201101817 六、發明說明： 【發明所屬之技術領域】 本發明係有關於一種晝面速率轉換(FRUC)，特別是有 關於一種具有亮度補償的畫面速率轉換之裝置及方法。 【先前技術】 液晶螢幕(LCD)是由許多像素組成。每個像素可驅動到 特定亮度的像素值。然而，因為液晶改變像素值時響應時 〇 間的延遲會導致LCD面板顯示的移動物體發生運動模糊。 為了克服運動模糊現象會應用晝面速率轉換的技術。通 常，FRUC是視訊解碼器中視訊内插的處理，可以用來增 加重建視訊的感知晝面速率。此外，FRUC需要實行運動 評估以發現兩個晝面之間所需要的運動向量，然後基於精 確的運動向量内插一個新晝面差入於兩晝面之間。不幸的 是，精確運動向量的評估是困難複雜且高成本。 為解決上述問題，有必要提供一種降低LCD運動模糊 ❹的方法及裝置。 【發明内容】 本發明提供一種晝面速率轉換之方法，包括步驟：接 收複數個連續輸入視訊晝面；偵測一目前晝面的亮度資 訊；以及根據目前晝面以及在目前晝面之前的一先前晝面 的亮度資訊產生一第一輸出晝面，以及根據目前晝面及在 目前晝面之後的一後續晝面的亮度資訊產生一第二輸出晝 面，其中於第一輸出晝面之後輸出第二輸出晝辱。 本發明提供一種接收複數個連續輸入晝面的晝面速率 201101817201101817 VI. Description of the Invention: [Technical Field] The present invention relates to a face rate conversion (FRUC), and more particularly to an apparatus and method for picture rate conversion with brightness compensation. [Prior Art] A liquid crystal display (LCD) is composed of a plurality of pixels. Each pixel can be driven to a pixel value of a particular brightness. However, the delay between the responses when the liquid crystal changes the pixel value causes the moving object displayed on the LCD panel to be blurred. In order to overcome the motion blur phenomenon, a technique of face rate conversion is applied. In general, FRUC is a video interpolation process in a video decoder that can be used to increase the perceived face rate of reconstructed video. In addition, FRUC needs to perform a motion assessment to find the required motion vector between the two faces, and then interpolate a new facet between the two faces based on the precise motion vector. Unfortunately, the assessment of accurate motion vectors is difficult and costly. In order to solve the above problems, it is necessary to provide a method and apparatus for reducing LCD motion blur. SUMMARY OF THE INVENTION The present invention provides a method for rate conversion of a facet, comprising the steps of: receiving a plurality of consecutive input video frames; detecting brightness information of a current face; and determining a picture based on the current picture and before the current picture The brightness information of the previous face produces a first output face, and a second output face is generated according to the brightness information of the current face and a subsequent face after the current face, wherein the output is output after the first output face The second output is insulting. The invention provides a face rate for receiving a plurality of consecutive input faces 201101817
=換器，包括：一亮度情測器，用以偵測一目前晝面的亮度 負訊，過驅動控制器，用以根據目前畫面以及在目前畫 面之前的—先前晝面的亮度資訊產生一第一輸出畫面，： 及根據目前晝面及在目前晝面之後的—後續晝面的亮度資 訊產生一第二輸出晝面，其中於第一輸出晝面之後輸出第 =輸出畫面。過驅動控制器包括一補償電路及一晝面產生 ，。補償電路根據目前晝面及—先前晝面的亮度資訊產生 f凋正資訊，以及根據目前晝面及一後續晝面的亮度資 =生第二調整資訊。畫面產生器根據第—調整資訊產生 第一輸出畫面以及根據第二調整資訊產生第二輪出晝面。 以上所述的晝錢率轉換方法及裝置以—輸人晝面速 早接收-連續輪人視訊4面以得到快於輪人畫面速率的一 率的新連續視訊晝面，然後以亮度過驅動改善新 ^的響應時間以有效地降低運動模糊。 【實施方式】 ' :本卷明之上述目的、特徵和優點能更明顯易 '重’下文特舉較佳實施例，並配合所關式，作詳細說 π㈣於少_亮度#細及過 像素的亮度在時_上的變化 應顯不 當像素沒達到目標值T。通常，The converter includes: a brightness detector for detecting a brightness response of the current face, and an overdrive controller for generating a brightness information according to the current picture and the previous face before the current picture. The first output picture, and generating a second output picture according to the brightness information of the current picture and the subsequent picture after the current picture, wherein the first output picture is output after the first output picture. The overdrive controller includes a compensation circuit and a facet generation. The compensation circuit