200906168 九、發明說明 【發明所屬之技術領域】 本發明係關於安裝在行動設備或類似物上之影像顯示 裝置，諸如用於顯市影像之液晶顯7^裝置，以及關於用於 該影像顯示裝置之影像顯示方法。 【先前技術】 當顯不裝置的輸入影像是暗時，該顯示裝置經由來自 光源之光的調變來顯示影像，藉由使用如圖2所示的輸入 影像中之框的最大値經由正規化增加輸入影像的亮度以及 藉由結合能夠射出與該框的最大値成正比的光之光源，用 於節省光源的電力消耗不會改變顯示影像的亮度之方法（ 見例如，J P 0 1 - 2 3 9 5 8 9 A )。 因爲該方法係基於使用自輸入影像擷取於最大値擷取 器1 2之整個框的最大値之正規化，所顯示影像的品質改 變相當小。然而，當有至少一具有接近至最大値的亮度之 像素時，較亮影像不可能被獲得，通常會導致小省電功效 。因此，例如，該方法主要使用於較高影像品質而不是較 低電力消耗所需之大螢幕電視接收器。 用於行動設備之顯示裝置被強力地認爲是用於中間輸 出的簡單顯示器，其中像素的數目及顯示影像尺寸係小於 諸如實際影像資料、硬拷貝或類似物之最後輸出形式的像 素的數目及顯示影像尺寸。因此，甚至影像品質多少被降 低，考慮到電池壽命，具有小電力消耗之顯示裝置有需求 -4- 200906168 性之趨勢。 因此，近年來，於行動設備的顯示裝置的例子，實施 不僅正規化還有枷馬轉換用於獲得藉由來自光源之光的同 步減小所結合之較亮影像以改善省電功效的方法已被熟知 該項技術的應用。更明確地，整個影像的平均亮度越暗， 伽馬轉換將影像轉換越亮，其將平均亮度値調整至接近至 目標亮度値。於此例中，因爲許多影像可藉由伽'馬轉換（ 見例如，JP 3430998 B)而轉換成較亮影像，電力消耗可 被有效地降低。 雖然依據基於輸入影像的平均亮度實施枷馬轉換的習 知方法可有效地降低光源的電力消耗，有藉由枷馬轉換將 較暗輸入影像轉換成較亮影像可減小其對比之問題。當實 施枷馬轉換同時抑制對比的降低時，有電力消耗的降低量 不可能足夠之問題。 於許多例子中，安裝在行動設備上之顯示裝置僅是用 於中間輸出之顯示器，其中像素的數目及顯示影像尺寸係 小於最後輸出形式之像素的數目及顯示影像尺寸。行動設 備上的顯示裝置確實強烈地需要電力消耗的減少。考慮具 有約2 5 6色之數位影像的實際動態範圍，因爲許多顯示影 像包括照明裝置及光澤物體，未飽和之影像的數目係小。 考慮到這些情況，最後輸出之最佳飽和量不必要等於中間 輸出之最佳飽和量。 【發明內容】 -5- 200906168 因此，本發明的目的實現影像顯示裝置及用於行動設 備之影像顯示方法，甚至當某些像素係極亮時，藉由允許 正規化的些微飽和’該影像顯示裝置及該影像顯示方法可 明顯地降低光源的電力消耗而不會明顯降低對比，藉此改 善整個平衡。 依據本發明，提供一種影像顯示裝置，包括：正規器 ’用於正規化輸入影像；正規化値計算器，用於接收輸入 影像且計算正規器中用於正規化的正規化値；調光器，其 基於獲得於正規化値計算器之正規化値來調整光的強度； 光調變顯示器’用於調變來自調光器之光以顯示正規於該 正規器之影像。 以相似方式，依據本發明，提供一種影像顯示方法， 包括：正規化步驟，使輸入影像正規化；正規化値計算步 驟’接收輸入影像且計算用於正規化步驟中的正規化之正 規化値；光源調整步驟，其基於獲得於正規化値計算器之 該正規化値來調整光的強度；光調變顯示步驟，調變該光 源調整步驟中之光以顯示正規化於正規化步驟之影像。 依據本發明，這係可能實現影像顯示裝置及影像顯示 方法’其中’因爲最佳値係使用如用於正規化輸入影像之 正規化値’甚至當輸入影像具有某些極亮的像素時，光源 的電力消耗係明顯地減少而無顯示器影像的對比之明顯降 低，藉此改善整個平衡。 【實施方式】 -6- 200906168 (實施例1 ) 現將參照圖1說明本發明的較佳實施例。 依據本發明之影像顯示裝置包括：用於正規化輸入影 像之正規器1、用於接收輸入影像及計算用於正規器1中 的正規化之正規化値之正規化値計算器2、基於正規化値 計算器2中所界定之正規化値來調整光的強度之調光器3 、用於調變來自調光器3的光以顯示正規器1所正規的影 像之光調變顯示器4、用於控制整個影像顯示裝置之控制 器5。 以相似方式，依據本發明之影像顯示方法係藉由以下 步驟而達到，其包括：正規化輸入影像的正規化步驟；接 收輸入影像且計算用於正規化步驟中的正規之正規化値的 正規化値計算步驟；基於正規化値計算步驟所獲得的正規 化値來調整光的強度的光源調整步驟；調變光源調整步驟 所獲得的光以顯示正規化步驟中所正規的影像的光調變顯 示步驟；及控制整個影像顯示裝置的控制步驟。 輸入影像例如爲光影像，且由二維地配置於橫向X 及縱向 Y之像素的亮度値L(X、Y)正規地表示。每一 亮度値L(X、Y)係表示光的強度之數位値，及有效値 的數目係有限。例如，於8位元指定，每一亮度値取〇、 1、2 ..... 2 5 5的2 5 6個値的一者。 以下’將詳細S兌明包括於依據本發明的影像顯示裝置 之各別單元及其影像顯示方法的各別步驟。 正規化値計算器接收輸入影像，且獲得考慮到影像品 -7- 200906168 質及電力消耗間整個平衡於正規器實施正規化所需之最佳 正規化値。這理所述之正規化値係取自輸入影像的亮度値 且線性地轉換成正規化中所允許最大値之値。換言之，大 於正規化値之輸入影像的亮度値成爲最大値或飽和。 如圖4所示，正規化値計算器因此包括計數器21、 累積器22及決定單元23。 輸入影像進入計數器21。每一亮度値L(X、Y)之 像素的數目被計數，藉此獲得每一亮度L之頻率分佈D( L)。具體而言，其容量在數目上等於亮度値L的有效値 之記憶體D(L)被準備。每一亮度値L的頻率D被初始 化於〇。當亮度値L(X、Y)被掃瞄於橫向X及縱向Y 時，如方程式1所示，每一記憶體D ( L ( X、Y ))的値 遞增以獲得頻率分佈D ( L )。 (方程式 1) D(L(X、Y) ) =D ( L ( X > Y ) ) +1 計數器2 1通常獲得影像的整個區的頻率分佈D ( L ) 。頻率分佈D ( L )亦可被獲得僅用於使對比最佳化之影 像的特定區。爲改變依據區的加權，例如，加權値的增加 可被實施，以取代加一的標準遞增。亮度差、拉普拉斯算 符（Lap lac ian )或類似物可以周圍像素來計算以改變將被 增加的値，藉此增加用於影像的有明顯改變的區之加權。 計數器2 1所獲得之頻率分佈D ( L )被傳送至累積器 22。累積器22累積於亮度的遞減順序用於每一亮度値L 之頻率分佈D(L)以獲得如下述的累積頻率分佈R(L) 或 R2 ( L)。 -8- 200906168200906168 IX. Description of the Invention [Technical Field] The present invention relates to an image display device mounted on a mobile device or the like, such as a liquid crystal display device for displaying a video, and for the image display device Image display method. [Prior Art] When the input image of the display device is dark, the display device displays the image via modulation from the light source, and normalizes by using the maximum frame of the frame in the input image as shown in FIG. Increasing the brightness of the input image and by combining a light source capable of emitting light proportional to the maximum pupil of the frame, the method for saving the power consumption of the light source does not change the brightness of the displayed image (see, for example, JP 0 1 - 2 3 9 5 8 9 A ). Since the method is based on the normalization of the maximum chirp taken from the entire frame of the maximum extractor 12 from the input image, the quality of the displayed image is quite small. However, when there is at least one pixel having a brightness close to the maximum chirp, a brighter image cannot be obtained, which usually results in a small power saving effect. Thus, for example, the method is primarily used for higher screen quality rather than the large screen television receivers required for lower power consumption. Display devices for mobile devices are strongly considered to be simple displays for intermediate output, where the number of pixels and display image size are smaller than the number and display