201127023 六、發明說明： 【發明所屬之技術領域】 本揭示内容是有關於一種影像處理裝置，且特別是有 關於一種消除雜訊的影像處理裝置。201127023 VI. Description of the Invention: [Technical Field] The present disclosure relates to an image processing apparatus, and more particularly to an image processing apparatus for eliminating noise.
I 【先前技術】 隨著科技的發達，數位影像的處理技術已並非單純靠 著電腦軟體來改善影像品質，而漸漸著手於利用硬體來直 ® 接對影像進行調整。透過軟硬體技術的結合，可以同時加 快處理速度與縮小硬體設備之體積。在擷取數位影像或是 影像資料傳輸的過程中，經常會因為許多的因素而使得影 像產生雜訊的干擾。例如電荷柄合元件（Charge-coupled Device，CCD)相機在擷取影像時，環境光度與感測器的溫 度都是造成影像中存在雜訊的重要因素。另一方面，如果 使用無線網路傳輸數位影像，也有機會因受到大氣中的干 擾而損壞，從而造成影像的品質變差。在品質不佳的影像 • 上進行分析處理，結果自然不會太理想。因此，為了改善 影像的品質，以便於電腦分析甚至人眼的辨識，影像中雜 訊的處理技術便成為一個很重要的課題。 【發明内容】 因此，本揭示内容之一技術態樣是在提供一種利用遮 罩加速濾除雜訊之影像處理裝置，來消除影像中的雜訊。 依據本揭示内容一實施方式，提出一種利用遮罩加速 201127023 濾除雜訊之影像處理裝置，包含一影像擷取裝置、/雙向 濾波器及一顯示裝置。影像擷取裝置用以取得至少〆待處 -!影像’此待處理影像包含至少-待處理晝素。雙甸濾波 ,，，用以處理待處理影像之待處理晝素。上述之雙向濾波 器包含-遮罩…計算單元及-模糊化單元。遮罩以待處 理畫素為中心，界定待處理晝素之多個周圍晝素，遮單在 水平方向的尺寸較垂直方向的尺寸大。計算單元用以根據 上，多個周圍晝素的資料，計算出一修正晝素資料。模糊 鲁/匕單元用以利用修玉畫素資料取代待處理晝素的資料，使 待處理衫像成為修玉影像。顯示裝置用以顯示修正影像。 、本揭示内容之另一技術態樣是在提供一種利用遮罩加 速濾除雜訊之影像處理方法，以利用遮罩來減少影像處理 所需的運算量。 依據本揭示内容另一實施方式，提出一種利用遮罩加 除雜訊之影像處理方法，包括：從一待處理影像中區 ，出一待處理晝素。以待處理晝素為中心，利用一遮罩界 疋待處理晝素之多個周圍晝素；其中’遮罩在水平方向的 ^寸較垂直方向的尺寸大。根據上述多個周圍晝素的* Ϊ素：Ϊ出一修炎畫素資料。以修正晝素資料取似寺處i = ’、'賁料，使得符處理影像成為一修正影像。最接 示修正影像。 後’ 依據本揭示内容又一實施方式，提出一種利用遮罩加 201127023 速濾除雜訊之數位相機，包括一影像擷取裝置、一影像處 理晶片以及一顯示裝置。影像擷取裝置係用以取得一待處 理影像；影像處理晶片係搭載一程式，以濾除待處理影像 之雜訊，進而產生一修正影像。最後，顯示裝置係用以顯 * 示前述之修正影像。具體而言，當影像處理晶片載入程式 後，能夠執行下列步驟：從待處理影像中區隔出一待處理 晝素。以待處理晝素為中心，利用一遮罩界定待處理畫素 之多個周圍畫素；其中，遮罩在水平方向的尺寸較垂直方 Φ 向的尺寸大。根據上述多個周圍晝素的資料，計算出一修 正晝素資料。以修正晝素資料取代待處理晝素的資料，使 得待處理影像成為上述之修正影像。 藉此，本揭示内容之上述諸實施方式，可以利用遮罩 減少影像處理所需的硬體需求。 【實施方式】 請參考第1圖，第1圖是本揭示内容一實施方式之利 • 用遮罩加速濾除雜訊之影像處理裝置的功能方塊圖。第1 圖中，影像處理裝置100包含一影像擷取裝置110、一雙 向濾波器120以及一顯示裝置130。影像擷取裝置110係 用以取得至少一待處理影像101，此待處理影像101包含 至少一待處理晝素102。雙向濾波器120，用以處理待處理 影像101之待處理晝素102。上述之雙向濾波器120包含 一遮罩121、一計算單元122及一模糊化單元123。遮罩 121以待處理晝素102為中心，界定待處理晝素102之多 個周圍晝素，遮罩121在水平方向的尺寸較垂直方向的尺 201127023 〇 言十 一 M 疋122用以根據上述多個周圍畫素的資料， 鼻出*" "f冬· ΓΡ 申 二旦素-貝料。模糊化單元123用以利用修正晝 取代待處理畫素102的資料，使待處理影像101成 •為n象：顯示骏置用以顯示修正影像。 值得注意的是’遮罩121之所以水平方向的尺寸較垂 直方向的尺寸大，是考量了一般數位影像處理晶月的運作 方式。具體請參考第2圖，第2圖是數位影像處理晶片戴 ^晝素資料的示意圖。一般數位影像處理晶片係以8x8視 籲®的格式大小’先水平後垂直掃描一晝面。以第2圖為例， 數位影像處理晶片水平掃描過第一列之後，才會垂直移動 到第一列繼續掃描。因此，遮罩121之設計係配合數位影 像處理晶片之硬體運作方式，可以使數位影像處理晶片在 相對較短的時間週期内，取得足夠進行運算的晝素資料 量’以合理化減少軟體運算量。 凊參考第3圖，第3圖是第1圖之遮罩的結構示意圖。 第3圖中，遮罩121可以設計為界定待處理晝素(S16)及其 • 同一列左右兩侧各七個晝素點（S9-S15，S17-S23)，位於待處 理晝素正上方第一列五個晝素點（S4_S8)，位於待處理晝素 正上方第一列二個畫素點（S1 -S3) ’位於待處理畫素正下 第一列五個晝素點（S24-S28)，以及位於待處理畫素正下方 第二列三個晝素點（S29-S31)。 具體來說，計算單元122係計算上述每一個畫素點與 待處理晝素之距離與晝素差異值，且利用一高斯函數來計 算每一個晝素點之權重，進而產生修正畫素資料來取代待 處理晝素。 201127023 舉例而言，計糞罝； si·)之位置表示為,,，待處理書辛 每-個晝素點與待處理晝素之距離表示為上為= =表母:個晝素點之晝素值表示為z，，待處理晝素之書 素值表不為Z7. ’則每—個晝素點與待處理晝素之I [Prior Art] With the development of technology, digital image processing technology has not only relied on computer software to improve image quality, but has gradually begun to use hardware to adjust the image. Through the combination of hardware and software technology, it can speed up processing and reduce the size of hardware devices at the same time. In the process of capturing digital images or video data transmission, there are often many factors that cause noise interference in the image. For example, when a charge-coupled device (CCD) camera captures an image, the ambient luminosity and the temperature of the sensor are both important