CN101945269B

CN101945269B - Image processing device and method thereof

Info

Publication number: CN101945269B
Application number: CN 200910142589
Authority: CN
Inventors: 黄朝宗
Original assignee: Novatek Microelectronics Corp
Current assignee: Novatek Microelectronics Corp
Priority date: 2009-07-03
Filing date: 2009-07-03
Publication date: 2013-08-21
Anticipated expiration: 2029-07-03
Also published as: CN101945269A

Abstract

The invention discloses an image processing device and a method thereof. The image processing device comprises a storage device, a first image data translation unit and a second image data translation unit, wherein the storage device is used for being written and reading first image data, each pixel value in each image data is provided with a first data format, and the first data format conforms to the preset data receiving format of the storage device; the first image data translation unit converts the second image data into the first image data, each pixel value in the second image data has a second data format, and the second data format does not conform to the preset data receiving format of the storage device; the second image data translation unit converts the first image data into the third image data, each pixel value in the third image data has a third data format, and the second data format does not conform to the preset data receiving format of the storage device.

Description

Image processing apparatus and method thereof

Technical field

The present invention relates to a kind of image processing apparatus and method thereof, and be particularly related to a kind of image processing apparatus and method thereof that is applied in image compression (coding) and (decoding) process of decompression.

Background technology

In the middle of the multimedia consumption electronic products, digital frame, video card or digital camera need carry out image usually to be handled, and wherein image compression standard JPEG (Joint Photographic ExpertsGroup) is that the image that present widely market is accepted is handled standard.Along with the development of technology, international relevant criterion is worked out and is organized the image compression standard JPEG-XR (JPEGExtent Range) that further works out a new generation.In image compression standard JPEG-XR, can support the view data of various color spaces (ColorSpace), for example sRGB color space or scRGB color space.

In the sRGB color space, the pixel value of composing images data can be 8 bit formats, and in the scRGB color space, the pixel value of composing images data can be 16 bit formats.Further, image compression standard JPEG-XR can adjust convergent-divergent sign (Scaled_Flag) in the process of handling, use the bit format that makes the handled pixel value of image compression standard JPEG-XR may not be certain to meet 8 multiples, for example 11,12,19,20 ... the bit format that waits.

Special like this data format can make in image compression standard JPEG-XR handles, cause the access memory device inefficent.That is to say, do not conform to based on the form of special like this data format with the access memory device, so in the process that image is handled, may waste storage space.

Summary of the invention

The invention provides a kind of image processing apparatus and method thereof, it can effectively improve the shortcoming that prior art is addressed.

The invention provides a kind of image processing apparatus, it comprises storage arrangement, first image data conversion unit and second image data conversion unit.Storage arrangement is in order to be written into and to be read first view data that comprises a plurality of first pixel values, and each first pixel value has first data format, and wherein first data format meets the tentation data reception form of storage arrangement.First image data conversion unit couples storage arrangement, second view data that comprises a plurality of second pixel values in order to reception, and carry out first the conversion and obtain first view data to write to storage arrangement, wherein each second pixel value has second data format, and second data format does not meet tentation data reception form.

Second image data conversion unit couples storage arrangement, in order to read first view data from storage arrangement, and carry out second conversion and obtain the 3rd view data that comprises a plurality of the 3rd pixel values, wherein each the 3rd pixel value has the 3rd data format, and the 3rd data format does not meet tentation data and receives form, and a plurality of second pixel value is in fact corresponding to a plurality of the 3rd pixel values.

The invention provides a kind of image processing method, it comprises: at first, provide first view data, wherein first view data has a plurality of first pixel values, each first pixel value has first data format, and the tentation data that first data format does not meet storage arrangement receives form.Then, execution first is transformed in first view data and obtains second view data to write to storage arrangement, wherein second view data has a plurality of second pixel values, each second pixel value has second data format, and the tentation data that second data format meets storage arrangement receives form.

At last, when second view data is read out from storage arrangement, execution second is transformed in second view data and obtains the 3rd view data that comprises a plurality of the 3rd pixel values, wherein each the 3rd pixel value has the 3rd data format, the 3rd data format does not meet tentation data and receives form, and described a plurality of first pixel values are in fact corresponding to described a plurality of the 3rd pixel values.

The present invention provides a kind of image processing apparatus in addition, and it comprises storage arrangement, has the view data computing converting unit of first arithmetic device and first image data conversion unit, and second image data conversion unit.Storage arrangement is in order to be written into and to be read first view data that comprises a plurality of first pixel values, and each first pixel value has first data format, and wherein first data format meets the tentation data reception form of storage arrangement.

First arithmetic device comprises second view data of a plurality of second pixel values in order to reception, and carry out correcting process to obtain the 3rd view data of a plurality of the 3rd pixel values, wherein each the 3rd pixel value is all in first preset range, each second pixel value has second data format, and second data format does not meet tentation data and receives form.First image data conversion unit couples first arithmetic device, in order to receiving the 3rd view data, and carries out first conversion and obtains first view data to write to storage arrangement.

Second image data conversion unit couples storage arrangement, in order to read first view data from storage arrangement, and carry out second conversion and obtain the 4th view data that comprises a plurality of the 4th pixel values, wherein each the 4th pixel value has the 3rd data format, the tentation data that the 3rd data format does not meet storage arrangement receives form, and described a plurality of second pixel values are in fact corresponding to described a plurality of the 4th pixel values.

The present invention provides a kind of image processing method in addition, it comprises: at first, provide first view data, wherein first view data has a plurality of first pixel values, each first pixel value has first data format, and the tentation data that first data format does not meet storage arrangement receives form.Then, to obtain second view data that comprises a plurality of second pixel values, wherein each second pixel value is all in preset range in first view data for the execution correcting process.

Afterwards, execution first is transformed in second view data and obtains the 3rd view data to write to storage arrangement, wherein the 3rd view data has a plurality of the 3rd pixel values, each the 3rd pixel value has the 3rd data format, and the tentation data that the 3rd data format meets storage arrangement receives form.At last, when the 3rd view data is read out from storage arrangement, execution second is transformed in the 3rd view data and obtains the 4th view data that comprises a plurality of the 4th pixel values, wherein each the 4th pixel value has the 4th data format, the 4th data format does not meet tentation data and receives form, and described a plurality of first pixel values are in fact corresponding to described a plurality of the 4th pixel values.

The present invention provides a kind of image processing apparatus in addition, and it comprises storage arrangement, first image data conversion unit, second image data conversion unit, has first arithmetic device and the 3rd image data conversion unit, and with the 4th image data conversion unit.Storage arrangement is in order to be written into and to be read first view data that comprises a plurality of first pixel values, and each first pixel value has first data format, and wherein first data format meets the tentation data reception form of storage arrangement.

First image data conversion unit couples storage arrangement, second view data that comprises a plurality of second pixel values in order to reception, and carry out first the conversion and obtain first view data to write to storage arrangement, wherein each second pixel value has second data format, and second data format does not meet tentation data and receives form.Second image data conversion unit couples storage arrangement, in order to read first view data from storage arrangement, and carry out second conversion and obtain the 3rd view data that comprises a plurality of the 3rd pixel values, wherein each the 3rd pixel value has the 3rd data format, the 3rd data format does not meet tentation data and receives form, and described a plurality of second pixel values are in fact corresponding to described a plurality of the 3rd pixel values.

