JP6481301B2 - Image processing apparatus and image processing program - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing program.

  In Patent Document 1, it is an object to suppress degradation in image quality due to compression processing, a quantization method is determined for each pixel, image data is quantized in units of pixels according to the determined method, and the quantized image is determined. Is stored in the same BTC (Block Truncation Coding) plane area of the image memory regardless of the quantization method, and identification for identifying the quantization method used corresponding to the quantized data It is disclosed to store data in an image memory.

JP 2009-094619 A

In the prior art, BTC plane area information is required for image compression.
An object of the present invention is to provide an image processing apparatus and an image processing program that perform fixed-length encoding of an image without requiring BTC plane region information.

The gist of the present invention for achieving the object lies in the inventions of the following items.
The invention of claim 1 is a dividing means for dividing an image into blocks, a quantizing means for quantizing the block by a plurality of methods, a feature extracting means for extracting features of the block, and the features corresponding to the features Selection means for selecting a quantization result by the quantization means, and encoding means for fixed-length encoding the quantization result selected by the selection means , wherein the quantization means has a resolution and a scale in the quantization result. Quantization is performed by a plurality of methods so that the key is in a trade-off relationship, and the feature extraction unit extracts the dynamic range of the block, the number of colors in the block, and the type of color distribution as features. An image processing apparatus characterized by this.

The invention of claim 2 further comprises extraction means for extracting a type of the quantization result, and the selection means selects the quantization result corresponding to the type and the feature. The image processing apparatus according to 1 .

The invention of claim 3 further comprises a calculation means for calculating a distance between the block and the quantization result, and the selection means selects the quantization result corresponding to the distance and the feature. the image processing apparatus according to claim 1 or 2, characterized in.

According to a fourth aspect of the present invention, there is provided a computer comprising: a dividing unit that divides an image into blocks; a quantizing unit that quantizes the block by a plurality of methods; a feature extracting unit that extracts features of the block; A selecting unit that selects a quantization result corresponding to the quantization unit; and a coding unit that performs fixed-length encoding on the quantization result selected by the selecting unit . Quantization is performed by a plurality of methods so that the resolution and gradation are in a trade-off relationship, and the feature extraction unit is characterized by the dynamic range of the block, the number of colors in the block, and the type of color distribution. Is an image processing program to be extracted .

  According to the image processing apparatus of the first aspect, the image can be fixed-length encoded without requiring the BTC plane area information.

According to the image processing apparatus of the second aspect, the quantization result corresponding to the type of the quantization result and the feature of the block can be selected.

According to the image processing apparatus of the third aspect, it is possible to select the quantization result corresponding to the distance between the block and the quantization result and the feature of the block.

According to the image processing program of the fourth aspect, the image can be fixed-length encoded without requiring the BTC plane area information.

It is a conceptual module block diagram about the structural example of 1st Embodiment. It is explanatory drawing which shows the process example by 1st Embodiment. It is explanatory drawing which shows the process example by 1st Embodiment. It is a flowchart which shows the process example by 1st Embodiment. It is a flowchart which shows the process example by 1st Embodiment. It is explanatory drawing which shows the example of a data structure of a characteristic and quantization ID corresponding table. It is a flowchart which shows the process example by 1st Embodiment. It is explanatory drawing which shows the system configuration example using this Embodiment. It is a conceptual module block diagram about the structural example of 2nd Embodiment. It is explanatory drawing which shows the process example by 2nd Embodiment. It is explanatory drawing which shows the example of a data structure of a feature and pattern correspondence table. It is a conceptual module block diagram about the structural example of 3rd Embodiment. It is explanatory drawing which shows the example of a data structure of a shape identifier and representative pixel value definition table. It is a notional module block diagram about the structural example of 4th Embodiment. It is a block diagram which shows the hardware structural example of the computer which implement | achieves this Embodiment.

