200952461 九、發明說明： 【發明所屬之技術領域】 本發明侧於-種影像_處理 種可調節縣品質之小波轉換_處理裝置特収關於一 【先前技術】 〇200952461 IX. Description of the invention: [Technical field of the invention] The present invention is directed to an image-processing type of wavelet-transformation that can adjust the quality of a county.
=曰的網路環境中，多嫖體的應用已經扮演了一個 重要的角色，對於影像_編啦的要求變得越來越嚴苛 更仍必須維持還原後的影像品質， 更進-步要絲像編瑪H能夠找多種特性，例如，還原 品質的控制，還原解析度的控制，具有感興趣區域卿In the network environment of 曰, the application of multi-corps has played an important role. The requirements for image _ editing become more and more severe, and the image quality after restoration must be maintained. Silk like Ma Ma can find a variety of characteristics, such as the control of the reduction quality, the control of the reduction resolution, with the region of interest
Regum Of barest)的檢索能力等等...。為了因應這些要求 ，近年來出現了許多不同的影像壓縮編碼標準。 由於-些吸引人的雜，候職已經被證明對於在 影像及視訊的編碼上是非常有用的，因此近年來所提出的 影像編碼器多是基於小波轉換。其中知名的編碼標準，如 EZWXEmbedded ZeiOtree Wavelets)是使用樹狀結構， SPIHT(Set Partitioned in Hierarchical Trees)是使用串列結 構 ’ SPECK(Set Partitioned Embedded Block Coder)則同時 使用了零區塊和串列結構，而零區塊與陣列之嵌入式小波 編碼（Embedded Wavelet Image Coding Based on Zero-Blocks and Array，EZBA)和改良型零區塊與陣列之缺 入式小波編碼（Improved Embedded Wavelet Image Coding Based on Zero-Blocks and Array, I-EZBA)使用零區塊和陣 列結構。 200952461Regum Of barest), etc. In response to these requirements, many different image compression coding standards have emerged in recent years. Due to the attractiveness of the miscellaneous, the waiting job has proven to be very useful for the encoding of video and video. Therefore, the image encoders proposed in recent years are mostly based on wavelet transform. Among them, the well-known coding standards, such as EZWXEmbedded ZeiOtree Wavelets, use a tree structure. SPIHT (Set Partitioned in Hierarchical Trees) uses a serial structure 'SPECK (Set Partitioned Embedded Block Coder), which uses zero block and serial structure. Embedded Wavelet Image Coding Based on Zero-Blocks and Array (EZBA) and Improved Zero-Block and Coordinate Based Coding Based on Zero (Improved Embedded Wavelet Image Coding Based on Zero) -Blocks and Array, I-EZBA) uses zero block and array structure. 200952461
上述所提到的編碼標準只支援失真度可調整，並不支 援解析度可調整以及RGI選取能力。為了解決這個問題， GuiXie提出了 S_SPECK編碼標準，它是由spE(：K所擴展 而來，使得編碼器能同時支援失真度可調整、解析度可 整及ROI選取能力等三項特性，並且不會增加太多^算複 雜度。然而，該S-SPECK編碼的壓縮核心所產生出來的編 碼位元較為分散，不一定會將屬於相同的單位代碼 (codeunit ’或是sub-band)的係數所編出來的碼放置在一起 ，又為了取得解析度可調整性必須將屬於相同單位代碼的 編碼位元放置在-起’因此需制用記㈣先將這些分散 的編碼位元收集在一起，之後再填入品質層中。因^，在 形成品質層的階段需要使用大量的記憶體空間，並且 碼串流中使用不少額外的記錄位元。 【發明内容】 本發明之主要目的係提供一種可調節影像品質之 波轉換編碼處縣置，其係可難失真度、娜解析 ^少-感興趣區域進行選取，以便大幅降低在編^序 中之一品質層形成階段所需的大量記鋪m以及在一 ，碼串流中使用較少量的額外位元，使得本發明具: 成本及運算複雜度之功效。 - 本發明之次要目的係提供一種可調節影像品 波轉換編碼處理裝置，其係可使一壓縮 、 瑀作蜚好口新％碼碼單兀•之壓縮編 碼作業及-品質層控制器之輸出作業同步 明具有提冑、_|鱗之姐。 Μ本發 200952461 根據本發明之可調節影像品質之小波轉換編碼處理 裝置’其包含—離散小波轉換單元、-小波係數分群單元 、-壓縮編解元及—品質層㈣器。該離散小波轉換單 ' %將—輸人影像進行二維小波機，並對應該輸入影像大 小產生一小波係數矩陣；該小波係數分群單元用以接收該 •小波係數矩陣，並將該小波係數矩陣内的各餘對應數個 感興趣區塊(ROI)進行係數重組；該壓縮編碼單元接收該重 ❹ 、组後的係數’並利用一零區塊編碼技術進行編碼；該品質 層控制器將塵縮編碼後的資料依據數個預設的位元預算及 數個解析度層級填入數個品質層格式。 【實施方式】 為讓本發明之上述及其他目的、特徵及伽能更明顯 易懂，下文特舉本發明之較佳實施例，並配合所附圖式， 作詳細說明如下： . 5月參照第1圖所示，本發明較佳實施例之可調節影像 ❹ 品質之小波轉換編碼處理裝置1，其係為一可調整感興趣 區域影像小波編碼器(Scalable R〇I Image Wavelet c〇dec， S-EZBA)，利用該編碼器之編碼格式以進行影像編碼。該 小波轉換編碼處理裝置i進一步包含一離散小波轉換單元 . 