generates the information of the faded information according to the brightness information of the current face and the previous face, and generates the second adjustment information according to the brightness of the current face and a subsequent face. The screen generator generates a first output screen according to the first adjustment information and generates a second round output surface according to the second adjustment information. The above-mentioned method and device for converting the money rate are as follows: the input side of the input speed is fast-received and the four sides of the continuous video are obtained to obtain a new continuous video picture faster than the rate of the picture of the wheel, and then the brightness is overdriven. Improve the response time of the new ^ to effectively reduce motion blur. [Embodiment] ': The above objects, features and advantages of the present disclosure can be more obvious and easy to 'heavy'. The following is a preferred embodiment, and in conjunction with the closed type, the details of π (four) in less _ brightness # fine and over pixel The change in brightness at time _ should be improper. The pixel does not reach the target value T. usually,
的起始像素的亮度是T1而非目標值τ。數值;二旦面J 丁羞後續的晝面中導致運動 · -;目才示 等致運動_。因此，當像素過驅動 4 201101817 ν υ，在下-晝面3的起始像㈣亮度會達 i目&值Τ *因此減少後續畫面中的模掏。 朵塑f,J ’過驅動方法需要即時以及精確的LCD面板 先曰應特性的資訊。過,_單元根據LCD面板光塑應的特 像素亮度的施加的值關示上㈣書 像素梵度。 一The luminance of the starting pixel is T1 instead of the target value τ. The value; the second face J Ding shy after the subsequent face caused the movement · -; Therefore, when the pixel is overdriven 4 201101817 ν υ, the brightness of the starting image (4) in the lower-side 3 will reach i & value Τ * thus reducing the modulus in subsequent pictures. The plastic f, J 'overdrive method requires instant and accurate LCD panel information on the characteristics of the LCD panel. The _cell is based on the value of the application of the brightness of the particular pixel of the LCD panel to indicate the (4) book pixel vane. One
第2圖係根據本發明的實施義示補償 的像素亮度之示意圖。如第2圖所示，輪入書面:有 _*600的可視尺寸以及畫面速率2跡位於輸入畫面匕 F„與Fn+1内的座標卿1〇〇的像素^與^分別且 有對應的亮度值Ln-]，Ln與Ln+l。位於輸入晝面&與。 内的座標100*100的像素Ρ，』Ρ，，η分別具有對應的亮度值〇 1^„與1/>像素匕的彩度資訊（色度與飽和度）等於像素1>\ 與Ρ，，η的彩度資訊。簡言之，輸入畫面（亦即FJ的彩度資 訊等於兩對應輸出晝面（亦即F，n與F，，n)的彩度資訊。^於 亮度Ln與Ln·〗之間的第一亮度差Wu是·5〇(Ln_Ln i)，並且 介於梵度Ln與Ln+1之間的第一亮度差Wl2是4〇(Ln_ Ln+i)。因此’売度L n與L’’n基於亮度差wLI與wL2分別 是50與120。根據上述，輸入晝面Fn根據第一與第二亮度 差WL1與WL2產生二對應輸出晝面^^與P，n。於另一實施 例’亮度偵測器偵測輸入晝面。偵測每一個像素的亮产以 及計算二連續輸入晝面之間每一個對應的亮度差。此外， 也考慮物體運動資訊以及邊緣對比。邊緣偵測識別是否像 素位於輸入晝面内的物件的邊緣。此外，執行運動評估以 評估二連續輸入畫面之間的移動物件的執跡（亦即運動向 201101817 量）。根據亮度差、邊緣資訊以及運動資訊補償輸出晝面内 的像素亮度。 第3圖根據本發明的實施例顯示像素的亮度在時間軸 上變化的連續輸入晝面以及相對應的輸出晝面。輸入晝面 Fn-Fn+3内位在座標（100，100)的像素亮度分別是150、100、 60及120。在補償亮度之後，輸出晝面F’^，F’n,F’’n，F’n+1, 卩'+1與1?’11+2的座標（100,100)的像素亮度分別是200、 50、120、40、20及160。輸入晝面Fn產生二對應輸出晝 〇 面F’n與F”n，其中二對應輸出晝面F’n與F”n的彩度資訊 是等於輸入晝面Fn的彩度資訊。同樣方式下，產生輸出晝 面F’n+2、F”n+2、F’n+3與F’’n+3。晝面數目增加兩倍，並且 因此晝面速率增加兩倍。二連續輸出晝面之間的亮度差也 會增強。 第4圖根據本發明的實施例顯示說明晝面速率轉換的 方法的流程圖。在步驟402，接收晝面速率在1/60的連續 輸入晝面。在步驟404，偵測輸入晝面的亮度資訊（光度、 © 強度）。產生目前晝面與先前晝面的亮度資訊之間的第一亮 度差資訊以及產生目前晝面與後續晝面的亮度資訊之間的 第二亮度差資訊。在步驟406,產生目前晝面與先前晝面 之間的第一運動資訊（運動相位）以及產生目前晝面與後續 晝面之間的第二運動資訊（運動相位）。在步驟408，產生 目前晝面的邊緣資訊。在步驟410，根據第一運動資訊、 邊緣資訊以及第一亮度差資訊產生第一調整資訊，以及根 據第二運動資訊，、邊緣資訊以及第二亮度差資訊產生第二 調整資訊。在步驟412，根據第一調整資訊產生第一輸出 201101817 晝面’以及根據第二調整資訊產生第二輸出晝面。此外， 藉由補償目前晝面的亮度資訊產生第一與第二輸出畫面的 亮度資訊，並且第一與第二輸出晝面的彩度資訊等於目前 晝面的彩度資訊。 ❹ 〇 舉例來說，在第2圖，輸入畫面1711_1與Fn之間的亮度 差是-50。目前晝面Fn的邊緣資訊會被偵測到。先前晝面 Fn-i與目刚晝面Fn之間的第一運動資訊會被偵測到。根據 第一亮度差、第一運動資訊以及邊緣資訊產生第一調整 值。根據第一調整值補償第一輸出晝面F，nW的像素p、亮 度。像素P，』彩度資訊等於像素Pn的彩度資訊。