Hereinafter, examples of various preferred embodiments for realizing the present invention will be described with reference to the drawings.
<< First Embodiment >>
FIG. 1 is a conceptual module configuration diagram of a configuration example according to the first embodiment.
The module generally refers to components such as software (computer program) and hardware that can be logically separated. Therefore, the module in the present embodiment indicates not only a module in a computer program but also a module in a hardware configuration. Therefore, the present embodiment is a computer program for causing these modules to function (a program for causing a computer to execute each procedure, a program for causing a computer to function as each means, and a function for each computer. This also serves as an explanation of the program and system and method for realizing the above. However, for the sake of explanation, the words “store”, “store”, and equivalents thereof are used. However, when the embodiment is a computer program, these words are stored in a storage device or stored in memory. It is the control to be stored in the device. Modules may correspond to functions one-to-one, but in mounting, one module may be configured by one program, or a plurality of modules may be configured by one program, and conversely, one module May be composed of a plurality of programs. The plurality of modules may be executed by one computer, or one module may be executed by a plurality of computers in a distributed or parallel environment. Note that one module may include other modules. Hereinafter, “connection” is used not only for physical connection but also for logical connection (data exchange, instruction, reference relationship between data, etc.). “Predetermined” means that the process is determined before the target process, and not only before the process according to this embodiment starts but also after the process according to this embodiment starts. In addition, if it is before the target processing, it is used in accordance with the situation / state at that time or with the intention to be decided according to the situation / state up to that point. When there are a plurality of “predetermined values”, the values may be different from each other, or two or more values (of course, including all values) may be the same. In addition, the description having the meaning of “do B when it is A” is used in the meaning of “determine whether or not it is A and do B when it is judged as A”. However, the case where it is not necessary to determine whether or not A is excluded.
In addition, the system or device is configured by connecting a plurality of computers, hardware, devices, and the like by communication means such as a network (including one-to-one correspondence communication connection), etc., and one computer, hardware, device. The case where it implement | achieves by etc. is included. “Apparatus” and “system” are used as synonymous terms. Of course, the “system” does not include a social “mechanism” (social system) that is an artificial arrangement.
In addition, when performing a plurality of processes in each module or in each module, the target information is read from the storage device for each process, and the processing result is written to the storage device after performing the processing. is there. Therefore, description of reading from the storage device before processing and writing to the storage device after processing may be omitted. Here, the storage device may include a hard disk, a RAM (Random Access Memory), an external storage medium, a storage device via a communication line, a register in a CPU (Central Processing Unit), and the like.

  The compression apparatus 100 of the image processing apparatus according to the present embodiment compresses an image 105. As shown in the example of FIG. 1, a block division module 110, a multiple quantization module 115, a feature extraction module 120, A selection module 125 and a fixed-length encoding module 130 are included. Further, the decompression device 150 of the image processing apparatus according to the present embodiment decodes the code 132 compressed by the compression apparatus 100 to generate an image 195, and is fixed as shown in the example of FIG. A long decoding module 155, an identification module 160, and an inverse quantization module 165 are provided.

The block division module 110 of the compression apparatus 100 is connected to the multiple quantization module 115 and the feature extraction module 120, receives the image 105, and passes the block 112 to the multiple quantization module 115 and the feature extraction module 120. The block division module 110 divides the image 105 into blocks. Accepting an image means, for example, reading an image with a scanner, a camera, etc., receiving an image from an external device via a communication line by fax, etc., a hard disk (in addition to what is built in a computer, via a network) And the like, and the like, and receiving print images from other information processing apparatuses. The image is a multivalued image (including a color image). One image may be received or a plurality of images may be received. Further, the contents of the image may be a document used for business, a pamphlet for advertisement, or the like. A block is rectangular and generally square, but need not be limited to this.
The example illustrated in FIG. 2A is a processing result of the block division module 110. For example, “gradation: 8 bits-resolution (subblock): 1 × 1” (hereinafter referred to as “8 bits−1 × 1”).

The multiple quantization module 115 is connected to the block division module 110 and the selection module 125, receives the block 112 from the block division module 110, and passes the quantization result 117 to the selection module 125. The multiple quantization module 115 quantizes the block 112 divided by the block division module 110 by a plurality of methods. Here, the “plurality of methods” are different. For example, the multiple quantization module 115 may perform the quantization by a plurality of methods so that the resolution and the gradation have a trade-off relationship in the quantization result. In this case, the quantization result 117 becomes a plurality of expressions obtained by trade-offing the block 112 by resolution and gradation. Here, the “trade-off relationship” means a trade-off relationship. To increase the resolution, it is necessary to lower the gradation. Conversely, to increase the gradation, the resolution must be decreased. It means the relationship of not becoming.
The examples shown in FIGS. 2B, 2 </ b> C, and 2 </ b> D show a plurality (three) of processing results by the plurality of quantization modules 115. For example, FIG. 2B shows 1 bit-1 × 1 (number of gradation bits: 1, resolution size: same size as the original pixel), and FIG. 2C shows 4 bits−2 × 2 (number of gradation bits: 4, resolution size: 2 (width) × 2 (height) size of the original pixel), FIG. 2D shows 8 bits-2 × 4 (floor The number of key bits is 8, and the resolution is 2 (width) × 4 (height) of the original pixel.

  The feature extraction module 120 is connected to the block division module 110 and the selection module 125, receives the block 112 from the block division module 110, and passes the feature 122 to the selection module 125. The feature extraction module 120 extracts the features of the block 112 divided by the block division module 110. Also, features are extracted for each block. For example, the feature extraction module 120 may extract the dynamic range of the block 112 as a feature. The dynamic range refers to the width of a pixel value in a target block (a width determined by the maximum value and the minimum value).

  The selection module 125 is connected to the multiple quantization module 115, the feature extraction module 120, and the fixed length encoding module 130. The selection module 125 receives the quantization result 117 from the multiple quantization module 115 and the feature 122 from the feature extraction module 120. “ID + quantization result” 127 is passed to the fixed-length encoding module 130. The selection module 125 selects the quantization result 117 by the multiple quantization module 115 corresponding to the feature 122 extracted by the feature extraction module 120. That is, the one corresponding to one block is selected from a plurality of quantization results. In addition to the quantization result, an ID (IDentification) that is information indicating a quantization method or a method of expressing the quantization result may be included as a result of the selection process.