First arithmetic device comprises the 4th view data of a plurality of the 4th pixel values in order to reception, and carry out a correcting process to obtain the 5th view data of a plurality of the 5th pixel values, wherein each the 5th pixel value is all in first preset range, each the 4th pixel value has the 4th data format, and the 4th data format does not meet tentation data and receives form.The 3rd image data conversion unit couples first arithmetic device, in order to receiving the 5th view data, and carries out the 3rd conversion and obtains first view data to write to storage arrangement.

The 4th image data conversion unit couples storage arrangement, in order to read first view data from storage arrangement, and carry out the 4th and change to obtain the 6th view data that comprises a plurality of the 6th pixel values, wherein each the 6th pixel value has the 6th data format, the 6th data format does not meet tentation data and receives form, and described a plurality of the 4th pixel values are in fact corresponding to described a plurality of the 6th pixel values.

Based on as can be known above-mentioned, when the pixel value in the view data does not meet the data format of memory device access, image processing apparatus provided by the present invention and method can convert view data earlier to the tentation data that meets storage arrangement and receive form, and after reading the view data of depositing in storage arrangement, the view data that reads is carried out another conversion, and it is corresponding in fact to use the view data that makes before and after the conversion.

Will be appreciated that above-mentioned general description and following embodiment only are exemplary and illustrative, it can not limit the scope that institute of the present invention desire is advocated.

Description of drawings

Fig. 1 is the calcspar according to the image processing apparatus of the present invention's one example embodiment.

Fig. 2 A is the calcspar of the image data processing unit of Fig. 1.

Fig. 2 B is another calcspar of the image data processing unit of Fig. 1.

Fig. 2 C is another calcspar of the image data processing unit of Fig. 1.

Fig. 3 is the calcspar of the image encoder of Fig. 1.

Fig. 4 is the flow chart of the image processing method of the present invention's one example embodiment.

Fig. 5 is the calcspar according to the image processing apparatus of another example embodiment of the present invention.

Fig. 6 A is the calcspar of the image data processing unit 510 of Fig. 5.

Fig. 6 B is another calcspar of the image data processing unit 510 of Fig. 5.

Fig. 7 is the calcspar of the image decoder 502 of Fig. 5.

Fig. 8 is the flow chart according to the image processing method of another example embodiment of the present invention.

Fig. 9 is the calcspar according to the image processing apparatus of another example embodiment of the present invention.

[main element symbol description]

100,500,900: image processing apparatus

104,108,508,512: image data conversion unit

504: view data computing converting unit

106,506: storage arrangement DA1～DA14: view data

102,510: image data processing unit

DA1Y, DA1U, DA1V, DA2Y, DA2U: pixel value

DA2V, DA3Y, DA3U, DA3V, DA4R: pixel value

DA4G, DA4B, DA7Y, DA7U, DA7V: pixel value

DA8Y, DA8U, DA8V, DA9Y, DA9U: pixel value

DA9V, DA12R, DA12G, DA12B, DA13R: pixel value

DA13G, DA13B: pixel value

110: image encoder 502: image decoder

STR1, STR2: bit stream ADJ1～ADJ6: arithmetic unit

302: overlapping transducer 304: quantizer

306: fallout predictor 308: the entropy encoder

702: entropy decoder 704: inverse predictor

706: inverse quantizer 708: contrary overlapping transducer

EnD1～EnD4: encoded data D eD1～DeD4: coded data

CT1～CT4: color converter

S400～S406, S800～S806: the step of image processing method

Embodiment

Implementation type according to image processing apparatus and method, example embodiment of the present invention can be divided into device, the method that is applied to image compression, the device that is applied to image decompressor (that is decoding) that are applied to image compression (that is coding), be applied to the method for image decompressor, and the device that is applied to image compression and decompression.Below explain as follows according to each implementation type.

Be applied to the device of image compression

Fig. 1 illustrates the calcspar into the image processing apparatus of the present invention's one example embodiment.Please refer to Fig. 1, image processing apparatus 100 comprises that image data processing unit 102, image

data conversion unit

104 and 108, storage arrangement 106 (for example are DRAM, but be not limited in this), and image encoder 110 (be the JPEG-XR encoder for example, but be not limited in this).Storage arrangement 106 can be written into and be read view data DA1, and wherein view data DA1 comprises a plurality of pixel values.

In this example embodiment, be example with pixel value DA1Y, pixel value DA1U and three pixel values of pixel value DA1V, wherein the data format of pixel value DA1Y, pixel value DA1U and pixel value DA1V is that a tentation data receives form, and this tentation data to receive form be the form that meets storage arrangement 106 accesses, the bit data form of for example 8,16 or 8 multiples.

On the other hand, in the product that image is handled (for example digital camera, but be not limited in this), have image data processing unit 102 usually, in order to receiving view data DA4, and carry out view data and handle, use output image data DA2.Wherein, view data DA2 comprises a plurality of pixel values.In this example embodiment, be example with pixel value DA2Y, pixel value DA2U and three pixel values of pixel value DA2V.Can comprise color conversion (Color Transform) or the processing of other related operations because the view data that image data processing unit 102 carries out is handled, so the data format of pixel value DA2Y, pixel value DA2U and pixel value DA2V does not meet aforesaid tentation data reception form usually.

The content that discloses according to prior art as can be known, if the data format that writes when not conforming to the form of access memory device 106, then may be wasted storage space in the process that image is handled.For fear of when view data DA2 write memory device 106 time waste storage space, this one exemplary embodiment is done Data Format Transform earlier with each pixel value among the view data DA2 and is received form with the tentation data that meets storage arrangement 106, writes in the storage arrangement 106 afterwards again.Clearer, image data conversion unit 104 is carried out first and is transformed in view data DA2, and acquisition view data DA1, use the tentation data reception form that makes the data format of pixel value DA1Y, pixel value DA1U and pixel value DA1V meet storage arrangement 106 accesses, write to storage arrangement 106 in order to view data DA1.

In addition, because view data DA2 first conversion of process image data conversion unit 104, when therefore the view data DA1 that stores when storage arrangement 106 is read out, the processing that must recover (recover) to it.Clearer, image data conversion unit 108 is carried out second conversion to view data DA1 behind storage arrangement 106 reads image data DA1, uses obtaining view data DA3.View data DA3 comprises a plurality of pixel values, and this example embodiment is example with pixel value DA3Y, pixel value DA3U and three pixel values of pixel value DA3V, and wherein the data format of pixel value DA3Y, pixel value DA3U and pixel value DA3V does not meet tentation data and receives form.Further, view data DA2 changes through second through first conversion and view data DA1, and a plurality of pixel values among the view data DA2 are in fact corresponding to a plurality of pixel values among the view data DA3.

On the other hand, the view data DA3 that image processing apparatus 100 obtains can do further processing, for example is played in screen, writes to storage device, or encode.Image processing apparatus 100 comprises image encoder 110 in this example embodiment.110 couples of view data DA3 of image encoder encode, and use output bit flow STR1.

According to above-mentioned, image data processing unit 102 receives view data DA4 and carries out view data and handle, and uses output image data DA2.For fear of the waste storage space, image data conversion unit 104 is carried out first and is transformed in view data DA2, and obtains view data DA1, and wherein the tentation data that each pixel value has met storage arrangement 106 accesses among the view data DA1 receives form.

