TW201947530A - Image cropping device and image cropping method - Google Patents
Image cropping device and image cropping method Download PDFInfo
- Publication number
- TW201947530A TW201947530A TW107115611A TW107115611A TW201947530A TW 201947530 A TW201947530 A TW 201947530A TW 107115611 A TW107115611 A TW 107115611A TW 107115611 A TW107115611 A TW 107115611A TW 201947530 A TW201947530 A TW 201947530A
- Authority
- TW
- Taiwan
- Prior art keywords
- boundary
- inspection range
- cropping
- information
- image
- Prior art date
Links
Abstract
Description
本發明是關於影像處理領域，尤其是關於影像處理中之影像裁剪裝置與影像裁剪方法。The present invention relates to the field of image processing, and more particularly to an image cropping device and an image cropping method in image processing.
某些節目或影片的兩側或四周會出現黑色長條，通常稱為黑邊(letterbox)，這是由於原始影像與螢幕尺寸規格不同所造成的。為了去除黑邊，傳統的作法係以一固定比例，例如3%，去除影像四周的部分，以得到一裁剪後之影像。然而，每張影像黑邊佔影片的比例均可能不同，導致裁剪後之影像還有剩餘黑邊，或是裁切掉部分內容，而無法準確地去除影像黑邊。Black strips appear on the sides or around some programs or movies, often called letterboxes, due to the difference between the original image and the screen size specifications. In order to remove black edges, the traditional method is to remove a part around the image at a fixed ratio, such as 3%, to obtain a cropped image. However, the proportion of each image's black border in the video may be different, resulting in a cropped image with residual black border, or cropping out some content, which cannot accurately remove the black border of the image.
因此，本發明之目在於提供一種可準確去除影像黑邊的影像裁剪裝置與方法。Therefore, an object of the present invention is to provide an image cropping device and method capable of accurately removing black edges of an image.
本發明揭露一種影像裁剪裝置，包含一裁剪資訊產生單元，對一輸入影像進行邊界偵測以產生一裁剪資訊；以及一裁剪單元，根據該裁剪資訊來裁剪該輸入影像，以產生一裁剪影像。The invention discloses an image cropping device, which includes a cropping information generating unit that performs boundary detection on an input image to generate cropping information; and a cropping unit that crops the input image according to the cropping information to generate a cropped image.
本發明另揭露一種影像裁剪方法，包含對一輸入影像進行邊界偵測以產生一裁剪資訊；以及根據該裁剪資訊來裁剪該輸入影像，以產生一裁剪影像。The invention further discloses an image cropping method, which includes performing boundary detection on an input image to generate a cropping information; and cropping the input image according to the cropping information to generate a cropped image.
有關本發明的特徵、實作與功效，茲配合圖式作較佳實施例詳細說明如下。Regarding the features, implementation, and effects of the present invention, the preferred embodiments with reference to the drawings are described in detail below.
為了準確地去除影像黑邊，本發明先透過對一輸入影像進行邊界偵測以產生一裁剪資訊，再根據該裁剪資訊來裁剪該輸入影像。請參閱第1圖與第2圖，第1圖為本發明實施例一影像裁剪裝置10之方塊圖，影像裁剪裝置10包含一裁剪資訊產生單元12與一裁剪單元14，第2圖為本發明實施例一影像裁剪方法20之流程圖。首先，裁剪資訊產生單元12對一輸入影像Fin 進行邊界偵測以產生一裁剪資訊Ic (步驟S202)，輸入影像Fin 例如可來自於一串流視訊內容或一廣播視訊內容，本發明不限視訊內容之來源；裁剪資訊Ic 例如包含一裁剪起始點位置(x, y)、一裁剪寬度Wc 以及一裁剪高度Hc ，如第3圖所示，第3圖為本發明一裁剪資訊Ic 之示意圖。接著，裁剪單元14根據裁剪資訊Ic 來裁剪輸入影像Fin ，以產生一裁剪影像Fc (步驟S204)，在一實施例中，裁剪單元14可根據裁剪起始點位置(x, y)、裁剪寬度Wc 以及裁剪高度Hc 可確定一裁剪範圍，並藉由保留輸入影像Fin 在該裁剪範圍中之內容來產生一裁剪影像Fc 。In order to accurately remove the black edges of the image, the present invention first performs boundary detection on an input image to generate a cropping information, and then crops the input image according to the cropping information. Please refer to FIG. 1 and FIG. 2. FIG. 1 is a block diagram of an image cropping device 10 according to an embodiment of the present invention. The image cropping device 10 includes a cropping information generating unit 12 and a cropping unit 14. FIG. 2 is the present invention. The flowchart of the image cropping method 20 according to the first embodiment. First, the cropping information generating unit 12 performs boundary detection on an input image F in to generate cropping information I c (step S202). The input image F in can be, for example, from a stream of video content or a broadcast video content. The present invention Unlimited sources of video content; the crop information I c includes, for example, a crop starting point position (x, y), a crop width W c, and a crop height H c , as shown in FIG. 3, which is the present invention A schematic diagram of cropped information I c . Then, the cropping unit 14 crops the input image F in according to the cropping information I c to generate a cropped image F c (step S204). In an embodiment, the cropping unit 14 may generate a cropping image starting point (x, y) , The cropping width W c and the cropping height H c can determine a cropping range, and generate a cropped image F c by retaining the content of the input image F in in the cropping range.
