200841309 九、發明說明： 【發明所屬之技術領域】 本發明係有關一種液晶顯示器之過驅動（〇verd「|ve)技術，特別 是關於-種可根據_影像馨輸出模式之獅性過购方法及其裝 【先前技術】200841309 IX. Description of the invention: [Technical field of the invention] The present invention relates to an overdrive (〇verd "|ve) technology of a liquid crystal display, and more particularly to a lion purchase method according to the image output mode And its predecessor
在液晶顯示ϋ巾，主要制職晶材質本身的雛，來達到顯示 目的。然，液晶反應時間如果過長影像就會料出現邊緣模糊 的現象，為杜祉現象通常會採祕_方法，轉締晶由目_ 不畫面之晝素灰階切換到下-觸示4面之晝素灰階所需的反應時 間0 —過驅動方法是再施加電壓過程中，將—個晝素之影像電壓設定 比實際高些，之後在鱗原本電壓，以_4素的大電壓產生較 大的力矩㈣·晶在較_時間喊變排顺換賴設定的 率，以藉此加速整個液晶的反應時間。如第彳_示，f知過驅 疋需要有-緩衝儲存單it 1G，用以儲存前—圖框（frame)的气號， 其儲存容量大且需要-直以目前所要顯示的灰階進行更新，所以: 使用靜態隨機存取記舰（SRAM)，以先進先出方式運作； 二 過驅動對照單元12制來儲存-過驅動對照表（〇D |QQk_叩伯 14，^設輸入影像為目前圖框N，則自緩衝儲存單元1〇傳輸出去的^ 會是前-圖框N-1，目前圖框N與前—圖框⑹所顯示的灰階^ 至過驅動對照單元12，以對照過驅動對照表14產生目前圖框二 過驅動訊號（灰階值）至資料驅動系統。其中，緩衝儲存單元1〇 = -整個圖框的晝素資料，因此所需用到記憶體容量相當的大。 麟’若細全灰階增賴式過驅鱗會導致_影像之移動物 1緣產生非職之腕出現，在日ff彡像巾可能會產生非預期之曰 色邊緣是常有的’例如夜晚時人臉邊緣料出現藍紫色邊緣，此乃: 6 200841309 混色時，如果全部晝面都使用加強過驅動方式，則常料致晝面的混 色比例出現問題，如第2 _示。因此’若能有條件的分離不备產生 比例問題的混色’便能進行有效的邊緣加強而不會產生非預期的色彩 偏移。 有鑑於此.，本發明係在提出一種選擇性過驅動方法及其裳置，以 改善存在於先前技術中之該等缺失。 【發明内容】 本發明之主要目_在提供—種選擇性過_方法及其裳置，直 係在過驅動對照單元中同時存放標準模式與增強模式之對照表，以根 據則後畫面之4素賴鮮模式或增賴式，財選擇性的進行有效 之動態影像邊緣加強而不會產生非預期的色彩偏移。 /本發明之另-目的係在提供—種聰性過驅動方法及其裝置，其 係在輸入顯示畫面至緩衝儲存單元或過驅動對照單元之前，先對顯示 晝面中之晝素灰階值進行降階，以輪出較少之位元數，進而使儲存於 緩衝儲存單兀之資料量減少，故可有效減少記憶體儲存容量。 —本發明係在提出-種選擇性過驅動方法，其係根據目前畫面中對 應每-畫素之畫素灰階值，得到—過驅動灰階值。此方法係先判斷目 ⑩前畫面之畫素之畫素灰階值以及前—晝面畫素之晝素灰階值是否 •為類灰色或類純色而需要進行增強模式過驅動，並據此產生一選擇訊 號；再·該目前晝面中之畫素之晝素灰階值以及前_晝面對應畫素 之晝素灰階值，並同時根據選擇訊號選擇性查閱至少一標準模式過驅 動對照表或加強模式過驅動對照表，以輸出對應該目前晝面之過 灰階值。 一 >此外丄本發明另提出一種選擇性過驅動裝置，其係包含一降階單 疋’對目面之晝素的晝素灰階值減少位元數；—緩衝儲存單元係 用來儲存前-畫面之晝素灰階值，並輕接一判斷單元，其係判斷每— 旦面之畫素的畫素灰階值是否為類灰色或類純色而需要進行畫素加 7 200841309 強 祕單元’減_齡單福觸單元，錄據判斷單 =之結果輸岐否要進行過軸之選擇訊號；以及—過驅動對昭 早兀則根據此誠性錢標轉式過 料 過驅動·表，以輸出對應該目前晝蚊過轉灰階值^強核式 料發明更提出—種驅動方法，用以驅動—畫素，此方法包 括先，收旦素之刖-晝面灰階值與目前晝面灰階值；再根據接收到之 二個畫面灰階值判斷其是否均為類純色與類灰色其中之… 以驅 到二判斷值；最後，根據此判斷值輸出—對應之過驅動灰階值于 動前述畫素。In the liquid crystal display wipes, the main body of the crystal material itself, to achieve the purpose of display. However, if the liquid crystal reaction time is too long, the image will appear to have edge blurring phenomenon. For the cuckoo phenomenon, the secret method will usually be adopted. The method of transferring the crystal from the target _ does not change the gray scale of the picture to the lower-touch 4 side. The reaction time required for the gray scale of the halogen is 0. The overdrive method is to set the image voltage of a halogen element higher than the actual value during the voltage application, and then generate the voltage at the original scale of the scale, and generate a large voltage of _4 prime. Larger torque (4)·The crystal is changed in the _ time to change the set rate, thereby accelerating the reaction time of the entire liquid crystal. As shown in the first _ _, f know that the drive needs to have a - buffer storage list it 1G, used to store the front - frame (frame) gas number, its storage capacity is large and needs - straight to the gray scale currently displayed Update, so: use static random access ship (SRAM) to operate in first-in-first-out mode; two-drive control unit 12 to