In addition because view data DA2 is through the processing of first conversion, therefore when image data conversion unit 108 behind storage arrangement 106 reads image data DA1, must carry out second conversion to view data DA1, use acquisition image restored data DA3.Wherein a plurality of pixel values among the view data DA2 are in fact corresponding to a plurality of pixel values among the view data DA3.Further, the pixel value among view data DA2 and the view data DA4 may comprise various different data formats.For in response to various different data formats, image data conversion unit 104 is carried out first conversion and also can be adjusted thereupon.

Fig. 2 A illustrates the calcspar into the image data processing unit of Fig. 1.Please merge with reference to Fig. 1 and Fig. 2 A, image data processing unit 102 comprises arithmetic unit ADJ1 and ADJ2 and color converter CT1.The view data DA4 that image data processing unit 102 receives comprises pixel value DA4R, pixel value DA4G and many pixel values of pixel value DA4B.In this example embodiment, suppose that pixel value DA4R, pixel value DA4G and pixel value DA4B are all 8 bit data forms, and the tentation data of storage arrangement 106 reception form is 8 bit data forms.

In view of this, in the process that image is handled, if image compression standard JPEG-XR adjusts convergent-divergent sign (Scaled_Flag) in the process of handling, Scaled_Flag=1 for example, then the handled pixel value of image compression standard JPEG-XR is not the bit format of 8 multiples.Clearer, because view data DA4 can pass through the processing of arithmetic unit ADJ1 to obtain view data DA5, each pixel value among the view data DA5 is not the bit format of 8 multiples, for example 11,12,19,20 bits ... the bit format that waits.

In this one exemplary embodiment, each pixel value multiply by first ratio value among the view data DA4 of arithmetic unit ADJ1, uses to obtain view data DA5.First ratio value can be 8, and each pixel value among the view data DA5 is all 11 bit data forms at this moment.Then, each pixel value deducts first numerical value among the view data DA5 of arithmetic unit ADJ2, uses to obtain view data DA6, and wherein first numerical value can be 1024.Afterwards, the view data DA6 of color converter CT1 carries out color conversion process, uses output image data DA2.Color conversion process can be the conversion rgb color space to the YUV color space, and this moment, pixel value DA2Y was 11 bit data forms, and pixel value DA2U and pixel value DA2V are all 12 bit data forms.

For fear of when view data DA2 write memory device 106 time waste storage space, image data conversion unit 104 is carried out first and is transformed in view data DA2, and wherein first conversion can comprise following various execution mode.For example: first conversion can be with desiring the processed pixels value divided by second ratio value, as DA1Y=DA2Y/8, and DA1U=DA2U/16, DA1V=DA2V/16.First conversion can be got desire processed pixels value and be rounded up divided by the result after second ratio value.First conversion can will be desired the processed pixels value divided by the unconditional carry of result after second ratio value.First conversion can unconditionally be rounded off divided by the result after second ratio value desiring the processed pixels value.First conversion can be cast out p low order position with desiring the processed pixels value, for example pixel value DA1Y casts out the result of 3 low order positions for pixel value DA2Y, pixel value DA1U casts out the result of 4 low order positions for pixel value DA2U, and pixel value DA1V casts out the result of 4 low order positions for pixel value DA2V.

Accordingly, the data format of pixel value DA1Y, pixel value DA1U and pixel value DA1V is all 8 bit data forms, that is meets the tentation data reception form of storage arrangement 106 accesses.

On the other hand, for the view data DA1 that storage arrangement 106 is read out recovers to handle, 108 of image data conversion unit are carried out second conversion to view data DA1, use obtaining view data DA3.Second conversion can comprise various execution modes.For example: second conversion can be on duty with second ratio value with desiring processed pixels, DA3Y=DA1Y*8 for example, DA3U=DA1U*16, DA3Y=DA1Y*16.Thus, a plurality of pixel values among the view data DA2 are in fact corresponding to a plurality of pixel values among the view data DA3.

Fig. 2 B illustrates another calcspar into the image data processing unit of Fig. 1.Please merge with reference to Fig. 1 and Fig. 2 B, image data processing unit 102 comprises arithmetic unit ADJ1 and color converter CT1.In this example embodiment, suppose the 16 bit data forms that are all of pixel value DA4R, pixel value DA4G and pixel value DA4B, and tentation data reception form is 16 bit data forms.

In view of this, in the process that image is handled, if Scaled_Flag=1, then each pixel value among the view data DA5 is not the bit format of 8 multiples.Clearer, each pixel value multiply by the 3rd ratio value among the view data DA4 of arithmetic unit ADJ1, uses to obtain view data DA5.First ratio value can be 8, and each pixel value among the view data DA5 is all 19 bit data forms at this moment.Then, the view data DA5 of color converter CT1 carries out color conversion process, uses output image data DA2.Color conversion process can be the conversion rgb color space to the YUV color space, and this moment, pixel value DA2Y was 19 bit data forms, and pixel value DA2U and pixel value DA2V are all 20 bit data forms.

For fear of the waste storage space, image data conversion unit 104 is carried out first and is transformed in view data DA2, wherein first conversion various execution modes as described above.In other words, first conversion will make the data format of pixel value DA1Y, pixel value DA1U and pixel value DA1V be all 16 bit data forms, that is meet the tentation data reception form of storage arrangement 106 accesses.Further, for the view data DA1 that storage arrangement 106 is read out recovers to handle, so 108 of image data conversion unit carry out second conversion to view data DA1, use obtaining view data DA3.Wherein second conversion can be according to aforesaid various execution modes, and thus, a plurality of pixel values among the view data DA2 are in fact corresponding to a plurality of pixel values among the view data DA3.

Fig. 2 C illustrates another calcspar into the image data processing unit of Fig. 1.Please merge with reference to Fig. 1 and Fig. 2 C, image data processing unit 102 comprises arithmetic unit ADJ2 and color converter CT1.In this example embodiment, Scaled_Flag=0 wherein supposes the 8 bit data forms that are all of pixel value DA4R, pixel value DA4G and pixel value DA4B, and tentation data to receive form be 8 bit data forms.

In view of this, each pixel value deducts first numerical value among the view data DA4 of arithmetic unit ADJ2, uses to obtain view data DA6, and wherein first numerical value can be 128.Afterwards, the view data DA6 of color converter CT1 carries out color conversion process, uses output image data DA2.Color conversion process can be the conversion rgb color space to the YUV color space, and this moment, pixel value DA2Y was 8 bit data forms, and pixel value DA2U and pixel value DA2V are all 9 bit data forms.

For fear of the waste storage space, image data conversion unit 104 is carried out first and is transformed in view data DA2, and wherein first conversion can comprise following various execution mode.First conversion can be desired the processed pixels value divided by second ratio value, DA1U=DA2U/2 for example, DA1V=DA2V/2, and DA1Y=DA2Y.First conversion can be got desire processed pixels value and be rounded up divided by the result after second ratio value.First conversion can will be desired the processed pixels value divided by the unconditional carry of result after second ratio value.First conversion can unconditionally be rounded off divided by the result after second ratio value desiring the processed pixels value.First conversion can be cast out p low order position with desiring the processed pixels value, for example pixel value DA1U casts out the result of 1 low order position for pixel value DA2U, pixel value DA1V casts out the result of 1 low order position for pixel value DA2V, does not cast out the result of low order position and pixel value DA1Y is pixel value DA2Y.

