CN108540798A - A kind of realization system and method for JPEG2000 images coding - Google Patents

A kind of realization system and method for JPEG2000 images coding Download PDF

Info

Publication number
CN108540798A
CN108540798A CN201810315318.7A CN201810315318A CN108540798A CN 108540798 A CN108540798 A CN 108540798A CN 201810315318 A CN201810315318 A CN 201810315318A CN 108540798 A CN108540798 A CN 108540798A
Authority
CN
China
Prior art keywords
channel
code
slope
target
coding
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810315318.7A
Other languages
Chinese (zh)
Other versions
CN108540798B (en
Inventor
雷杰
沈林浩
李云松
黄碧莹
孙洁婷
赵东升
于露露
解静
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan Rongchuang Microelectronic Co ltd
Original Assignee
Xidian University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xidian University filed Critical Xidian University
Publication of CN108540798A publication Critical patent/CN108540798A/en
Application granted granted Critical
Publication of CN108540798B publication Critical patent/CN108540798B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/124—Quantisation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/13—Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/146—Data rate or code amount at the encoder output
    • H04N19/147—Data rate or code amount at the encoder output according to rate distortion criteria
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/154—Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/63—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/85—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The present invention proposes a kind of realization system and method for JPEG2000 images coding, it is intended to which under the premise of not influencing image compression encoding performance, realize effectively reduces the consumption of rate-distortion optimization storage resource under arbitrary bit rate, and reduces and calculate the time, realizes that step is：Preprocessing module pre-processes input picture；Wavelet transform module carries out wavelet transform to pretreatment image, obtains wavelet sub-band coefficient；Quantization modules quantify wavelet sub-band coefficient, obtain quantization code block；Tier1 modules carry out bit-plane coding, MQ codings and slope to quantization code block and calculate, and obtain compressed bit stream, code check and the slope in all effective channels；The target bit rate R that code stream interception module is encoded according to the code check and slope in all effective channels and this imagetarget, the compressed bit stream in all effective channels is intercepted parallel, obtains interception code stream；Code stream organization module carries out code stream organization to interception code stream, obtains output image.