factors in the presence of noise in the image. On the other hand, if a digital image is transmitted over a wireless network, there is also a chance that it will be damaged by atmospheric interference, resulting in poor image quality. Analytical processing on poor quality images • The results are naturally not ideal. Therefore, in order to improve the quality of images for computer analysis and even human eye recognition, the processing technology of noise in images has become an important issue. SUMMARY OF THE INVENTION Accordingly, it is a technical aspect of the present disclosure to provide an image processing apparatus that utilizes a mask to accelerate filtering of noise to eliminate noise in an image. According to an embodiment of the present disclosure, an image processing apparatus for accelerating noise of 201127023 by using a mask is provided, including an image capturing device, a bidirectional filter, and a display device. The image capture device is configured to obtain at least the image to be processed. The image to be processed contains at least a pixel to be processed. Shuangdian Filter, ,, to process the pixels to be processed. The bidirectional filter described above includes a -mask...calculation unit and a fuzzification unit. The mask is centered on the pixel to be processed, and defines a plurality of surrounding pixels of the pixel to be processed, and the size of the mask in the horizontal direction is larger than the size in the vertical direction. The calculation unit is configured to calculate a modified pixel data based on the data of the plurality of surrounding pixels. The fuzzy Lu/匕 unit is used to replace the data of the element to be processed with the data of the jade image, so that the image of the shirt to be treated becomes a jade image. The display device is used to display the corrected image. Another aspect of the present disclosure is to provide an image processing method for filtering noise by using a mask to reduce the amount of computation required for image processing by using a mask. According to another embodiment of the present disclosure, an image processing method for removing noise by using a mask is provided, including: processing a pixel to be processed from a region to be processed. Focusing on the pixel to be processed, a mask boundary is used to treat a plurality of surrounding pixels of the pixel; wherein the size of the mask in the horizontal direction is larger than that in the vertical direction. According to the above-mentioned plurality of surrounding alizarins: a inflammatory pixel data. The corrected 昼 资料 取 取 i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i The corrected image is most displayed. According to still another embodiment of the present disclosure, a digital camera using a mask and a 201127023 speed filter to remove noise is provided, including an image capturing device, an image processing chip, and a display device. The image capture device is configured to obtain a pending image; the image processing chip is loaded with a program to filter out the noise of the image to be processed, thereby generating a corrected image. Finally, the display device is used to display the aforementioned corrected image. Specifically, when the image processing chip is loaded