  The fixed-length encoding module 130 is connected to the selection module 125, receives “ID + quantization result” 127 from the selection module 125, and outputs a code 132. The fixed length encoding module 130 performs fixed length encoding on the quantization result selected by the selection module 125. The fixed-length encoding module 130 may perform fixed-length encoding on the “ID + quantization result” 127 selected by the selection module 125. Since it is fixed length coding, the code 132 for one block has a fixed length. For the fixed-length encoding, an existing technique may be used. Examples of processing results by the fixed-length encoding module 130 are shown in FIGS. Note that the example shown in FIG. 2E shows a bit representation of one block of the original image, and 16 pixels of 8 bits (total 128 bits) are continuous. The example shown in FIG. 2 (F) shows a bit representation corresponding to FIG. 2 (B). As a result of the fixed-length encoding process, 16 representative values 1 pixel 1 bit are continuous, and ID 2 bits are added as (ID 0:00) (18 bits in total). It should be noted that the number of ID bits may be equal to or greater than the number of bits that can represent the number of types. In this example, since there are three types of ID0, ID1, and ID2, 2 bits are used. The example shown in FIG. 2 (G) shows a bit representation corresponding to FIG. 2 (C). As a result of the fixed-length encoding process, four representative values of one pixel and 4 bits are consecutive. 2 bits are added as (ID1: 01) (18 bits in total). The example shown in FIG. 2 (H) shows a bit representation corresponding to FIG. 2 (D). As a result of the fixed-length encoding process, two representative values of one pixel 8 bits are continuous, and the ID 2 bits are added as (ID2: 10) (18 bits in total). It should be noted that the positional relationship between the result of the fixed-length encoding process and the ID may be other than shown (for example, inserted at the head).

The fixed-length decoding module 155 of the decompression device 150 is connected to the identification module 160, accepts the code 132, and passes “ID + quantization result” 157 to the identification module 160. The fixed length decoding module 155 decodes the code 132. Of course, decoding corresponds to fixed length coding. The decoding result is a quantization result. When the above-described ID is included in the code 132, “ID + quantization result” 157 is a decoding result.
The identification module 160 is connected to the fixed length decoding module 155 and the inverse quantization module 165, receives the “ID + quantization result” 157 from the fixed length decoding module 155, and passes the quantization result 162 to the inverse quantization module 165. . The identification module 160 extracts the ID described above from “ID + quantization result” 157.

The inverse quantization module 165 is connected to the identification module 160, receives the quantization result 162 from the identification module 160, and outputs an image 195. The inverse quantization module 165 inversely quantizes the quantization result 162 and restores the original image. If the above-mentioned ID is included, the quantization result 162 is inversely quantized according to the ID to restore the original image. The inverse quantization method here corresponds to a quantization method for generating the quantization result selected by the selection module 125.
In the example shown in FIGS. 2I, 2J, and 2K, the inverse quantization module 165 selects one of them based on the identification result by the identification module 160 and uses it for the inverse quantization. 2I corresponds to FIG. 2B, FIG. 2J corresponds to FIG. 2C, and FIG. 2K corresponds to FIG. 2D.
The example illustrated in FIG. 2L is a processing result by the inverse quantization module 165.
In this way, a fixed-length expression traded off between resolution and gradation is selected according to the block characteristics. Thus, for example, the ID is 2 bits, the quantization result is 16 bits, and a total of 18 bits [block] (16 × 8/18 = 7.11) is encoded as the code 132 (FIG. 2 described above). (See examples shown in (F), (G), (H)).
The inverse quantization module 165 outputs the restored image 195. To output an image is, for example, printing on a printing device such as a printer, displaying on a display device such as a display, transmitting an image on an image transmission device such as a fax, or image to an image storage device such as an image database. , Storing in a storage medium such as a memory card, passing to another information processing apparatus, and the like.