Description

A kind of realization system and method for JPEG2000 images coding
Technical field
The invention belongs to technical field of image processing, and in particular to a kind of realization system of JPEG2000 images coding and side Method can be used for the image compression encoding of various digital devices.
Background technology
With the development and application of multimedia and internet, traditional image compression algorithm has been unable to meet Vehicles Collected from Market With the requirement of practical application, thus International Standards Organization formulated the new standard of still image compression in November, 2000 JPEG2000.The new standard uses the rate-distortion optimized truncation built-in code block encryption algorithm (EBCOT) based on wavelet transformation technique, Achieve preferable compression of images effect.The coded system of standard JPEG2000 is as shown in Figure 1, wherein EBCOT points are compiled for Tier1 Code device and Tier2 encoder two parts.Tier1 is made of embedded bit-plane coding and MQ codings, complete context formed and MQ is encoded；Tier2 is calculated by slope, code stream interception and code stream organization form.When EBCOT is encoded, each wavelet sub-band is divided into Code block independently makees Tier1 codings as unit of code block.Different code blocks generates length not equal code stream, mirror as unit of channel Different to the importance for restoring picture quality in it, compression processing after Tier2 is carried out using rate-distortion optimization technology completes code stream Rate control and tissue.
Rate-distortion optimization in standard JPEG2000 is a kind of post-processing approach, completes the code check control of image coding, makes Obtaining compressed bit stream not only has optimal overall code check performance, and has optimal arbitrary point of cut-off code check performance.
Standard JPEG2000 carries out rate-distortion optimization using PCRD algorithms, which can appoint according to Embeded coding code stream Meaning, which is blocked and decoded, has the characteristics that certain mass image, is intercepted as unit of code block, can be in given target bit rate Under conditions of keep compression picture quality best.But PCRD algorithms are the global optimization procedures in entire image, it is needed to whole The each channel of all code blocks of width image carries out interception point search, therefore, it is necessary to first each to all code blocks by Tier1 encoders Channel is encoded, and stores code check and the distortion in all channels, then the code check according to all channels and mistake by Tier2 encoders The slope in all channels is really calculated, and carries out repeatedly interception and completes code check control.Obviously, rate-distortion optimization needs to consume a large amount of Time and storage resource are calculated, this is the bottleneck that JPEG2000 is realized.
In order to reduce calculating time and the storage resource of rate-distortion optimization, the prior art is mainly estimated by all kinds of code checks The method of meter carries out primary pre- interception so that Tier1 encoders only encode portion of channel, reduce at Tier1 Calculation amount at Tier1 and amount of storage, for example, Authorization Notice No. is CN103118259B, a kind of entitled " JPEG2000 images The Chinese patent of coding method ", discloses a kind of JPEG2000 coding methods, and this method is right in JPEG2000 standard bases Tier1 codings are improved, initial value and cycle criterion according to the minimum rate-distortion slope of first encoding block as thresholding The two features, the coding pass to being set as one state encode, and the coding pass for being set as " 0 " state abandons.It is logical Cycle criterion is crossed, the redundancy encoding channel of a large amount of Tier1 can be blocked, while also reducing Tier2 and searching for optimal rate-distortion slope The range of threshold value and optimal truncation point, therefore calculating time and the storage money of JPEG2000 images coding can be significantly decreased Source.But it has a defect that, the number and code that Tier1 needs the number of active lanes encoded is reduced by code check pre-estimation and pre- interception Rate (bpp) is related, and in low bit- rate, a large amount of channels are rejected during the pre- interception of Tier1, to significantly reduce storage Amount and calculating time, but in high code check, only a small amount of channel is rejected during the pre- interception of Tier1, therefore reduce Resource and time are extremely limited, in addition, by intercepting the channel given up in advance and PCRD algorithms were intercepted in Tier2 at Tier1 The channel given up in journey is not the relationship completely included, leads to the useful code stream for having lost a part, therefore can influence image Compressed encoding performance.
Invention content
It is an object of the invention to overcome above-mentioned the shortcomings of the prior art, it is proposed that a kind of JPEG2000 images coding Realization system and method, it is intended under the premise of not influencing image compression encoding performance, realization effectively reduced under arbitrary bit rate The consumption of rate-distortion optimization storage resource, and reduce and calculate the time.
To achieve the above object, the technical solution that the present invention takes is：
A kind of realization system of JPEG2000 images coding, including preprocessing module, wavelet transform module, quantization mould Block and EBCOT modules, wherein：
Preprocessing module, for the pretreatment by carrying out piecemeal, level shift and component transformation to input picture, to obtain Take pretreatment image；
Wavelet transform module is used for by carrying out wavelet transform to pretreatment image, to obtain wavelet sub-band Coefficient；
Quantization modules, for by quantifying to wavelet sub-band coefficient, to obtain quantization code block；
EBCOT modules, including Tier1 modules and Tier2 modules, for being carried out in rate-distortion optimized truncation to quantization code block Embedding code block coding；
The Tier1 modules, including bit-plane coding module, MQ coding modules and slope computing module, wherein：
Bit-plane coding module is used for by carrying out bit-plane coding to quantization code block, every to obtain all code blocks The binary character and context in a channel, while calculating the distortion in each channel of all code blocks；
MQ coding modules are used for through the binary character and context progress MQ codings to each channel of all code blocks, To obtain the compressed bit stream in each channel of all code blocks, and by counting to obtain the code check in each channel of all code blocks；