of pixels in the final output form such as actual image data, hard copy or the like. Image size. Therefore, even the image quality is somewhat reduced. Considering the battery life, display devices with small power consumption have a tendency to be -4-200906168. Therefore, in recent years, in the example of the display device of the mobile device, a method of not only normalizing but also Hummer conversion for obtaining a brighter image combined with the reduction of the light from the light source to improve the power saving effect has been implemented. Be familiar with the application of this technology. More specifically, the darker the average brightness of the entire image, the brighter the image is converted by the gamma conversion, which adjusts the average brightness 接近 to near the target brightness 値. In this example, since many images can be converted into brighter images by gamma-ray conversion (see, for example, JP 3430998 B), power consumption can be effectively reduced. Although the conventional method of performing the Hummer conversion based on the average brightness of the input image can effectively reduce the power consumption of the light source, the conversion of the darker input image into a brighter image by the Hummer conversion can reduce the contrast. When the Hummer conversion is implemented while suppressing the decrease in contrast, there is a problem that the reduction in power consumption is not sufficient. In many instances, the display device mounted on the mobile device is only a display for intermediate output, wherein the number of pixels and display image size is less than the number of pixels in the final output format and the displayed image size. The display device on the mobile device does strongly require a reduction in power consumption. Consider the actual dynamic range of a digital image with about 256 colors, since many display images include illumination devices and glossy objects, the number of images that are not saturated is small. With these conditions in mind, the optimal saturation of the final output does not have to be equal to the optimum saturation of the intermediate output. SUMMARY OF THE INVENTION -5-200906168 Therefore, the object of the present invention is to realize an image display device and an image display method for a mobile device, even when some pixels are extremely bright, by allowing a slight saturation of the normalization The device and the image display method can significantly reduce the power consumption of the light source without significantly reducing the contrast, thereby improving the overall balance. According to the present invention, there is provided an image display apparatus comprising: a normalizer 'for normalizing an input image; a normalized chirp calculator for receiving an input image and calculating a normalization for normalization in a normalizer; a dimmer It adjusts the intensity of the light based on the normalization obtained by the normalization 値 calculator; the light modulation display 'is used to modulate the light from the dimmer to display an image normal to the normalizer. In a similar manner, in accordance with the present invention, an image display method is provided, comprising: a normalization step to normalize an input image; a normalization 値 calculation step 'receiving an input image and calculating a normalization for normalization in the normalization step値a light source adjusting step of adjusting the intensity of the light based on the normalized 値 obtained by the normalized 値 calculator; and a light modulating display step of modulating the light in the light source adjusting step to display an image normalized to the normalizing step . According to the present invention, it is possible to implement an image display device and an image display method 'where' because the best system is used for normalizing the input image, even when the input image has some extremely bright pixels, the light source The power consumption is significantly reduced without a significant reduction in the contrast of the display image, thereby improving the overall balance. [Embodiment] -6-200906168 (Embodiment 1) A preferred embodiment of the present invention will now be described with reference to FIG. The image display device according to the present invention comprises: a normalizer for normalizing