into the program, the following steps can be performed: a pixel to be processed is separated from the image to be processed. Focusing on the pixel to be processed, a mask is used to define a plurality of surrounding pixels of the pixel to be processed; wherein the size of the mask in the horizontal direction is larger than the size of the vertical direction Φ. Based on the data of the above-mentioned multiple peripheral elements, a revised data is calculated. The data of the element to be processed is replaced by the modified morpheme data, so that the image to be processed becomes the above-mentioned corrected image. Thus, the above embodiments of the present disclosure can utilize masks to reduce the hardware requirements required for image processing. [Embodiment] Please refer to Fig. 1. Fig. 1 is a functional block diagram of an image processing apparatus for accelerating noise filtering by a mask according to an embodiment of the present disclosure. In the first embodiment, the image processing apparatus 100 includes an image capturing device 110, a two-way filter 120, and a display device 130. The image capturing device 110 is configured to obtain at least one image to be processed 101. The image to be processed 101 includes at least one pixel 102 to be processed. The bidirectional filter 120 is configured to process the pixel 102 to be processed of the image 101 to be processed. The bidirectional filter 120 described above includes a mask 121, a calculation unit 122, and a blurring unit 123. The mask 121 is centered on the pixel 102 to be processed, and defines a plurality of surrounding pixels of the pixel 102 to be processed. The size of the mask 121 in the horizontal direction is larger than the vertical direction of the 201127023 〇11 M 疋 122 for Information on multiple surrounding pixels, nose out *""f winter· 申 申二旦素-贝料. The blurring unit 123 is configured to replace the data of the pixel 102 to be processed with the correction 昼, so that the image to be processed 101 is n-image: the display is used to display the corrected image. It is worth noting that the reason why the size of the mask 121 is larger than the size in the vertical direction is to consider the operation mode of the general digital image processing crystal. For details, please refer to FIG. 2, which is a schematic diagram of the digital image processing chip wearing data. Generally, the digital image processing chip is scanned horizontally in an 8x8 format. Taking Figure 2 as an example, after the digital image processing wafer is horizontally scanned through the first column, it will move vertically to the first column to continue scanning. Therefore, the design of the mask 121 is compatible with the hardware operation mode of the digital image processing chip, so that the digital image processing chip can obtain a sufficient amount of data for the operation in a relatively short period of time to rationalize the software operation amount. . Referring to Fig. 3, Fig. 3 is a schematic view showing the structure of the mask of Fig. 1. In Fig. 3, the mask 121 can be designed to define the pixels to be treated (S16) and the seven pixel points (S9-S15, S17-S23) on the left and right sides of the same column, directly above the pixel to be processed. The first