store-overdrive comparison table (〇D |QQk_叩伯14,^ set input image For the current frame N, the self-buffering storage unit 1〇 is transmitted to the front-frame N-1, and the current frame N and the front-frame (6) display the gray level ^ to the overdrive comparison unit 12, The current overdrive signal (grayscale value) is generated by the comparison overdrive table 14 to the data drive system, wherein the buffer storage unit 1 - = - the entire frame of the data, so the memory capacity is required It is quite big. If the lining of the whole gray-scale grading over the scale will cause the _image of the moving object to produce a non-professional wrist, the FF 彡 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能 可能Some of them, for example, at night, the edges of the face appear blue-violet edges, this is: 6 200841309 color mixing At the same time, if all the kneading surfaces are used in the enhanced overdrive mode, it is often caused by the problem of the color mixing ratio of the kneading surface, as shown in the second figure. Therefore, 'if the conditional separation is not prepared for the color mixture of the proportional problem' Effective edge enhancement without unintended color shifting. In view of this, the present invention is directed to a selective overdrive method and its placement to improve such defects present in the prior art. The main purpose of the present invention is to provide a selective over-method and its skirting, and directly store the comparison table of the standard mode and the enhanced mode in the overdrive control unit, according to the 4th of the subsequent picture. The fresh mode or the add-on type, the effective dynamic image edge enhancement is performed without causing unintended color shift. / The other object of the present invention is to provide a smart overdrive method and device thereof. Before the input display screen is sent to the buffer storage unit or the overdrive control unit, the pixel gray scale value in the display plane is first reduced to rotate a smaller number of bits, thereby storing the The amount of data stored in the rushed storage unit is reduced, so that the memory storage capacity can be effectively reduced. - The present invention proposes a selective overdrive method based on the pixel scale value corresponding to each pixel in the current picture. Obtaining the overdrive grayscale value. This method first determines whether the pixel grayscale value of the pixel before the head 10 and the prime grayscale value of the front-surface pupil are gray-like or solid-like. The enhanced mode is overdriven, and a selection signal is generated accordingly; and the pixel grayscale value of the pixel in the current surface and the pixel grayscale value of the pixel corresponding to the front surface are selected according to the selection signal. Checking at least one standard mode overdrive comparison table or enhanced mode overdrive comparison table to output an overtone value corresponding to the current face. A > In addition, the present invention further provides a selective overdrive device, which includes a descending order 疋 '''''''''''''''''''''''''''''''''''''''''''' Judging the painting of each face Whether the grayscale value is gray-like or solid-like color and needs to be added to the pixel. 