the input image, a normalization calculator for receiving the input image, and a normalization for calculating the normalization in the regularizer 2, based on the regularity a dimming device 3 for adjusting the intensity of the light, and a light modulation display 4 for modulating the light from the dimmer 3 to display a normal image of the normal device 1 A controller 5 for controlling the entire image display device. In a similar manner, the image display method according to the present invention is achieved by the steps of: normalizing the input image to normalize; receiving the input image and calculating the normalized normalized 値 in the normalization step a enthalpy calculation step; a light source adjustment step of adjusting the intensity of the light based on the normalization 获得 obtained by the normalization 値 calculation step; and modulating the light obtained by the light source adjustment step to display the light modulation of the image normalized in the normalization step a display step; and a control step of controlling the entire image display device. The input image is, for example, a light image, and is represented by a luminance 値L (X, Y) which is two-dimensionally arranged in the pixels of the horizontal X and the longitudinal direction Y. Each brightness 値L(X, Y) is a digital 値 indicating the intensity of light, and the number of effective 値 is limited. For example, in the 8-bit designation, each of the luminances is one of 2 5 6 値 of 〇, 1, 2 ..... 2 5 5 . Hereinafter, the respective steps of the respective units included in the image display device according to the present invention and the image display method thereof will be described in detail. The normalized 値 calculator receives the input image and takes into account the best normalization required for the normalization of the normalization of the quality and power consumption. This normalization is taken from the brightness of the input image and linearly converted to the maximum allowed in the normalization. In other words, the brightness of the input image larger than normalized becomes the maximum 饱和 or saturation. As shown in FIG. 4, the normalized chirp calculator thus includes a counter 21, an accumulator 22, and a decision unit 23. The input image enters the counter 21. The number of pixels of each luminance 値L (X, Y) is counted, whereby the frequency distribution D(L) of each luminance L is obtained. Specifically, the memory D(L) whose capacity is equal in number to the effective 値L of the luminance 値L is prepared. The frequency D of each luminance 値L is initialized to 〇. When the luminance 値L(X, Y) is scanned in the lateral direction X and the longitudinal direction Y, as shown in Equation 1, the 値 of each memory D (L (X, Y)) is incremented to obtain a frequency distribution D ( L ) . (Equation 1) D(L(X, Y)) = D ( L ( X > Y ) ) +1 Counter 2 1 usually obtains the frequency distribution D ( L ) of the entire region of the image. The frequency distribution D ( L ) can also be obtained for a particular region of the image that is only used to optimize the contrast. To change the weighting of the basis area, for example, an increase in the weighted 値 can be implemented to replace the standard increment of one. The luminance difference, Laplacian (Lap lac ian) or the like can be calculated by surrounding pixels to change the enthalpy to be increased, thereby increasing the weighting of the regions for significant changes in the image. The frequency distribution D ( L ) obtained by the counter 21 is transmitted to the accumulator 22. The accumulator 22 accumulates in descending order of luminance for the frequency distribution D(L) of each luminance 値L to obtain a cumulative frequency distribution R(L) or R2(L) as described below. -8- 200906168