基於像 素接著像素的基礎產生第一輸出畫面F，n。在此方法之下， 根據第二調整資訊以及目前畫面匕的彩度資訊亦產生第二 輪出裊。像’匕可經由以下函數求得： w ~a (Ln~Ln_i) + b*M + c*E P\ = P„*{\ + d*W) ί t號M表示第—運動資訊，符號E表示邊緣資訊，符 〜表不第—調整資訊，符號a到d表示對應的係數。 為輪人晝面根據二相鄰輸人畫面產生二對應的輸 屮:輪出晝面數目是輸入晝面數目的兩倍。因此，輸 的率也會增加。較佳地，纽較輸人晝面之間像素 鄰畫面之間像素的亮度差會增強。同時， 根據視訊定並且用於較佳亮度。然後 件）以到、* ^、异法施加過驅動電壓到驅動硬體（像素^ 達目標像素數值，根據新書面速率由先 、◊母—個像素的亮度轉態會顯示於顯示裝置。 201101817 第5圖,Jtt用於貫行本發明的 速率轉換器5GG包含運動料〜I·㈣旦面 ., 建動°平估益510、邊緣偵測器520、亮 度偵測器530以及過驅動柝 Φ ^ . ^ to 勒徑制盗540。過驅動控制器54〇 匕3旦 1 542以及補償電路544。運動評估器 Ο ❹ 510、邊_測器52G、亮度偵測器53G以及過驅動控制器 540接收輸入晝面。運動評估器51()產生介於目前畫面盘 先前晝面之間的第一運動資訊(運動向量），以及目前晝面 與後續晝面之間的第二運動資訊（運動向量）。亮度摘測器 530偵測目前畫面的亮度資訊（光度或強度）。過驅動控制 益540根據目前畫面與先前晝面的第一運動資訊、邊緣資 訊以及免度貢訊產生第-輸出畫面，以及根據目前畫面與 後續晝面的第二運動資訊、邊緣資訊以及亮度資訊產生第 二輸出晝面’其中第二輸出畫面在第—輸出晝面之後輸出。 過驅動控制器540包括補償電路542以及畫面產生器 544。補償電路542根據目前畫面與先前晝面的第一運動資 訊、邊緣資訊以及亮度資訊產生第一調整資訊，以及根據 第二運動資訊、邊緣資訊以及亮度資訊產生第二調整資 訊m，補償電路542 #由參考目前晝面與先前晝面 的焭度之間的差異資訊產生第一調整資訊。補償電路542 藉由參考目前晝面與後續晝面的亮度之間的差異資訊產生 第二調整資訊。晝面產生器544根據第一調整資訊產生第 一輸出晝面，以及根據第二調整資訊產生第二輸出晝面， 其中藉由補償目前晝面的亮度資訊產生第一與第二輸出畫 面的竞度資訊，並且第一與第二輪出晝面的彩度資訊（色度 與飽和度）等於目前畫面的彩度資訊。 201101817 以第2圖舉例決定輸出像素。運動評估器5i〇產生介 於先前晝面Fn_]與目前畫面匕之間的第一運動資訊。邊緣 偵測益520產生目前晝面&的邊緣資訊。亮度侧器別 分別摘測先前晝面^與目前晝面&之内的像素 的亮度。補償電路542根據第—移動資訊、邊緣資訊以及 像素U與^的亮度產生第一調整值。此外，補償電路⑷ 參考像素Ln_ALn的亮度之間亮度差W。晝面產生器⑽ 根據第-調整值以及像素Pn的彩度資訊產生輸出畫面p 〇之内的像素P，n，然後基於像素接著像素操作產生第一輸出 晝面F n °詳§之’根據第一調整值產生像素p，n的亮度， 並且像素P，n的彩度資訊等於像素Pn的彩度資訊。因此， 產生第輸出晝面p’n之内的每一個像素。同樣方式之下， 根據第二調整資訊以及目前畫面Fn的彩度資訊產 出畫面F’’n。 制 因為輸人畫面（亦即畫面Fn)可能產生二對應的輸出畫 ❹面（亦即畫φ卩’„與F”n)，所以輸出畫面的數目可能增加並 且畫面速率也會較快。此外，輸出晝面内的像素的亮度會 被補仏但是像素的彩度資訊（色度與餘和度）等於對應的輸 入晝面。對於二連續的輸出晝面，相較二連續輸入晝面， 像素的亮度差會增強。因此，輸出晝面具有不同於輸入視 訊晝面的晝面速率以及亮度差資訊。 最後，熟此技藝者可體認到他們可以輕易地使用揭露 的觀念以及特定實施例為基礎而變更及設計可以實施同樣 目的之其他結構且不脫離本發明以及申請專利範圍。 201101817 【圖式簡單說明】 第1圖基於少驅動亮度響應以及過驅動亮度響應顯示 像素的亮度在時間軸上的變化； 第2圖係根據本發明的實施例顯示補償過的輸出晝面 的像素亮度之示意圖； 第3圖根據本發明的實施例顯示像素的亮度在時間轴 上變化的連續輸入晝面以及相對應的輸出晝面； 第4圖根據本發明的實施例顯示說明晝面速率轉換的 〇 方法的流程圖；以及 第5圖顯示用於實行本發明的晝面速率轉換器。 【主要元件符號說明】 4〇2、404、406、408、410、412〜流程步驟 500〜晝面速率轉換器 510〜運動評估器 q 520〜邊緣偵測器 530〜亮度偵測器 540〜過驅動控制器 542〜補償電路 544〜晝面產生器 10Figure 2 is a schematic illustration of pixel brightness compensated in accordance with an implementation of the present invention. As shown in Figure 2, the round entry is written: there are _*600 visual size and the picture rate 2 traces are located in the input screen 匕F„ and Fn+1 in the coordinates of the pixel 1 and ^ respectively and have corresponding The brightness values Ln-], Ln and Ln+l. The pixels Ρ, Ρ, η of the coordinates 100*100 located in the input pupil & and respectively have corresponding brightness values 〇1^„ and 1/> The chroma information (chroma and saturation) of the pixel 等于 is equal to the chroma information of the pixel 1 >\ and Ρ,, η. In short, the input screen (that is, the chroma information of the FJ is equal to the chroma information of the two corresponding output pupils (ie, F, n, F, and n). ^ The first between the luminances Ln and Ln· The luminance difference Wu is ·5 〇 (Ln_Ln i), and the first luminance difference W12 between the vanishing degrees Ln and Ln+1 is 4 〇 (Ln_Ln+i). Therefore, 'the LL n and L'' n is based on the luminance differences wLI and wL2 are 50 and 120 respectively. According to the above, the input pupil plane Fn generates two corresponding output pupils ^ and P, n according to the first and second luminance differences WL1 and WL2. In another embodiment' The brightness detector detects the input side, detects the brightness of each pixel and calculates the corresponding brightness difference between the two consecutive input sides. In addition, the object motion