Slope computing module is used for the distortion according to each channel of all code blocks and code check, calculates all code blocks and each lead to The slope in road, and by rejecting the singular point in all channels, to obtain all code blocks each effective channel；
The Tier2 modules, including code stream interception module and code stream organization module, wherein：
Code stream interception module, for according to all code blocks each effective channel code check and slope and this image encode Target bit rate Rtarget, intercepted parallel by the compressed bit stream to each effective channel of all code blocks, to obtain interception code Stream；
Code stream organization module is used for by carrying out code stream organization to interception code stream, to obtain output image.
A kind of realization system of above-mentioned JPEG2000 images coding, the Tier1 modules, bit-plane coding mould therein Block and MQ coding modules often complete a channel coding, i.e., the distortion in the channel and code check are input to slope computing module, real Existing channel coding carries out while calculating with channel slope.
A kind of implementation method of JPEG2000 images coding, includes the following steps：
(1) preprocessing module pre-processes input picture：
Preprocessing module is split input picture, obtains one or more image blocks, and to the pixel of image block Level shift is carried out, " 0 " symmetrical image block is obtained, then color component transformation is carried out to " 0 " symmetrical image block, obtains pre- place Manage image；
(2) wavelet transform module carries out wavelet transform to pretreatment image, obtains wavelet sub-band coefficient；
(3) quantization modules quantify wavelet sub-band coefficient, obtain quantization code block；
(4) Tier1 modules obtain the compressed bit stream, code check and slope in each effective channel of all code blocks：To quantifying code block It carries out bit-plane coding, MQ codings and slope to calculate, specifically includes following steps：
(4a) bit-plane coding module carries out bit-plane coding to quantization code block, and obtained all code blocks are each The binary character in channel inputs MQ coding modules with context；The distortion in n-th of code block, i-th of channel is calculated simultaneouslyAnd Result of calculation is inputted into slope computing module；
(4b) MQ coding modules carry out MQ codings to the binary character in each channel of all code blocks with context, obtain institute There is a compressed bit stream in each channel of code block, and by the code check in n-th of code block, i-th of the channel countedSlope is inputted to calculate Module；
(4c) slope computing module receives the distortion in n-th of code block, i-th of channelAnd code checkAccording toWithCalculate the slope in n-th of code block, i-th of channelThe set of slopes in all channels is obtained, and is rejected in all channels Singular point, each channel is at most rejected forward three times, by the compressed bit stream, code check in each effective channel of all code blocks after rejecting With slope input code flow interception module；
(5) code stream interception module according to all code blocks each effective channel code check and slope and this image coding Target bit rate Rtarget, the compressed bit stream in each effective channel of all code blocks is intercepted parallel, obtains interception code stream：
(5a) initializes the bound of all effective channel set of slopes, enables slope lower limit SminIt is 0, slope upper limit SmaxFor Maximum value in all effective channel set of slopes；
(5b) calculates the slope threshold S when previous interceptionthrs=(Smin+Smax)/2, and possible two are intercepted next time A slope threshold Snext1=(Smin+Sthrs)/2 and Snext2=(Sthrs+Smax)/2；
(5c) is with three slope threshold Sthrs、Snext1And Snext2All effective channels are intercepted parallel, obtain three The effective channel set intercepted out and three corresponding interception code checks：
(i) slope threshold S is usedthrsAll effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channel With SthrsIt is compared, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th of code block i-th A effective channel, and using all effective channels of reservation as slope threshold SthrsThe effective channel set intercepted out, then should Effectively the code check in channel is added in set, obtains slope threshold SthrsCorresponding interception code check Rthrs；
(ii) slope threshold S is usednext1All effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channelWith Snext1It is compared, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th yard I-th of effective channel of block, and using all effective channels of reservation as slope threshold Snext1The effective channel set intercepted out, then The code check in effective channel in the set is added, slope threshold S is obtainednext1Corresponding interception code check Rnext1；
(iii) slope threshold S is usednext2All effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channelWith Snext2It is compared, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th of code block I-th of effective channel, and using all effective channels of reservation as slope threshold Snext2The effective channel set intercepted out, then will The code check in effective channel is added in the set, obtains slope threshold Snext2Corresponding interception code check Rnext2；
(5d) is by slope threshold S in step (5c)thrsCorresponding interception code check RthrsWith target bit rate RtargetIt is compared, If Rthrs=Rtarget, then it is S to enable target slopes thresholdingthrs, go to step (5g)；Otherwise, if Rthrs<Rtarget, then step is gone to (5e), if Rthrs>Rtarget, then step (5f) is gone to；
(5e) is if the difference for intercepting possible two slope thresholds next time meets Snext2-Snext1≤ 1, then enable target oblique Rate thresholding is Snext1, go to step (5g)；If not, by slope threshold S in step (5c)next1Corresponding interception code check Rnext1 With target bit rate RtargetIt is compared, if Rnext1=Rtarget, then it is S to enable target slopes thresholdingnext1, go to step (5g)；It is no Then, if Rnext1<Rtarget, then S is enabledminFor 0, SmaxFor Snext1, step (5b) is gone to, if Rnext1>Rtarget, then S is enabledminFor Snext1、SmaxFor Sthrs, go to step (5b)；
(5f) is if the difference for intercepting possible two slope thresholds next time meets Snext2-Snext1≤ 1, then enable target oblique Rate thresholding is Snext2, go to step (5g)；If not, by slope threshold S in step (5c)next2Corresponding interception code check Rnext2 With target bit rate RtargetIt is compared, if Rnext2=Rtarget, then it is S to enable target slopes thresholdingnext2, go to step (5g)；It is no Then, if Rnext2<Rtarget, then S is enabledminFor Sthrs、SmaxFor Snext2, step (5b) is gone to, if Rnext2>Rtarget, then S is enabledminFor Snext2、SmaxIt is constant, go to step (5b)；
