]
[Means for Solving the Problems]
The present invention provides, in an image processing apparatus, an image enlargement / reduction unit capable of performing an enlargement / reduction process on image data, a color space conversion unit capable of performing a color space conversion process on image data, When the input / output image data is subjected to the color space conversion processing by the color space conversion means after the enlargement / reduction processing by the image enlargement / reduction means, and after the color space conversion processing by the color space conversion means, The image processing apparatus further includes a control unit that controls the image enlargement / reduction unit and the color space conversion unit so that the processing is performed in a processing order in which the processing time is short when the enlargement / reduction processing is performed. By performing the processing in the order in which the processing time is shortened in this way, it is possible to provide an image processing apparatus capable of performing image processing at high speed.
[0016]
When the control unit controls the processing order according to the processing time, it is possible to take into account the enlargement / reduction ratio in the image enlargement / reduction processing unit and the number of color components in the color space before and after the conversion processing by the color space conversion unit. When the image data is compressed in the JPEG-DCT format, the processing order is controlled in consideration of the color space at the time of compression of the image data decompressed by the decompression means while maintaining the color space at the time of compression. be able to. At this time, when a specific color component is thinned out at the time of compression, the processing order can be controlled in consideration of the resolution of each color component of the image data while keeping the resolution different for each color component. Thus, depending on the color components, the processing is performed with the low resolution, and the processing time can be reduced.
[0017]
Further, the image enlargement / reduction unit may be configured to be capable of performing enlargement / reduction processing by a plurality of enlargement / reduction methods having different processing loads and processing image qualities. In this case, the control means controls the scaling method used in the image scaling means and the processing order. At this time, for example, the enlargement / reduction technique and the processing order can be controlled based on the image quality or processing speed desired as an output image. For example, the scaling method can be selected for each color component based on the visual characteristics of the color components of the image data. Alternatively, the information presenting means presents guide information about the processing time and the image quality after processing by a combination of the processing order or the further scaling method, and controls the processing order according to the processing order instructed by the operator or the further scaling method. Alternatively, it can be configured to further select a scaling method. Alternatively, it is possible to control the processing order or select a scaling method in accordance with the image quality reproducibility and the output size / resolution of the image output means for outputting the processed image data. In this way, optimal image processing can be performed according to the desired image quality, speed, image quality reproduction capability of the image output means, and the like.
[0018]
BEST MODE FOR CARRYING OUT THE INVENTION
FIG. 1 is a block diagram showing a first embodiment of the image processing apparatus of the present invention. In the figure, 1 is an image storage unit, 2 is an image input unit, 3 is an enlargement / reduction processing unit, 4 is a color conversion processing unit, 5 is a control unit, and 6 is an instruction input unit. The image storage unit 1 stores digitized image data to be processed. The retained image data may be in an uncompressed format, or may be compressed in various formats, including the JPEG-DCT format.
[0019]
The image input unit 2 reads out image data stored in the image storage unit 1. At this time, the compressed image data can be read out while performing decompression processing, and has a function as decompression means. Further, when thinning is performed by color components as in the JPEG-DCT format, reading at a different resolution depending on the color components can be performed while the thinning is performed.
[0020]
The enlargement / reduction processing unit 3 can perform enlargement / reduction processing of image data by one or more enlargement / reduction methods having different image quality and processing speed (processing load). Specific examples of some scaling methods will be described later.
[0021]
The color conversion processing unit 4 has a function of converting a color space (for example, RGB, YCbCr, CMYK, etc.) assumed for input image data into a color space (for example, RGB, CMYK, etc.) required as an output. This function is realized by various methods such as a combination of a matrix operation and a lookup table process, and a three-dimensional direct lookup table process with interpolation. Since the details of these processes are described in detail in documents relating to various color processes, detailed descriptions thereof are omitted here.