information and edge contrast are also considered. The pixel is located at the edge of the object in the input pupil plane. In addition, motion estimation is performed to evaluate the obstruction of the moving object between the two consecutive input pictures (ie, the amount of motion to 201101817). The output is compensated according to the luminance difference, edge information, and motion information. Pixel brightness within the facet. Figure 3 shows a pixel in accordance with an embodiment of the present invention The brightness of the continuous input face and the corresponding output face on the time axis. The pixel brightness of the input face Fn-Fn+3 at coordinates (100, 100) is 150, 100, 60 and 120 respectively. After compensating for the brightness, the pixel brightness of the coordinates (100, 100) of the output faces F'^, F'n, F''n, F'n+1, 卩'+1 and 1?'11+2 are 200, respectively. 50, 120, 40, 20 and 160. The input face Fn generates two corresponding output faces F'n and F"n, wherein the chroma information of the two corresponding output faces F'n and F"n is equal to the input 昼The saturation information of the face Fn. In the same way, the output face F'n+2, F"n+2, F'n+3 and F''n+3 are produced. The number of facets is increased by two times, and therefore 昼The face rate is increased by a factor of 2. The difference in brightness between the two successive output faces is also enhanced. Figure 4 shows a flow chart illustrating a method of face rate conversion in accordance with an embodiment of the present invention. 1/60 continuous input face. In step 404, the brightness information (luminosity, © intensity) of the input face is detected. The first difference between the brightness information of the current face and the previous face is generated. The brightness difference information and the second brightness difference information between the current face and the subsequent facet brightness information. In step 406, the first motion information (motion phase) between the current face and the previous face is generated and the current is generated The second motion information (motion phase) between the face and the subsequent face. At step 408, the edge information of the current face is generated. In step 410, the first motion information, the edge information, and the first brightness difference information are generated. Adjusting the information, and generating second adjustment information according to the second motion information, the edge information, and the second brightness difference information. At step 412, a first output 201101817 ’ face is generated based on the first adjustment information and a second output face is generated based on the second adjustment information. In addition, the brightness information of the first and second output pictures is generated by compensating the brightness information of the current face, and the chroma information of the first and second output faces is equal to the current color information of the face. ❹ 〇 For example, in Fig. 2, the luminance difference between the input screens 1711_1 and Fn is -50. At the moment, the edge information of the Fn will be detected. The first motion information between the previous face Fn-i and the front face Fn is detected. A first adjustment value is generated based on the first luminance difference, the first motion information, and the edge information. The pixel p and the luminance of the first output pupil F, nW are compensated according to the first adjustment