例如，當以8位元表示亮度値L時，累積頻率分佈R (255 )於初始條件被設定於〇。然後，頻率分佈d ( L ) 係以自2 5 4的亮度値L之順序加入方程式2來累積以獲得 累積頻率分佈R(L)。仔此獲得之累積頻率分佈R(L) 對應於在藉由亮度値L使輸入影像正規化的例子中飽和之 像素的數目。 (方程式 2) R ( L) =R ( L+l ) +D ( L+1 )，其中 R( 255 )=〇 例如，當方程式3所示，累積器22可以自254遞減 順序進一步實施累積以獲得累積頻率分佈R2 ( L )。所獲 得累積頻率分佈R2 ( L )係以自更大亮度値L的順序累積 飽和像素的數目所獲得之値且對應於飽和的程度。 (方程式 3) R2 ( L) =R2 ( L+l ) +R ( L)，其中 R( 255 )=0 使用方程式2及3之頻率分佈D(L)的累積顯示一 實例，獲得顯不飽和程度之飽和量的任何方法可被使用。 累積器22中獲得作爲飽和量之累積頻率分佈R(L) 或R2 ( L )被傳送至決定單元2 3。例如，如方程式4所 示，決定單元2 3以自亮度L的最大値之順序以預定容許 飽和量“A”比較獲得作爲飽和量之累積頻率分佈R ( l )或 R2 ( L )以獲得最大亮度値L作爲正規化値Μ，獲得作爲 飽和量之累積頻率分佈R(L)或R2(L)在最大亮度値 L不會超過容許飽和量“A”。 (方程式 4) M = Max [L: {R(L)或 R2(L) }<A] 於此例中，僅累積頻率分佈R ( L )及R2 ( L )的一 者可被使用。 -9- 200906168 預定常數値可被使用作爲方程式4中的容許飽和量 “A”。且’依據亮度値L改變之値亦可被使用。例如，容 許飽和量“A”可與亮度値L成正比地設定。具體地說，當 亮度値L係大時’假設是亮影像，容許飽和量“ a，，自動地 增加。當亮度値L係小時，假設是暗影像，容許飽和量 “ A ”係自動地減小。因此’正規化値μ可基於對應於影像 的亮度之最佳容許飽和量“Α”。 當以8位元表示亮度値L時，如方程式（5 )所示， 例如，於決定單元23實施校正以增加方程式4所獲得之 正規化値，最佳正規化値可被獲得用於明顯地降低正規器 1中的飽和量。 (方程式 5) M = 25 5 -0.8x ( 25 5-Μ ) 以上已說明於計數器21、累積器22及決定單元23 中分開處理所有亮度値L之例子作爲實例。然而，爲每一 亮度値L，可以自最大亮度値L的順序藉由計數器、累積 器及決定單元實施一系列處理，且可在正規化値Μ被獲 得時完成正規化値計算器2的處理。 例如，於有些像素非常亮之例子，諸如夜視影像的例 子，正規化値Μ變成小値，因此劣化包括夜視的輝度。 爲解決此種問題，藉由降低最大像素亮度値達例如，3 0 % 所獲得之亮度値被設定作爲正規化値。選擇性地，自累積 頻率分佈達到預定値之亮度値減小20%所獲得之亮度値被 設定作爲正規化値。因此，正規化値可對應於影像而動態 地設定。 -10- 200906168 正規化値亦可經由包括計數步驟、累積步驟及決定步 驟的正規化値計算步驟藉由相同方法來計算。 正規器1基於正規化値計算器2獲得之正規化値Μ 來正規化每一像素之輸入影像1的每一像素的亮度値L ( X、Υ )以獲得正規化影像F ( Χ、Υ )。例如’方程式6 表示8位元的例子。注意到’藉由正規化所獲得之大於 2 5 5的値被設定於2 5 5。在增加〇 . 5之後’經由函數“ i n t ” 擷取之整數部份係捨入操作以獲得整數値。 (方程式 6) F (X、Y) =int{ L (X、Υ) X255/M + 0.5} 正規化被實施以獲得與正規化値M實質地成反比之 更亮影像。 對應於一框之影像通常被要求藉由正規化値計算益2 來計算正規化値。當僅提供輸入影像一次時，對應於一框 之影像可被儲存於正規器1，且所存影像可在正規化値Μ 的計算之後被正規化。當可提供相同輸入影像二次時，可 自最先提供的輸入影像藉由正規化値計算器2予以獲得正 規化値Μ，且然後可藉由正規器1正規化第二次提供之輸 入影像。當連續提供移動影像時，可自對應於前一框的輸 入影像基於正規化値計算器2所獲得之正規化値Μ藉由 正規器1正規化對應於現在框之輸入影像。 正規化亦可經由正規化步驟藉由相同方法來實現。 調光器3產生顯示用光且將所產生光發射至光調變顯 示器4。調光器3依據正規器1所轉換的亮影像來調整光 的強度。正規化影像係藉由正規器1轉換成與正規化値Μ -11 - 200906168 實質地成反比的亮影像。因此，光的強度可與正規化値Μ 成正比地藉由調光器3來改變。因此，影像可被顯示而不 會實質地改變表面亮度。 顯示用光亦可以相同方法改變光的強度經由光源調整 步驟來實現。 光調變顯示器4依據來自正規器丨的正規化影像F( X、Y)來調整透射比或反射比以調變來自調光器3之光 ，藉此顯示影像。於此實施例，使用液晶裝置調變光以顯 示影像之實例被說明。然而，本發明未受限於此實例。 光調變顯示器亦可使用相同方法經由顯示用之光學調 變顯示步驟來實現。 控制器5控制例如整個影像顯示裝置的狀態及順序。 於控制步驟，整個影像顯示裝置的狀態及順序被同樣 地控制。 如上述’依據此實施例，用於實施最佳正規化之正規 化値係考慮影像品質及電力消耗間的平衡而藉由正規化値 計算器而獲得。正規化可藉由正規器來實施以增加與正規 化値成反比之影像的亮度。來自光源之光量可藉由調光器 調整至與正規化値成正比之光量。光源的電力消耗因此可 被減小而不會實質地改變表面顯示影像的亮度。換言之， 當輸入影像的目標影像包括極亮的相同像素時，諸如照明 光的影像像素及光澤的影像像素，極亮的像素係飽和的。 因此’雖然影像品質因爲飽和而些微劣化，可明顯降低光 源的電力消耗。 -12- 200906168 用於顯示單色影像之影像顯示裝置的實例被說明。甚 至於彩色影像的例子，例如，頻率分佈係藉由計數器自所 有彩色構成成份而產生，或頻率分佈係在轉換成單色影像 後而產生。因此，當如單色影像的例子之相同方法被使用 於其它組成單元時，電力消耗可被明顯地減小而不會明顯 地改變顯示影像的亮度及影像品質。 藉由與正規化値成正比之調光器3來調整光的強度之 例子被說明作爲實例。於此實例中，現示影像的未飽和區 的亮度未被改變。因此，當飽和區變更暗時，整個框的亮 度傾向變些微暗。當包括飽和區之整個框的亮度調整至實 質地恆定値時，例如，如圖5所示，輸入影像及自正規器 1送出之影像被傳送至光量計算器7。正規器1的前與後 的影像間之平均亮度値係藉由光量計算器7獲得。與藉由 正規器1增加之影像的亮度之比成正比之光量値被傳送至 調光器3。因此，影像可被更準確地顯示而不會改變表面 亮度。當平均亮度値改變之比將被獲得時，例如，在正規 化後的頻率分佈係基於正規化値計算器2的計數器21所 獲得的頻率分佈來估算，以及二個頻率分佈被使用。能夠 獲得亮度改變比之任何單元可被使用。 飽和造成之暗化程度對應於使用於正規化値計算器2 的決定單元23之容許飽和量。因此，自調光器射出之光 的強度可基於容許飽和量精密地調整。 (實施例2 ) -13- 200906168 實施例1說明使用最佳正規化値藉由正規化所獲彳辱$ 亮影像以減小光源的電力消耗之裝置及方法。於接著說曰月 本發明的實施例2，如實施例1所述之裝置及方法係結# 有使用習知之亮度轉換來減小光源的電力消耗之結構，& 同時解決兩結構的問題。因此，影像品質的降低可被最小 化以進一步減少電力消耗。 因此，如圖3所示，依據實施例2之影像顯示裝置另 包括亮度轉換器6及依據本發明的實施例1的影像顯示裝 置中之光量計算器7。 除了依據實施例1的影像顯示裝置之外，依據實施例 2之影像顯示方法另包括亮度轉換步驟及光量計算步驟。 以下’將詳細S兌明另包括貫施例2之各別單兀及步驟 。其它單元及步驟實質地與實施例1的單元與步驟相同。 亮度轉換器6藉由諸如枷馬（gamma)轉換的亮度轉 換將正規器1所正規的影像轉換成亮影像，且將亮影像輸 出至光調變顯示器4。 關於亮度轉換的特性，例如，基於對應於電力消耗的 減少程度的恆定値之枷馬轉換被使用。替代地，基於使整 個影像的平均値接近於預設目標値的伽馬値之枷馬轉換被 使用。用於增加影像的亮度而不會明顯地改變其表面之任 何枷馬轉換可被使用。 亮度轉換器6在亮度轉換前後藉由例如方程式7來計 算增加影像的亮度之比R。 (方程式7) R=(在轉換後的平均影像亮度）/ (在轉換 -14- 200906168 前之平均影像亮度） 該比可基於影像的亮度的統計圖及亮度轉換的特性而 容易獲得。基於整個影像的平均亮度獲得該比的任何方法 可被利用。 如於正規器1的例子，當需要儲存對應於一框之影像 以藉由亮度轉換器6來獲得亮度轉換的特性及亮度用比R ’存於正規器1的影像可被共同使用。於此例中，較佳獲 得對應於使用於正規器1的正規化値之亮度轉換的特性及 亮度用的比R。 亮度轉換亦可經由亮度轉換步驟藉由相同方法來實現 〇 光量計算器7基於正規化値計算器2所獲得之正規化 値Μ及亮度轉換器6所獲得之亮度用的比R藉由例如方 程式8來計算組合光量値。（方程式8 )(組合光量値）For example, when the luminance 値L is expressed in 8-bit, the cumulative frequency distribution R (255) is set to 〇 at the initial condition. Then, the frequency distribution d ( L ) is added to Equation 2 in order from the luminance 値 L of 2 4 4 to obtain a cumulative frequency distribution R(L). The cumulative frequency distribution R(L) obtained here corresponds to the number of pixels saturated in the example of normalizing the input image by the luminance 値L. (Equation 2) R ( L) = R ( L + l ) + D ( L +1 ), where R ( 255 ) = 〇 For example, as shown in Equation 3, the accumulator 22 may further perform accumulation in 254 decreasing order. The cumulative frequency distribution R2 (L) is obtained. The obtained cumulative frequency distribution R2 ( L ) is obtained by accumulating the number of saturated pixels from the order of larger luminance 値 L and corresponds to the degree of saturation. (Equation 3) R2 ( L) = R2 ( L + l ) + R ( L), where R ( 255 ) = 0 Using the cumulative display of the frequency distribution D (L) of Equations 2 and 3, an example is obtained to obtain desaturation Any method