[0022]
The control unit 5 controls processing of each unit. In particular, the case where the color space conversion processing is performed by the color conversion processing unit 4 after the enlargement / reduction processing by the enlargement / reduction processing unit 3 for the image data to be processed and the case where the color space conversion processing by the color conversion processing unit 4 is performed The enlargement / reduction processing unit 3 and the color conversion processing unit 4 are controlled so that the processing is performed in the processing order in which the processing time becomes shorter when the enlargement / reduction processing is performed by the processing unit 3. In addition, the enlargement / reduction processing unit 3 controls selection of an enlargement / reduction technique for performing enlargement / reduction processing. The processing time is compared in consideration of the enlargement / reduction ratio, the number of color components, and the resolution of each color component in the case of being compressed by the JPEG-DCT method.
[0023]
The instruction input unit 6 inputs an instruction selected from a plurality of options relating to image quality or processing time of an output image by an operator, and transmits the instruction to the control unit 5.
[0024]
Next, an outline of a scaling method that can be executed by the scaling processing unit 3 will be described with some examples. In the following description, it is assumed that the image is enlarged. However, in some methods, the reduction processing can be performed as it is, or a separate reduction method may be provided.
[0025]
FIG. 2 is an explanatory diagram of the enlargement process by the nearest neighbor method. According to the nearest neighbor method, as shown in FIG. 2, the pixel value Pout of the output image shown in FIG. 2A is a point obtained by inversely mapping the center point on the input image (shown by a black circle in FIG. 2B). ) Is a method of setting the pixel value Pin on the input image closest to (). Specifically, it is obtained from the pixel value Pin of the input image and the magnification (rx, ry) in the horizontal / vertical directions by the following equation (1). Note that floor (a) is a function for obtaining an integer value obtained by truncating a decimal part of a.
Pout (X, Y) = Pin (floor ((X + 0.5) / rx), floor ((Y + 0.5) / ry)) (Equation 1)
When processing is performed as in Expression 1, an output image can be obtained by subjecting an input image to a nearest-neighbor enlargement process at a magnification (rx, ry).
[0026]
In Equation 1, the calculation is performed for each point of the output. However, in practice, a method called DDA (Digital Difference Analyzer) is used to add values based on the values one point before each of X and Y. Processing can be speeded up by performing calculations using only (Expression 2).
cx = (X + 0.5) / rx
cy = (Y + 0.5) / rx
dx = 1 / rx
dy = 1 / ry
Pout (X + 1, Y + 1) = Pin (floor (cx + dx), floor (cy + dy)) (Equation 2)
[0027]
FIG. 3 is an explanatory diagram of the enlarging process by the linear interpolation method. In the linear interpolation method, as shown in FIG. 3, the pixel value Pout of the output image is obtained by linearly interpolating the pixel values of four points near the point where the center point of Pout is inversely mapped on the input image for each axis. It becomes a pixel value. The inverse mapping point can be calculated only by addition using DDA as in the nearest neighbor enlarging method. For example, when the input image is enlarged twice by using the linear interpolation method, when (4, 2) on the output image is inversely mapped on the input image according to the above-described equation 1, (2.25, 1.25) is obtained. Become. In this case, since the pixel position of the input image is internally divided into 3: 1 in both the X direction and the Y direction, when the interpolation process is performed in the X direction,
P '= (Pin1 + 3 · Pin3) / 4
P ″ = (Pin2 + 3 · Pin4) / 4
When interpolation processing is performed in the Y direction using these,
P′out = (P ′ + 3 · P ″) / 4
It becomes. Here, the interpolation processing is performed using four neighboring points, but the interpolation processing may be performed using more neighboring points.
[0028]
FIG. 4 is an explanatory diagram of the enlargement processing by the multi-edge pattern method. The multi-edge pattern method is, for example, a high-quality enlargement method proposed in Japanese Patent Application No. 2002-76896. The details are described in Japanese Patent Application No. 2002-76896, and will be briefly described here. As shown in FIG. 4, first, in S11, a small image block including a target pixel block and peripheral pixels is cut out. In S12, the edge angle is estimated from each of the target pixel block and the area block including the peripheral pixels. In S13, an enlarged pattern prepared in advance corresponding to the edge angle estimated in S12 is selected, and an enlarged block area is generated and replaced in S14 based on the enlarged pattern. As a result, a high-quality enlarged image with sharp edges can be obtained.