value. The pixel P, 』 chroma information is equal to the chroma information of the pixel Pn. The first output picture F,n is generated based on the basis of the pixel followed by the pixel. Under this method, a second round of exit is also generated based on the second adjustment information and the chroma information of the current screen. For example, '匕 can be obtained by the following function: w ~a (Ln~Ln_i) + b*M + c*EP\ = P„*{\ + d*W) ί t M means the first motion information, symbol E Indicates the edge information, the symbol ~ the table does not the first - adjust the information, the symbols a to d indicate the corresponding coefficient. For the wheel, the two corresponding inputs are generated according to the two adjacent input screens: the number of rounded faces is the input face The number of times is doubled. Therefore, the rate of loss will also increase. Preferably, the brightness difference of the pixels between the pixels adjacent to the picture is increased, and is determined according to the video and used for better brightness. Then, the device applies the driving voltage to the driving hardware (the pixel reaches the target pixel value according to the new method), and the brightness transition state of the pixel is displayed on the display device according to the new writing rate. 201101817 In Fig. 5, Jtt is used for the speed converter 5GG of the present invention to include a moving material 〜I·(四) 旦面. Build a flat estimate 510, an edge detector 520, a brightness detector 530, and an overdrive 柝Φ ^ . ^ to stalk 540. Overdrive controller 54 〇匕 3 1 1 542 and compensation circuit 544. Motion estimator Ο The 510 510, the edge detector 52G, the brightness detector 53G, and the overdrive controller 540 receive the input face. The motion estimator 51() generates the first motion information (motion vector) between the previous facets of the current picture disk. And the second motion information (motion vector) between the current face and the subsequent face. The brightness finder 530 detects the brightness information (luminosity or intensity) of the current picture. The overdrive control 540 is based on the current picture and the previous picture. The first motion information, the edge information and the exemption information generate a first-output picture, and generate a second output picture according to the second motion information, the edge information and the brightness information of the current picture and the subsequent picture. The output picture is output after the first output side. The overdrive controller 540 includes a compensation circuit 542 and a picture generator 544. The compensation circuit 542 generates the first picture based on the current picture and the first motion information, edge information, and brightness information of the previous picture. Adjusting the information, and generating second adjustment information m according to the second motion information, the edge information, and the brightness information, the compensation circuit 542 # The difference information between the face and the previous face produces the first adjustment information. The compensation circuit 542 generates the second adjustment information by referring to the difference information between the brightness of the current face and the subsequent face. The face generator 544 Generating a first output face according to the first adjustment information, and generating a second output face according to the second adjustment information, wherein the competition information of the first and second output images is generated by compensating the brightness information of the current face, and The chroma information (chroma and saturation) of the first and second rounds is equal to the chroma information of the current screen. 