The principle of the embodiment will be described using the example of FIG.
The inventor's research has revealed that the original image can be sufficiently reproduced if the target blocks are traded off as they are along the resolution and gradation axes. In the present embodiment, this is used to perform fixed-length representation in which resolution (size of sub-block to be expressed flatly) and gradation (number of bits) are traded off for each block.
The first line in the example of FIG. 3 shows an original picture block. This is a 128-bit code amount. Since it is not encoded (because the amount of information has not been reduced), characters, gray characters (characters with gradation), and gradations (including photos etc.) are reproduced as shown in the first line example. become.
The second row in the example of FIG. 3 shows a quantization example in which the original image block in the first row has high resolution and low gradation (monochrome binary). This is a code amount of 16 bits (1 bit gradation × 16). In this case, there is no problem with image quality in reproduction when applied to characters. However, when applied to gray characters, there are many cases where the color changes and a part of the erasure occurs. In addition, when applied to gradation, gradation steps often occur.
The third line in the example of FIG. 3 illustrates a quantization example in which the original image block in the first line has a medium gradation with medium resolution. This is a code amount of 16 bits (4 bit gradation × 4). In this case, there is no image quality problem in reproduction when applied to gray characters. However, when applied to characters, collapse often occurs. In addition, when applied to gradation, gradation steps often occur.
The fourth line in the example of FIG. 3 shows an example of quantization that achieves a high gradation with low resolution for the original block in the first line. This is a code amount of 16 bits (8-bit gradation × 2). In this case, there is no problem with image quality in reproduction when applied to gradation. However, when applied to characters, collapse often occurs. In addition, when applied to gray characters, there are many cases where collapse occurs.
In other words, none of the text, gray text, and gradation can be handled by quantization. However, high resolution low gradation, medium resolution medium gradation, and low resolution high gradation quantization. If you do it, you can deal with image quality either way. These are used properly according to the characteristics of the block. Of course, quantization other than the example shown in FIG. 3 (high resolution, low gradation, medium resolution, medium gradation, low resolution, high gradation) may be employed.

FIG. 4 is a flowchart illustrating a processing example according to the first exemplary embodiment.
In step S402, the block division module 110 receives the image 105.
In step S404, the block division module 110 blocks the image 105.
In step S406, the multiple quantization module 115 quantizes the block 112 into a plurality of representations traded off in resolution and gradation. A detailed processing example will be described later with reference to FIG.
In step S408, the feature extraction module 120 extracts the feature 122 of the block 112.
In step S <b> 410, the selection module 125 selects from a plurality of quantization results 117 according to the feature 122. A detailed processing example will be described later with reference to FIG.
In step S412, the fixed length encoding module 130 performs an encoding process.
In step S414, the fixed-length encoding module 130 transmits the code 132 to the decompression device 150.

FIG. 5 is two types of flowcharts showing an example of processing according to the first embodiment (multiple quantization module 115). In this example, medium resolution quantization is performed with medium gradation.
The example shown in FIG. 5A is a processing result by the block division module 110, and shows one block of the original image. The left flowchart (step S522, step S524) uses subsampling / quantization, and the right flowchart (step S532, step S534) uses average value / quantization.
In step S522, sub-sampling processing is performed. The example shown in FIG. 5B shows the processing result in step S522. Pixels are selected from the example shown in FIG.
In step S524, the gradation is simply quantized. The example shown in FIG. 5C shows the processing result of step S524. The pixel values in the example shown in FIG. 5C are quantized.
In step S532, the average value of the sub blocks is calculated. The example shown in FIG. 5D shows the processing result in step S532. The average value of four pixels is calculated from the example shown in FIG.
In step S534, the gradation is simply quantized. The example shown in FIG. 5E shows the processing result in step S534. This is a quantized pixel value of the example shown in FIG.
Low resolution and high gradation quantization can be performed by the same processing.
For high-resolution and low-gradation quantization, an existing binarization process may be performed.

FIG. 6 is an explanatory diagram showing an example of the data structure of the feature / quantization ID correspondence table 600. The feature / quantization ID correspondence table 600 includes a D range column 610 and an ID column 620. The D range column 610 stores a dynamic range. The ID column 620 stores an ID (information indicating a quantization method or a quantization result expression method) corresponding to the dynamic range.
The first line of the feature / quantization ID correspondence table 600 indicates high resolution / low gradation (black and white) when the dynamic range is 255 (see FIG. 2B). When the dynamic range is 16 to 254, medium resolution / medium gradation is shown (see FIG. 2C). The third line shows low resolution / high gradation when the dynamic range is less than 16. (See FIG. 2D). Of course, this numerical value is an example, and three IDs are also examples. Further, when the feature extraction module 120 extracts a feature other than the dynamic range, an ID is associated with the feature value.
The selection module 125 searches the D range column 610 corresponding to the feature 122 and extracts the ID column 620 corresponding thereto. One of the plurality of quantization results 117 is selected according to the ID.

FIG. 7 is a flowchart illustrating a processing example according to the first exemplary embodiment (decompression apparatus 150).
In step S702, the fixed length decoding module 155 receives the code 132.
In step S704, the fixed length decoding module 155 performs a decoding process.
In step S706, the identification module 160 identifies the ID of inverse quantization.
In step S708, the inverse quantization module 165 performs inverse quantization. Using the above-described feature / quantization ID correspondence table 600, inverse quantization according to the ID is performed.
In step S710, the inverse quantization module 165 outputs the image 195.