7 200841309 Strong secret unit 'minus_single single-touch unit, the record judgment order = the result of the input is not to select the axis; and - According to this honesty, the price of the driver is based on this honest money, and the output is corresponding to the current 昼 mosquito over-rotation gray value ^ strong nuclear material invention is further proposed - a driving method for driving - Pixel, the method includes first, the 灰-刖 灰 灰 昼 昼 昼 昼 昼 昼 昼 ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; 灰 灰 灰 灰 灰 灰 灰 灰 灰 灰 灰 灰 灰 灰 灰 灰 灰... to drive to the second judgment value; finally, according to the judgment value output - the corresponding overdrive gray scale value is used to move the aforementioned pixels.
底下藉由賊實施娜合__式詳加說明，t更容易瞭解本 發明之目的、技術内容、特點及其所達成之功效。 【實施方式】 ' 一本^明揭示之選擇性過驅動方法及其裝置，其係於過驅動對照單 儿中同時存放有鮮赋舰麟縣與增_式雜賊照表，以 根據目前畫面與前—畫面之晝素的畫素灰階值選擇二者其中之一，以 便有選擇性魏行极之_影像邊緣加強或紐加強而不會如先前 技術般產生非預期的色彩偏移。 在進行過驅動時，係根據目前畫面中對應每一畫素之畫素灰階 值得到過驅動灰階值，第3圖係為本發明在進行選擇性過驅動方 法之流程圖，如圖所示，首先如步驟S10所示，先對每一畫面之畫素 ^晝素灰階值標記其是否為需要進行影像加強（過驅動）之類灰色或 頒純色，此種屬於人眼敏感的的色彩，在判斷該畫素之畫素灰階值滿 足下列至少其中之一條件時，即表示為類灰色： (1) 在該晝素中之三原色中，該三原色皆為〇 ;以及 (2) 在該晝素中之三原色中，任意二該原色之灰階值的差值小於 k ’其中該k小於一第一預定灰階差值kth，其中該第一預定灰 階差值kth的決定方法，可根據人眼實驗或其他模擬方式以得 8 200841309 知，例如 10$kth$20。 1 另外’在判斷該畫素之晝素灰階值係滿足下列至少其中之一條件 時，即表示類純色： (1) 在3畫素中之三原色中，其中二原色為〇 ;以及 (2) 在该晝素中之三原色中，其中二該原色之灰階值小於p，其中 „ 遠p小於一第二預定灰階差值Pth，其中第二預定灰階差值pth • 的決定方法，可根據人眼實驗或其他模擬方式以得知，例如10 雜h^20 〇 其中’上述之三原色係為紅、綠、藍（R、G、B)，畫素之畫素灰 P白值若判蚧為類灰色或類純色時，標示該畫素之1位元數位訊號為1， 若不是類灰色或類純色，則標示畫素之1位元數位訊號為〇。藉由上 述方式即可得知目前4面之畫权畫素総伽及前—畫面對應畫素 之畫素灰階值為疋否為類灰色或類純色，也就是〇或1。 其次，如步驟S12所示，利用目前晝面中之畫素之畫素灰階值以 及前一畫面對應晝素之畫素灰階值，判斷是否同時為類灰色或類純 色，亦即，二者之畫素灰階值之組合輸出選擇訊號為糾時，即如步驟 S14所示，係選擇一加強模式過驅動對照表；相對地，若二者之畫素 φ 灰階值中有至少一組不是類灰色或類純色時，其組合輸出選擇訊號為 ° ⑻、Q1 ' 1G時’即如步驟S16所示選擇標準模式過驅動對照表；最 ， 後如步驟S18，再將根據對應之對照表所細出來的過驅動灰階值傳 輸出去。 接續，以第4圖所示之選擇性過驅動裝置之架構示意圖來完整說 明本發明之技術内容。 、—參閱第4圖所示，-選擇性過驅動裝置係接收晝素於複數畫面之 複數晝素灰階值，並可據以輸出複數個過驅動灰階值。此選擇性過驅 動裝置係包括-降階單it 20,其係輸人之目前晝面之畫素的畫素灰階 值進行降階，同時參閱第5圖所示，以不均勻對照的方式將8 _減 9 200841309 ，5 bits，以減少位元數。並有一判斷單元22，係根據前述原則來判 =輸入之目前晝面之晝素的晝素灰階值是否為類灰色或類純色而需要 晝素加強，進而輸出位元數只有]bjt的數位訊號i或〇，]代表該晝 素灰階值為類灰色或類純色，〇代表該畫素灰階值不是屬於人眼所敏 感的類灰色或類純色。 。一緩衝儲存單元24係耦接降階單元2〇與判斷單元22,緩衝儲存 單元24可使用靜態隨機存取記憶體（SRAM)，用以儲存經過降階單 70 2 〇減少位兀數後輸出之畫面的該些畫素灰階值並與判斷單元2 2輸 出之1位元數位訊號合併成具有可供辨識是否為類灰色或類純色之畫 素灰卩ό值，並傳輸至一過驅動選擇對照單元24中之選擇單元;且 因已事先減少位元數，所以缓衝儲存單元24所需之健存空間容量可以 有效減少。 當目丽晝面之畫素之晝素灰階值經過降階單元2〇減少位元數之 後會傳送至過驅鱗擇對照單元24之過·鶴對照單元2δ，此過驅動 對照單元28内係同時儲存有至少—標準模式過驅動對照表以及加強 模式過驅動_表，過_對照單元28會先根獅目前畫面中之畫素 之畫素灰階值以及丽-畫面對應畫素之該晝素灰階值同時查閱標準模 式過驅動對照表以及加賴式過驅動對照表，並傳輸該二對照表查閱 出的二組過驅動灰階值至一多工器3〇。同時，目前畫面之畫素之畫素 灰階值經過判斷單it 22後之結果（i位元數位織）除了會傳送至緩 衝暫存單元24之外，亦會傳送至獅單元％巾，此時，暫存於緩衝 暫存單元24中之前—晝面對應晝素之畫素灰階值之供_數位訊號 亦會傳运·^擇單it 26 ’ In接於緩衝儲存單元24與判斷22的選 擇單元26會同時根據目前晝面中之晝素之晝素灰階值以及前一畫面 對應畫素之畫素灰階值，觸二者是否同時為類灰色或類純色，進而 輸出4擇訊號給多工器3G，多工器3G再根據選擇訊號輸出其中一 組過驅動灰階值，換言之，當選擇訊號為H，多工器3〇係輸出加強 200841309 模式之過驅動灰階值，當選擇訊號為〇〇 出標準模式之過驅動灰階值，以藉此針對晝寸’夕^^、30係輪 性來決定適當的處理方式。 一不同的旦素灰階值特 其中，本發明之選擇性過驅動裝置 輕接_單it 2G與緩_存單元24，二’元A其係 面之晝素的畫素灰階值是否相同，若相直與前-畫 前畫面之畫素灰階值，無須再進行杳夺，接輪出原輪入之該目 對照單元28中查閱對照表，以輸出"^ 亍於過驅動 前述實施例係將選擇訊號傳送給多U7以值中 動灰階值；除此之外，選擇單元26係“=:之—組過驅 動對照單元28，如第6圖所示，先選苴中擇^虎餘先傳送至過.