11、一小波係數分群單元12、一壓縮編碼單元13及一品 質層控制器14。 如第1圖所示，該小波轉換編碼處理裝置丨之離散小 波轉換單元11用以接收一欲壓縮的影像，以進行二維小波 轉換產生一與原影像大小相同的小波係數矩陣；該小波 ~ 8—— 200952461 係數分群單元12用以接收該小波係數矩陣，經過一係數分 群子單元121及一 ROI區塊分群子單元122的重新編排和 集合’以進行該小波係數矩陣之係數重組，使該編碼系統 1取得感興趣區域(Region of Interest, ROI)之檢索能力及解 析度的可調整性’其中該係數分群子單元121用以將該小 波係數矩陣之係數分群，以形成數個r〇I區塊，而該R〇I 區塊分群子單元122則將該數個ROI區塊區分成數個單位 代碼(Codeunit);該壓縮編碼單元13可接收該重組後的係 數’並利用I-EZBA編碼技術進行編碼；該品質層控制器 Η可依照使用者的需求，在設定不同的位元預算以及解析 度層級後’將壓縮編碼後的位元填入該S-EZBA之編碼系 統1所制定的格式’形成能符合使用者要求的位元串流， 並藉由一解碼裝置’使該影像還原時選擇所要還原的影像 品質和解析度。 上述I-EZBA編碼利用直接萃取特徵值的方式完成影 像之特徵值萃取處理，相較於習用該EZBA編碼需建立係 數對應圖的編碼方式，該LEZBA編碼因不需要重要係數 對應圖的建立，而可達成減少記憶體使用量的目的。 為了取得ROI檢索能力，可以將該欲壓縮的影像所有 的小波係數重組成一空間導向樹，並對該空間導向樹獨立 的進行編碼。如第2圖所示，其揭示該空間導向樹呈現出 在不同次頻帶 LL3、LH3、HL3、HH3、LH2、HL2、HH2 、LH1、HL1及HH1的父子係數關係，所有的係數被組織 成樹狀結構，並且樹根位在最低頻次頻帶LL3中；第2圖 200952461 之各次頻帶中的係數在空間上與該原始的影像中的某一個 區域係相互對應，即一個空間導向樹的父子係數，在其所 ' 屬的次頻帶中，會對應到該原始影像中的相同位置。因此 . ，該原始影像中一個方形區域可以由單獨一個空間導向樹 的係數進行重建。其中若第2圖中最低頻次頻帶LL3的大 小為4*4，即16個係數，因此共有16個空間導向樹，可 以用來重建出16個對應的R〇i影像。 ❹ 該原始影像經過小波轉換可分解出不同的次頻帶，κ 階段的分解可以產生3Κ+1個次頻帶及K+1層解析度層級 。請參照第3圖所示，其揭示一個經過三階段(κ=3)小波轉 換的次頻帶結構及四層解析度層級R〇、R1、幻及R3的 關係，其中最低的解析度層級R0由該最低頻的次頻帶LL3 所組成；次低的解析度層級R1則是再額外包含三個高頻 次頻帶HL3、LH3及HH3，接下來的解析度層級，則以相 似的方法，增加組成的次頻帶。當需要某一個解析度的影 ❾ 像時，只要將不包含在此解析度内的次頻帶排除，以剩下 的次頻帶進行還原即可。例如，當需要該解析度時， 可以將該三個次頻帶HU、HH1及LH1排除，而不對該三 個次頻帶HL1、HH1及LH1中的係數進行編碼。因此^ 了取得騎度的可罐性，敎鮮内則、波倾依照解 才斤度層級分成不同的子集合，並對該子集合獨立進行壓縮 編碼。 综上所述，本發明較佳實施例之小波係數分群單元12 為了達成同時ROI檢索能力和解析度可調整性，其必須先 200952461 將該小波係數矩陣之係數組織成為該數個空間導 後進一步將空間導向樹中的係數依照所屬的“度 將相同解析度層級的係數集合起來以形成如第4 ^、:’ 數個單位代碼(Codeunit)CO、Cl、C2，簡言之，八之 ❹ 〇 建能力係可藉由該小波係數矩陣解析出的數個空間導=重 所獲得，然後再由該空間導向樹區分出該數個單^代碼^ Cl、C2 ’以便各該r〇i獲得解析度的可調整性。其中該 些單位代碼CG、a、C2是壓賴碼的基本單位。如第^ 圖所示，因為要進行該ΜΖΒΑ之零區塊壓縮編瑪作業的 最小單位至少要是一個2x2的區塊，因此c〇至少也要是一 個2x2的區塊，所以縱使在3階段的小波轉換可以區= 4個解析度層級的次頻帶，不過為了屈就零區塊壓縮編碼 j最小單位，所以將位在第一層解析度R〇〔即最低頻次頻 ▼〕，跟第二層解析度R1的3個係數共同組成該單位代碼 C0，因此在空間導向樹中只分出3個解析度層級，所以該 單位代碼co是由空間導向樹中，位在次頻帶LL3和hl3 、LH3、HH3中的係數所組成，該單位代碼C1由空間導 向樹中的次頻帶HL2、LH2、HH2所組成，該單位代碼C2 由空間導向樹中的次頻帶HL1、LH1、HH1所組成。 在元成該小波係數矩陣之係數重組後，即可利用該壓 縮編碼單元13對其進行後續的壓縮編碼作業。請參照第5 圖所示，在該壓縮編碼單元13中，本發明對區塊中 的係數之壓縮編碼作業係利用Z字型的編碼順序，即對該 單位代碼由低解析度到高解析度順序編碼，而其編碼位元 200952461 的輸出如所示’由高位元平㈣錄元平面，在每 個位元平面中由該低解析度單位代碼⑶到高解析度單位 代碼C2 ’如此’可方便解碼時能夠輕易地分辨出屬於不同 解析度層級的編碼位70。由於本發明_用χ棚a進行編 碼作業，因此相較於習用編碼技術利用係數與門捏值對應 錄的結果’具有降佩娜算的複雜度，並節省在比較過程 中大量記憶體之使用。 ❹ ❹ 請參照第7a及7b圖所示，其揭示本發明之小波轉換 編碼處理裝置1所制定的品質層格式2，當各該r〇i區塊 進行壓縮㈣之後，可將錄㈣編敬元分配到該格式 2中。其中該品質層格式2設有數個品質層，以本發明較 佳實施例為例，該品質層格式2包含一第一品質層Q〇、一 第二品質$ Q卜一第三品質層q2、一第四品質層Q3及 一第五品質層Q4，其用以使該R0I區塊進行重建時能夠 藉由該品質層Q0至Q4之選擇以進行解碼，進而可以選擇 該ROI區塊影像品質的重建。如第8圖所示，該數個品質 層Q0至Q4可以位元率來作為分界，例如當需要〇 25bpp (bit per pixel，每像素位元數)的品質時，即可選擇該第一 品質層Q0和第二品質層Q1進行解碼。 請再參照第7a圖所示’在該品質層格式2之第一品質 層Q0中包含一額外位元（overhead bit)及一交錯編瑪位元 (interleaved bit)23兩個部份，在該第一品質層Q〇的額外位 元中進一步設有一第一欄位21及一第二欄位22，該第一 欄位用以紀錄品質層的長度，該第二欄位22用以紀錄該 ——12 — 200952461 ROI區塊所能提供的解析度層級的數量，此外，如第几圖 所示’包含該第二品質層Q1之後的所有品質層Q2至q4 之額外位元僅設有該第—攔位21及該交錯編碼位元。 該交錯編碼位元23可供該壓縮編碼單元13之輸出直接填 入此部份’直到該品質層的長度達到該第—攔位21所記錄 的長度為止，在進行下—個品f層的位元分配時同樣將 該第-攔位21先輸出’然後延續上—個 位元23,將後續的編瑪位元填入該品質層二= 23部份，直到位元預算到達為止。 請參照第9圖所示，其揭示該小波轉換編碼處理裝置 1之壓縮編碼單元13的電路方麵，其巾該電路方塊圖係 適合用以處理8*8像素大小的R〇I區塊。該壓縮編碼單元 13的電路方塊圖係包含一單位代碼讀取器131、一係數處 理模組=2、- 16位元暫存器133、一符號記憶元件134 、一決策編碼器135、一係數/符號合成模組136、一符號 旗標137及一門檻產生器138。 