[0029]
The multi-edge pattern method involves complicated steps such as peripheral pixel reference, edge angle estimation, pattern selection, enlarged block generation, and replacement processing. Therefore, compared to the above-described nearest neighbor method, linear interpolation method, and the like, this method has high image quality but requires much processing time.
[0030]
As other enlargement / reduction methods, for example, many methods such as the methods proposed in the above-mentioned Patent Document 1, Patent Document 2, and Japanese Patent Application No. 2002-79388 can be provided. In the following, in order to simplify the description, it is assumed that three scaling methods of the above-described nearest neighbor method, linear interpolation method, and multi-edge pattern method are provided, and the nearest neighbor method is method 1 and the linear interpolation method is method 2 , The multi-edge pattern method is referred to as method 3.
[0031]
Next, an outline of the operation of the control unit 5 will be described. As described above, the control unit 5 determines whether the color space conversion processing by the color conversion processing unit 4 is performed after the scaling processing by the scaling processing unit 3 on the image data to be processed, When the enlargement / reduction processing by the enlargement / reduction processing unit 3 is performed after the conversion processing of the color space by the processing, the enlargement / reduction processing unit 3 and the color conversion processing unit 4 are controlled so that the processing is performed in the processing order in which the processing time is shortened. . Before describing such control, the relationship between the processing times in the enlargement processing storage unit 3 and the color conversion processing unit 4 will be described.
[0032]
FIG. 5 is an explanatory diagram of a relationship between processing times in the enlargement processing storage unit 3 and the color conversion processing unit 4. FIG. 5A shows the relationship between the magnification and the processing time in the scaling processing by the scaling processing unit 3, and FIG. 5B shows the relationship between the number of pixels and the processing time in the color conversion processing unit 4. I have. In the enlargement / reduction processing performed by the enlargement / reduction processing unit 3, although there are differences depending on the above-described enlargement / reduction methods, generally, the magnification is large and the processing time is required as the size of the image to be generated is large. Therefore, the graph becomes ascending to the right as shown in FIG. In many cases, the value of the pixel after the enlargement is determined for each color component, and thus the processing time is related to the number of color components. For example, when the color space of the image data has three color components such as RGB and YCbCr, and when the color space has four color components such as CMYK, as shown in FIG. The larger the number, the longer the processing time.
[0033]
On the other hand, in the color space conversion processing performed by the color conversion processing unit 4, since processing is performed for each pixel, the processing time generally depends on the number of pixels. Therefore, as shown in FIG. 5B, the processing time becomes longer as the number of pixels increases.
[0034]
Based on these relationships, a case where the color space conversion processing is performed by the color conversion processing unit 4 after the enlargement / reduction processing by the enlargement / reduction processing unit 3 and a case where the enlargement / reduction processing unit 3 is performed after the color space conversion processing by the color conversion processing unit 4 are performed. It is sufficient to check which processing time is shorter when the enlargement / reduction processing is performed. For example, in the example shown in FIG. 5, it is assumed that the magnification is α, and the number of pixels increases from β to γ by scaling processing with the magnification α. It is also assumed that the number of color components increases from three to four by the color space conversion process. In this case, if the enlargement / reduction processing is performed first, the number of color components of the image data is 3, so the processing time is t3, and the subsequent color space conversion processing by the color conversion processing unit 4 is after the enlargement / reduction processing. The number of pixels is γ, and the processing time is required by tγ. Conversely, when the color space conversion processing is first performed by the color conversion processing unit 4, the number of pixels is β and the processing time is tβ. After that, when the enlargement / reduction processing is performed, the number of color components becomes 4 by the color space conversion processing, so that the processing time is required only by t4. Therefore, it can be seen that it is sufficient to simply compare t3 + tγ and tβ + t4 and select the one with a shorter processing time.
[0035]
Actually, in addition to this, if the processing time is different depending on the enlargement / reduction method, or if the compression method is changed by the JPEG-DCT method, the resolution for each color component or the enlargement / reduction method for each color component is changed. The processing time may be predicted and compared in consideration of various factors such as the processing time in each scaling method. In the case of a scaling method or a color conversion process that does not depend on an image as in the above-described scaling methods 1 to 3, a considerably accurate prediction value can be obtained from the size and magnification of the input image data. .