of the degree of saturation can be used. The cumulative frequency distribution R(L) or R2(L) obtained as the saturation amount in the accumulator 22 is transmitted to the decision unit 23. For example, as shown in Equation 4, the decision unit 2 3 obtains the cumulative frequency distribution R ( l ) or R 2 ( L ) as the saturation amount by the predetermined allowable saturation amount "A" in order from the maximum 値 of the luminance L to obtain the maximum. The luminance 値L is normalized 値Μ, and the cumulative frequency distribution R(L) or R2(L) obtained as the saturation amount does not exceed the allowable saturation amount “A” at the maximum luminance 値L. (Equation 4) M = Max [L: {R(L) or R2(L) } < A] In this example, only one of the cumulative frequency distributions R ( L ) and R2 ( L ) can be used. -9- 200906168 The predetermined constant 値 can be used as the allowable saturation amount "A" in Equation 4. And ’ can also be used depending on the brightness 値L. For example, the allowable saturation amount "A" can be set in proportion to the luminance 値L. Specifically, when the brightness 値L is large, 'assumed to be a bright image, the allowable saturation amount “a,” is automatically increased. When the brightness 値L is small, it is assumed to be a dark image, and the allowable saturation amount “A” is automatically reduced. Therefore, the 'normalized 値μ can be based on the optimum allowable saturation amount "Α" corresponding to the brightness of the image. When the brightness 値L is expressed in 8 bits, as shown in the equation (5), for example, in the decision unit 23 By performing a correction to increase the normalized enthalpy obtained in Equation 4, the optimal normalization 値 can be obtained to significantly reduce the amount of saturation in the normalizer 1. (Equation 5) M = 25 5 - 0.8x (25 5-Μ An example in which all the luminances 値L are separately processed in the counter 21, the accumulator 22, and the decision unit 23 has been described above. However, for each luminance 値L, the counter, the accumulator can be used from the order of the maximum luminance 値L. And the decision unit performs a series of processing, and can perform the normalization 値 calculator 2 processing when the normalization 値Μ is obtained. For example, in some examples where the pixels are very bright, such as the example of night vision images, the normalization becomes Xiao Yan, so In order to solve such a problem, by reducing the maximum pixel brightness up to, for example, 30% of the obtained brightness 値 is set as the normalized 値. Optionally, the self-accumulated frequency distribution reaches a predetermined level. The brightness 値 obtained by reducing the brightness 20 by 20% is set as the normalization 値. Therefore, the normalization 値 can be dynamically set corresponding to the image. -10- 200906168 The normalization 値 can also include the counting step, the accumulation step, and the decision. The normalization step of the step is calculated by the same method. The normalizer 1 normalizes the brightness of each pixel of the input image 1 of each pixel based on the normalization obtained by the normalization calculator 2 (X) , Υ) to obtain the normalized image F (Χ, Υ). For example, 'Equation 6 represents an example of 8-bit. Note that the 値 obtained by normalization is greater than 2 5 5 is set at 2 5 5 . After adding 〇. 5, the integer part captured by the function "int" is rounded to obtain an integer 値. (Equation 6) F (X, Y) = int{ L (X, Υ) X255/M + 0.5 } Normalization is implemented to obtain and formalize 値M The texture is inversely brighter than the image. The image corresponding to a frame is usually required to calculate the normalization 正规 by normalizing 値 calculating benefit 2. When only the input image is provided once, the image corresponding to a frame can be stored in the regular Device 1, and the stored image can be normalized after the normalization 値Μ calculation. When the same input image can be provided twice, the input image provided from the first can be normalized by the normalization 値 calculator 2 Μ, and then the input image provided by the second time can be normalized by the normalizer 1. When the moving image is continuously provided, the input image corresponding to the previous frame can be normalized based on the normalized 値 calculator 2正规 Normalize the input image