FIG. 8 is an explanatory diagram showing two system configuration examples using this embodiment.
The example shown in FIG. 8A shows an example used for a printer. The drawing device 810, the compression device 100, the storage device 820, the decompression device 150, and the printer 830 are configured.
The drawing device 810 is connected to the compression device 100. The drawing device 810 generates an image to be printed.
The compression device 100 is connected to a drawing device 810 and a storage device 820. The compression device 100 compresses an image.
The storage device 820 is connected to the compression device 100 and the decompression device 150. The storage device 820 stores a code that is a result of compression by the compression device 100.
The expansion device 150 is connected to the storage device 820 and the printer 830. The decompressing device 150 decompresses the code in the storage device 820 and restores the image.
The printer 830 is connected to the expansion device 150. The printer 830 prints the image restored by the decompression device 150.
For example, when the printer 830 is a continuous-book printer, the roll paper cannot be stopped, and therefore data must be continuously transmitted without delaying the printing speed. Therefore, the guarantee of the compression rate and the compression / decompression speed are required. There are two types of compression (encoding): variable-length encoding and fixed-length encoding, but a high resolution (for example, about several [GB / s]) in a high resolution (for example, 1200 dpi) continuous printer. Therefore, guaranteeing the compression rate in variable length coding is not realistic. In other words, since the code amount is not determined unless variable length coding is tried, it is difficult to guarantee the compression rate, and it is also difficult to cut out the code when decompressing. For this reason, high-speed printers often perform fixed-length encoding processing.

The example shown in FIG. 8B shows an example used for image transmission.
The compression device 100A, the compression device 100B, the expansion device 150A, the expansion device 150B, the expansion device 150C, and the expansion device 150D are connected via a communication line 890, respectively. The communication line 890 may be wireless, wired, or a combination thereof, for example, the Internet as a communication infrastructure. Examples of the decompression device 150 include a mobile terminal, a PC, a printer, and the like, and reproduce an image compressed by the compression device 100.

<< Second Embodiment >>
FIG. 9 is a conceptual module configuration diagram of a configuration example according to the second embodiment.
The compression apparatus 900 includes a block division module 110, a multiple quantization module 115, a pattern extraction module 920, a feature extraction module 120, a selection module 125, and a fixed length encoding module 130. In addition, the same code | symbol is attached | subjected to the site | part of the same kind as the above-mentioned embodiment, and the overlapping description is abbreviate | omitted (hereinafter the same). An expansion device corresponding to the compression device 900 has the same configuration as the expansion device 150 according to the first embodiment.
The compression apparatus 900 according to the second embodiment uses a quantized pixel value pattern in order to increase the accuracy of ID selection. For example, ID = 1 is an expression in which resolution is more important than ID = 2 (at the expense of gradation), but depending on the situation, the resolution may not be high. In order to prevent this, the compression apparatus 900 selects using a combination of a dynamic range and a pattern (histogram) after quantization.

The multiple quantization module 115 is connected to the block division module 110, the selection module 125, and the pattern extraction module 920, receives the block 112 from the block division module 110, and sends the quantization result 117 to the pattern extraction module 920 and the selection module 125. hand over.
The pattern extraction module 920 is connected to the multiple quantization module 115 and the selection module 125, receives the quantization result 117 from the multiple quantization module 115, and passes the quantized pattern 922 to the selection module 125. The pattern extraction module 920 extracts the type (hereinafter also referred to as a pattern) of the quantization result 117 obtained by the multiple quantization module 115.
The selection module 125 is connected to the multiple quantization module 115, the pattern extraction module 920, the feature extraction module 120, and the fixed length encoding module 130, and the quantization result 117 is received from the multiple quantization module 115 and the pattern extraction module 920 is used. The quantized pattern 922 receives the feature 122 from the feature extraction module 120 and passes “ID + quantization result” 127 to the fixed-length encoding module 130. The selection module 125 selects the quantized pattern 922 extracted by the pattern extraction module 920 and the quantization result 117 corresponding to the feature 122 extracted by the feature extraction module 120.

FIG. 10 is an explanatory diagram illustrating a processing example according to the second exemplary embodiment.
As a pattern, pixel values in the quantization result can be classified as one color, two colors, three colors, and four colors. Furthermore, for the two colors, the whole is divided equally (the example on the right side of “not improved” in the example of FIG. 10) and the checkered pattern (the diagonal values are equivalent, in the example of FIG. 10) Two-color checkered).
The example of FIG. 10A shows an example of a pattern for selecting ID1 when it is determined only by the feature 122 as in the first embodiment. Therefore, in comparison with the case where ID2 is applied, if “ID1 (see FIG. 2C) has a higher resolution than ID2 (see FIG. 2D)”, the case of “not improved” , “Possibility of improvement” indicates that there is a case of “improvement”. That is, since ID1 is divided into four and ID2 is divided into two, in general, ID1 should have a high resolution. However, in the case where the whole is one color, the resolution is not high even if ID1 is used for the one that divides the whole into two equal parts (two examples of “not improved” in the example of FIG. 10), and ID2 Is clearly advantageous in terms of gradation. In addition, two colors (one that divides the whole into two equal parts, excluding the checkered pattern, the example on the left side of “potentially improved” in the example of FIG. 10) are highly likely to improve, while 3 The color (the example on the right side of “possibly improved” in the example of FIG. 10) is unlikely to improve. In the case of a checkered pattern and four colors, it improves. Therefore, as shown in the example of FIG. 10B, the threshold value S is divided into those to which ID1 and ID2 are applied.