[0036]
FIG. 6 is an explanatory diagram of a specific example of operation control on the scaling processing unit 3 and the color conversion processing unit 4 in the control unit 5. Here, the control unit 5 receives attribute information (image size, color space information, and the like) of the processing target image from the image input unit 2 and transmits selection instruction information regarding image quality / processing time from the instruction input unit 6. Shall be The selection instruction information may be, for example, such that “high image quality / low speed”, “standard image quality”, and “low image quality / high speed” are assigned to “0”, “1”, and “2”, respectively. Depending on the number, more pieces of selection information may be divided. In FIG. 6, for simplicity, the description will be made assuming that the selection instruction information is any one of "0", "1", and "2". Here, it is assumed that RGB, YCbCr, CMYK can be processed as the input / output color space, and the image size (or magnification ratio) after scaling and the output color space are stored in the control unit 5 in advance or the instruction input unit 6 or It has been instructed by some means that has not been done. The enlargement / reduction ratio is larger than 1, that is, the enlargement process is performed.
[0037]
When “0” is instructed to the control unit 5 from the instruction input unit 6 as the selection instruction information (mode “0”), the control unit 5 operates as follows. First, the attribute information (image size, color space, etc.) of the input image stored in the image storage unit 1 is obtained using the image input unit 2. As a result, when the color space of the input image is equal to the required output color space, the method 3 that requires the highest image quality and processing time among the plurality of scaling methods provided in the scaling processing unit 3 is used. The magnification is set by selecting only one, and the image input unit 2 reads out the image data in the image storage unit 1 while performing the enlargement processing by the enlargement / reduction processing unit 3. At this time, when the image data is compressed, the image data is read out while performing the decompression process. When the resolution is different due to thinning out by the color component, the magnification is adjusted so that the image size after the enlargement becomes equal for each color component. Make settings. In this case, the color conversion processing by the color conversion processing unit 4 is unnecessary.
[0038]
When the input color space and the output color space are not equal, a three-component image is converted to the same three-component different color space from RGB to YCbCr or vice versa, or a four-color image is converted from CMYK to RGB or YCbCr. When converting a component into a three-component color space, it is more advantageous to perform scaling processing on image data in the three-component color space. It is more advantageous to perform the color space conversion process before the number of pixels is increased by the enlargement process. Therefore, the color space of the image data is converted by operating the color conversion processing unit 4 on the image input from the image input unit 2 first. At this time, for example, when the Cb and Cr components are thinned out from the Y component, the processing is performed using the pixel values adjacent to the thinned pixels. After that, in the same manner as described above, only one of the plurality of enlargement / reduction methods stored in the enlargement / reduction processing unit 3 which has the highest image quality and the longest processing time is selected, the magnification is set, the enlargement processing is performed, and the output is performed. Get an image. The control unit 5 outputs the processing result to, for example, a display device or a printer device (not shown) outside the system.
[0039]
In the case where the input color space is composed of three components such as RGB and YCbCr and the output color space is composed of four components such as CMYK, first, the image input from the image input unit 2 is provided in the scaling unit 3. A single method 3 is selected from a plurality of enlargement / reduction methods, a magnification is set, an enlargement process is performed, and the color conversion processing unit 4 is operated on the enlarged image to change the color space of the image. Convert to obtain an output image. Here, the reason why the enlargement / reduction processing was performed prior to the color space conversion processing is that the difference in the enlargement processing time between the four-component image and the three-component image using the method 3 is based on the color conversion processing before and after the enlargement. This is because it is determined that the difference is larger than the time difference. As described above, since the enlargement / reduction processing time and the color space conversion processing time also depend on the magnification (the number of pixels), it may be determined that the color conversion processing is performed first depending on the magnification. The determination is made by the control unit 5.
[0040]
Further, as described above, when the image data is compressed in a color space such as YCbCr or Lab by the JPEG-DCT compression, the Cb and Cr components and the a and b components are thinned out depending on the compression conditions, and the Y and L components are thinned out. The resolution may be lower. In this case, the control unit 5 determines the order of the color conversion and the enlargement in consideration of the difference in the processing time due to the enlargement process of different magnification depending on the color component.