該單位代碼讀取器131讀取存放在一外部記憶體的 ROI區塊的係數；該單位代碼讀取器131將該係數送至該 係數處理模組132，該係數處理模組132將該係數的符號 位元擷取出來並存入該符號記憶元件134，藉由該係數處 理模組132擷取出二進制位元平面值，並輸入該16位元暫 存器133，接著由該決策編碼器135讀取係數的二進制值 進行重要係數或區塊的判斷並輸出第一階段的編碼位元， 再由該係數/符號合成模組136將編碼位元和相對應的符號 —13 — 200952461 位二，併，送出第二階段的編碼位元，即壓縮編竭的 二f該品，㈣11 14依照所要求的位元預算和C 又日級，將編碼位元填入到各該品質層中，並立即 3疋輪出直齡元預算到達為止，#該品質層控制器; 單位代碼之闕位元的輸出，會發出—訊號通知 該單位代碼讀取II 131，以讀取τ—解位代碼的係數。 如上所述，本發明之小波轉換編碼處理裝置i在分配The coding standards mentioned above only support distortion adjustment, and do not support resolution adjustment and RGI selection. In order to solve this problem, GuiXie proposed the S_SPECK coding standard, which is extended by spE(:K, so that the encoder can support three characteristics of distortion adjustment, resolution and ROI selection, and not It will increase too much computational complexity. However, the coding bits generated by the S-SPECK coded compression kernel are more scattered, and the coefficients belonging to the same unit code (codeunit ' or sub-band) are not necessarily included. The coded codes are placed together, and in order to obtain the resolution adjustability, the coded bits belonging to the same unit code must be placed at the beginning. Therefore, the coded bits are collected (4), and then the scattered coded bits are collected together. Refilling the quality layer. Because of the need to use a large amount of memory space in the stage of forming the quality layer, a large number of additional recording bits are used in the code stream. SUMMARY OF THE INVENTION The main object of the present invention is to provide a The wave-converting code of the image quality can be adjusted at the county level, and the system can be difficult to be distorted, and the narrative is less - the region of interest is selected to greatly reduce one of the sequences. The large number of tiles required for the quality layer formation phase and the use of a smaller number of extra bits in the code stream result in the invention having the cost and computational complexity. - The secondary objective of the present invention is to provide An adjustable image wave conversion coding processing device, which can improve the compression coding operation of a compression, 瑀 新 新 % % - - - - - - - - - - - - - - 品质 品质 品质 品质 品质The sister of the scale. Μ本发200952461 The wavelet transform coding processing device of the adjustable image quality according to the present invention includes a discrete wavelet transform unit, a wavelet coefficient grouping unit, a compression coding element and a quality layer (four). The discrete wavelet transform single '%--the input image is subjected to a two-dimensional wavelet machine, and a wavelet coefficient matrix is generated for the input image size; the wavelet coefficient grouping unit is configured to receive the wavelet coefficient matrix, and the wavelet coefficient matrix Each of the plurality of corresponding blocks of interest (ROI) is subjected to coefficient recombination; the compression coding unit receives the coefficient ' after the group, and uses a zero block coding technique Line coding; the quality layer controller fills the dust-coded data into a plurality of quality layer formats according to a plurality of preset bit budgets and a plurality of resolution levels. [Embodiment] To make the above and other