column of five pixel points (S4_S8), located in the first column of the pixel before the pixel to be processed (S1 - S3) 'located in the first column below the pixel to be processed five points (S24 -S28), and the three pixel points (S29-S31) in the second column directly below the pixel to be processed. Specifically, the calculation unit 122 calculates the distance between each pixel point and the pixel to be processed and the pixel difference value, and uses a Gaussian function to calculate the weight of each pixel point, thereby generating corrected pixel data. Replace the halogen to be treated. 201127023 For example, the position of the sputum sputum; si·) is expressed as,,,, the distance between the 昼 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 每 = = = = = The value of the prime value is expressed as z, and the table of the prime value of the element to be processed is not Z7. 'Each each pixel point and the pixel to be processed
，表4^4。接下來’利用高斯函數賦予每」個畫 各點之距轉重參數表示為*，且高斯函數賦予每一個晝 …點之晝素值權重參數表示為_。最後，遮罩内之所有畫 重：和為,，且每一個晝素點與其相對應之權重： 采積W和為…心與？及前述各參數的關係可以整理如下. -(dis2) -(dif2), Table 4^4. Next, the Gaussian function is used to assign a weight parameter to each of the points of each painting, and the Gaussian function assigns a weight parameter of each pixel value to _. Finally, all the paintings in the mask are: sum, and each pixel points have their corresponding weights: the accumulation of W and the... heart and? And the relationship between the above parameters can be organized as follows. -(dis2) -(dif2)
W ：£ 2*vdis2 if： ^ 2*vdif ……⑴ w ~~(dis2) -(dif1、 (g 2*νΛί2 * 泛 2*νίΛ/2 ) i=l s y ··,··· (2) ，後，計算單元122可藉由繼的比例關係來產生修正晝素 貝料。具體而言’ π為遮罩内所有晝素點的權重總和，户為 ，罩内晝素點乘以本身權重的總和，因此，使擁有權重越 尚的畫素點對於修正的影響程度越高。 換句話說，計算單元m先根據周圍晝素與待處理晝 素 之距離與晝素差異值，分別以高斯函數決定這些周 圍旦素的權重。然後’計算單元122再根據這些周圍晝素 的權重’加權平均這些朋晝素的晝素值，以獲得修^ 素資料。 '旦 m 值得注意的是在第1圖中，影像處理裝置100内部更 設計有—轉換單元111及-區隔單元112。轉換單元lu δ 201127023 及區隔單元112可設計於影像擷取裝置110内或雙向濾波 器120内，當然亦可單獨存在於兩者之間。轉換單元111 係用以處理影像擷取裝置110所攝入或取得之待處理影像 101，使其轉換為適切的資料格式；而區隔單元112則係用 以從待處理影像101中區隔出待處理晝素102。因為對一 張影像來說，如果所有的晝素都一視同仁地進行處理，會 造成色塊邊緣或細節處反而模糊而不清楚。因此，區隔單 元112可以從影像101中區隔出非邊緣晝素群以及非細節 畫素群，以逐一作為待處理晝素102，來進行處理。換句 話說，只有當畫素不是位於影像之線條邊緣，也不是影像 之局部細節的時候，此晝素方適合被當作待處理晝素102。 區隔單元112先將待處理影像101區分為多個視窗， 再分別計算這些視窗之晝素值變異數；然後在這些視窗之 畫素值變異數中，取最低者作為一門檻值。接下來，區隔 單元112以一檢測晝素為中心，劃定一區域視窗；再計算 此區域視窗之晝素值變異數。如果區域視窗之晝素值變異 數小於門檻值，則判定檢測畫素為待處理晝素102。 舉例而言，區隔單元112可分割待處理影像101為一 8x8視窗，以視窗不重疊方式計算各視窗内晝素值之變異 數，取最小變異數為一門檻值B。接下來，區隔單元112 再任選一晝素為檢測點，取5x5視窗來計算檢測點之一檢 測點變異數V。若檢測點變異數V小於門檻基B，則定義 檢測點為非邊緣或細節晝素群。此時，檢測點便被當作待 處理晝素102來進行處理。 換句話說，區隔單元112將影像101以8x8=64個晝素 201127023 為一組’在各組不重疊的前提下，分割 計算各m;::?最小變異數作：值:逐: 下來，區隔早兀ui開始住選一個畫素 接 .其周邊ΜΑ個畫素為基準，計算出目禪2 ’以 .*果目標畫素的，，如，則：、不是= 像之線條邊緣’也=衫像之局部細節的機率 、办 而可被選作為待處理晝素102。 羊便很回’因 當然，若考量設計上的餘裕度，區隔單元 嗖定一調整參數T ;然後，若‘& 也可被 • ^值B乘以調整參數T，亦/述檢測點變異數V小於門 或細節畫素群。具體而言，、J7 V<BT ’則定義其為非邊緣 範圍越多，但相對的影* τ值越大則經濾波器處理的 此可針對不同影像特細節與邊緣部份保存就較少，因 另一方面，由认、布求進行調整。 &影像ini > 在處理上比影像1〇1 、 的党度/色度/濃度資料(γυν) 以，轉換單元Ul係^三原色資料(RGB)更具代表性；所 轉換為亮度/色度/濃度將影像101之三原色資料(rgb) 更進一步的說，由於^貝料0〇JV)，再饋入區隔單元112。 所以，轉換單元更可^眼對於亮度的敏銳度遠高於色彩； 單元112來進行處王只提供影像1〇1之亮度資料(Y)予區隔 精簡硬體設備的需^即可。上述兩種做法都有利於進一步 值得一提的是， 單元ηι及區隔單-上述影像顧取裝置110若不包含轉換 如電荷輕合元件二Γ112,可逕由一般光學鏡頭實現之，例 _取裝置 m ^rge-coupled Device，CCD)鏡頭；若影 迎非直接攝錄一實物之光學影像，而是用 201127023 來擷取其他裴置肉 -影像處理晶片；此;轉::元貝=擷取裝置⑽可為 内。另-方面:=形式’内建於影像擷取裝置二 (DSP)來實現之；卜。