On the other hand, for the view data DA1 that storage arrangement 106 is read recovers to handle, so 108 of image data conversion unit carry out second conversion to view data DA1, use obtaining view data DA3.Second conversion can comprise various execution modes.Second conversion can be on duty with second ratio value with desiring processed pixels, DA3U=DA1U*2 for example, DA3Y=DA1Y*2, and DA3Y=DA1Y.Further, a plurality of pixel values among the view data DA2 are in fact corresponding to a plurality of pixel values among the view data DA3.

The view data DA3 that obtains when image processing apparatus 100 can do further processing, for example is played in screen, writes to storage device, or encode.In this enforcement demonstration example, the execution mode of image encoder 110 can adopt various Standard of image compression.Image encoder 110 can be the JPEG-XR encoder.

Fig. 3 illustrates the calcspar of the image encoder of Fig. 1.Please merge with reference to Fig. 1 and Fig. 3, image encoder 110 comprises color converter CT2, overlapping transducer 302, quantizer 304, fallout predictor 306, and entropy coder 308.Wherein, the view data DA3 of color converter CT2 carries out color conversion, uses outputting encoded data EnD4.302 couples of coded data EnD4 of overlapping transducer carry out the overlapping conversion (Two-level Lapped Transform) of second order, and outputting encoded data EnD1.

304 couples of coded data EnD1 of quantizer quantize (Quantization), and outputting encoded data EnD2.306 couples of coded data EnD2 of fallout predictor predict (Prediction), and output encoder EnD3 data.308 couples of coded data EnD3 of entropy coder carry out entropy coding (EntropyEncoding), use output bit flow STR1.In other one exemplary embodiment of the present invention, image encoder 110 can be according to the practical design demand, and omits color converter CT2, can encode to obtain bit stream STR1 to view data DA3 equally by this.

Be applied to the method for image compression

According to the explanation of above-mentioned Fig. 1, Fig. 4 illustrates the flow chart into the image processing method of the present invention's one example embodiment.Please merge with reference to Fig. 1～Fig. 4, the image processing method of this example embodiment comprises: at first, view data DA2 is provided (step S400), wherein view data DA2 has a plurality of second pixel values, each second pixel value has second data format, and second data format does not meet the tentation data reception form of storage arrangement 106.

Then, execution first is transformed in view data DA2 and obtains view data DA1 to write to storage arrangement 106 (step S402), wherein view data DA1 has a plurality of first pixel values, each first pixel value has first data format, and first data format meets tentation data reception form.

Afterwards, when view data DA1 is read out from storage arrangement 106, execution second is transformed in view data DA1 and obtains the view data DA3 (step S404) that comprises a plurality of the 3rd pixel values, wherein each the 3rd pixel value has the 3rd data format, and the 3rd data format does not meet tentation data and receives form, and a plurality of first pixel values are in fact corresponding to a plurality of the 3rd pixel values.At last, output image data DA3 is to image encoder 110 (step S406).

In this one exemplary embodiment, the execution mode that first described in step S402 and S404 conversion and second conversion can exemplify with reference to above-described embodiment is so just repeat no more at this.

Be applied to the device of image decompressor

Fig. 5 illustrates and is the calcspar according to the image processing apparatus of another example embodiment of the present invention.Please refer to Fig. 5, image processing apparatus 500 comprises that image decoder 502 (for example is the JPEG-XR decoder, but be not limited in this), view data computing converting unit 504, storage arrangement 506 (for example be DRAM, but be not limited in this), image data conversion unit 508, and image data processing unit 510, wherein view data computing converting unit 504 comprises arithmetic unit ADJ3 and image data conversion unit 512.

Storage arrangement 506 can be written into and be read view data DA7, and wherein view data DA7 comprises a plurality of pixel values.In this example embodiment, be example with pixel value DA7Y, pixel value DA7U and three pixel values of pixel value DA7V, wherein the data format of pixel value DA7Y, pixel value DA7U and pixel value DA7V meets tentation data reception form, and it is the form that meets storage arrangement 506 accesses that this tentation data receives form, the bit data form of for example 8,16 or 8 multiples.

In the product that image is handled, (digital camera for example, but be not limited in this), usually has image decoder 502, in order to receive bit stream STR2, and carry out the image decoding and use output image data DA8, wherein view data DA8 comprises a plurality of pixel values, for example pixel value DA8Y, pixel value DA8U and pixel value DA8V.

If image decoder 502 adopts the technology of image compression standard JPEG-XR, then the tentation data that each pixel value does not meet storage arrangement 506 among the view data DA8 usually receives form.The content that discloses according to prior art as can be known, if the data format that writes when not conforming to the form of access memory device 506, then may be wasted storage space in the process that image is handled.For fear of when view data DA8 write memory device 506 time waste storage space, the view data computing converting unit 504 of this one exemplary embodiment is converted to view data DA8 the view data DA7 that the tentation data that meets storage arrangement 506 receives form earlier.

Carrying out in the process of image compression standard JPEG-XR, view data DA8 may be because of quantizing or the computing of re-quantization (Inverse Quantization) and can't corresponding actual color space.Therefore, this one exemplary embodiment can be used output image data DA11 by arithmetic unit ADJ3 so that view data DA8 is carried out correcting process.This correcting process can be saturated (Saturation) and handles.In other words, if when certain pixel value among the view data DA8 surpasses first preset range, then this pixel value will carry out correcting process, uses each pixel value of making among the view data DA11 all in first preset range.

The correcting process of carrying out about described arithmetic unit ADJ3, more specifically, suppose that the pixel value DA8Y among the view data DA8 is 19 bit data forms, pixel value DA8Y is positive integer and 0≤DA8Y≤524287, wherein more than or equal to 0 to the positive integer smaller or equal to 524287 be first preset range.The pixel value DA8Y that image decoder 502 obtains in this one exemplary embodiment may be less than 0, or greater than 524287.This moment, arithmetic unit ADJ3 carried out correcting process to the pixel value DA8Y that surpasses first preset range, made each pixel value among the DA11 that arithmetic unit ADJ3 exports all in first preset range.Described first preset range is not restricted to more than or equal to 0 to smaller or equal to 524287 positive integer.

Further, through after the correcting process, each pixel value of view data DA11 usually and do not meet tentation data and receive form, therefore image data conversion unit 512 is carried out first and is transformed in view data DA11, and obtain view data DA7, wherein the data format of pixel value DA7Y, pixel value DA7U and pixel value DA7V is that tentation data receives form.

In addition, because view data DA7 has passed through correcting process and first conversion process, therefore when view data DA7 was read out from storage arrangement 506, this one exemplary embodiment was recovered view data DA7 to handle.Clearer, 508 couples of view data DA7 that read out from storage arrangement 506 of image data conversion unit carry out second conversion, use obtaining view data DA9.

This example embodiment is example with pixel value DA9Y, pixel value DA9U and three pixel values of pixel value DA9V, and wherein a plurality of pixel values of view data DA9 are in fact corresponding to a plurality of pixel values among the view data DA8.Further, the view data DA9 that image data conversion unit 508 obtains can do further processing, for example is played in screen, writes to storage device, or carry out other processing.In this example embodiment, image processing apparatus 500 comprises image data processing unit 510, in order to receiving view data DA9, and carries out view data and handles, and uses output image data DA10.

According to above-mentioned, image decoder 502 receives bit stream STR2, and decodes and use output image data DA8.For fear of the waste storage space, view data computing converting unit 504 is carried out correcting process and first and is transformed in view data DA8, and obtain view data DA7, wherein the tentation data that each pixel value meets storage arrangement 506 accesses among the DA7 receives form.