The compressed bit stream in effective channel that (5g) intercepts out target slopes thresholding merges, and obtains interception code stream；
(6) code stream organization module carries out code stream organization to interception code stream, obtains output image.
Compared with prior art, the present invention having the following advantages that：
First, slope computing module is advanced to Tier1 by the present invention from Tier2, into row of channels while channel coding Slope calculates, and to reduce the time loss of channel coding and slope calculating, and avoids Tier1 and stores all channels mistakes Genuine demand, to reduce the consumption of Tier1 storage resources, compared with prior art, the rate under arbitrary bit rate that realizes is distorted The effective reduction for optimizing time and resource consumption, improves code efficiency.
Second, the present invention takes the mode intercepted parallel when 2 code streams of Tier intercept, and interception number is reduced half, Interception speed by about one time is improved, the time loss of rate-distortion optimization is further reduced.
Third, the present invention realize that the principle of rate-distortion optimization and PCRD algorithms are completely the same, ensure that the property of image coding It can be suitable with standard JPEG2000.
Description of the drawings
Fig. 1 is the system block diagram of standard JPEG2000；
Fig. 2 is the structural schematic diagram of realization system of the present invention；
Fig. 3 is the flow chart of implementation method of the present invention.
Specific implementation mode
Below in conjunction with the drawings and specific embodiments, present invention is further described in detail.
With reference to Fig. 2, a kind of realization system of JPEG2000 images coding, including the mold changing of preprocessing module, discrete wavelet transformer Block, quantization modules and EBCOT modules, wherein：
Preprocessing module, for the pretreatment by carrying out piecemeal, level shift and component transformation to input picture, to obtain Take pretreatment image；
Wavelet transform module is used for by carrying out wavelet transform to pretreatment image, to obtain wavelet sub-band Coefficient；
Quantization modules, for by quantifying to wavelet sub-band coefficient, to obtain quantization code block；
EBCOT modules, including Tier1 modules and Tier2 modules, for being carried out in rate-distortion optimized truncation to quantization code block Embedding code block coding；
The Tier1 modules, including bit-plane coding module, MQ coding modules and slope computing module, wherein：
Bit-plane coding module is used for by carrying out bit-plane coding to quantization code block, every to obtain all code blocks The binary character and context in a channel, while calculating the distortion in each channel of all code blocks；Often obtain the two of a channel The binary character in the channel and context are just inputted MQ coding modules by hex notation and context, while it is logical to calculate this The distortion in road inputs slope computing module；
MQ coding modules are used for through the binary character and context progress MQ codings to each channel of all code blocks, To obtain the compressed bit stream in each channel of all code blocks, and by counting to obtain the code check in each channel of all code blocks；Often receive As soon as to the binary character and context in channel, starts a MQ coding, obtain the compressed bit stream in the channel and storage, then The byte number in the channel compressions code stream is counted, the code check input slope computing module in the channel is obtained；
Slope computing module is used for the distortion according to each channel of all code blocks and code check, calculates all code blocks and each lead to The slope in road, and by rejecting the singular point in all channels, to obtain all code blocks each effective channel；One is often received to lead to The distortion in road and code check just calculate a slope, the channel slope are obtained, by first three channel of the channel slope and the channel Slope compare, be unsatisfactory for slope monotone decreasing channel be singular point, reject singular point, all channels retained are to have Imitate channel；
In above-mentioned Tier1 modules, bit-plane coding module and MQ coding modules often complete a channel coding, i.e., should The distortion in channel is input to slope computing module with code check, realizes the synchronous progress that channel coding is calculated with channel slope；Slope Computing module is located at Tier2 modules in standard JPEG2000, after slope computing module is transferred to Tier1 modules, in bit Plane coding and MQ can carry out slope calculating while coding, on the one hand shorten three's total working time, another side exists No longer memory channel is needed to be distorted after Tier1；
The Tier2 modules, including code stream interception module and code stream organization module, wherein：
Code stream interception module, for according to all code blocks each effective channel code check and slope and this image encode Target bit rate Rtarget, intercepted parallel by the compressed bit stream to each effective channel of all code blocks, to obtain interception code Stream；Parallel interception is exactly to be carried out at the same time to intercept twice, and the iterations intercepted in standard JPEG2000 are reduced half；
Code stream organization module is used for by carrying out code stream organization to interception code stream, to obtain output image.
With reference to figure 3, a kind of implementation method of JPEG2000 images coding includes the following steps：
(1) preprocessing module pre-processes input picture：
Preprocessing module is split input picture, obtains one or more image blocks, and to the pixel of image block Level shift is carried out, " 0 " symmetrical image block is obtained, then color component transformation is carried out to " 0 " symmetrical image block, obtains pre- place Manage image；
The input picture of the present embodiment is the RGB coloured pictures of 8 bit-depths of 512*512, and preprocessing module is by input picture It is divided into an image block, i.e. input picture itself, and all pixels point is subtracted 27Offset, make pixel about " 0 " Symmetrically, it then is converted by reversible component and input picture is transformed to yuv space from rgb space, obtain pretreatment image, reversible point Change of variable formula is as follows：