拉，、；& 120可利用—離散訊號處理器 單元123可採用軟體編程的方式，内；糊化 内，或利用雙向清、±遷於雙向濾波器120 然，若轉換單12°㈣之硬體電路來實現之。當 器120内，、則意介U及區隔單元112係被設計在雙向遽波 ° 、/、亦可為軟體編程或硬體電路。 。月，’、》、、第4圖，第4圖是本揭示内容一實施方式之 用遮罩加速;慮除雜訊的影像處理方法的步驟流程圖。影 處理方法包括下列步驟。首先，如步驟210所示，從」待 處理影像中區隔出一待處理畫素。然後，如步驟2 2 0所示， 以待處理晝素為中心，利用一遮罩界定待處理晝素之多個 周圍畫素，其中，遮罩在水平方向的尺寸較垂直方向的尺 寸大。接下來，如步驟230所示，根據上述多個周圍晝素 的資料’計算出一修正畫素資料。然後’如步驟240所示， 以修正晝素資料取代待處理畫素的資料，使得待處理影像 成為一修正影像。最後，如步驟250所示’顯示修正影像。 在第4圖中，進行步驟210前，更可先執行一步驟201， 此步驟201係將影像之三原色資料(RGB)轉換為亮度/色度/ 濃度資料(YUV)，再進行步驟210。如前所述’步驟201亦 可僅提取影像之亮度資料(Y)來進行步驟210。 請參考第5圖，第5圖是本揭示内容一實施方式之利 用遮罩加速濾除雜訊的數位相機的結構示意圖。第5圖 201127023 中，數位相機300包括包括一影像擷取 虛搜曰Η _ 又置310、一影像 處理曰日片320以及-顯示裝置33()。影像_取裝置3 用以取得一待處理影像；影像處理晶片32〇係 'j 式，以濾除待處理影像之雜訊，進而產生一修正^傻。， 後，顯示裝置330係用以顯示前述之修正影像。具體而士， 當影像處理晶片320载入程式後，能夠執行下列步驟广從 待處理影像中區隔出一待處理晝素。以待處理^素為中 心，利用一遮罩321界定待處理畫素之多個周圍^素了其 中，遮罩321在水平方向的尺寸較垂直方向的尺;大。j艮 據上述多個周圍晝素的資料，計算出一修正畫素資料。以 修正晝《料取代待處理畫料㈣，使得待處二 為上述之修正影像。 /佩 接下來’本揭示内容以上述諸實施方式實際處理樣本 影像，並比較其效果如下：請參考附件附件—中第认 及2A圖是未受雜訊污染的影像。第1B及2B圖是受到高 斯雜訊污染的影像。第1C及2C圖分別是第圖= φ 本實施方式處理後的影像。在應用本實施方式產生第 及2C圖時，係先將晝素值512x512的第1B及2b圖以8χ8 視窗大小切割為4096組，進而找出門檻值β==Π8 ;且設定 調整參數Τ=8。 ’ & 附件-中第3Α圖為受到高斯雜訊污染的 β 圖是第3Α圖經平均濾波器處理後的影像。第3c 圖經本實施方式處理後的影像。第4Α、4Β及^別 是第3Α、3Β及3C圖的局部放大圖。在應用本實施二式 產生第3C圖時，係先將帛3Α圖以8χΜ見窗大小切割為 [S] 12 201127023 彻〇組，進而找出門檻值B=11;且設定調整參數τ=3〇。 附件-中第5圖是以Can〇n公司所生產的咖. 位相機，在光圈設定F11，快門時間1/6 3梢湖影像。此影像任取三個視窗進行局部 較本實施方式與習知之5x5平均濾、波 附件一中第6AmA圖為原始影像。 及8B圖分別是原始影像經平均據波器處理後的 ;C:7C及8C圖分別是原始影像經本實施方式處理後的^W : £ 2*vdis2 if: ^ 2*vdif ......(1) w ~~(dis2) -(dif1, (g 2*νΛί2 *gene 2*νίΛ/2 ) i=lsy ··,··· (2) Then, the calculating unit 122 can generate the modified pixel material by the following proportional relationship. Specifically, π is the sum of the weights of all the pixel points in the mask, and the household, the pixel point in the mask is multiplied by its own weight. Therefore, the degree of influence of the pixel point with the weight is higher on the correction. In other words, the calculation unit m firstly takes the distance between the surrounding element and the pixel to be processed and the value of the element, respectively, in Gauss The function determines the weight of these surrounding densities. Then the 'computing unit 122' weights the average value of these pheromones according to the weights of these surrounding elements to obtain the repair data. 'Don m is worth noting In the figure, the image processing device 100 is further provided with a conversion unit 111 and a segmentation unit 112. The conversion unit lu δ 201127023 and the segmentation unit 112 can be designed in the image capturing device 110 or in the bidirectional filter 120, of course. It can also exist separately between the two. The conversion unit 111 is used to process images. The image 101 to be processed taken or taken by the device 110 is converted into an appropriate data format; and the segmentation unit 112 is used to separate the pixel 102 to be processed from the image 101 to be processed. In the case of an image, if all the pixels are treated equally, the edges or details of the patches may be blurred and unclear. Therefore, the segmentation unit 112 may separate the non-edge pixel groups from the image 101 and Non-detailed pixel groups are processed one by one as the pixel 102 to be processed. In other words, this element is suitable only when the pixel is not at the edge of