aspects of the present invention The objects, features, and gamma energy are more apparent and understood. The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. FIG. Adjustable image 品质 quality wavelet transform coding processing device 1 is an adjustable region of interest image wavelet encoder (Scalable R〇I Image Wavelet c〇dec, S-EZBA), using the encoder encoding format For performing image coding, the wavelet transform coding processing apparatus i further includes a discrete wavelet transform unit. 11. A wavelet coefficient grouping unit 12, a compression coding unit 13, and a quality layer controller 14. As shown in FIG. 1, the wavelet transform unit 11 of the wavelet transform coding processing device is configured to receive a compressed image to perform a two-dimensional wavelet transform to generate a wavelet coefficient matrix of the same size as the original image; 8 - 200952461 The coefficient grouping unit 12 is configured to receive the wavelet coefficient matrix, and perform a re-arrangement and set of a coefficient grouping sub-unit 121 and a ROI block grouping sub-unit 122 to perform coefficient recombination of the wavelet coefficient matrix. The encoding system 1 obtains the retrievability of the region of interest (ROI) and the adjustability of the resolution. The coefficient sub-unit 121 is used to group the coefficients of the wavelet coefficient matrix to form a plurality of r〇I. a block, and the R〇I block sub-unit 122 divides the plurality of ROI blocks into a plurality of code units; the compression coding unit 13 can receive the recombined coefficients and utilize I-EZBA coding. The technology is coded; the quality layer controller can fill in the compressed coded bits after setting different bit budgets and resolution levels according to the user's needs. The S-EZBA based coding system format of an established 'formed bitstreams can meet the requirements of the user, and by a decoding means' the selection of the resolution and image quality when the image is to be restored restored. The above I-EZBA coding uses the method of directly extracting the feature values to complete the feature value extraction processing of the image, and the coding mode of the coefficient correspondence map needs to be established compared to the conventional EZBA coding, and the LEZBA coding does not require the establishment of the correspondence map of the important coefficients. The purpose of reducing the amount of memory used can be achieved. In order to obtain the ROI search capability, all wavelet coefficients of the image to be compressed can be reconstructed into a spatial steering tree, and the spatial steering tree is independently encoded. As shown in Fig. 2, it is revealed that the spatially oriented tree exhibits a parent-child coefficient relationship in different sub-bands LL3, LH3, HL3, HH3, LH2, HL2, HH2, LH1, HL1, and HH1, and all coefficients are organized into a tree shape. Structure, and the root of the tree is in the lowest frequency band LL3; the coefficients in each frequency band of the second picture 200952461 spatially correspond to one of the original images, that is, the parent-child coefficient of a spatially oriented