關於裁剪資訊產生單元12如何對輸入影像Fin 進行邊界偵測以產生裁剪資訊Ic ，請參閱第4圖，第4圖為本發明實施例一裁剪資訊產生單元12之方塊圖，在一實施例中，裁剪資訊產生單元12包含一邊界偵測單元122以及一邊界資訊產生單元124。邊界偵測單元122藉由對輸入影像Fin 進行邊界偵測來以產生一邊界資訊Ib ，邊界資訊Ib 例如包含垂直邊界x1 、x2 與上下邊界y1 、y2 ，如第5圖所示，第5圖為本發明一邊界資訊Ib 之示意圖。接著，邊界資訊產生單元124根據邊界資訊Ib 產生裁剪資訊Ic ，在一實施例中，邊界資訊產生單元124藉由比較垂直邊界x1 、x2 之大小來決定裁剪起始點位置之x座標，藉由比較水平邊界y1 、y2 之大小來決定裁剪起始點位置之y座標，以產生裁剪資訊Ic 中之起始點位置(x, y)；舉例來說，若以輸入影像Fin 之左上角為原點，往右x為正，往左x為負，往下y為正，往上y為負，資訊產生單元124將取垂直邊界x1 、x2 中較小者作為裁剪起始點位置之x座標，並取水平邊界y1 、y2 中較小者作為裁剪起始點位置之y座標。請注意，在不同的座標系統下，資訊產生單元124可取垂直邊界x1 、x2 中較大者作為裁剪起始點位置之x座標，並可取水平邊界y1 、y2 中較大者作為裁剪起始點位置之y座標。此外，資訊產生單元124另根據垂直邊界x1 、x2 間之距離決定裁剪資訊Ic 之裁剪寬度Wc ，並根據水平邊界y1 、y2 間之距離決定裁剪資訊Ic 之裁剪高度Hc ；舉例來說，若x1 =80、x2 =980、y1 =50、y2 =450，則裁剪寬度Wc =900，裁剪高度Hc =400。Regarding how the cropping information generating unit 12 performs boundary detection on the input image F in to generate cropping information I c , please refer to FIG. 4. FIG. 4 is a block diagram of the cropping information generating unit 12 according to an embodiment of the present invention. In the example, the cropping information generating unit 12 includes a boundary detecting unit 122 and a boundary information generating unit 124. The boundary detection unit 122 generates boundary information I b by performing boundary detection on the input image F in . The boundary information I b includes, for example, vertical boundaries x 1 and x 2 and upper and lower boundaries y 1 and y 2 , as described in Section 5. As shown in the figure, FIG. 5 is a schematic diagram of a boundary information I b according to the present invention. Next, the boundary information generating unit 124 generates cropping information I c according to the boundary information I b . In one embodiment, the boundary information generating unit 124 determines the position of the cropping starting point x by comparing the sizes of the vertical boundaries x 1 and x 2 . Coordinates, by comparing the sizes of the horizontal boundaries y 1 and y 2 to determine the y coordinates of the cropping starting point position to generate the starting point position (x, y) in the cropping information I c ; for example, if the input The upper left corner of the image F in is the origin, x is positive to the right, x is negative to the left, y is positive to the bottom, and y is negative to the top. The information generating unit 124 will take the smaller of the vertical boundaries x 1 and x 2 . This is taken as the x-coordinate of the cropping start point position, and the smaller of the horizontal boundaries y 1 , y 2 is taken as the y-coordinate of the cropping start position. Please note that under different coordinate systems, the information generating unit 124 may take the larger of the vertical boundaries x 1 and x 2 as the x-coordinate of the cropping starting position, and may take the larger of the horizontal boundaries y 1 and y 2 as The y-coordinate of the crop starting position. In addition, the information generating unit 124 further determines the cropping width W c of the cropping information I c according to the distance between the vertical boundaries x 1 and x 2 , and determines the cropping height H of the cropping information I c according to the distance between the horizontal boundaries y 1 and y 2 . c ; for example, if x 1 = 80, x 2 = 980, y 1 = 50, y 2 = 450, then the crop width W c = 900 and the crop height H c = 400.