[0041]
The reason why the enlargement was performed using the same method 3 for each color is that when the enlargement is performed by a different method, the color bleeding caused by the difference in the enlargement method that occurs at a boundary portion of a different color is suppressed. This is for performing high-quality enlargement. Of course, the processing time is increased by the use of the enlargement / reduction technique for performing the high-quality enlargement processing, so that the processing speed is reduced.
[0042]
Next, the case where “1” is instructed to the control unit 5 from the instruction input unit 6 as the selection instruction information (mode “1”) will be described similarly. First, attribute information (image size, color space, etc.) of an input image stored in the image storage unit 1 is acquired from the image input unit 2. As a result, if the color space of the input image is equal to the required output color space, the following processing is performed.
1) In the case of RGB
In RGB, G>R> B has a large influence on vision in the order of G>R> B. Therefore, G is the method 3, R is the method 2, and B is the method 1, and enlargement processing is performed.
2) For YCbCr
In YCbCr, the effect on vision is Y >> Cb, Cr. Therefore, Y performs processing in method 3, and Cb and Cr perform processing in method 2 (or method 1).
3) For CMYK
CMYK has a large effect on vision as K> C and M> Y. Therefore, K performs processing according to method 3, C and M perform processing according to method 2, and Y performs processing according to method 1.
[0043]
At this time, if the input image is compressed, read it out while performing decompression processing. If the resolution differs due to thinning out depending on the color component, the magnification is set so that the image size after expansion becomes equal for each color component. Make the settings for Note that the color space conversion processing by the color conversion processing unit 4 is not performed.
[0044]
When the input color space and the output color space are not equal, a three-component image is converted to the same three-component different color space from RGB to YCbCr or vice versa, or a four-color image is converted from CMYK to RGB or YCbCr. In the case where the color space is converted from the components into a three-component color space, the color space of the image data is first converted by the color conversion processing unit 4 for the image input from the image input unit 2 in the same manner as in the mode 0 described above. After the conversion, the same enlargement processing as in the above 1) to 3) is performed based on the subsequent color space. The control unit 5 outputs the processing result to, for example, a display device or a printer device (not shown) outside the system.
[0045]
When the input color space is composed of three components such as RGB and YCbCr and the output color space is composed of four components such as CMYK, first, the image input from the image input unit 2 is processed according to the color space. The same enlargement processing as in 1) to 3) is performed, and the color conversion processing unit 4 converts the color space of the image data to the enlarged image to obtain an output image. Here, the reason why the enlargement / reduction processing is performed before the color conversion processing is that the control unit 5 performs the enlargement of the four-component image and the three-component image when the method 3 is used, as in the case of the mode “0” described above. This is because it is determined that the difference between the processing times is larger than the difference between the color conversion processing times before and after the enlargement. As described above, since the enlargement / reduction processing time and the color space conversion processing time also depend on the magnification, it may be determined that the color conversion processing is performed first depending on the magnification. The determination is made by the control unit 5 as in the case of the mode “0”.
[0046]
In this processing, the processing order of the enlargement / reduction processing and the color space conversion processing is not changed as compared with the case where the mode “0” is designated, but the enlargement processing is performed in a mixed method with the method 2 or the method 1 which is faster. Therefore, the overall processing time is faster than in the case of the mode “0”. However, regarding image quality, visually important components are sharply expanded, while other components are slightly blurred or block-shaped, so color boundaries are different at different color boundaries. A phenomenon such as bleeding occurs and the image quality is slightly reduced.
[0047]
Next, the case where “2” is instructed to the control unit 5 from the instruction input unit 6 as the selection instruction information (mode “2”) will be described similarly. First, attribute information (image size, color space, etc.) of an input image stored in the image storage unit 1 is acquired from the image input unit 2. As a result, when the color space of the input image is equal to the required output color space, one of the methods 1 having the highest speed and the lowest image quality is selected from among the plurality of scaling methods provided in the scaling unit 3. Is selected, the magnification is set, and the image input unit 2 reads the input image from the image storage unit 1 while performing the enlargement processing by the selected enlargement / reduction method of the method 1. At this time, when the input image is compressed, readout is performed while performing decompression processing. When the resolution is different due to thinning out by color components, the magnification is adjusted so that the image size after expansion becomes equal for each color component. Make settings.