tree, In the sub-band to which it belongs, it corresponds to the same position in the original image. Therefore, a square area in the original image can be reconstructed from the coefficients of a single spatially oriented tree. If the size of the lowest frequency band LL3 in Fig. 2 is 4*4, that is, 16 coefficients, there are 16 spatial steering trees, which can be used to reconstruct 16 corresponding R〇i images. ❹ The original image is decomposed into different sub-bands by wavelet transform, and the decomposition of the κ phase can generate 3Κ+1 sub-bands and K+1 layer resolution levels. Referring to FIG. 3, it discloses a sub-band structure after three-stage (κ=3) wavelet transform and a relationship of four layers of resolution levels R〇, R1, illusion and R3, wherein the lowest resolution level R0 is The lowest frequency sub-band LL3 is composed; the second lowest resolution level R1 is additionally including three high-frequency sub-bands HL3, LH3, and HH3, and the next resolution level is increased in a similar manner. Sub-band. When a resolution image of a certain resolution is required, the subband not included in the resolution is excluded, and the restoration is performed in the remaining subband. For example, when the resolution is required, the three sub-bands HU, HH1, and LH1 may be excluded without encoding the coefficients in the three sub-bands HL1, HH1, and LH1. Therefore, the canability of the riding degree is obtained, and the wave is divided into different sub-sets according to the level of the solution, and the sub-set is independently compressed and encoded. In summary, in order to achieve simultaneous ROI retrieval capability and resolution adjustability, the wavelet coefficient grouping unit 12 of the preferred embodiment of the present invention must first organize the coefficients of the wavelet coefficient matrix into the plurality of spatial guides after further 200952461. The coefficients in the spatial steering tree are grouped according to the degree of the same degree of resolution to form the 4th, :' number of unit codes (Codeunit) CO, Cl, C2, in short, eight of them The built-in capability can be obtained by a plurality of spatial derivatives=resolved by the wavelet coefficient matrix, and then the spatial guide tree distinguishes the plurality of single codes ^Cl, C2' so that each r〇i is obtained. The adjustability of the resolution, wherein the unit codes CG, a, and C2 are basic units of the squaring code. As shown in the figure, the minimum unit for performing the zero block compression marshalling operation of the ΜΖΒΑ is at least A 2x2 block, so c〇 is at least a 2x2 block, so even if the wavelet transform in the 3 stage can be the sub-band of the 4 resolution level, the zero-block compression coding j is the smallest. Therefore, the first layer resolution R〇 (ie, the lowest frequency frequency ▼) is combined with the three coefficients of the second layer resolution R1 to form the unit code C0, so only three parsings are separated in the spatial steering tree. Degree level, so the unit code co is composed of the coefficients in the sub-band LL3 and hl3, LH3, HH3 in the space-oriented tree, and the unit code C1 is