In addition, change through correcting process and first based on view data DA8, therefore when image data conversion unit 508 behind storage arrangement 506 reads image data DA7, this one exemplary embodiment is carried out second conversion to view data DA7, uses acquisition image restored data DA9.Wherein, a plurality of pixel values among the view data DA9 are in fact corresponding to a plurality of pixel values among the view data DA8.In addition, image decoder 502 exportable different data formats, in response to various different data formats, view data computing converting unit 504 is carried out first conversion and also can be adjusted thereupon.

Fig. 6 A illustrates the calcspar of the image data processing unit 510 of Fig. 5.Please merge with reference to Fig. 5 and Fig. 6 A, in this example embodiment, the tentation data reception form of supposing storage arrangement 506 is 8 bit data forms, and pixel value DA8Y is 11 bit data forms, pixel value DA8U and pixel value DA8V are all 12 bit data forms, and its neither tentation data that meets storage arrangement 506 receives form.

For fear of when view data DA8 write memory device 506 time waste storage space, image data conversion unit 512 is carried out first and is transformed in view data DA11, and wherein first conversion can comprise following various execution mode.For example: first conversion can be with desiring the processed pixels value divided by second ratio value, as DA7Y=DA11Y/8, and DA7U=DA11U/16, DA7V=DA11V/16.First conversion can be got desire processed pixels value and be rounded up divided by the result after second ratio value.First conversion can will be desired the processed pixels value divided by the unconditional carry of result after second ratio value.First conversion can unconditionally be rounded off divided by the result after second ratio value desiring the processed pixels value.First conversion can be cast out p low order position with desiring the processed pixels value, for example pixel value DA7Y casts out the result of 3 low order positions for pixel value DA11Y, pixel value DA7U casts out the result of 4 low order positions for pixel value DA11U, and pixel value DA7V casts out the result of 4 low order positions for pixel value DA11V.

Accordingly, the data format of pixel value DA7Y, pixel value DA7U and pixel value DA7V is all 8 bit data forms, that is meets the tentation data reception form of memory device access.

For the view data DA7 that storage arrangement 506 is read out recovers to handle, 508 couples of view data DA7 of image data conversion unit carry out second conversion.Second conversion can comprise following various execution mode.For example: second conversion can be on duty with first ratio value with desiring processed pixels, DA9Y=DA7Y*8 for example, DA9U=DA7U*16, DA9Y=DA7Y*16, this moment, pixel value DA9Y was 11 bit data forms, pixel value DA9U and pixel value DA9V are 12 bit data forms, and wherein a plurality of pixel values of view data DA9 are in fact corresponding to a plurality of pixel values among the view data DA8.

In this one exemplary embodiment, according to the data format of many each pixel values among the view data DA9, view data DA9 can do further processing, for example is played in screen, writes to storage device, or carry out other processing.Please refer to Fig. 6 A, the image data processing unit 510 of this example embodiment comprises color converter CT3 and arithmetic unit ADJ4～ADJ6.The view data DA9 of color converter CT3 carries out color conversion process, uses output image data DA12.Color conversion process can be conversion YUV color space to rgb color space.In this example embodiment, view data DA12 comprises a plurality of pixel values, and is example with pixel value DA12R, pixel value DA12G and three pixel values of pixel value DA12B.Pixel value DA12R, pixel value DA12G and pixel value DA12B are all 11 bit data forms at this moment.

Each pixel value adds first numerical value among the view data DA12 of arithmetic unit ADJ4, and output image data DA13, and wherein first numerical value can be 1024.The view data DA13 of arithmetic unit ADJ5 carries out correcting process, makes each pixel value among the output image data DA14 all in second preset range.Each pixel value uses obtaining view data DA10 divided by one the 3rd ratio value among the view data DA14 of arithmetic unit ADJ6, and wherein the 3rd ratio value can be 8.

About the correcting process that described arithmetic unit ADJ5 carries out, more specifically, suppose that view data DA13 comprises a plurality of pixel values, be example with pixel value DA13R, pixel value DA13G and three pixel values of pixel value DA13B.Suppose that pixel value DA13R is 11 bit data forms, pixel value DA13R is positive integer and 0≤DA13R≤2047, wherein more than or equal to 0 to the positive integer smaller or equal to 2047 be second preset range.The pixel value DA13R that arithmetic unit ADJ4 obtains in this one exemplary embodiment may be less than 0, or greater than 2047.This moment, arithmetic unit ADJ5 carried out correcting process to the pixel value DA13R that surpasses second preset range, made each pixel value among the DA14 that arithmetic unit ADJ5 exports all in second preset range.Described second preset range is not restricted to more than or equal to 0 to smaller or equal to 2047 positive integer.

Fig. 6 B illustrates another calcspar of the image data processing unit 510 of Fig. 5.Please merge with reference to Fig. 5 and Fig. 6 B, in this example embodiment, suppose that the tentation data reception form of storage arrangement 506 is 16 bit data forms, pixel value DA8Y is 19 bit data forms, pixel value DA8U and pixel value DA8V are all 20 bit data forms, and its neither tentation data that meets receives form.

For fear of when view data DA8 write memory device 506 time waste storage space, image data conversion unit 512 is carried out first and is transformed in view data DA11, wherein first conversion various execution modes as described above.In other words, first conversion will make the data format of pixel value DA7Y, pixel value DA7U and pixel value DA7V be all 16 bit data forms, that is meet the tentation data reception form of storage arrangement 506 accesses.

Further, for the view data DA7 that storage arrangement 506 is read recovers to handle, 508 couples of view data DA7 of image data conversion unit carry out second conversion.Wherein second conversion can be according to aforesaid various execution modes.This moment, pixel value DA9Y was 19 bit data forms, and pixel value DA9U and pixel value DA9V are 20 bit data forms, and wherein a plurality of pixel values of view data DA9 are in fact corresponding to a plurality of pixel values among the view data DA8.

In addition, according to the data format of many each pixel values among the view data DA9, view data DA9 can do further processing.Please refer to Fig. 6 B, image data processing unit 510 comprises color converter CT3, arithmetic unit ADJ5 and arithmetic unit ADJ6 in this example embodiment.The view data DA9 of color converter CT3 carries out color conversion process, uses output image data DA12.Color conversion process can be conversion YUV color space to rgb color space.Pixel value DA12R, pixel value DA12G and pixel value DA12B are all 19 bit data forms at this moment.Further, the view data DA12 of arithmetic unit ADJ5 carries out correcting process, makes each pixel value among the output image data DA14 all in the 3rd preset range.Each pixel value uses obtaining view data DA10 divided by one the 3rd ratio value among the view data DA14 of arithmetic unit ADJ6, and wherein the 3rd ratio value can be 8.

The correcting process of carrying out about described arithmetic unit ADJ5, more specifically, suppose that pixel value DA12R is 19 bit data forms, pixel value DA12R be integer and-262144≤DA12R≤262143, wherein more than or equal to-262144 to the integer smaller or equal to 262143 be the 3rd preset range.The pixel value DA12R that color converter CT3 obtains in this one exemplary embodiment may be less than-262144, or greater than 262143.This moment, arithmetic unit ADJ5 carried out correcting process to the pixel value DA12R that surpasses the 3rd preset range, made each pixel value among the DA12 that arithmetic unit ADJ5 exports all in the 3rd preset range.Described the 3rd preset range is not restricted to more than or equal to-262144 to smaller or equal to 262143 integer.