[0048]
When the input color space and the output color space are not equal, unlike the above-described modes “0” and “1”, the conversion to any color space requires the color conversion processing to be performed before the enlargement. Do. Specifically, the color conversion processing unit 4 performs a color space conversion process on the image data input from the image input unit 2, and the image data after the color space conversion process is provided in the scaling processing unit 3. Only one of the fastest and lowest image quality methods 1 is selected from the plurality of enlargement / reduction methods described above, the magnification is set, and the enlargement processing is performed. Next, the control unit 5 outputs the processing result to, for example, a printer (not shown) outside the system.
[0049]
Here, unlike the modes “0” and “1”, the reason why the processing is always performed prior to the enlarging processing when the color conversion processing is necessary is that the control unit uses the method 1 which is a high-speed enlarging / reducing method. This is because No. 5 determined that the difference between the enlargement processing times of the four-component image and the three-component image when the method 1 was used was smaller than the difference between the color conversion processing times before and after the enlargement. The image quality is lower than that of the modes “0” and “1” because the low-image-quality / high-speed system 1 is used. As described above, since the enlargement / reduction processing time and the color space conversion processing time also depend on the magnification, it may be determined that the color conversion processing is performed later depending on the magnification. The determination is made by the control unit 5 as in the case of the modes “0” and “1”.
[0050]
As described above, the processing according to the user's request is executed by selecting the scaling method and changing the processing order of the color space conversion processing and the scaling processing according to the selection of the image quality / processing time. be able to. Of course, FIG. 6 described above is an example of the determination by the control unit 5, and the processing order and the scaling method applied to each color component are not limited to this example.
[0051]
Here, for simplicity of explanation, the selection instruction information is described as being narrowed down to three types of “0”, “1”, and “2”. However, for example, in mode “1”, instead of method 2 as the enlargement method, By using the method 1, it is possible to provide a mode in which the image quality and the processing time are intermediate between “1” and “2”. If more enlargement processing modules are prepared in the enlargement / reduction processing unit 3, it is possible to control the balance between image quality and processing time in smaller units. Further, in the above description, the case where the enlargement / reduction processing unit 3 mainly performs the enlargement processing is described, but the same applies to the case where the reduction processing is performed.
[0052]
Further, the control unit 5 predicts and calculates the enlargement / reduction processing time and the color space conversion processing time before performing the processing on the image data. In addition, for example, a table as shown in FIG. In addition, it is also possible to adopt a configuration in which the processing order or the enlargement / reduction technique is selected by referring to the table.
[0053]
FIG. 7 is a block diagram showing a second embodiment of the image processing apparatus of the present invention. In the figure, the same parts as those in FIG. Reference numeral 7 denotes an information presentation unit. The information presenting unit 7 presents information to the operator, and the operator can give an instruction from the instruction input unit 6 while referring to the information presented on the information presenting unit 7. In particular, in this embodiment, according to the control from the control unit 5, the processing time and the post-processing time in combination with the processing order of the scaling process and the color space conversion process and the selection of the scaling method provided in the scaling process unit 3 Information about the image quality is presented. The processing time and image quality can be predicted based on the input image size, enlargement magnification, and input / output color space information. In particular, in the case of the scaling method that does not depend on the content of the image as in the above-described methods 1 to 3, the predicted value of the processing time can be obtained fairly accurately from the input image size and the magnification. As for the image quality, if a level of the output image quality is selected in advance by a subjective evaluation experiment or the like, the level value can be presented.
[0054]
By presenting information such as the processing time and image quality in each of these modes, an instruction from the instruction input unit 6 by the operator can be assisted. The operator may select a processing order or a scaling method (such as mode selection) from the instruction input unit 6 based on the presented information.
[0055]
When the number of modes is as small as three as in the example described in the first embodiment, the operator can easily select the mode. However, as the number of modes increases, there may be a case where the operator loses the selection. In such a case, as in the second embodiment, information indicating how much processing time and image quality can be obtained in each mode is presented to the operator by the information presenting unit 7. This can assist the operator's selection.