represented by the sub-bands HL2, LH2, HH2 in the space-oriented tree. The unit code C2 is composed of the sub-bands HL1, LH1, and HH1 in the space-oriented tree. After the coefficients of the matrix of the wavelet coefficients are reorganized, the compression coding unit 13 can be used to perform subsequent compression and coding operations. Referring to FIG. 5, in the compression coding unit 13, the compression coding operation of the coefficients in the block of the present invention utilizes a Z-type coding order, that is, the low resolution to high resolution of the unit code. Degree sequential encoding, and the output of its encoding bit 200952461 is as shown by 'high-order flat (four) recording element plane, in this bit-plane from the low-resolution unit code (3) to the high-resolution unit code C2 'so' Square At the time of decoding, the coded bits 70 belonging to different resolution levels can be easily distinguished. Since the present invention _ uses the shed a for the coding operation, the result of the correspondence between the coefficient and the gate pinch is compared with the conventional coding technique. The complexity of the calculation, and the use of a large amount of memory during the comparison process. ❹ ❹ Please refer to Figures 7a and 7b, which reveals the quality layer format 2 of the wavelet transform coding processing apparatus 1 of the present invention. After each of the r〇i blocks is compressed (4), the recorded (4) edited elements can be assigned to the format 2. The quality layer format 2 is provided with a plurality of quality layers, which is exemplified by a preferred embodiment of the present invention. The layer format 2 includes a first quality layer Q〇, a second quality $Qb, a third quality layer q2, a fourth quality layer Q3, and a fifth quality layer Q4 for reconstructing the R0I block. The quality layer Q0 to Q4 can be selected for decoding, and the reconstruction of the image quality of the ROI block can be selected. As shown in FIG. 8, the plurality of quality layers Q0 to Q4 can be used as a boundary by a bit rate. For example, when a quality of 25 bpp (bit per pixel) is required, the first quality can be selected. Layer Q0 and second quality layer Q1 are decoded. Referring to FIG. 7a, the first quality layer Q0 of the quality layer format 2 includes an extra bit and an interleaved bit 23. Further, a first field 21 and a second field 22 are further disposed in the extra bit of the first quality layer Q, the first field is used to record the length of the quality layer, and the second field 22 is used to record the ——12 — 200952461 The number of resolution levels that can be provided by the ROI block. In addition, as shown in the figure, the extra bits of all quality layers Q2 to q4 after the second quality layer Q1 are only provided. First - the intercept 21 and the interleaved coded bit. The interleaved coded bit 23 can be used to directly fill the output of the compression coding unit 13 until the length of the quality layer reaches the length recorded by the first block 21, and the next layer f When the bit is allocated, the first block 21 is also outputted first and then the last bit 23 is continued, and the