關於邊界偵測單元122如何藉由對輸入影像Fin 進行邊界偵測來以產生邊界資訊，請參閱第6圖，第6圖為本發明一邊界偵測方法之流程圖60。首先，邊界偵測單元122會判斷一檢查範圍是否包含邊界(步驟601)，並判斷該檢查範圍是否可縮減(步驟602)。其中該檢查範圍可為輸入影像Fin 中小於輸入影像Fin 之一矩形範圍，在一實施例中，當偵測輸入影像Fin 之左邊界時，檢查範圍可預設為輸入影像Fin 左方1/4的區域；在其他實施例中，當偵測輸入影像Fin 之右邊界、上邊界或下邊界時，檢查範圍可分別預設為輸入影像Fin 右方、上方或下方1/4的區域。在一實施例中，邊界偵測單元122透過判斷該檢查範圍中之最大像素値是否對應於黑色，來判斷檢查範圍是否包含邊界。當該檢查範圍中之最大像素値對應於黑色時，代表該檢查範圍中所有像素値均對應於為黑色，因此可判定檢查範圍未包含邊界；反之，判定該檢查範圍未包含邊界。在一實施例中，當該檢查範圍中之最大像素值為0時，可被判斷對應於黑色；反之，可被判斷未對應於黑色；在另一實施例中，當該檢查範圍中之最大像素值落在一接近0的範圍，例如0~5，亦可被判斷對應於黑色。請注意，上述均為RGB全域範圍 (Full Range) 系統的實施例，在YUV有限範圍 (Limited Range)系統的實施例中，最大像素值可為一Y値，換句話說，當該檢查範圍中之最大Y値為16時，可被判斷對應於黑色。在任意範圍的色彩空間，皆可定義不同的值為黑色。在一實施例中，當該檢查範圍之寬度的一半大於或等於1時，代表該檢查範圍可縮減；反之，代表該檢查範圍不可縮減。在另一實施例中，當該檢查範圍之高度的一半大於或等於1時，代表該檢查範圍可縮減；反之，代表該檢查範圍不可縮減。For how the boundary detection unit 122 performs boundary detection on the input image F in to generate boundary information, please refer to FIG. 6, which is a flowchart 60 of a boundary detection method according to the present invention. First, the boundary detection unit 122 determines whether an inspection range includes a boundary (step 601), and determines whether the inspection range can be reduced (step 602). Wherein the check range can be less than the input image F to F in the input image in one rectangular area, in one embodiment, when the detected input image F in the left margin, inspection range can be preset in an input image left F 1/4 region side; in other embodiments, the input image when the detected F in the right hand boundary or lower boundary on the check range can be respectively preset for the input image F in the right, above, or below 1 / 4 area. In one embodiment, the boundary detection unit 122 determines whether the inspection range includes a boundary by judging whether the largest pixel 値 in the inspection range corresponds to black. When the largest pixel 値 in the inspection range corresponds to black, it means that all pixels 该 in the inspection range correspond to black, so it can be determined that the inspection range does not include a boundary; otherwise, it is determined that the inspection range does not include a boundary. In one embodiment, when the maximum pixel value in the inspection range is 0, it can be judged that it corresponds to black; otherwise, it can be judged that it does not correspond to black; in another embodiment, when the largest pixel in the inspection range is The pixel value falls in a range close to 0, such as 0 ~ 5, and can also be determined to correspond to black. Please note that the above are all embodiments of the RGB Full Range system. In the embodiment of the YUV Limited Range system, the maximum pixel value may be a Y 値. In other words, when the inspection range is When the maximum Y 値 is 16, it can be judged that it corresponds to black. In any range of color spaces, different values can be defined as black. In one embodiment, when half of the width of the inspection range is greater than or equal to 1, it means that the inspection range can be reduced; otherwise, it means that the inspection range cannot be reduced. In another embodiment, when half of the height of the inspection range is greater than or equal to 1, it means that the inspection range can be reduced; otherwise, it means that the inspection range cannot be reduced.