the line of the image or the local details of the image. The pixel 102 is to be processed. The segmentation unit 112 first divides the image 101 to be processed into a plurality of windows, and then calculates the pixel value variation of the windows respectively; and then among the pixel value variations of the windows, the lowest one As a threshold value, next, the segmentation unit 112 delimits an area window centering on a detection element; and then calculates the pixel value variation of the area window. If the area window has a pixel value If the difference is less than the threshold value, the pixel is determined to be the pixel 102 to be processed. For example, the segmentation unit 112 can divide the image to be processed 101 into an 8×8 window, and calculate the pixel value in each window in a window non-overlapping manner. For the number of mutations, the minimum variability is a threshold B. Next, the segmentation unit 112 selects a sputum as the detection point, and takes a 5x5 window to calculate the detection point variation V of one of the detection points. If V is smaller than the threshold base B, the detection point is defined as a non-edge or detail pixel group. At this time, the detection point is treated as the pixel 102 to be processed. In other words, the segmentation unit 112 sets the image 101 to 8x8. =64 昼素 201127023 is a group of 'in the premise of each group does not overlap, the division calculates each m;::? the smallest variability number: value: by: down, the interval 兀 ui began to choose a pixel The surrounding pixels are used as a benchmark to calculate the target pixels of the 2's.., if, then:, not = the edge of the line like 'the edge of the line', the probability of the local details of the shirt, Can be selected as the pending halogen 102. The sheep will return very much 'because of course, if you consider the margin of design, the segmentation unit determines an adjustment parameter T; then, if '& can also be multiplied by ^ value B to adjust the parameter T, also / checkpoint The variation number V is smaller than the gate or detail pixel group. Specifically, J7 V < BT ' defines it as the non-edge range, but the larger the shadow * τ value is, the filter processed this can be saved for different image details and edge parts. On the other hand, it is adjusted by recognition and demand. &image ini > in the processing than the image 1〇1, the party degree / chroma / concentration data (γ υ ν), the conversion unit Ul ^ three primary color data (RGB) is more representative; converted to brightness / color The degree/concentration further compares the three primary color data (rgb) of the image 101 to the segmentation unit 112 because it is 0〇JV). Therefore, the conversion unit can make the sharpness of the brightness far higher than the color; the unit 112 can perform the brightness data (Y) of the image 1〇1 to distinguish the needs of the hardware device. The above two methods are beneficial to further mention that the unit ηι and the segmentation unit-the above image capturing device 110 can be implemented by a general optical lens if it does not include a conversion such as a charge-and-light-synthesis element 112. Take the device m ^rge-coupled Device, CCD) lens; if the image is not directly recorded