subsequent numerator bit is filled into the quality layer 2 = 23 until the bit budget reaches. Referring to FIG. 9, the circuit aspect of the compression coding unit 13 of the wavelet transform coding processing apparatus 1 is disclosed. The circuit block diagram is suitable for processing an R〇I block of 8*8 pixel size. The circuit block diagram of the compression coding unit 13 includes a unit code reader 131, a coefficient processing module=2, a 16-bit temporary register 133, a symbol memory element 134, a decision encoder 135, and a coefficient. / symbol synthesis module 136, a symbol flag 137 and a threshold generator 138. The unit code reader 131 reads the coefficient of the ROI block stored in an external memory; the unit code reader 131 sends the coefficient to the coefficient processing module 132, and the coefficient processing module 132 sets the coefficient. The symbol bit is retrieved and stored in the symbol memory component 134. The coefficient processing module 132 extracts the binary bit plane value and inputs the 16-bit scratchpad 133, and then the decision encoder 135. Reading the binary value of the coefficient to perform the judgment of the important coefficient or the block and outputting the coding bit of the first stage, and then the coefficient/symbol synthesis module 136 sets the coding bit and the corresponding symbol -13 - 200952461 And sending the coded bit of the second stage, that is, compressing and compiling the product, and (4) 11 14 filling the coded bit into each quality layer according to the required bit budget and C and the date level, and Immediately after the 3rd round of the straight age budget arrives, #the quality layer controller; the output of the unit code of the unit code will be sent - the signal informs the unit code to read II 131 to read the coefficient of the τ-displacement code. As described above, the wavelet transform coding processing apparatus i of the present invention is allocating
Q 編碼位元職品質層時，可以讓壓縮編碼和品質層之輸出 的作業同步進行，不需要等到該品質層的位元預算到達才 進行填入的動作，並且不需要使用記憶體空間將屬於相同 早位代碼之位置的編瑪位元收集在一起，以減少記憶體的 ，用’進而使本發明具有降低成本、降低運算複雜度及提 南編碼效率之功效。 雖然本發明已利用上述較佳實施例揭示，然其並非用 以限定本發明，任何熟習此技藝者在不脫離本發i之精神 和範圍之内’相對上述實施例進行各種更動與修改仍屬本 發明所保護之技術料，因此本發明之保護麵當.視後附 之申請專利範圍所界定者為準。 200952461 【圖式簡單說明】 第1圖：本發明較佳實施例之可調節影像品質之小波轉 換編碼處理裝置之編碼流程示意圖。 第2圖：本發明較佳實施例之可調節影像品質之小波轉 換編碼處理裝置之空間導向樹示意圖。 第3圖：本發明較佳實施例之可調節影像品質之小波轉 換編碼處理裝置之解析度層級示意圖。 第4圖：本發明較佳實施例之可調節影像品質之小波轉 換編碼處理裝置之各單位代碼之組成示意圖。 第5圖：本發明較佳實施例之可調節影像品質之小波轉 換編碼處理裝置之Z字型的編碼示意圖。 第6圖：本發明較佳實施例之可調節影像品質之小波轉 換編碼處理裝置之編碼位元輸出順序之示意圖。 第7a圖：本發明較佳實施例之可調節影像品質之小波 轉換編碼處理裝置之品質層格式示意圖。 第7b圖：本發明較佳實施例之可調節影像品質之小波 轉換編碼處理裝置之品質層格式示意圖。 第8圖：本發明較佳實施例之可調節影像品質之小波轉 換編碼處理裝置以位元率作為數個品質層分界之示意圖。 第9圖：本發明較佳實施例之可調節影像品質之小波轉 換編碼處理裝置之壓縮編碼單元的電路方塊圖。 【主要元件符號說明】 1 小波轉換編碼處理裝置11 離散小波轉換單元 —15 — 200952461 12 小波係數分群單元 122 ROI區塊分群子單元 131單位代碼讀取器 133 16位元暫存器 135決策編碼器 137符號旗標 14 品質層控制器 21 第一欄位 23 交錯編碼位元 121係數分群子單元 13 壓縮編碼單元 132係數處理模組 134符號記憶元件 136係數/符號合成模組 138門檻產生器 2 編碼格式 22 第二欄位When the Q-coded bit quality layer is used, the operations of the compression coding and the output of the quality layer can be synchronized, and the operation of filling in the bit-level budget of the quality layer is not required, and the memory space is not required to be used. The numerator bits of the position of the same early code are collected together to reduce the memory, and the invention has the effect of reducing cost, reducing computational complexity and improving coding efficiency. The present invention has been disclosed in the above-described preferred embodiments, and it is not intended to limit the scope of the present invention. The technical material protected by the present invention is therefore defined by the scope of the appended claims. 