For example, specifically, the selection module 125 makes a determination using the feature / pattern correspondence table 1100. FIG. 11 is an explanatory diagram showing an example of the data structure of the feature / pattern correspondence table 1100. The feature / pattern correspondence table 1100 has a D range column 1110, a 1 color column 1120, a 2 color ID2 same column 1130, a 2 color column 1140, a checkered column 1150, a 3 color column 1160, and a 4 color column 1170. The D range column 1110 stores a dynamic range. The one color column 1120 stores a pattern for one color. The two-color ID2 same column 1130 stores a pattern in the case of two-color ID2 same (the two colors are divided into equal parts). The two-color column 1140 stores a pattern for two colors. The checkered column 1150 stores a pattern in the case of a checkered pattern. A three-color column 1160 stores a pattern for three colors. The 4-color column 1170 stores a pattern for four colors. That is, the table is used to select an ID based on the feature 122 and the pattern 922 after quantization.
In the example of the feature / pattern correspondence table 1100, when the dynamic range is 16 to 254 (in the first embodiment, ID1 is selected by the feature / quantization ID correspondence table 600), the ID1 depends on the pattern. Or it will be divided into ID2.

選択モジュール１２５は、複数量子化モジュール１１５、距離計算モジュール１２２０、特徴抽出モジュール１２０、固定長符号化モジュール１３０と接続されており、複数量子化モジュール１１５より量子化結果１１７を、距離計算モジュール１２２０より距離１２２２を、特徴抽出モジュール１２０より特徴１２２を受け取り、固定長符号化モジュール１３０に「ＩＤ＋量子化結果」１２７を渡す。選択モジュール１２５は、距離計算モジュール１２２０によって算出された距離１２２２と特徴抽出モジュール１２０によって抽出された特徴１２２に対応する量子化結果１１７を選択する。例えば、具体的には、第１の実施の形態と同等の処理を行い、同じ解像度のＩＤが選択された場合は、ＩＤ２とＩＤ３の場合の距離Ｅを比較して、小さい値となるＩＤを選択すればよい。 << Third Embodiment >>
FIG. 12 is a conceptual module configuration diagram of an exemplary configuration according to the third embodiment.
The compression device 1200 includes a block division module 110, a multiple quantization module 115, a distance calculation module 1220, a feature extraction module 120, a selection module 125, and a fixed length encoding module 130.
The compression apparatus 1200 according to the third embodiment is configured to change the shape expressing the resolution even with the same gradation. As shown in the example of FIG. 13, ID3 is added to the ID of the first embodiment. The distance is used to identify ID2 and ID3. Increasing the variation can improve the image quality.
The multiple quantization module 115 is connected to the block division module 110, the selection module 125, and the distance calculation module 1220, receives the block 112 from the block division module 110, and sends the quantization result 117 to the distance calculation module 1220 and the selection module 125. hand over.
The distance calculation module 1220 is connected to the block division module 110, the multiple quantization module 115, and the selection module 125. The distance calculation module 1220 receives the block 112 from the block division module 110 and the quantization result 117 from the multiple quantization module 115. The distance 1222 is passed to 125. The distance calculation module 1220 calculates the distance between the block 112 divided by the block division module 110 and the quantization result 117 by the multiple quantization module 115. For example, the distance of the multi-valued image is calculated using Expression (1). Here, P represents the pixel value of the original image, Q represents the pixel value of the quantization result, and the index i represents the coordinate position of the pixel.

The selection module 125 is connected to the multiple quantization module 115, the distance calculation module 1220, the feature extraction module 120, and the fixed length encoding module 130, and the quantization result 117 from the multiple quantization module 115 and the distance calculation module 1220. The feature 122 is received from the feature extraction module 120 as the distance 1222, and “ID + quantization result” 127 is passed to the fixed-length encoding module 130. The selection module 125 selects the quantization result 117 corresponding to the distance 1222 calculated by the distance calculation module 1220 and the feature 122 extracted by the feature extraction module 120. For example, specifically, the same processing as in the first embodiment is performed, and when an ID with the same resolution is selected, the distance E in the case of ID2 and ID3 is compared, and an ID that becomes a small value is obtained. Just choose.

You may combine 2nd Embodiment and 3rd Embodiment. That is, the distance calculation module 1220 may be added to the compression apparatus 900.
In this case, it may be determined by the shape identifier / representative pixel value definition table 1300 whether or not the processing by the distance calculation module 1220 is performed. FIG. 13 is an explanatory diagram showing an example of the data structure of the shape identifier / representative pixel value definition table 1300. The shape identifier / representative pixel value definition table 1300 has an ID column 1310 and a shape column 1320. The ID column 1310 stores an ID. The shape column 1320 stores the quantization shape in the ID. In other words, it is determined whether there are different shapes in the same number of quantization results (for example, ID2 and ID3), and if there are (for example, there are ID2 and ID3), the distance calculation module. The processing of 1220 may be performed, and the selection module 125 may make a determination including the distance 1222 as well. If not, the selection module 125 may make a determination based on the feature 122 and the quantized pattern 922 without performing processing in the distance calculation module 1220.