The image decoder 502 of image processing apparatus 500 configurations can adopt various Standard of image compression, uses the function of finishing the image decoding.Image decoder 502 can be the JPEG-XR decoder.Fig. 7 illustrates the calcspar into the image decoder 502 of Fig. 5.Please refer to Fig. 7, image decoder 502 comprises entropy decoder 702, inverse predictor 704, inverse quantizer 706, contrary overlapping transducer 708, and color converter CT4.Wherein, 702 couples of bit stream STR2 of entropy decoder carry out entropy decoding (EntropyDecoding), use output decoded data DeD1.

704 couples of decoded data DeD1 of inverse predictor carry out inverse prediction (Inverse Prediction), use output decoded data DeD2.706 couples of decoded data DeD2 of inverse quantizer carry out re-quantization (InverseQuantization), use output decoded data DeD3.Contrary 708 couples of decoded data DeD3 of overlapping transducer carry out the contrary conversion (Two-level Lapped Inverse Transform) that overlaps of second order, use output decoded data DeD4.Color converter CT4 is, and decoded data DeD4 carries out color converter, uses output image data DA8.

In other one exemplary embodiment of the present invention, image decoder 502 can be according to the various demands of actual design, and then omit color converter CT4, can decode to obtain view data DA8 to bit stream STR2 equally by this.

Be applied to the method for image decompressor

According to the explanation of above-mentioned Fig. 5, Fig. 8 illustrates the flow chart into the image processing method of another example embodiment of the present invention.Please merge with reference to Fig. 5～Fig. 8, the image processing method of this example embodiment comprises: at first, provide view data DA8 (step S800) by image decoder 502, wherein view data DA8 has a plurality of the 8th pixel values, each the 8th pixel value has the 8th data format, and the 8th data format does not meet the tentation data reception form of storage arrangement 506.

Then, to obtain the view data DA11 (step S802) that comprises a plurality of the 11 pixel values, wherein each the 11 pixel value is all in preset range in view data DA8 for the execution correcting process.Afterwards, carrying out first is transformed in view data DA11 and obtains view data and write to storage arrangement 506 (step S804) with DA7, wherein this view data DA7 has a plurality of the 7th pixel values, each the 7th pixel value has the 7th data format, and the 7th data format meets the tentation data reception form of storage arrangement 506.

At last, when view data DA7 is read out from storage arrangement 506, execution second is transformed in view data DA7 and obtains the view data DA9 (step S806) that comprises a plurality of the 9th pixel values, wherein each the 9th pixel value has the 9th data format, the 9th data format does not meet this tentation data and receives form, and a plurality of the 8th pixel values are in fact corresponding to a plurality of the 9th pixel values.

In this one exemplary embodiment, the execution mode that first described in step S804 and S806 conversion and second conversion can exemplify with reference to above-described embodiment is so just repeat no more at this.

Be applied to the device of image compression and decompression

In the product that image is handled, generally include encoder, decoder and a plurality of image data processing unit, use making product function more.Fig. 9 illustrates the calcspar into the image processing apparatus of another example embodiment of the present invention.Please refer to Fig. 9, the device that image processing apparatus 900 is applied to image compression comprises image data processing unit 102, image data conversion unit 104, image data conversion unit 108, image encoder 110 and storage arrangement 106.On the other hand, the device that image processing apparatus 900 is applied to image decompressor comprises image decoder 502, view data computing converting unit 504, image data conversion unit 508, image data processing unit 510 and storage arrangement 106, and wherein view data computing converting unit 504 also comprises arithmetic unit ADJ3 and image data conversion unit 512.In other words, storage arrangement 106 can share in the application of image compression and decompression.

When image processing apparatus 900 is applied to image compression, the execution mode of image data processing unit 102, image data conversion unit 104, image data conversion unit 108, image encoder 110, storage arrangement 106, view data DA1～DA3, view data DA7 and bit stream STR1 can be according to above-mentioned explanation for Fig. 1, use and implement image processing apparatus 900, just repeat no more at this.

When image processing apparatus 900 is applied to image decompressor, the execution mode of image decoder 502, view data computing converting unit 504, image data conversion unit 508, image data processing unit 510, storage arrangement 106, view data DA1, view data DA4～DA6, view data DA8 and bit stream STR2 can be according to above-mentioned explanation for Fig. 5, use and implement image processing apparatus 900, just repeat no more at this.

Because image processing apparatus 900 comprises image compression (coding) and the device of decompress (decoding), therefore image processing apparatus 900 image compression of carrying out and the correlation step of decompression, all can be with reference to above-mentioned Fig. 4 and the disclosed image processing method of Fig. 8, and can instruct and then know by inference according to above-mentioned example embodiment, just repeat no more at this.

In sum, image processing apparatus provided by the present invention has a plurality of image data conversion unit.When the pixel value in the view data does not meet the data format of memory device access, image data conversion unit can be converted into view data the tentation data that meets storage arrangement and receive form, uses and avoids waste storage space when view data write memory device.And after reading the view data of depositing in storage arrangement, the view data that reads is carried out another conversion, it is corresponding in fact to use the view data that makes before and after a plurality of conversions.

Though the present invention with example embodiment openly as above; so it is not in order to limit the present invention; those skilled in the art; without departing from the spirit and scope of the present invention; when the content that can instruct according to above-mentioned example embodiment, disclose or hint is done a little change and retouching, so protection scope of the present invention is as the criterion when looking the appended claims person of defining.

Claims

1. image processing apparatus comprises:

One storage arrangement, in order to be written into and to be read one first view data that comprises a plurality of first pixel values, each first pixel value has one first data format, and wherein this first data format meets the tentation data reception form of this storage arrangement;

One first image data conversion unit, couple this storage arrangement, one second view data that comprises a plurality of second pixel values in order to reception, and carry out one first the conversion and obtain this first view data to write to this storage arrangement, wherein each second pixel value has one second data format, and this second data format does not meet this tentation data and receives form;

One second image data conversion unit, couple this storage arrangement, in order to read this first view data from this storage arrangement, and carry out one second conversion and obtain one the 3rd view data that comprises a plurality of the 3rd pixel values, wherein each the 3rd pixel value has one the 3rd data format, and the 3rd data format does not meet this tentation data and receives form, and these second pixel values are in fact corresponding to these the 3rd pixel values; And

One image data processing unit, couple this first image data conversion unit, one the 4th view data that comprises a plurality of the 4th pixel values in order to reception, and carry out a view data and handle to obtain this second view data, wherein this image data processing unit comprises: a first arithmetic device, in order to receiving the 4th view data, and each the 4th pixel value be multiply by one first ratio value to obtain one the 5th view data that comprises a plurality of the 5th pixel values; One second arithmetic device couples this first arithmetic device, in order to receiving the 5th view data, and each the 5th pixel value is deducted one first numerical value to obtain one the 6th view data that comprises a plurality of the 6th pixel values; And a color transducer, couple this second arithmetic device, in order to receiving the 6th view data, and carry out color conversion process of the same colour to obtain this second view data.

2. image processing apparatus as claimed in claim 1, wherein this first conversion comprises cast out each second pixel value p low order position, p is the integer more than or equal to 0.

3. image processing apparatus as claimed in claim 2, wherein p is 0,1,3 or 4.