驅 再直接輸出對應的該過驅動灰階值。〜固對照表進行查閱後， 根據二狀觸_，本發日収具有另—實施 达二 刖旦Χί^值與一目岫畫面灰階值；接續如步驟S22， 满前-畫面灰階值與目前畫面灰階值衫均為類純色與類灰色盆中 二值，其中當前一畫面灰階值與目前畫面灰階值同 ί純色時，如步驟S24輸出之判斷值係選擇查閱一加強 騎照表；反之，前—晝面灰階值與目前晝面灰階值豈中至 灰色或類純色時，如步驟S26所^^ 擇二4 ‘輪式過驅騎照表；最後如步驟S28所示，會根據不同 之判斷值’以輸丨對應之加強赋過雜灰離或是鮮模式過驅動 灰階值，以驅動該畫素。 、，其中，、嶋畫面灰與目前畫面灰階值是否為賴色或類灰色 之斷方式係與前述實補姻，不再贅述。當然，此實施例亦更可 對珂一晝面灰階值與目前晝面灰階值先進行降階步驟，以減少其位元 數。另一方面，在接收前一畫面灰階值與一目前畫面灰階值之後，更 11 200841309 可先判斷該前一畫面灰階值與該目前畫面灰階值是否相同，若相同， 即可直接輸出原輸入之目前晝面灰階值，不用在進行前述步驟。 綜上所述，本發明係在過驅動對照單元中同時存放標準模弋與妗 強模式之對照表，以便根據不同需求選擇標準模式或增強模=二 選擇性的進行有效動態影像邊緣加強。再者，本發明於暫存查 ‘口 先對畫面中之畫素灰階值有效減少位元數，使儲存於緩衝儲存 資料量減少，故可有效減少記憶體儲存容量。 早7°之 以上所述之實施例僅係為說明本發明之技術思想及特點，1 在使熱習此項技藝之人士能夠瞭解本發明之内 二目的 以之限林發日狀細請，即大驗本發鴨揭示之^所二不能 寻變化或修飾，仍應涵蓋在本發明之專利範圍内。 乍之均 【圖式簡單說明】 第1圖為習知過驅動單元的示意圖。 第2圖為晝面產生混色比例問題的示意圖。 本發明在進行選雜過鶴方法之流程圖。 發明之選擇性過驅動裝置的架構示意圖。 第對照表示意圖。 第7圖為本發二:=選的架構示意圖。 ¥ ± Φ ^ 〜丹貝苑例流程圖。 【主要7C件符號說明】 10緩衝儲存單元 19 14過驅動對照I 過驅動對照單元 20降階單元 24過驅動選擇對 3 ϋ斷單元 28過驅動對昭26選擇單元 切野知早凡 30多工哭 32預選擇單元 夕⑽ 12Under the thief to implement Nahe __ detailed description, it is easier to understand the purpose, technical content, characteristics and the achieved effects of the present invention. [Embodiment] A selective overdrive method and apparatus thereof disclosed in the present invention are stored in an overdrived comparison single child and simultaneously stored with a fresh Fulin County and an increased _ type of thief according to the current picture. One of the pixel-level grayscale values of the front-picture element is selected so as to have a selective edge enhancement or enhancement of the image without causing an unexpected color shift as in the prior art. When the driver is driven, the overdrive grayscale value is obtained according to the pixel grayscale value corresponding to each pixel in the current picture, and the third figure is a flow chart of the selective overdrive method of the present invention. First, as shown in step S10, firstly, the pixel gray value of each picture is marked as whether it is gray or solid color that needs image enhancement (overdrive), which is sensitive to the human eye. The color, when determining that the pixel gray scale value of the pixel satisfies at least one of the following conditions, is expressed as gray-like: (1) in the three primary colors of the pixel, the three primary colors are all 〇; and (2) In the three primary colors of the element, the difference between the gray level values of any two of the primary colors is less than k ', wherein the k is smaller than a first predetermined gray level difference kth, wherein the first predetermined gray level difference kth is determined According to the human eye experiment or other simulation methods, we can get 8 200841309, for example, 10$kth$20. 