<< Fourth Embodiment >>
FIG. 14 is a conceptual module configuration diagram of a configuration example according to the fourth embodiment. In the above-described embodiment, the selection is made after the quantization. However, the selection process may be performed before the quantization, and the quantization may be performed by the selected quantization method.
The compression device 1400 includes a block division module 1410, a feature extraction module 1415, a selection module 1420, a quantization module 1425, and a fixed length encoding module 1430.
The block division module 1410 is connected to the feature extraction module 1415 and the quantization module 1425, receives the image 1405, and passes the block 1412 to the feature extraction module 1415 and the quantization module 1425. The block division module 1410 performs the same processing as the block division module 110. That is, the image 1405 is divided into blocks.
The feature extraction module 1415 is connected to the block division module 1410 and the selection module 1420, receives the block 1412 from the block division module 1410, and passes the feature 1417 to the selection module 1420. The feature extraction module 1415 performs the same processing as the feature extraction module 120. That is, the feature of the block 1412 is extracted.
The selection module 1420 is connected to the feature extraction module 1415 and the quantization module 1425, receives the feature 1417 from the feature extraction module 1415, and passes the ID 1422 to the quantization module 1425. The selection module 1420 selects a quantization technique (ID) corresponding to the feature 1417 extracted by the feature extraction module 1415. A plurality of quantization methods are selected so that the resolution and gradation have a trade-off relationship in the quantization result. A selection process equivalent to that of the selection module 125 described above is performed. However, the selection target is not a quantization result but a quantization method.

The quantization module 1425 is connected to the block division module 1410, the selection module 1420, and the fixed length encoding module 1430. The quantization module 1425 receives the block 1412 from the block division module 1410 and the ID 1422 from the selection module 1420 and receives the fixed length encoding module 1430. "ID + quantization result" 1427 is passed. The quantization module 1425 quantizes the block 1412 divided by the block division module 1410 by the quantization method (ID 1422) selected by the selection module 1420.
The fixed-length encoding module 1430 is connected to the quantization module 1425, receives “ID + quantization result” 1427 from the quantization module 1425, and outputs a code 1432. The fixed length encoding module 1430 performs processing equivalent to that of the fixed length encoding module 130. That is, “ID + quantization result” 1427 by the quantization module 1425 is fixed-length encoded, and a code 1432 is output.

  A hardware configuration example of the image processing apparatus according to the present embodiment will be described with reference to FIG. The configuration shown in FIG. 15 is configured by a personal computer (PC), for example, and shows a hardware configuration example including a data reading unit 1517 such as a scanner and a data output unit 1518 such as a printer.

  A CPU (Central Processing Unit) 1501 includes various modules described in the above-described embodiments, that is, a block division module 110, a multiple quantization module 115, a feature extraction module 120, a selection module 125, a fixed length encoding module 130, Fixed-length decoding module 155, identification module 160, inverse quantization module 165, pattern extraction module 920, distance calculation module 1220, block division module 1410, feature extraction module 1415, selection module 1420, quantization module 1425, fixed-length encoding module The control unit executes processing according to a computer program describing an execution sequence of each module such as 1430.

  A ROM (Read Only Memory) 1502 stores programs used by the CPU 1501, calculation parameters, and the like. A RAM (Random Access Memory) 1503 stores programs used in the execution of the CPU 1501, parameters that change as appropriate during the execution, and the like. These are connected to each other by a host bus 1504 including a CPU bus or the like.

  The host bus 1504 is connected to an external bus 1506 such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge 1505.

  A keyboard 1508 and a pointing device 1509 such as a mouse are input devices operated by an operator. The display 1510 includes a liquid crystal display device or a CRT (Cathode Ray Tube), and displays various types of information as text or image information.

  An HDD (Hard Disk Drive) 1511 includes a hard disk, drives the hard disk, and records or reproduces a program executed by the CPU 1501 and information. The hard disk stores an image 105, a block 112, a quantization result 117, a feature 122, an “ID + quantization result” 127, a code 132, an “ID + quantization result” 157, a quantization result 162, an image 195, and the like. Further, various computer programs such as various other data processing programs are stored.

  The drive 1512 reads out data or a program recorded on a removable recording medium 1513 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and the data or program is read out as an interface 1507 and an external bus 1506. , To the RAM 1503 connected via the bridge 1505 and the host bus 1504. The removable recording medium 1513 can also be used as a data recording area similar to the hard disk.

  The connection port 1514 is a port for connecting the external connection device 1515 and has a connection unit such as USB and IEEE1394. The connection port 1514 is connected to the CPU 1501 and the like via an interface 1507, an external bus 1506, a bridge 1505, a host bus 1504, and the like. The communication unit 1516 is connected to a communication line and executes data communication processing with the outside. The data reading unit 1517 is a scanner, for example, and executes document reading processing. The data output unit 1518 is, for example, a printer, and executes document data output processing.