4. image processing apparatus as claimed in claim 1, wherein this first conversion comprises each second pixel value divided by one second ratio value.

5. image processing apparatus as claimed in claim 4, wherein this second ratio value is 1,2,8 or 16.

6. image processing apparatus as claimed in claim 1, wherein this second conversion comprises each first pixel value be multiply by one second ratio value.

7. image processing apparatus as claimed in claim 6, wherein this second ratio value is 1,2,8 or 16.

8. image processing apparatus as claimed in claim 1, wherein this first ratio value is 8.

9. image processing apparatus as claimed in claim 1, wherein this first numerical value is 128 or 1024.

10. image processing apparatus as claimed in claim 1 also comprises an image encoder, couples this second image data conversion unit, obtains a bit stream in order to receive the 3rd view data and to encode, to use.

11. image processing apparatus as claimed in claim 10, wherein this image encoder comprises:

One overlapping transducer couples this second image data conversion unit, in order to the 3rd view data is carried out the overlapping conversion of second order, uses and obtains a first coding data;

One quantizer couples this overlapping transducer, in order to this first coding data is quantized, use to obtain one second coded data;

One fallout predictor couples this quantizer, obtains one the 3rd coded data in order to this second coded data is predicted, used; And

One entropy coder couples fallout predictor, in order to the 3rd coded data is carried out the entropy coding, uses and obtains this bit stream.

12. image processing apparatus as claimed in claim 10, wherein this image encoder comprises:

One color transducer couples this second image data conversion unit, in order to the 3rd view data carried out a color conversion process, to use obtaining one the 4th coded data;

One overlapping transducer couples this color converter, in order to the 4th coded data is carried out the overlapping conversion of second order, uses and obtains a first coding data;

One entropy coder couples this fallout predictor, in order to the 3rd coded data is carried out the entropy coding, uses and obtains this bit stream.

13. image processing apparatus as claimed in claim 10, wherein this image encoder is the JPEG-XR encoder.

14. an image processing method comprises:

One first view data is provided, and wherein this first view data has a plurality of first pixel values, and each first pixel value has one first data format, and the tentation data that this first data format is not inconsistent the unification storage arrangement receives form;

Carrying out one first is transformed in this first view data and obtains one second view data to write to this storage arrangement, wherein this second view data has a plurality of second pixel values, each second pixel value has one second data format, and this second data format meets this tentation data and receives form; And

When this second view data is read out from this storage arrangement, carrying out one second is transformed in this second view data and obtains one the 3rd view data that comprises a plurality of the 3rd pixel values, wherein each the 3rd pixel value has one the 3rd data format, the 3rd data format does not meet this tentation data and receives form, these first pixel values are in fact corresponding to these the 3rd pixel values

Wherein, provide the step of this first view data to comprise: one the 4th view data that comprises a plurality of the 4th pixel values is carried out a view data handle to obtain this first view data, and this view data processing to comprise: each the 4th pixel value be multiply by one first ratio value to obtain one the 5th view data that comprises a plurality of the 5th pixel values; Each the 5th pixel value is deducted one first numerical value to obtain one the 6th view data that comprises a plurality of the 6th pixel values; And the 6th view data carried out a color conversion process to obtain this first view data.

15. image processing method as claimed in claim 14, wherein this first conversion comprises the low order position of the p that casts out each first pixel value, and wherein p is the integer more than or equal to 0.

16. image processing method as claimed in claim 15, wherein p is 0,1,3 or 4.

17. image processing method as claimed in claim 14, wherein this first conversion comprises each first pixel value divided by one second ratio value.

18. image processing method as claimed in claim 17, wherein this second ratio value is 1,2,8 or 16.

19. image processing method as claimed in claim 14, wherein this second conversion comprises each second pixel value be multiply by one second ratio value.

20. image processing method as claimed in claim 19, wherein this second ratio value is 1,2,8 or 16.

21. image processing method as claimed in claim 14 also comprises:

The 3rd view data to an image encoder is provided.

22. an image processing apparatus comprises:

One view data computing converting unit comprises:

One first arithmetic device, one second view data that comprises a plurality of second pixel values in order to reception, and carry out a correcting process to obtain one the 3rd view data of a plurality of the 3rd pixel values, wherein each the 3rd pixel value is all in one first preset range, each second pixel value has one second data format, and this second data format does not meet this tentation data and receives form; And

One first image data conversion unit couples this first arithmetic device, in order to receiving the 3rd view data, and carries out one first conversion and obtains this first view data to write to this storage arrangement;

One second image data conversion unit, couple this storage arrangement, in order to read this first view data from this storage arrangement, and carry out one second and change to obtain one the 4th view data that comprises a plurality of the 4th pixel values, wherein each the 4th pixel value has one the 3rd data format, the 3rd data format does not meet this tentation data and receives form, and these second pixel values are in fact corresponding to these the 4th pixel values; And

One image data processing unit, couple this second image data conversion unit, in order to receive the 4th view data, and carry out a view data and handle to obtain one the 5th view data that comprises a plurality of the 5th pixel values, wherein this image data processing unit comprises: a color transducer, in order to receiving the 4th view data, and carry out color conversion process of the same colour to obtain one the 6th view data that comprises a plurality of the 6th pixel values; One second arithmetic device couples this color converter, in order to receiving the 6th view data, and each the 6th pixel value is added that one first numerical value is to obtain one the 7th view data that comprises a plurality of the 7th pixel values; One the 3rd arithmetic unit couples this second arithmetic device, in order to receiving the 7th view data, and carries out a correcting process to obtain one the 8th view data that comprises a plurality of the 8th pixel values, and wherein each the 8th pixel value is all in one second preset range; And one the 4th arithmetic unit, couple the 3rd arithmetic unit, in order to receiving the 8th view data, and to each the 8th pixel value divided by one first ratio value to obtain the 5th view data.

23. image processing apparatus as claimed in claim 22, wherein this first conversion comprises the low order position of the p that casts out each the 3rd pixel value, and wherein p is the positive integer greater than 0.

24. image processing apparatus as claimed in claim 23, wherein p is 3 or 4.

25. image processing apparatus as claimed in claim 22, wherein this first conversion comprises each the 3rd pixel value divided by one second ratio value.

26. image processing apparatus as claimed in claim 25, wherein this second ratio value is 8 or 16.

27. image processing apparatus as claimed in claim 22, wherein this second conversion comprises each first pixel value be multiply by one second ratio value.

28. image processing apparatus as claimed in claim 27, wherein this second ratio value is 8 or 16.

29. image processing apparatus as claimed in claim 22, wherein this first numerical value is 1024.

30. image processing apparatus as claimed in claim 22, wherein this first ratio value is 8.

31. image processing apparatus as claimed in claim 22 also comprises an image decoder, couples this view data computing converting unit, in order to a bit stream is decoded to obtain this second view data.

32. image processing apparatus as claimed in claim 31, wherein this image decoder comprises:

One entropy decoder in order to this bit stream is carried out the entropy decoding, is used and is obtained one first decoded data;

One inverse predictor couples this entropy decoder, in order to this first decoded data is carried out inverse prediction, use obtaining one second decoded data;

One inverse quantizer couples this inverse predictor, in order to this second decoded data is carried out re-quantization, use obtaining one the 3rd decoded data; And

One contrary overlapping transducer couples this inverse quantizer, in order to the 3rd decoded data is carried out the contrary conversion that overlaps of second order, uses and obtains this second view data.