200952461 [Simplified description of the drawings] Fig. 1 is a diagram showing the coding flow of the wavelet transform coding processing apparatus capable of adjusting image quality according to a preferred embodiment of the present invention. Fig. 2 is a schematic diagram showing a spatial steering tree of a wavelet transform coding processing apparatus capable of adjusting image quality according to a preferred embodiment of the present invention. Figure 3 is a diagram showing the resolution level of a wavelet transform coding processing apparatus capable of adjusting image quality in accordance with a preferred embodiment of the present invention. Fig. 4 is a view showing the composition of each unit code of the wavelet transform coding processing apparatus capable of adjusting image quality according to a preferred embodiment of the present invention. Fig. 5 is a diagram showing the coding of a zigzag type of a wavelet transform coding processing apparatus capable of adjusting image quality according to a preferred embodiment of the present invention. Figure 6 is a diagram showing the output order of the coding bits of the wavelet transform coding processing apparatus of the image quality control according to the preferred embodiment of the present invention. Fig. 7a is a diagram showing the quality layer format of the wavelet transform coding processing apparatus capable of adjusting image quality according to a preferred embodiment of the present invention. Figure 7b is a diagram showing the quality layer format of the wavelet transform coding processing apparatus capable of adjusting image quality according to a preferred embodiment of the present invention. Figure 8 is a diagram showing the wavelet transform coding processing apparatus for adjusting image quality according to a preferred embodiment of the present invention. The bit rate is used as a schematic diagram of a plurality of quality layer boundaries. Figure 9 is a circuit block diagram of a compression coding unit of a wavelet transform coding processing apparatus capable of adjusting image quality in accordance with a preferred embodiment of the present invention. [Major component symbol description] 1 Wavelet transform coding processing device 11 Discrete wavelet transform unit - 15 - 200952461 12 Wavelet coefficient grouping unit 122 ROI block grouping subunit 131 unit code reader 133 16 bit register 135 decision encoder 137 symbol flag 14 quality layer controller 21 first field 23 interlaced coded bit 121 coefficient grouping subunit 13 compression coding unit 132 coefficient processing module 134 symbol memory element 136 coefficient / symbol synthesis module 138 threshold generator 2 coding Format 22 second field