[0056]
FIG. 8 is a block diagram showing a third embodiment of the image processing device of the present invention. In the figure, the same parts as those in FIGS. Reference numeral 8 denotes a display device, and 9 denotes a printer device. In the third embodiment, the display device 8 and the printer device 9 are explicitly shown as output devices. If information such as the image quality reproducibility and resolution of these output devices is held in the control unit 5, it is possible to perform processing at higher speed without performing useless high image quality enlarging processing that cannot be reproduced visually.
[0057]
For example, the display device 8 is capable of multi-value display and has a low resolution. For this reason, if the operator intends to view at high magnification, the image quality difference is likely to be visually recognized. For example, the mode “1” is designated in the example shown in FIG. 6 and the color space is set to YCbCr image data. When performing the scaling process, the Y component is processed by the method 3 and the Cb and Cr components are processed by the method 2 according to the instruction. However, the printer 9 is screen-processed and has a high resolution, so that it is rarely seen in detail. Therefore, for example, the Y component can be processed by method 3, and the Cb and Cr components can be processed by method 1. In this way, it is possible to automatically switch to a faster method within a range of image quality deterioration that cannot be visually recognized by the operator.
[0058]
FIG. 9 is an explanatory diagram of an example of a computer program and a storage medium storing the computer program when the functions of the image processing apparatus of the present invention are realized by the computer program. In the figure, 21 is a program, 22 is a computer, 31 is a magneto-optical disk, 32 is an optical disk, 33 is a magnetic disk, 34 is a memory, 41 is a magneto-optical disk device, 42 is an optical disk device, and 43 is a magnetic disk device.
[0059]
Some or all of the functions of the image input unit 2 and the enlargement / reduction processing unit 3 and the functions of the color conversion processing unit 4, the control unit 5, and the instruction input unit 6 described in each of the embodiments of the present invention described above are implemented by a computer. It can also be realized by a program 21 that can be executed. In this case, the program 21 and data used by the program can be stored in a computer-readable storage medium. A storage medium is a type of signal corresponding to a change in energy such as magnetism, light, electricity, etc., caused to a reading device provided in a hardware resource of a computer in accordance with a description content of a program. Thus, the program description can be transmitted to the reading device. For example, there are a magneto-optical disk 31, an optical disk 32 (including a CD and a DVD), a magnetic disk 33, a memory 34 (including an IC card, a memory card, and the like). Of course, these storage media are not limited to portable types.
[0060]
The programs 21 are stored in these storage media, and these storage media are mounted in, for example, the magneto-optical disk device 41, the optical disk device 42, the magnetic disk device 43, or a memory slot (not shown) of the computer 22, so that the computer 21 The program 21 can be read to execute the functions of the image processing apparatus of the present invention or the image processing method. Alternatively, a storage medium may be mounted on the computer 22 in advance, the program 21 may be transferred to the computer 22 via, for example, a network, and the program 21 may be stored in the storage medium and executed. The image data storage unit 2 can use a memory in the computer 22, an attached magnetic disk device, or another storage medium.
[0061]
Of course, some of the functions of the present invention may be configured by hardware, or all functions may be configured by hardware. In addition to being realized by one computer as shown in FIG. 9, the present invention can be implemented by distributing functions to a plurality of computers via, for example, a LAN, the Internet, or the like. A configuration that realizes the function of the image processing apparatus described above may be used.
[0062]
【The invention's effect】
As is apparent from the above description, according to the present invention, by appropriately controlling the processing order of the scaling process and the color space conversion process, and by selecting and using an appropriate method from a plurality of scaling methods, An image processing device capable of high-speed processing can be provided. In addition, it is possible to perform a scaling process and a color space conversion process according to a request for image quality / processing time. Further, there is an effect that image processing capable of performing high-speed processing can be realized without performing excessive image quality improvement processing due to the performance of the output destination.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a first embodiment of an image processing apparatus according to the present invention.
FIG. 2 is an explanatory diagram of an enlarging process by the nearest neighbor method.
FIG. 3 is an explanatory diagram of an enlarging process by a linear interpolation method.