33. image processing apparatus as claimed in claim 31, wherein this image decoder comprises:

One inverse quantizer couples this inverse predictor, in order to this second decoded data is carried out re-quantization, use obtaining one the 3rd decoded data;

One contrary overlapping transducer couples this inverse quantizer, in order to the 3rd decoded data is carried out the contrary conversion that overlaps of second order, uses and obtains one the 4th decoded data; And

One color transducer couples this contrary overlapping transducer, in order to the 4th decoded data is carried out color conversion, uses and obtains this second view data.

34. image processing apparatus as claimed in claim 31, wherein this image decoder is the JPEG-XR decoder.

35. an image processing method comprises:

Carry out a correcting process in this first view data to obtain one second view data that comprises a plurality of second pixel values, wherein each second pixel value is all in a preset range;

Carrying out one first is transformed in this second view data and obtains one the 3rd view data to write to this storage arrangement, wherein the 3rd view data has a plurality of the 3rd pixel values, each the 3rd pixel value has one the 3rd data format, and the 3rd data format meets this tentation data and receives form;

When the 3rd view data is read out from this storage arrangement, carrying out one second is transformed in the 3rd view data and obtains one the 4th view data that comprises a plurality of the 4th pixel values, wherein each the 4th pixel value has one the 4th data format, the 4th data format does not meet this tentation data and receives form, and these first pixel values are in fact corresponding to these the 4th pixel values; And

The 4th view data is carried out a view data handle to obtain one the 5th view data that comprises a plurality of the 5th pixel values, wherein this view data is handled and is comprised: the 4th view data is carried out a color conversion process to obtain one the 6th view data that comprises a plurality of the 6th pixel values; Each the 6th pixel value is added that one first numerical value is to obtain one the 7th view data that comprises a plurality of the 7th pixel values; The 7th view data is carried out a correcting process to obtain one the 8th view data that comprises a plurality of the 8th pixel values, and wherein each the 8th pixel value is all in one second preset range; And to each the 8th pixel value divided by one first ratio value to obtain the 5th view data.

36. image processing method as claimed in claim 35, wherein this first conversion comprises the low order position of the p that casts out each second pixel value, and wherein p is the positive integer greater than 0.

37. image processing method as claimed in claim 36, wherein p is 3 or 4.

38. image processing method as claimed in claim 35, wherein this first conversion comprises each second pixel value divided by one second ratio value.

39. image processing method as claimed in claim 38, wherein this second ratio value is 8 or 16.

40. image processing method as claimed in claim 35, wherein this second conversion comprises each the 3rd pixel value be multiply by one second ratio value.

41. image processing method as claimed in claim 40, wherein this second ratio value is 8 or 16.

42. image processing method as claimed in claim 35 also comprises:

By an image decoder so that this first view data to be provided.

43. an image processing apparatus comprises:

One second image data conversion unit, couple this storage arrangement, in order to read this first view data from this storage arrangement, and carry out one second conversion and obtain one the 3rd view data that comprises a plurality of the 3rd pixel values, wherein each the 3rd pixel value has one the 3rd data format, and the 3rd data format does not meet this tentation data and receives form, and these second pixel values are in fact corresponding to these the 3rd pixel values;

One view data computing converting unit comprises:

One first arithmetic device, one the 4th view data that comprises a plurality of the 4th pixel values in order to reception, and carry out a correcting process to obtain one the 5th view data that comprises a plurality of the 5th pixel values, wherein each the 5th pixel value is all in one first preset range, each the 4th pixel value has one the 4th data format, and the 4th data format does not meet this tentation data and receives form; And

One the 3rd image data conversion unit couples this first arithmetic device, in order to receiving the 5th view data, and carries out one the 3rd conversion and obtains this first view data to write to this storage arrangement;

One the 4th image data conversion unit, couple this storage arrangement, in order to read this first view data from this storage arrangement, and carry out one the 4th and change to obtain one the 6th view data that comprises a plurality of the 6th pixel values, wherein each the 6th pixel value has one the 6th data format, the 6th data format does not meet this tentation data and receives form, and these the 4th pixel values are in fact corresponding to these the 6th pixel values; And

One image data processing unit, couple the 4th image data conversion unit, in order to receive the 6th view data, and carry out a view data and handle to obtain one the 7th view data that comprises a plurality of the 7th pixel values, wherein this image data processing unit comprises: a color transducer, in order to receiving the 6th view data, and carry out color conversion process of the same colour to obtain one the 8th view data that comprises a plurality of the 8th pixel values; One second arithmetic device couples this color converter, in order to receiving the 8th view data, and each the 8th pixel value is added that one first numerical value is to obtain one the 9th view data that comprises a plurality of the 9th pixel values; One the 3rd arithmetic unit couples this second arithmetic device, in order to receiving the 9th view data, and carries out a correcting process to obtain 1 the tenth view data that comprises a plurality of the tenth pixel values, and wherein each the tenth pixel value is all in one second preset range; And one the 4th arithmetic unit, couple the 3rd arithmetic unit, in order to receiving the tenth view data, and to each the tenth pixel value divided by one first ratio value to obtain the 7th view data.

44. image processing apparatus as claimed in claim 43, wherein this first conversion comprises the low order position of the p that casts out each second pixel value, and p is the integer more than or equal to 0.

45. image processing apparatus as claimed in claim 44, wherein p is 0,1,3 or 4.

46. image processing apparatus as claimed in claim 43, wherein this first conversion comprises each second pixel value divided by one second ratio value.

47. image processing apparatus as claimed in claim 46, wherein this second ratio value is 1,2,8 or 16.

48. image processing apparatus as claimed in claim 43, wherein this second conversion comprises each first pixel value be multiply by one second ratio value.

49. image processing apparatus as claimed in claim 48, wherein this second ratio value is 1,2,8 or 16.

50. image processing apparatus as claimed in claim 43, wherein the 3rd conversion comprises the low order position of the p that casts out each the 5th pixel value, and wherein p is the positive integer greater than 0.

51. image processing apparatus as claimed in claim 50, wherein p is 3 or 4.

52. image processing apparatus as claimed in claim 43, wherein the 3rd conversion comprises each the 5th pixel value divided by one second ratio value.

53. image processing apparatus as claimed in claim 52, wherein this second ratio value is 8 or 16.

54. image processing apparatus as claimed in claim 43, wherein the 4th conversion comprises each first pixel value be multiply by one second ratio value.

55. image processing apparatus as claimed in claim 54, wherein this second ratio value is 8 or 16.

56. an image processing apparatus comprises:

One view data computing converting unit comprises:

One image data processing unit, couple this second image data conversion unit, in order to receive the 4th view data, and carry out a view data and handle to obtain one the 5th view data that comprises a plurality of the 5th pixel values, wherein this image data processing unit comprises: a color transducer, in order to receiving the 4th view data, and carry out color conversion process of the same colour to obtain one the 6th view data that comprises a plurality of the 6th pixel values; One second arithmetic device couples this color converter, in order to receiving the 6th view data, and carries out a correcting process to obtain one the 7th view data that comprises a plurality of the 7th pixel values, and wherein each the 7th pixel value is all in one second preset range; And one the 3rd arithmetic unit, couple this second arithmetic device, in order to receiving the 7th view data, and to each the 7th pixel value divided by one first ratio value to obtain the 5th view data.

57. image processing apparatus as claimed in claim 56, wherein this first ratio value is 8.