corresponding to the current frame by the normalizer 1. Normalization can also be achieved by the same method via formalization steps. The dimmer 3 generates display light and emits the generated light to the light modulation display 4. The dimmer 3 adjusts the intensity of the light in accordance with the bright image converted by the normalizer 1. The normalized image is converted by the normalizer 1 into a bright image that is substantially inversely proportional to the normalized 値Μ -11 - 200906168. Therefore, the intensity of the light can be changed by the dimmer 3 in proportion to the normalization. Therefore, the image can be displayed without substantially changing the surface brightness. The display light can also be changed in the same way by changing the intensity of the light via the light source adjustment step. The light modulation display 4 adjusts the transmittance or the reflectance in accordance with the normalized image F (X, Y) from the normal device 以 to modulate the light from the dimmer 3, thereby displaying the image. In this embodiment, an example in which a liquid crystal device is used to modulate light to display an image is explained. However, the invention is not limited to this example. The optical modulation display can also be implemented by the optical modulation display step for display using the same method. The controller 5 controls, for example, the state and sequence of the entire image display device. In the control step, the state and order of the entire image display device are similarly controlled. As described above, according to this embodiment, the normalization for implementing the optimal normalization is obtained by normalizing the calculator in consideration of the balance between image quality and power consumption. Normalization can be implemented by a normalizer to increase the brightness of the image that is inversely proportional to the normalization. The amount of light from the source can be adjusted by the dimmer to the amount of light proportional to the normalized enthalpy. The power consumption of the light source can thus be reduced without substantially changing the brightness of the surface display image. In other words, when the target image of the input image includes the same bright pixels, such as the image pixels of the illumination light and the glossy image pixels, the extremely bright pixels are saturated. Therefore, although the image quality is slightly deteriorated due to saturation, the power consumption of the light source can be significantly reduced. -12- 200906168 An example of an image display device for displaying a monochrome image is explained. Even in the case of color images, for example, the frequency distribution is generated by the counter from all color components, or the frequency distribution is generated after being converted into a monochrome image. Therefore, when the same method as the example of the monochrome image is used for other constituent units, the power consumption can be remarkably reduced without significantly changing the brightness and image quality of the display image. An example of adjusting the intensity of light by the dimmer 3 in proportion to the normalization is explained as an example. In this example, the brightness of the unsaturated region of the present image is not changed. Therefore, when the saturation region changes dark, the brightness of the entire frame tends to become slightly dark. When the brightness of the entire frame including the saturation region is adjusted to be substantially constant, for example, as shown in Fig. 5, the input image and the image sent from the normalizer 1 are transmitted to the light amount calculator 7. The average brightness between the front and rear images of the normalizer 1 is obtained by the light amount calculator 7. The amount of light 値 which is proportional to the ratio of the brightness of the image increased by the normalizer 1 is transmitted to the dimmer 3. Therefore, the image can be displayed more accurately without changing the surface brightness. When the ratio of the average luminance 値 change is obtained, for example, the normalized frequency distribution is estimated based on the frequency distribution obtained by the counter 21 of the normalized chirp calculator 2, and two frequency distributions are used. Any unit capable