(2) wavelet transform module carries out wavelet transform to pretreatment image, obtains wavelet sub-band coefficient；
Wavelet transform be by pretreatment image be decomposed into low-frequency band coefficient LL and multiple high frequency band coefficient LH, HL and HH, wherein LL are expressed as being both horizontally and vertically smooth sub-band coefficients, and it is smooth, Vertical Square that LH, which is expressed as horizontal direction, To the sub-band coefficients for being details, it is smooth sub-band coefficients that it is details, vertical direction that HL, which is expressed as horizontal direction, and HH is expressed as water Gentle vertical direction is all the sub-band coefficients of details；Therefore low-frequency band coefficient LL is the frequency band that image energy is concentrated, high frequency band system Number HL, LH, HH are mainly edge, texture and profile, wavelet coefficient very littles；
(3) quantization modules quantify wavelet sub-band coefficient, obtain quantization code block；
Quantization is that the precision for the wavelet sub-band coefficient that will be obtained after wavelet transform is adjusted, i.e., by adjusting quantization The size of step-length come change indicate a wavelet sub-band coefficient needed for bit number, quantify wavelet sub-band coefficient formula it is as follows：
Wherein, qb[m, n] is the wavelet coefficient after subband b quantizations, sb[m, n] is the wavelet coefficient of subband b, ΔbFor subband The quantization step of b, the present embodiment take Δb=10, sign (sb[m, n]) indicate wavelet sub-band coefficient sbThe sign symbol of [m, n]；
(4) Tier1 modules obtain the compressed bit stream, code check and slope in each effective channel of all code blocks：To the code of quantization Block carries out bit-plane coding, MQ codings and slope and calculates, and specifically includes following steps：
(4a) bit-plane coding module carries out bit-plane coding to quantization code block, and obtained all code blocks are each The binary character in channel inputs MQ coding modules with context；The distortion in n-th of code block, i-th of channel is calculated simultaneouslyAnd Result of calculation is inputted into slope computing module；
(4b) MQ coding modules carry out MQ codings to the binary character in each channel of all code blocks with context, obtain institute There is a compressed bit stream in each channel of code block, and by the code check in n-th of code block, i-th of the channel countedSlope is inputted to calculate Module；
(4c) slope computing module receives the distortion in n-th of code block, i-th of channelAnd code checkAccording toWithCalculate the slope in n-th of code block, i-th of channelThe set of slopes in all channels is obtained, and is rejected in all channels Singular point, each channel is at most rejected forward three times, by the compressed bit stream, code check in each effective channel of all code blocks after rejecting With slope input code flow interception module；
Slope computing module first receives the distortion in n-th of code block, i-th of channelIt willCaching waits for MQ to encode mould After block completes the MQ codings to n-th of code block, i-th of channel, slope computing module receives the code in n-th of code block, i-th of channel again RateIt is calculated by the following formula the slope in n-th of code block, i-th of channel
Since channel slope is served only for comparison operation in interception, the present embodiment is to avoid floating-point division operation and floating-point It indicates, using above formula instead of in standard JPEG2000Calculation formula；
Slope computing module is calculatedAfterwards, channel before needing to judge n-th of code block, i-th of channel whether be Singular point is then rejected if singular point；Since singular point at most continuously occurs three, the present embodiment does not retrieve n-th All channels before i-th of channel of a code block only retrieve forward three channels, i.e., at most reject singular point three times forward, have Body process is as follows：
(4c1) is by the slope in i-th of channel of n-th of code blockWith the slope in (i-1)-th channelCompare, ifThen (i-1)-th channel is singular point, rejects (i-1)-th channel, and enable the distortion in i-th of channelCode checkAccording toI-th is recalculated to lead to The slope in roadIt goes to (4c2)；Otherwise, i-th of channel is rejected forward singular point and is terminated；
(4c2) is by the slope in i-th of channel of n-th of code blockWith the slope in the i-th -2 channelsCompare, ifThen the i-th -2 channels are singular point, reject the i-th -2 channels, and enable the distortion in i-th of channelCode checkAccording toRecalculate i-th channel SlopeIt goes to (4c3)；Otherwise, i-th of channel is rejected forward singular point and is terminated；
(4c3) is by the slope in i-th of channel of n-th of code blockWith the slope in the i-th -3 channelsCompare, ifThen the i-th -3 channels are singular point, reject the i-th -3 channels, and enable the distortion in i-th of channelCode checkAccording toI-th is recalculated to lead to The slope in roadSingular point is rejected forward and is terminated in i-th of channel；Otherwise, i-th of channel is rejected forward singular point and is terminated；
All channels not being removed are effective channel, and the singular point for being removed, channel distortion and code check It has been added on effective channel, it is clear that the compressed bit stream of singular point is also just included into effective channel, and final Tier1 modules will Compressed bit stream, code check and the slope in all effective channels are input to code stream interception module and carry out code stream interception；
(5) code stream interception module according to all code blocks each effective channel code check and slope and this image coding Target bit rate Rtarget, the compressed bit stream in each effective channel of all code blocks is intercepted parallel, obtains interception code stream；
Interception process needs to be traversed for the slope in all effective channels, therefrom searches out target slopes thresholding, uses target slopes The code check for the compressed bit stream that thresholding is truncated to is consistent with target bit rate, to realize code check control；