當該檢查範圍包含邊界且可縮減時，縮減該檢查範圍(步驟603)，並再次重新判斷縮減後的檢查範圍是否包含邊界(步驟601)，並判斷縮減後的該檢查範圍是否可縮減(步驟602)。舉例來說，該檢查範圍可縮減為原來的一半。在一實施例中，當偵測輸入影像Fin 之左邊界時，縮減後的檢查範圍為原檢查範圍的左半部分；在其他實施例中，當偵測輸入影像Fin 之右邊界、上邊界或下邊界時，縮減後的檢查範圍可分別為原檢查範圍的右半部分、上半部分或下半部分。When the inspection range includes a boundary and can be reduced, reduce the inspection range (step 603), and re-judge again whether the reduced inspection range includes a boundary (step 601), and determine whether the reduced inspection range can be reduced (step 602). For example, the inspection scope can be reduced by half. In one embodiment, when the detected input image F in the left margin, the inspection range is reduced to the left half of the original inspection region; in other embodiments, when the detected F in the right hand boundary of the input image, the At the boundary or lower boundary, the reduced inspection range can be the right half, the upper half, or the lower half of the original inspection range, respectively.
當該檢查範圍不包含邊界且可繼續縮減時，移動該檢查範圍(步驟604)，並再次重新判斷移動後的檢查範圍是否包含邊界(步驟601)，並判斷移動後的該檢查範圍是否可縮減(步驟602)。舉例來說，該檢查範圍之移動距離可為該檢查範圍之寬度或高度。在一實施例中，當偵測輸入影像Fin 之左邊界時，該檢查範圍可向右移動該檢查範圍之寬度；在其他實施例中，當偵測輸入影像Fin 之右邊界、上邊界或下邊界時，該檢查範圍可分別向左移動該檢查範圍之寬度、向下移動該檢查範圍之高度或向上移動該檢查範圍之高度。When the inspection range does not include a boundary and can be further reduced, move the inspection range (step 604), and re-determine whether the inspection range after the movement includes the boundary again (step 601), and determine whether the inspection range after the movement can be reduced (Step 602). For example, the moving distance of the inspection range may be the width or height of the inspection range. In one embodiment, when the left boundary of the input image F in is detected, the inspection range may be shifted to the right by the width of the inspection range; in other embodiments, when the right boundary and the upper boundary of the input image F in are detected At the lower boundary, the inspection range can be shifted to the left by the width of the inspection range, downward by the height of the inspection range, or upward by the height of the inspection range.
以偵測輸入影像之一左邊界為例，在此實施例中，檢查範圍S0 可預設為輸入影像Fin 左方1/4的區域。若檢查範圍S0 包含邊界且可縮減，則檢查範圍S0 會被縮減為檢查範圍S1 ，即檢查範圍S0 的左半部分 (步驟603)；接著，若檢查範圍S1 不包含邊界且可縮減，則檢查範圍S0 會被向右移動為檢查範圍S2 (步驟604)，移動距離為檢查範圍S1 之寬度；接著，若檢查範圍S2 包含邊界且可縮減，則檢查範圍S0 會被縮減為檢查範圍S3 (步驟603)，即檢查範圍S2 的左半部分；以此類推，直到某一次的檢查範圍不可縮減而輸出邊界資訊Ib 為止。Taking the detection of a left boundary of an input image as an example, in this embodiment, the inspection range S 0 may be preset as an area 1/4 to the left of the input image F in . If the inspection range S 0 includes a boundary and can be reduced, the inspection range S 0 will be reduced to the inspection range S 1 , that is, the left half of the inspection range S 0 (step 603); then, if the inspection range S 1 does not include a boundary and Can be reduced, the inspection range S 0 is moved to the right to the inspection range S 2 (step 604), and the movement distance is the width of the inspection range S 1 ; then, if the inspection range S 2 includes a boundary and can be reduced, the inspection range S 0 will be reduced to the inspection range S 3 (step 603), that is, the left half of the inspection range S 2 ; and so on, until a certain inspection range cannot be reduced and the boundary information I b is output.