of obtaining a brightness change ratio can be used. The degree of darkening caused by saturation corresponds to the allowable saturation amount of the decision unit 23 used for the normalization 値 calculator 2. Therefore, the intensity of the light emitted from the dimmer can be precisely adjusted based on the allowable saturation amount. (Embodiment 2) -13- 200906168 Embodiment 1 describes an apparatus and method for reducing the power consumption of a light source by using an optimum normalization and insulting a bright image by normalization. Next, in the second embodiment of the present invention, the apparatus and method as described in the first embodiment have a structure in which the conventional light-converting is used to reduce the power consumption of the light source, and the problems of the two structures are simultaneously solved. Therefore, the degradation of image quality can be minimized to further reduce power consumption. Therefore, as shown in Fig. 3, the image display device according to the second embodiment further includes a brightness converter 6 and a light amount calculator 7 in the image display device according to the first embodiment of the present invention. In addition to the image display device according to the first embodiment, the image display method according to the second embodiment further includes a brightness conversion step and a light amount calculation step. The following 'details' will include the separate steps and steps of the second example. The other units and steps are substantially the same as those of the embodiment 1. The brightness converter 6 converts the image normalized by the normalizer 1 into a bright image by luminance conversion such as gamma conversion, and outputs the bright image to the light modulation display 4. Regarding the characteristics of the luminance conversion, for example, a constant chirp conversion based on the degree of reduction corresponding to the power consumption is used. Alternatively, a Hummer conversion based on gamma which makes the average 値 of the entire image close to the preset target 被 is used. Any Hummer conversion that is used to increase the brightness of the image without significantly changing its surface can be used. The luminance converter 6 calculates the ratio R of the luminance of the added image by, for example, Equation 7 before and after the luminance conversion. (Equation 7) R = (average image brightness after conversion) / (average image brightness before conversion -14 - 200906168) This ratio can be easily obtained based on the statistical graph of the brightness of the image and the characteristics of the brightness conversion. Any method of obtaining this ratio based on the average brightness of the entire image can be utilized. As in the example of the normalizer 1, when it is necessary to store an image corresponding to a frame to obtain a luminance conversion by the luminance converter 6, and an image in which the brightness ratio R' is stored in the normalizer 1 can be used in common. In this example, it is preferable to obtain the characteristics of the luminance conversion corresponding to the normalization 使用 used in the normalizer 1 and the ratio R for the luminance. The brightness conversion can also be achieved by the same method as the brightness calculation step 7 based on the normalization obtained by the normalization calculator 2 and the ratio R of the brightness obtained by the brightness converter 6 by, for example, an equation. 8 to calculate the combined amount of light 値. (Equation 8) (combined light quantity 値)
= M/R 於此例中，說明藉由除法獲得組合光量値之實例。當 來自亮度轉換器6之亮度用的比R對應於其倒數時，比R 可藉由乘法來獲得。 光量亦可經由使用相同方法所執行的光量計算步驟來 計算。 如上述，依據實施例2的影像顯示裝置，用於實施最 佳正規化之正規化値係考慮到影像品質及電力消耗間的平 衡藉由正規化値計算器2而獲得。正規化係藉由正規器1 實施以增加與正規化値成反比之影像的亮度。再者’影像 -15- 200906168 的亮度係使用亮度轉換器6藉由亮度轉換來增加。組合光 量値係基於正規化値及在亮度轉換器6所實施的轉換前與 後之亮度比而藉由光量計算器7來計算。來自調光器3之 光量被調整成與組合光量値成正比。來自調光器3之光係 基於亮度轉換器6所獲得之亮影像來調變以顯示該影像在 光調變顯示器4上。 接著，依據實施例2，當藉由正規化造成之影像的飽 和量及藉由亮度轉換造成之對比降低量的每一者係保持於 容許範圍內，電力消耗係與正規化及亮度轉換的組合成正 比地減少，於正規化及亮度轉換的每一者中，影像被轉換 成亮影像。因此，光源的電力消耗可被明顯地降低而不會 明顯地改變表面顯示影像的表面及其影像品質。 爲了將包括飽和區之整個框的亮度調整至如於實施例 1的例子之實質恆定値，如圖6所示，光量計算器7將與 正規器1及亮度轉換器6間的影像亮度增加比成反比之値 輸出至調光器3，影像亮度增加比係基於提供給正規器1 之影像的亮度及自亮度轉換器6送出之影像的亮度而獲得 。因此’影像可更準確地顯示而不會改變表面亮度。甚至 於此種例’如於實施例1的例子，亮度改變比係基於頻率 分佈的改變而獲得。能夠獲得亮度改變比之任何單元可被 利用。 