In order to quickly traverse, target gate slope is found using dichotomy, when initial, takes slope lower limit SminFor 0, Slope upper limit SmaxFor the maximum value in all effective channel set of slopes, section intermediate value S is takenthrs=(Smin+Smax)/2 are initial Slope threshold carries out first time interception, and all effective channels more than or equal to slope threshold can be retained, intercepted when for the first time Cheng Hou, the code check in all effective channels retained of adding up obtain interception code check Rthrs, compare RthrsWith target bit rate Rtarget, If Rthrs=Rtarget, then SthrsFor target slopes thresholding, if Rthrs<Rtarget, then illustrate that set slope threshold is bigger than normal, under order The slope threshold once intercepted is (Smin+Sthrs)/2, if Rthrs>Rtarget, then illustrate that set slope threshold is less than normal, under order The slope threshold once intercepted is (Sthrs+Smax)/2；Such loop iteration, after repeatedly intercepting, interception code check reaches target Code check or the set of slopes for having traversed all effective channels, interception terminate；
Since only there are two may for the slope threshold that on the basis of when the slope threshold of previous interception, intercepts next time Value, therefore, the present embodiment simultaneously with the slope threshold for working as previous interception and two possible slope thresholds intercepting next time into Row interception can be obtained when previous interception by the operation once intercepted and intercept next time intercepting twice as a result, realizing simultaneously Row interception；The present embodiment obtains one when next time the interception code check of previous interception and two intercept after primary parallel interception Possibility intercept code check, first according to the comparison result when previous interception code check and target bit rate, determine two possibility next time Which is really to intercept code check next time in interception code check, further according to the comparison knot for intercepting code check and target bit rate next time Fruit, more new slope bound, are as follows：
(5a) initializes the bound of all effective channel set of slopes, enables slope lower limit SminIt is 0, slope upper limit SmaxFor Maximum value in all effective channel set of slopes；
(5b) calculates the slope threshold S when previous interceptionthrs=(Smin+Smax)/2, and possible two are intercepted next time A slope threshold Snext1=(Smin+Sthrs)/2 and Snext2=(Sthrs+Smax)/2；
(5c) is with three slope threshold Sthrs、Snext1And Snext2All effective channels are intercepted parallel, obtain three The effective channel set intercepted out and three corresponding interception code checks：
(i) slope threshold S is usedthrsAll effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channel With progress SthrsCompare, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th of code block i-th A effective channel obtains slope threshold S after the judgement for completing all effective channelsthrsThe effective channel set intercepted out, and will The code check in effective channel is added in the set, obtains slope threshold SthrsCorresponding interception code check Rthrs；
(ii) slope threshold S is usednext1All effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channelWith Snext1It is compared, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th yard I-th of effective channel of block obtains slope threshold S after the judgement for completing all effective channelsnext1The effective channel set intercepted out It closes, and the code check in effective channel in the set is added, obtain slope threshold Snext1Corresponding interception code check Rnext1；
(iii) slope threshold S is usednext2All effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channelWith Snext2It is compared, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th yard I-th of effective channel of block obtains slope threshold S after the judgement for completing all effective channelsnext2The effective channel set intercepted out It closes, and the code check in effective channel in the set is added, obtain slope threshold Snext2Corresponding interception code check Rnext2；
(5d) is by slope threshold S in step (5c)thrsCorresponding interception code check RthrsWith target bit rate RtargetIt is compared, If Rthrs=Rtarget, then it is S to enable target slopes thresholdingthrs, go to step (5g)；Otherwise, if Rthrs<Rtarget, then step is gone to (5e), if Rthrs>Rtarget, then step (5f) is gone to；
(5e) is if the difference for intercepting possible two slope thresholds next time meets Snext2-Snext1≤ 1, then enable target oblique Rate thresholding is Snext1, go to step (5g)；If not, by slope threshold S in step (5c)next1Corresponding interception code check Rnext1 With target bit rate RtargetIt is compared, if Rnext1=Rtarget, then it is S to enable target slopes thresholdingnext1, go to step (5g)；It is no Then, if Rnext1<Rtarget, then S is enabledminFor 0, SmaxFor Snext1, step (5b) is gone to, if Rnext1>Rtarget, then S is enabledminFor Snext1、SmaxFor Sthrs, go to step (5b)；
Snext2-Snext1≤ 1 shows the slope threshold phase that the slope threshold intercepted parallel next time will parallel be intercepted with this Together, illustrate that the slope in all effective channels has traversed, this parallel interception intercepts for last time；
(5f) is if the difference for intercepting possible two slope thresholds next time meets Snext2-Snext1≤ 1, then enable target oblique Rate thresholding is Snext2, go to step (5g)；If not, by slope threshold S in step (5c)next2Corresponding interception code check Rnext2 With target bit rate RtargetIt is compared, if Rnext2=Rtarget, then it is S to enable target slopes thresholdingnext2, go to step (5g)；It is no Then, if Rnext2<Rtarget, then S is enabledminFor Sthrs、SmaxFor Snext2, step (5b) is gone to, if Rnext2>Rtarget, then S is enabledminFor Snext2、SmaxIt is constant, go to step (5b)；
The compressed bit stream in effective channel that (5g) intercepts out target slopes thresholding merges, and obtains interception code stream；
(6) code stream organization module carries out code stream organization to interception code stream, obtains output image；
Code stream organization module completes code stream organization and packing, output by code stream compressed format as defined in standard JPEG2000 Final encoding code stream realizes image compression encoding.
Unspecified part of the present invention belongs to common sense well known to those skilled in the art.
Above description is only the specific example of the present invention, does not constitute any limitation of the invention.Obviously for this It, all may be without departing substantially from the principle of the invention, structure after having understood the content of present invention and principle for the professional in field In the case of, any modifications and variations in form and details are carried out, but these modifications and variations based on inventive concept are still Within the claims of the present invention.