請再次參閱第6圖，當該檢查範圍不包含邊界且不可繼續縮減時，輸出邊界資訊Ib (步驟605)。在一實施例中，當偵測輸入影像Fin 之左邊界時，與檢查範圍右側相鄰之一條線會被輸出作為一垂直邊界；舉例來說，若檢查範圍係對應於x=9時，邊界偵測單元122會輸出x=10作為一垂直邊界。在另一實施例中，當偵測輸入影像Fin 之右邊界時，與檢查範圍左側相鄰之一條線會被輸出作為一垂直邊界。在另一實施例中，當偵測輸入影像Fin 之上邊界時，與檢查範圍下側相鄰之一條線會被輸出作為一水平邊界。在另一實施例中，當偵測輸入影像Fin 之下邊界時，與檢查範圍上側相鄰之一條線會被輸出作為一水平邊界。Please refer to FIG. 6 again, when the inspection range does not include a boundary and cannot be further reduced, the boundary information I b is output (step 605). In an embodiment, when the left boundary of the input image F in is detected, a line adjacent to the right side of the inspection range is output as a vertical boundary. For example, if the inspection range corresponds to x = 9, The boundary detection unit 122 outputs x = 10 as a vertical boundary. In another embodiment, when the right boundary of the input image F in is detected, a line adjacent to the left side of the inspection range is output as a vertical boundary. In another embodiment, when the upper boundary of the input image F in is detected, a line adjacent to the lower side of the inspection range is output as a horizontal boundary. In another embodiment, when the lower boundary of the input image F in is detected, a line adjacent to the upper side of the inspection range is output as a horizontal boundary.
請再次參閱第6圖，當該檢查範圍包含邊界且不可繼續縮減時，輸出邊界資訊Ib (步驟606)。在一實施例中，當偵測輸入影像Fin 之左邊界或右邊界時，檢查範圍會被輸出作為一垂直邊界；舉例來說，若檢查範圍係對應於x=9時，邊界偵測單元122會輸出x=9作為一垂直邊界。在另一實施例中，當偵測輸入影像Fin 之上邊界或下邊界時，檢查範圍會被輸出作為一水平邊界；舉例來說，若檢查範圍係對應於y=9時，邊界偵測單元122會輸出y=9作為一水平邊界。Referring to FIG. 6 again, when the inspection range includes a boundary and cannot be further reduced, the boundary information I b is output (step 606). In an embodiment, when the left or right boundary of the input image F in is detected, the inspection range is output as a vertical boundary. For example, if the inspection range corresponds to x = 9, the boundary detection unit 122 will output x = 9 as a vertical boundary. In another embodiment, when the upper or lower boundary of the input image F in is detected, the inspection range is output as a horizontal boundary; for example, if the inspection range corresponds to y = 9, the boundary detection The unit 122 outputs y = 9 as a horizontal boundary.
透過上述的邊界偵測方式，可迅速並準確地找到輸入影像之邊界，作為裁剪輸入影像之依據，進而達到準確地去除影像黑邊的技術效果。Through the above-mentioned boundary detection method, the boundary of the input image can be found quickly and accurately as a basis for cropping the input image, thereby achieving the technical effect of accurately removing black edges of the image.
以上所述僅為本發明的實施方式，並非因此限制本發明的專利範圍，凡是利用本發明說明書及附圖內容所作的等效結構或等效流程變換，或直接或間接運用在其他相關的技術領域，均同理包括在本發明的專利保護範圍內。The above description is only an embodiment of the present invention, and thus does not limit the scope of the patent of the present invention. Any equivalent structure or equivalent process transformation made by using the description and drawings of the present invention, or directly or indirectly applied to other related technologies The same applies to the fields of patent protection of the present invention.