FIG. 4 is an explanatory diagram of an enlargement process by a multi-edge pattern method.
FIG. 5 is an explanatory diagram of a relationship between processing times in an enlargement processing storage unit 3 and a color conversion processing unit 4;
FIG. 6 is an explanatory diagram of a specific example of operation control for the enlargement / reduction processing unit 3 and the color conversion processing unit 4 in the control unit 5;
FIG. 7 is a block diagram illustrating a second embodiment of the image processing apparatus according to the present invention.
FIG. 8 is a block diagram illustrating a third embodiment of the image processing apparatus according to the present invention.
FIG. 9 is an explanatory diagram of an example of a computer program when the functions of the image processing apparatus of the present invention are implemented by the computer program, and an example of a storage medium storing the computer program.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Image memory part, 2 ... Image input part, 3 ... Enlargement / reduction processing part, 4 ... Color conversion processing part, 5 ... Control part, 6 ... Instruction input part, 7 ... Information presentation part, 8 ... Display device, 9 ... Printer Apparatus, 21 program, 22 computer, 31 magneto-optical disk, 32 optical disk, 33 magnetic disk, 34 memory, 41 magneto-optical disk apparatus, 42 optical disk apparatus, 43 magnetic disk apparatus.

Claims (10)

Image enlargement / reduction means capable of performing enlargement / reduction processing on image data, color space conversion means capable of performing color space conversion processing on image data, and The processing is performed in the case where the color space conversion processing is performed by the color space conversion means after the enlargement / reduction processing by the image enlargement / reduction means, and when the enlargement / reduction processing by the image enlargement / reduction means is performed after the color space conversion processing by the color space conversion means. An image processing apparatus comprising: a control unit that controls the image enlargement / reduction unit and the color space conversion unit so that processing is performed in a processing order in which time is short. 2. The image processing apparatus according to claim 1, wherein the control unit controls a processing order according to the processing time in consideration of a scaling factor. 3. The control unit according to claim 1, wherein the control unit controls the processing order according to the processing time in consideration of the number of color components in a color space before and after the conversion process by the color space conversion unit. An image processing apparatus according to the item. Further, the image processing apparatus further comprises a decompression means for decompressing the compressed image data. The decompression means, when the image data is compressed in the JPEG-DCT format, converts the decompressed image data in the color space at the time of compression. 4. The image processing apparatus according to claim 1, wherein the control unit controls the processing order in consideration of a color space at the time of compression as a color space of the image data. 5. An image processing apparatus according to the item. Further, the image processing apparatus further comprises a decompression means for decompressing the compressed image data. The decompression means, when the image data is compressed in the JPEG-DCT format, when a specific color component is thinned out during compression, 5. The image processing apparatus according to claim 1, wherein the control unit controls the processing order in consideration of a resolution for each color component of the image data. The image processing apparatus according to claim 1. The image enlargement / reduction unit can perform enlargement / reduction processing using a plurality of enlargement / reduction methods having different processing loads and processing image quality, and the control unit controls the enlargement / reduction method used in the image enlargement / reduction unit and the processing order. The image processing apparatus according to any one of claims 1 to 5, wherein The image processing apparatus according to claim 6, wherein the control unit controls the scaling method and the processing order based on a desired image quality or processing speed as an output image. 7. The method according to claim 6, wherein the control unit selects the scaling method for each color component based on a visual characteristic of a color component of the image data to be scaled by the image scaling unit. 8. The image processing device according to 7. An information presenting means for presenting information to the operator, and an instruction means for instructing the operator are provided, and the control means serves as a guide for the processing time and the image quality after processing by the combination of the processing order or the scaling method. 9. The method according to claim 1, wherein information is presented by said information presenting means, and control of a processing order or selection of a further enlargement / reduction technique is performed in accordance with a processing order designated by said instruction means or a further enlargement / reduction technique. The image processing device according to any one of the preceding claims. 2. The control unit according to claim 1, wherein the control unit controls the processing order or further selects an enlargement / reduction method according to an image quality reproducing capability and an output size / resolution of an image output unit that outputs the processed image data. An image processing apparatus according to claim 8.

Images