  Note that the hardware configuration of the image processing apparatus shown in FIG. 15 shows one configuration example, and the present embodiment is not limited to the configuration shown in FIG. 15, and the modules described in the present embodiment are executed. Any configuration is possible. For example, some modules may be configured with dedicated hardware (for example, Application Specific Integrated Circuit (ASIC), etc.), and some modules are in an external system and connected via a communication line In addition, a plurality of systems shown in FIG. 15 may be connected to each other via communication lines so as to cooperate with each other. Further, it may be incorporated in a copying machine, a fax machine, a scanner, a printer, a multifunction machine (an image processing apparatus having any two or more functions such as a scanner, a printer, a copying machine, and a fax machine).

The program described above may be provided by being stored in a recording medium, or the program may be provided by communication means. In that case, for example, the above-described program may be regarded as an invention of a “computer-readable recording medium recording the program”.
The “computer-readable recording medium on which a program is recorded” refers to a computer-readable recording medium on which a program is recorded, which is used for program installation, execution, program distribution, and the like.
The recording medium is, for example, a digital versatile disc (DVD), which is a standard established by the DVD Forum, such as “DVD-R, DVD-RW, DVD-RAM,” and DVD + RW. Standard “DVD + R, DVD + RW, etc.”, compact disc (CD), read-only memory (CD-ROM), CD recordable (CD-R), CD rewritable (CD-RW), Blu-ray disc ( Blu-ray (registered trademark) Disc), magneto-optical disk (MO), flexible disk (FD), magnetic tape, hard disk, read-only memory (ROM), electrically erasable and rewritable read-only memory (EEPROM (registered trademark)) )), Flash memory, Random access memory (RAM) SD (Secure Digital) memory card and the like.
The program or a part of the program may be recorded on the recording medium for storage or distribution. Also, by communication, for example, a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a wired network used for the Internet, an intranet, an extranet, or a wireless communication It may be transmitted using a transmission medium such as a network or a combination of these, or may be carried on a carrier wave.
Furthermore, the program may be a part of another program, or may be recorded on a recording medium together with a separate program. Moreover, it may be divided and recorded on a plurality of recording media. Further, it may be recorded in any manner as long as it can be restored, such as compression or encryption.

DESCRIPTION OF SYMBOLS 100 ... Compression apparatus 105 ... Image 110 ... Block division module 112 ... Block 115 ... Multiple quantization module 117 ... Quantization result 120 ... Feature extraction module 122 ... Feature 125 ... Selection module 127 ... ID + quantization result 130 ... Fixed-length encoding Module 132 ... Code 150 ... Expansion device 155 ... Fixed length decoding module 157 ... ID + quantization result 160 ... Identification module 162 ... Quantization result 165 ... Inverse quantization module 195 ... Image 810 ... Drawing device 820 ... Storage device 830 ... Printer 890 ... Communication line 900 ... Compressor 920 ... Pattern extraction module 922 ... Pattern after quantization 1200 ... Compressor 1220 ... Distance calculation module 1222 ... Distance 1400 ... Compressor 1405 ... Image 1410 ... Block division module 1412 ... Block 1415 ... Feature extraction module 1417 ... Feature 1420 ... Selection module 1422 ... ID
1425: Quantization module 1427 ... ID + quantization result 1430 ... Fixed-length encoding module 1432 ... Code

Claims (4)

A dividing means for dividing the image into blocks;
Quantization means for quantizing the block in a plurality of ways;
Feature extraction means for extracting features of the block;
Selection means for selecting a quantization result by the quantization means corresponding to the feature;
Encoding means for fixed-length encoding the quantization result selected by the selection means ,
The quantization means quantizes by a plurality of methods so that the resolution and gradation are in a trade-off relationship in the quantization result,
The image processing apparatus is characterized in that the feature extraction means extracts the dynamic range of the block, the number of colors in the block, and the type of color distribution as features . An extraction unit for extracting the type of the quantization result;
The image processing apparatus according to claim 1, wherein the selection unit selects the quantization result corresponding to the type and the feature. A calculation means for calculating a distance between the block and the quantization result;
It said selection means, the image processing apparatus according to claim 1 or 2, characterized in that selecting the quantization result corresponding to the characteristic and the distance. Computer
A dividing means for dividing the image into blocks;
Quantization means for quantizing the block in a plurality of ways;
Feature extraction means for extracting features of the block;
Selection means for selecting a quantization result by the quantization means corresponding to the feature;
Functioning as encoding means for fixed-length encoding the quantization result selected by the selection means ,
The quantization means quantizes by a plurality of methods so that the resolution and gradation are in a trade-off relationship in the quantization result,
The feature extraction means is an image processing program that extracts the dynamic range of the block, the number of colors in the block, and the type of color distribution as features .

Images