10‧‧‧影像裁剪裝置10‧‧‧Image cropping device
12‧‧‧裁剪資訊產生單元12‧‧‧ Crop information generation unit
14‧‧‧裁剪單元14‧‧‧ cropping unit
122‧‧‧邊界偵測單元122‧‧‧Boundary Detection Unit
124‧‧‧邊界資訊產生單元124‧‧‧Boundary information generating unit
Fin‧‧‧輸入影像F in ‧‧‧ input image
Ic‧‧‧裁剪資訊I c ‧‧‧ Crop Information
Fc‧‧‧裁剪影像F c ‧‧‧ crop image
(x, y)‧‧‧裁剪起始點位置(x, y) ‧‧‧ Cropping start position
Wc‧‧‧裁剪寬度W c ‧‧‧ crop width
Hc‧‧‧裁剪高度H c ‧‧‧cut height
Ib‧‧‧邊界資訊I b ‧‧‧ border information
x1、x2‧‧‧垂直邊界x 1 , x 2 ‧‧‧ vertical border
y1、y2‧‧‧水平邊界y 1 , y 2 ‧‧‧ horizontal border
S0、S1、S2、S3‧‧‧檢查範圍S 0 , S 1 , S 2 , S 3 ‧‧‧ Inspection range
S202~S204‧‧‧步驟S202 ~ S204‧‧‧step
S601~S606‧‧‧步驟S601 ~ S606‧‧‧ steps
［圖1］顯示本發明實施例一影像裁剪裝置之方塊圖； ［圖2］顯示本發明實施例一影像裁剪方法之流程圖； ［圖3］顯示本發明一裁剪資訊之示意圖； ［圖4］顯示本發明實施例一裁剪資訊產生單元之方塊圖； ［圖5］顯示本發明一邊界資訊之示意圖；以及 ［圖6］顯示本發明一邊界偵測方法之流程圖。[Figure 1] A block diagram showing an image cropping device according to a first embodiment of the present invention; [Figure 2] A flowchart showing an image cropping method according to a first embodiment of the present invention; [Figure 3] a schematic diagram showing a cropping information according to the present invention; [Figure 4] ] A block diagram showing a cropped information generating unit according to an embodiment of the present invention; [Fig. 5] a schematic diagram showing a boundary information of the present invention; and [Fig. 6] a flowchart showing a boundary detection method of the present invention.
Claims (16)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW107115611A TW201947530A (en) | 2018-05-08 | 2018-05-08 | Image cropping device and image cropping method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW107115611A TW201947530A (en) | 2018-05-08 | 2018-05-08 | Image cropping device and image cropping method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| TW201947530A true TW201947530A (en) | 2019-12-16 |
Family
ID=69582855
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW107115611A TW201947530A (en) | 2018-05-08 | 2018-05-08 | Image cropping device and image cropping method |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TW201947530A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI787134B (en) * | 2022-05-26 | 2022-12-11 | 國立臺灣大學 | A gpu-accelerated data processing method for rapid noise-suppressed contrast enhancement |
-
2018
- 2018-05-08 TW TW107115611A patent/TW201947530A/en unknown
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI787134B (en) * | 2022-05-26 | 2022-12-11 | 國立臺灣大學 | A gpu-accelerated data processing method for rapid noise-suppressed contrast enhancement |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019091271A1 (en) | Human face detection method and human face detection system | |
| US8498444B2 (en) | Blob representation in video processing | |
| US10484601B2 (en) | Image processing apparatus, image processing method, and storage medium | |
| TWI439788B (en) | System and method for projection correction | |
| US8355079B2 (en) | Temporally consistent caption detection on videos using a 3D spatiotemporal method | |
| WO2014170886A1 (en) | System and method for online processing of video images in real time | |
| US9727984B2 (en) | Electronic device and method for processing an image | |
| JP2013505669A5 (en) | ||
| US9953422B2 (en) | Selective local registration based on registration error | |
| CN106204441B (en) | Image local amplification method and device | |
| US7620241B2 (en) | Artifact reduction in a digital video | |
| CN101122998B (en) | Image interpolation method and device based on direction detection | |
| US8752967B2 (en) | Projection system and image processing method thereof | |
| US20120320433A1 (en) | Image processing method, image processing device and scanner | |
| JP2005172559A (en) | Method and device for detecting line defect on panel | |
| US11170252B2 (en) | Face recognition method and computer system thereof | |
| WO2017128646A1 (en) | Image processing method and device | |
| TW201947530A (en) | Image cropping device and image cropping method | |
| WO2013088657A1 (en) | Projecting projector device, optical anti-glare method, and optical anti-glare program | |
| JP3659426B2 (en) | Edge detection method and edge detection apparatus | |
| JP2011050054A (en) | Method and device for generating display mask, and computer readable storage medium | |
| WO2014033885A1 (en) | Image processing device, image processing method, and image processing program | |
| US20110158516A1 (en) | Image classification methods and systems | |
| US20150213585A1 (en) | Image processing apparatus, image processing method, and storage medium | |
| CN113610091A (en) | Intelligent identification method and device for air switch state and storage medium |