by a physical optical image, use 201127023 to capture other placed meat-image processing chips; this; turn:: yuanbei = The picking device (10) can be internal. Another-side: = form is built into Image Capture Device 2 (DSP) to implement it; Pull,,; & 120 can be utilized - the discrete signal processor unit 123 can be in the form of software programming, within the gelatinization, or by using two-way clearing, ± moving to the bidirectional filter 120, if the conversion is 12 ° (four) Hardware circuit to achieve it. In the device 120, the interface U and the segmentation unit 112 are designed to be bidirectional chopping, /, and can also be software programming or hardware circuits. . Month, ',', 4, and 4 are flowcharts showing the steps of the image processing method for noise reduction in accordance with an embodiment of the present disclosure. The shadow processing method includes the following steps. First, as shown in step 210, a pixel to be processed is distinguished from the image to be processed. Then, as shown in step 2 2 0, a plurality of surrounding pixels of the pixel to be processed are defined by a mask centering on the pixel to be processed, wherein the size of the mask in the horizontal direction is larger than the size in the vertical direction. Next, as shown in step 230, a corrected pixel data is calculated based on the data of the plurality of surrounding pixels. Then, as shown in step 240, the data of the pixel to be processed is replaced with the modified pixel data, so that the image to be processed becomes a corrected image. Finally, as shown in step 250, the corrected image is displayed. In FIG. 4, before step 210, a step 201 is further performed. This step 201 converts the three primary color data (RGB) of the image into luminance/chroma/density data (YUV), and then proceeds to step 210. Step 210 can also be performed by extracting only the luminance data (Y) of the image as described above. Please refer to FIG. 5. FIG. 5 is a schematic structural diagram of a digital camera using a mask to accelerate filtering of noise according to an embodiment of the present disclosure. In Fig. 5, in 201127023, the digital camera 300 includes an image capture virtual search _ again 310, an image processing 曰 320, and a display device 33 (). The image capturing device 3 is used to obtain a to-be-processed image; the image processing chip 32 is configured to filter out the noise of the image to be processed, thereby generating a correction. Then, the display device 330 is used to display the aforementioned corrected image. Specifically, after the image processing chip 320 loads the program, the following steps can be performed to widely separate a pixel to be processed from the image to be processed. With the mask as the center, a mask 321 is used to define a plurality of surrounding pixels of the pixel to be processed, and the size of the mask 321 in the horizontal direction is larger than that in the vertical direction; j艮 Based on the data of the above-mentioned multiple surrounding elements, a corrected pixel data is calculated. Replace the material to be processed (4) with the modified material, so that the second image is the above corrected image. Next, the present disclosure actually processes the sample images in the