Claims (7)

1. a kind of realization system of JPEG2000 images coding, including preprocessing module, wavelet transform module, quantization modules With EBCOT modules, wherein：
Preprocessing module, it is pre- to obtain for the pretreatment by carrying out piecemeal, level shift and component transformation to input picture Handle image；
Wavelet transform module is used for by carrying out wavelet transform to pretreatment image, to obtain wavelet sub-band coefficient；
Quantization modules, for by quantifying to wavelet sub-band coefficient, to obtain quantization code block；
EBCOT modules, including Tier1 modules and Tier2 modules, for carrying out rate-distortion optimized truncation inline code to quantization code block Block coding；
It is characterized in that：
The Tier1 modules, including bit-plane coding module, MQ coding modules and slope computing module, wherein：
Bit-plane coding module, for by carrying out bit-plane coding to quantization code block, each being led to obtaining all code blocks The binary character and context in road, while calculating the distortion in each channel of all code blocks；
MQ coding modules are used for by the binary character and context progress MQ codings to each channel of all code blocks, to obtain The compressed bit stream in each channel of all code blocks is taken, and by counting to obtain the code check in each channel of all code blocks；
Slope computing module is used for the distortion according to each channel of all code blocks and code check, calculates each channel of all code blocks Slope, and by rejecting the singular point in all channels, to obtain all code blocks each effective channel；
The Tier2 modules, including code stream interception module and code stream organization module, wherein：
Code stream interception module, for according to all code blocks each effective channel code check and slope and this image coding mesh Coding rate Rtarget, intercepted parallel by the compressed bit stream to each effective channel of all code blocks, to obtain interception code stream；
Code stream organization module is used for by carrying out code stream organization to interception code stream, to obtain output image.
2. a kind of realization system of JPEG2000 images coding according to claim 1, which is characterized in that the Tier1 Module, bit-plane coding module therein and MQ coding modules often complete a channel coding, i.e., by the distortion in the channel with Code check is input to slope computing module, realizes and is carried out while channel coding is calculated with channel slope.
3. a kind of implementation method of JPEG2000 images coding, which is characterized in that include the following steps：
(1) preprocessing module pre-processes input picture：
Preprocessing module is split input picture, obtains one or more image blocks, and carry out to the pixel of image block Level shift obtains " 0 " symmetrical image block, then carries out color component transformation to " 0 " symmetrical image block, obtains pretreatment figure Picture；
(2) wavelet transform module carries out wavelet transform to pretreatment image, obtains wavelet sub-band coefficient；
(3) quantization modules quantify wavelet sub-band coefficient, obtain quantization code block；
(4) Tier1 modules obtain the compressed bit stream, code check and slope in each effective channel of all code blocks：Quantization code block is carried out Bit-plane coding, MQ codings and slope calculate, and specifically include following steps：
(4a) bit-plane coding module carries out bit-plane coding, and each channel of all code blocks that will be obtained to quantization code block Binary character and context input MQ coding modules；The distortion in n-th of code block, i-th of channel is calculated simultaneouslyAnd it will Result of calculation inputs slope computing module；
(4b) MQ coding modules carry out MQ codings to the binary character in each channel of all code blocks with context, obtain all codes The compressed bit stream in each channel of block, and by the code check in n-th of code block, i-th of the channel countedInput slope computing module；
(4c) slope computing module receives the distortion in n-th of code block, i-th of channelAnd code checkAccording toWithMeter Calculate the slope in n-th of code block, i-th of channelThe set of slopes in all channels is obtained, and is rejected unusual in all channels Point, each channel at most reject forward three times, by all code blocks after rejecting each the compressed bit stream, code check in effective channel and tiltedly Rate input code flow interception module；
(5) code stream interception module according to all code blocks each effective channel code check and slope and this image coding target Code check Rtarget, the compressed bit stream in each effective channel of all code blocks is intercepted parallel, obtains interception code stream：
(5a) initializes the bound of all effective channel set of slopes, enables slope lower limit SminIt is 0, slope upper limit SmaxIt is all Maximum value in the set of slopes of effective channel；
(5b) calculates the slope threshold S when previous interceptionthrs=(Smin+Smax)/2, and possible two slopes are intercepted next time Thresholding Snext1=(Smin+Sthrs)/2 and Snext2=(Sthrs+Smax)/2；
(5c) is with three slope threshold Sthrs、Snext1And Snext2All effective channels are intercepted parallel, obtain three interceptions The effective channel set gone out and three corresponding interception code checks：
(i) slope threshold S is usedthrsAll effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channelWith SthrsIt is compared, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th code block i-th Effective channel, and using all effective channels of reservation as slope threshold SthrsThe effective channel set intercepted out, then by the collection Effectively the code check in channel is added in closing, and obtains slope threshold SthrsCorresponding interception code check Rthrs；
(ii) slope threshold S is usednext1All effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channelWith Snext1It is compared, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th of code block i-th A effective channel, and using all effective channels of reservation as slope threshold Snext1The effective channel set intercepted out, then should Effectively the code check in channel is added in set, obtains slope threshold Snext1Corresponding interception code check Rnext1；
(iii) slope threshold S is usednext2All effective channels are intercepted, i.e., by the slope in n-th of code block, i-th of effective channelWith Snext2It is compared, ifThen retain n-th of code block, i-th of effective channel；Otherwise, give up n-th of code block i-th A effective channel, and using all effective channels of reservation as slope threshold Snext2The effective channel set intercepted out, then should Effectively the code check in channel is added in set, obtains slope threshold Snext2Corresponding interception code check Rnext2；
(5d) is by slope threshold S in step (5c)thrsCorresponding interception code check RthrsWith target bit rate RtargetIt is compared, if Rthrs=Rtarget, then it is S to enable target slopes thresholdingthrs, go to step (5g)；Otherwise, if Rthrs<Rtarget, then step is gone to (5e), if Rthrs>Rtarget, then step (5f) is gone to；
(5e) is if the difference for intercepting possible two slope thresholds next time meets Snext2-Snext1≤ 1, then enable target slopes thresholding For Snext1, go to step (5g)；If not, by slope threshold S in step (5c)next1Corresponding interception code check Rnext1With target Code check RtargetIt is compared, if Rnext1=Rtarget, then it is S to enable target slopes thresholdingnext1, go to step (5g)；Otherwise, if Rnext1<Rtarget, then S is enabledminFor 0, SmaxFor Snext1, step (5b) is gone to, if Rnext1>Rtarget, then S is enabledminFor Snext1、Smax For Sthrs, go to step (5b)；
(5f) is if the difference for intercepting possible two slope thresholds next time meets Snext2-Snext1≤ 1, then enable target slopes thresholding For Snext2, go to step (5g)；If not, by slope threshold S in step (5c)next2Corresponding interception code check Rnext2With target Code check RtargetIt is compared, if Rnext2=Rtarget, then it is S to enable target slopes thresholdingnext2, go to step (5g)；Otherwise, if Rnext2<Rtarget, then S is enabledminFor Sthrs、SmaxFor Snext2, step (5b) is gone to, if Rnext2>Rtarget, then S is enabledminFor Snext2、 SmaxIt is constant, go to step (5b)；
The compressed bit stream in effective channel that (5g) intercepts out target slopes thresholding merges, and obtains interception code stream；
(6) code stream organization module carries out code stream organization to interception code stream, obtains output image.
4. a kind of implementation method of JPEG2000 images coding according to claim 3, which is characterized in that step (2) institute The wavelet transform module stated carries out wavelet transform to pretreatment image, refers to that pretreatment image is decomposed into low-frequency band Coefficient LL and multiple high frequency band coefficient LH, HL and HH.
5. a kind of implementation method of JPEG2000 images coding according to claim 3, which is characterized in that step (3) institute The quantization modules stated quantify wavelet sub-band coefficient, refer to the essence for the wavelet sub-band coefficient that will be obtained after wavelet transform Degree is adjusted, i.e., by adjusting the size of quantization step come change indicate a wavelet sub-band coefficient needed for bit number.
6. a kind of implementation method of JPEG2000 images coding according to claim 3, which is characterized in that step (4c) institute The slope in n-th of the calculating stated, i-th of code block channelCalculation formula is：
Wherein,It is channel distortion,It is channel code check, n indicates that n-th of code block, i indicate i-th of channel.
7. a kind of implementation method of JPEG2000 images coding according to claim 3, which is characterized in that step (4c) institute Singular point in all channels of rejecting stated, each channel are at most rejected forward three times, and following steps are specifically included：
(4c1) is by the slope in i-th of channel of n-th of code blockWith the slope in (i-1)-th channelCompare, ifThen (i-1)-th channel is singular point, rejects (i-1)-th channel, and enable the distortion in i-th of channelCode checkAccording toI-th is recalculated to lead to The slope in roadIt goes to (4c2)；Otherwise, i-th of channel is rejected forward singular point and is terminated；
(4c2) is by the slope in i-th of channel of n-th of code blockWith the slope in the i-th -2 channelsCompare, ifThen the i-th -2 channels are singular point, reject the i-th -2 channels, and enable the distortion in i-th of channelCode checkAccording toRecalculate i-th channel SlopeIt goes to (4c3)；Otherwise, i-th of channel is rejected forward singular point and is terminated；
(4c3) is by the slope in i-th of channel of n-th of code blockWith the slope in the i-th -3 channelsCompare, ifThen the i-th -3 channels are singular point, reject the i-th -3 channels, and enable the distortion in i-th of channelCode checkAccording toRecalculate i-th of channel SlopeSingular point is rejected forward and is terminated in i-th of channel；Otherwise, i-th of channel is rejected forward singular point and is terminated.
CN201810315318.7A 2017-12-29 2018-04-10 JPEG2000 image coding implementation system and method Active CN108540798B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711476424 2017-12-29
CN2017114764245 2017-12-29