1 In addition, when it is judged that the pixel gray value of the pixel satisfies at least one of the following conditions, it indicates a solid color: (1) among the three primary colors of the three pixels, wherein the two primary colors are 〇; and (2) Among the three primary colors of the element, two of the primary colors have a gray scale value smaller than p, wherein „ far p is smaller than a second predetermined gray scale difference Pth, wherein the second predetermined gray scale difference pth • is determined, It can be known according to human eye experiments or other simulation methods, for example, 10 miscellaneous h^20 〇 where 'the above three primary colors are red, green, and blue (R, G, B), and the pixel prime gray P white value When the color is gray or solid, the 1-bit digit signal indicating the pixel is 1. If it is not gray-like or solid-like, the 1-bit digital signal of the pixel is 〇. It is known that the current gray scale value of the current picture of the four-dimensional picture gamma and the front-picture corresponding pixel is 类 gray or solid color, that is, 〇 or 1. Next, as shown in step S12, use At present, the gray scale value of the pixel in the picture and the gray scale value of the pixel corresponding to the previous picture are judged to be At the same time, it is gray-like or solid-like, that is, the combined output selection signal of the two gray scale values is corrected, that is, as shown in step S14, a boost mode overdrive comparison table is selected; When at least one of the grayscale values is not gray-like or solid-like, the combined output selection signal is ° (8), Q1 '1G', ie, the standard mode overdrive comparison table is selected as shown in step S16; Then, as in step S18, the overdrive gray scale value which is detailed according to the corresponding comparison table is transmitted. Continuation, the technical content of the present invention is fully illustrated by the schematic diagram of the selective overdrive device shown in FIG. - As shown in Figure 4, the selective overdrive device receives the complex pixel grayscale values of the complex picture and outputs a plurality of overdrive grayscale values. The selective overdrive device The system includes a reduced order single it 20, which is reduced by the pixel gray scale value of the current face of the input person, and is also shown in FIG. 5, which is 8 _ 9 in a non-uniform comparison manner. , 5 bits to reduce the number of bits. A judging unit 22 judges whether the pixel gray value of the pixel of the current face is gray-like or solid-like according to the foregoing principle, and requires a pixel enhancement, and then outputs a bit signal with a number of bits only bjt i or 〇,] represents that the gray scale value of the element is gray-like or solid-like, and 〇 represents that the gray scale value of the pixel is not a gray-like or solid-like color sensitive to the human eye. A buffer storage unit 24 is coupled and lowered. The level unit 2〇 and the determining unit 22, the buffer storage unit 24 can use a static random access memory (SRAM) for storing the pixel gray scales of the picture outputted after reducing the number of bits by the reduced order unit 70 2 〇 The value is combined with the 1-bit digital signal outputted by the determining unit 2 2 into a pixel having a pixel gray value that can be recognized as gray-like or solid-like, and transmitted to a selection unit in the overdrive selection control unit 24; Moreover, since the number