above embodiments, and compares the effects as follows: Please refer to the attached attachments - the middle and the 2A are images that are not contaminated by noise. Figures 1B and 2B are images contaminated by Gaussian noise. The first and second graphs of Fig. 1C and Fig. 2C are the images processed in the present embodiment. When the first and second graphs are generated by applying the first embodiment, the first and second graphs of 512x512 are cut into 4096 groups by 8χ8 window size, and then the threshold value β==Π8 is found; and the adjustment parameter 设定= is set. 8. The figure in '& Annex' shows that the β-graph contaminated by Gaussian noise is the image processed by the averaging filter in Figure 3. Fig. 3c shows an image processed by the present embodiment. Sections 4, 4, and 4 are partial enlarged views of the 3rd, 3rd, and 3C diagrams. When the third embodiment is generated by applying the second formula, the 帛3Α map is first cut into the [S] 12 201127023 〇 〇 group by the 8 χΜ window, and then the threshold value B=11 is found; and the adjustment parameter τ=3 is set. Hey. Attachment - Picture 5 is a coffee machine produced by Can〇n, with F11 set at the aperture and 1/6 of the shutter time. The image is taken in three windows for local comparison with the 5x5 averaging filter of the present embodiment and the conventional method, and the 6AmA image of the wave attachment 1 is the original image. And 8B are the original images processed by the average data filter; C: 7C and 8C are the original images processed by this embodiment ^
Ff定發:月f以諸實施方式揭露如上，然其並非用以 限=本發明，任何熟習此技藝者，在不脫離本發明之精神 内，當可作各種之更動與潤飾， 範圍當視後附之申請專利範圍所界定者為準。發月之保遵 【圖式簡單說明】 施例容:上述和其他目的、特徵、優點與實 又I，，、貝易M，所附圖式之說明如下： ^1圖疋本揭不内容一實施方式之利用 雜訊的影像處理裳置的功能方塊圖。 也慮除 =2,數位影像處理晶片截取畫素資料的示意圖。 第3圖疋第1圖之遮罩的結構示意圖。 雜訊====如崎峨除 第5圖是太姐-& 雜訊的數位相_；^=施方式之利用遮罩加速遽除 S1 13 201127023 【主要元件符號說明】 100 :影像處理裝置 111 :轉換單元 101 :影像 120 :雙向濾波器 122 :計算單元 130 :顯示裝置 300 :數位相機 330 :顯示裝置 110、310 :影像擷取裝置 112 :區隔單元 102 :待處理晝素 121、321 :遮罩 123 :模糊化單元 201-250 :步驟 320 :影像處理晶片Ff: The present invention is disclosed in the above embodiments, but it is not intended to limit the present invention. Anyone skilled in the art can make various changes and refinements without departing from the spirit of the present invention. The scope defined in the appended patent application shall prevail. The monthly warranty is as follows: Simple description: The above and other purposes, features, advantages and realities I,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, A functional block diagram of an image processing skirt using noise in an embodiment. Also taking into account = 2, the digital image processing chip captures the schematic of the pixel data. Fig. 3 is a schematic view showing the structure of the mask of Fig. 1. Noise ==== If the rugged figure 5 is the digital phase of the sister-& noise _; ^= The method uses the mask to accelerate the removal of S1 13 201127023 [Main component symbol description] 100 : Image processing device 111: conversion unit 101: image 120: bidirectional filter 122: calculation unit 130: display device 300: digital camera 330: display device 110, 310: image capture device 112: segmentation unit 102: pending pixels 121, 321 : Mask 123: Blurring unit 201-250: Step 320: Image Processing Wafer