Publications (2)

Publication Number Publication Date
CN108540798A true CN108540798A (en) 2018-09-14
CN108540798B CN108540798B (en) 2020-06-05

Family

ID=63479793

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810315318.7A Active CN108540798B (en) 2017-12-29 2018-04-10 JPEG2000 image coding implementation system and method

Country Status (1)

Country Link
CN (1) CN108540798B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110430430A (en) * 2019-08-08 2019-11-08 浙江大华技术股份有限公司 Image processing method and device, storage medium, electronic device
CN110855996A (en) * 2019-09-30 2020-02-28 中国船舶重工集团公司第七0九研究所 Image coding and decoding and network transmission method and device based on FPGA
CN111669588A (en) * 2020-05-27 2020-09-15 西安电子科技大学 Ultra-high definition video compression coding and decoding method with ultra-low time delay

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070217699A1 (en) * 2006-03-17 2007-09-20 Irani Firdaus Mahiar Digital images
CN101360242A (en) * 2008-09-22 2009-02-04 西安电子科技大学 Tag-tree encoding method based on JPEG2000
CN102790882A (en) * 2012-07-25 2012-11-21 南京信息工程大学 Encoding method of remote sensing image
CN103118259A (en) * 2013-02-22 2013-05-22 南京信息工程大学 JPEG2000 image coding method
CN103152568A (en) * 2013-02-27 2013-06-12 西安理工大学 JPEG2000 (Joint Photographic Experts Group 2000) coding method for real time cutoff of code stream and implementation system thereof

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070217699A1 (en) * 2006-03-17 2007-09-20 Irani Firdaus Mahiar Digital images
CN101360242A (en) * 2008-09-22 2009-02-04 西安电子科技大学 Tag-tree encoding method based on JPEG2000
CN102790882A (en) * 2012-07-25 2012-11-21 南京信息工程大学 Encoding method of remote sensing image
CN103118259A (en) * 2013-02-22 2013-05-22 南京信息工程大学 JPEG2000 image coding method
CN103152568A (en) * 2013-02-27 2013-06-12 西安理工大学 JPEG2000 (Joint Photographic Experts Group 2000) coding method for real time cutoff of code stream and implementation system thereof

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110430430A (en) * 2019-08-08 2019-11-08 浙江大华技术股份有限公司 Image processing method and device, storage medium, electronic device
CN110430430B (en) * 2019-08-08 2021-07-06 浙江芯昇电子技术有限公司 Image processing method and device, storage medium and electronic device
CN110855996A (en) * 2019-09-30 2020-02-28 中国船舶重工集团公司第七0九研究所 Image coding and decoding and network transmission method and device based on FPGA
CN110855996B (en) * 2019-09-30 2021-10-22 中国船舶重工集团公司第七0九研究所 Image coding and decoding and network transmission method and device based on FPGA
CN111669588A (en) * 2020-05-27 2020-09-15 西安电子科技大学 Ultra-high definition video compression coding and decoding method with ultra-low time delay

Also Published As

Publication number Publication date
CN108540798B (en) 2020-06-05

Similar Documents

Publication Publication Date Title
CN101049006B (en) Image coding method and apparatus, and image decoding method and apparatus
CN105120293B (en) Image collaboration coding/decoding method and device based on CPU and GPU
CN108540798A (en) A kind of realization system and method for JPEG2000 images coding
CN1767655A (en) Multi view point video image parallax difference estimating method
CN108921910B (en) JPEG coding compressed image restoration method based on scalable convolutional neural network
CN104994382B (en) A kind of optimization method of perception rate distortion
CN102186076B (en) Image compression method and image compression device for real-time code rate pre-allocation
CN106385584B (en) The adaptively sampled coding method of distributed video compressed sensing based on spatial correlation
CN102281446B (en) Visual-perception-characteristic-based quantification method in distributed video coding
CN106993190B (en) Software-hardware synergism coding method and system
CN101917622B (en) 14bit wide image compression hardware coder
CN102036073B (en) Method for encoding and decoding JPEG2000 image based on vision potential attention target area
CN103118259B (en) JPEG2000 image coding method
CN106031168B (en) Self-adaptive processing with the video flowing for reducing color-resolution
CN109922339A (en) In conjunction with the image coding framework of multi-sampling rate down-sampling and super-resolution rebuilding technology
CN104780383B (en) A kind of 3D HEVC multi-resolution video coding methods
CN108810534A (en) Method for compressing image based on direction Lifting Wavelet and improved SPIHIT under Internet of Things
JP4111259B2 (en) Encoding apparatus, encoding method, software program, table data, and recording medium
CN103152568A (en) JPEG2000 (Joint Photographic Experts Group 2000) coding method for real time cutoff of code stream and implementation system thereof
CN106210727B (en) Video hierarchical code stream coding method and system based on neural network processor array
CN106303527B (en) Video hierarchical code stream coding method and system of time division multiplexing neural network processor
CN1267858C (en) JPEG 2000 speed control method through real time cut-off
CN102769744B (en) Satellite image encoding method
CN112468826B (en) VVC loop filtering method and system based on multilayer GAN
WO2022100173A1 (en) Video frame compression method and apparatus, and video frame decompression method and apparatus

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20220929

Address after: Building A5, CLP Software Park, No. 18, Jianshan Road, High-tech Zone, Changsha City, Hunan Province, 410205

Patentee after: HUNAN RONGCHUANG MICROELECTRONIC Co.,Ltd.

Address before: 710071 Taibai South Road, Yanta District, Xi'an, Shaanxi Province, No. 2

Patentee before: XIDIAN University

TR01 Transfer of patent right