of bits has been reduced in advance, the storage space capacity required for the buffer storage unit 24 can be effectively reduced. When the grayscale value of the pixel of the eye surface is reduced by the reduced order unit 2, the number of bits is transferred to the overrunning control unit 2δ of the overdrive scale control unit 24, and the overdrive control unit 28 At the same time, at least the standard mode overdrive comparison table and the enhanced mode overdrive _ table are stored, and the over _ comparison unit 28 firstly displays the pixel gray scale value of the pixel in the current picture of the lion and the corresponding picture pixel of the 丽-picture. The gradation gray scale value is simultaneously referred to the standard mode overdrive comparison table and the garner overdrive comparison table, and the two sets of overdrive gray scale values consulted by the two comparison tables are transmitted to a multiplexer 3 。. At the same time, the result of the current picture gray scale value of the picture is judged by the single it 22 (i bit number weaving), in addition to being transmitted to the buffer temporary storage unit 24, it is also transmitted to the lion unit% towel. When it is temporarily stored in the buffer temporary storage unit 24, the _-bit signal corresponding to the pixel gray scale value of the pixel is also transmitted. The selection unit it 26 'In is connected to the buffer storage unit 24 and the judgment 22 The selection unit 26 simultaneously selects the gray scale value of the pixel in the current picture and the pixel gray value of the corresponding pixel of the previous picture, and touches whether the two are both gray-like or solid-like, and then output 4 The signal is sent to the multiplexer 3G, and the multiplexer 3G outputs one set of overdrive grayscale values according to the selection signal. In other words, when the selection signal is H, the multiplexer 3 输出 output enhances the overdrive grayscale value of the 200841309 mode. When the selection signal is the overdrive driving grayscale value of the standard mode, the appropriate processing method is determined for the '^^^, 30 system roundness. A different denier gray scale value, wherein the selective overdrive device of the present invention is lightly connected to the single-it 2G and the buffer unit 24, and the two-element A has the same pixel gray scale value of the surface of the surface. If the vertical and the front-picture picture are the gray scale values of the picture, there is no need to recapture, and the reference table 28 of the original wheel is taken out to check the comparison table to output the "^ 过In the embodiment, the selection signal is transmitted to the multi-U7 to the value of the gray-scale value; in addition, the selection unit 26 is "=:--the group is over-driven to the control unit 28, as shown in FIG. Select ^虎余 first to pass to. Drive and then directly output the corresponding overdrive grayscale value. ~ After the solid reference table for review, according to the second touch _, the daily harvest has another - implementation up to two dan dan ί ^ The value and