201101817 Take the example of Figure 2 to determine the output pixels. The motion estimator 5i generates first motion information between the previous face Fn_] and the current frame 。. Edge Detection Benefit 520 produces the edge information of the current Face & The brightness side device separately measures the brightness of the pixels in the previous face and the current face & The compensation circuit 542 generates a first adjustment value based on the first movement information, the edge information, and the brightness of the pixels U and ^. Further, the compensation circuit (4) refers to the luminance difference W between the luminances of the pixels Ln_ALn. The facet generator (10) generates pixels P, n within the output picture p 根据 according to the first adjustment value and the chroma information of the pixel Pn, and then generates a first output face F n ° based on the pixel followed by the pixel operation. The first adjustment value produces the brightness of the pixels p, n, and the chroma information of the pixels P, n is equal to the chroma information of the pixel Pn. Therefore, each pixel within the first output pupil plane p'n is generated. In the same manner, the picture F''n is generated based on the second adjustment information and the chroma information of the current picture Fn. Since the input picture (ie, picture Fn) may produce two corresponding output pictures (ie, φ卩' and F"n), the number of output pictures may increase and the picture rate will be faster. In addition, the brightness of the pixels in the output pupil will be compensated for, but the chroma information (chroma and margin) of the pixel is equal to the corresponding input plane. For two consecutive output pupils, the luminance difference of the pixels is enhanced compared to the two consecutive input pupil planes. Therefore, the output pupil has a different face velocity and luminance difference information than the input video face. In the end, it is obvious to those skilled in the art that they can easily use the concept of the disclosure and the specific embodiments to change and design other structures that can perform the same purpose without departing from the scope of the invention and the scope of the claims. 201101817 [Simplified Schematic] FIG. 1 is based on the change in the brightness of the display pixel based on the less-driving luminance response and the over-driving luminance response; FIG. 2 is a diagram showing the pixels of the compensated output pupil according to an embodiment of the present invention. Schematic diagram of luminance; FIG. 3 shows a continuous input pupil whose luminance of a pixel changes on a time axis and a corresponding output pupil plane according to an embodiment of the present invention; FIG. 4 shows an example of a face rate conversion according to an embodiment of the present invention. A flowchart of the trickle method; and FIG. 5 shows a face rate converter for practicing the present invention. [Main component symbol description] 4〇2, 404, 406, 408, 410, 412~ flow step 500~ face rate converter 510~ motion estimator q 520~ edge detector 530~ brightness detector 540~ Drive controller 542 to compensation circuit 544 to face generator 10