【圖式簡單說明】 圖1係顯示依據本發明的實施例丨之影像顯示裝置的 -16- 200906168 方塊圖； 圖2係顯示習知影像顯示裝置的方塊圖； 圖3係顯示依據本發明的實施例2之影像顯示裝置的 方塊圖； 圖4係顯示正規化値計算器的結構之方塊圖； 圖5係顯示依據本發明的實施例1之影像顯示裝置的 實例之方塊圖； 圖6係顯示依據本發明的實施例2之影像顯示裝置的 實例之方塊圖。 【主要元件符號說明】 X :橫向 Y :縱向 L ( X、Y ):亮度値 D ( L):記憶體 L :亮度 D :頻率 R(L):累積頻率分佈 R2 ( L):累積頻率分佈 Μ :正規化値 A :飽和量 I :輸入影像 F ( X、Y ):正規化影像 R :比 -17 - 200906168 1 :正規器 2 :正規化値計算器 3 :調光器 4 :光調變顯示器 5 _·控制器 6 :亮度轉換器 7 :光量計算器 1 2 :最大値擷取器 2 1 :計數器 22 :累積器 2 3 :決定單元 -18= M/R In this example, an example in which the combined light amount 値 is obtained by division is explained. When the ratio R for the luminance from the luminance converter 6 corresponds to its reciprocal, the ratio R can be obtained by multiplication. The amount of light can also be calculated via the light amount calculation step performed using the same method. As described above, according to the image display device of the second embodiment, the normalization for implementing the optimum normalization is obtained by normalizing the UI 2 in consideration of the balance between image quality and power consumption. Normalization is implemented by the normalizer 1 to increase the brightness of the image that is inversely proportional to the normalization. Furthermore, the brightness of the image -15-200906168 is increased by the brightness conversion using the brightness converter 6. The combined light amount is calculated by the light amount calculator 7 based on the normalized chirp and the luminance ratio before and after the conversion performed by the luminance converter 6. The amount of light from the dimmer 3 is adjusted to be proportional to the combined amount of light 値. The light from the dimmer 3 is modulated based on the bright image obtained by the brightness converter 6 to display the image on the light modulation display 4. Then, according to Embodiment 2, each of the saturation amount of the image caused by the normalization and the contrast reduction amount by the luminance conversion is maintained within the allowable range, and the combination of the power consumption system and the normalization and the luminance conversion In proportion to the reduction, in each of the normalization and brightness conversion, the image is converted into a bright image. Therefore, the power consumption of the light source can be significantly reduced without significantly changing the surface of the surface display image and its image quality. In order to adjust the brightness of the entire frame including the saturation region to substantially constant 値 as in the example of Embodiment 1, as shown in FIG. 6, the light amount calculator 7 increases the image brightness ratio between the normalizer 1 and the luminance converter 6. The output is inversely proportional to the dimmer 3, and the image brightness increase ratio is obtained based on the brightness of the image supplied to the normalizer 1 and the brightness of the image sent from the luminance converter 6. Therefore, the image can be displayed more accurately without changing the surface brightness. Even in this example, as in the example of Embodiment 1, the luminance change ratio is obtained based on the change in the frequency distribution. Any unit capable of obtaining a brightness change ratio can be utilized. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an image display apparatus according to an embodiment of the present invention; FIG. 2 is a block diagram showing a conventional image display apparatus; FIG. 3 is a block diagram showing a conventional image display apparatus; Figure 4 is a block diagram showing the structure of a normalized UI calculator; Figure 5 is a block diagram showing an example of an image display device according to Embodiment 1 of the present invention; A block diagram showing an example of an image display apparatus according to Embodiment 2 of the present invention. [Description of main component symbols] X: Horizontal Y: Longitudinal L (X, Y): Brightness 値 D (L): Memory L: Brightness D: Frequency R (L): Cumulative frequency distribution R2 (L): Cumulative frequency distribution Μ : Normalization 値 A : Saturation amount I : Input image F ( X, Y ): Normalized image R : Ratio -17 - 200906168 1 : Normalizer 2 : Normalization 値 Calculator 3 : Dimmer 4 : Light tone Variable display 5 _·controller 6: brightness converter 7: light quantity calculator 1 2 : maximum picker 2 1 : counter 22 : accumulator 2 3 : decision unit -18