the one-screen grayscale value; the continuation step S22, the full-front-picture grayscale value and the current picture grayscale value shirt are both the solid color and the gray-like basin, wherein the current picture grayscale value and the current picture When the gray scale value is the same as the ί solid color, the judgment value outputted in step S24 is selected to consult a strengthened riding table; When the gray scale value of the face is compared with the current gray scale value 岂 medium to gray or solid color, as shown in step S26, the second 4' wheeled overdrive riding table is selected; finally, as shown in step S28, according to different The judgment value 'drives the grayscale value by the reinforcement corresponding to the gray or the fresh mode to drive the pixel. , , , , , , , , , , , , , , , , , , , , , , , , , The gray break mode is the same as the above, and will not be described again. Of course, this embodiment can also reduce the number of bits in the first step of the gray scale value and the current gray scale value. On the other hand, after receiving the previous screen grayscale value and a current screen grayscale value, 11 201141309 may first determine whether the previous screen grayscale value is the same as the current screen grayscale value, and if they are the same, Directly outputting the current grayscale value of the original input, the foregoing steps are not performed. In summary, the present invention simultaneously stores a comparison table of standard mode and bare mode in the overdrive comparison unit, so as to select according to different needs. Standard mode or enhanced mode = two selective The effective dynamic image edge enhancement is performed. Furthermore, the present invention effectively reduces the number of pixels in the pixel grayscale value in the temporary storage check to reduce the amount of data stored in the buffer, thereby effectively reducing the memory storage capacity. The embodiments described above at 7° or more are merely illustrative of the technical idea and features of the present invention, and 1 is intended to enable those skilled in the art to understand the second purpose of the present invention. That is to say, the big test of the duck reveals that it cannot be changed or modified, and should still be covered by the patent of the present invention. 乍之均 [Simple description of the drawing] Fig. 1 is a schematic diagram of a conventional driving unit. 2 is a schematic diagram of the problem of the color mixing ratio of the kneading surface. The present invention is a flow chart of the method for selecting the hybrid crane. The schematic diagram of the structure of the selective overdrive device of the invention. Figure 7 is a schematic diagram of the architecture of the second: = selection. ¥ ± Φ ^ ~ Danbei Court example flow chart. [Main 7C symbol description] 10 buffer storage unit 19 14 overdrive comparison I overdrive comparison unit 20 reduction unit 24 overdrive selection pair 3 单元 interrupt unit 28 overdrive pair 昭 26 selection unit cut wild knowledge early 30 gong Crying 32 pre-selection unit evening (10) 12