TWI495354B - Image refinement apparatus and image refinement method - Google Patents
Image refinement apparatus and image refinement method Download PDFInfo
- Publication number
- TWI495354B TWI495354B TW099134103A TW99134103A TWI495354B TW I495354 B TWI495354 B TW I495354B TW 099134103 A TW099134103 A TW 099134103A TW 99134103 A TW99134103 A TW 99134103A TW I495354 B TWI495354 B TW I495354B
- Authority
- TW
- Taiwan
- Prior art keywords
- foreground
- image
- unit
- block
- background
- Prior art date
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Description
本發明係關於一種影像優化裝置及影像優化方法。The invention relates to an image optimization device and an image optimization method.
隨著三維(three-dimension，3D)顯示技術的發展，3D顯示裝置等相關產品也逐漸在市面上嶄露頭角，並造成一股風潮。With the development of three-dimensional (3D) display technology, related products such as 3D display devices have gradually emerged in the market and caused a wave of trends.
以現階段而言，由於缺乏可提供3D顯示裝置使用之3D影像內容，因此，由2D影像轉3D影像之技術，在近年來已經成為業者及學術界共同探討與待解決之問題。其中，現有的2D影像轉3D影像的方法中，一般是先產生2D影像的深度圖，再利用原始2D影像與深度圖來產生3D播放器所支援的3D影像。然而，習知之影像處理裝置係著重於以全自動之方式進行影像處理與提升影像處理之速度，但在一昧追求處理速度之前提下，習知之影像處理裝置反而忽略了所生成之影像的品質。At this stage, due to the lack of 3D video content that can be used in 3D display devices, the technology of converting 2D images to 3D images has become a problem that the industry and academia have jointly discussed and solved in recent years. Among the existing methods for converting 2D images to 3D images, a depth map of 2D images is generally generated first, and then the original 2D images and depth maps are used to generate 3D images supported by the 3D player. However, the conventional image processing apparatus focuses on the speed of image processing and image processing in a fully automatic manner, but before the pursuit of processing speed, the conventional image processing apparatus neglects the quality of the generated image. .
因此，如何提供一種影像優化裝置及影像優化方法，使其能夠提升所生成之影像的品質，並維持影像處理之速度，已成為重要課題之一。Therefore, how to provide an image optimization device and an image optimization method to improve the quality of the generated image and maintain the speed of image processing has become one of the important topics.
有鑑於上述課題，本發明之目的為提供一種能夠提升所生成之影像的品質，並維持影像處理之速度的影像優化裝置及影像優化方法。In view of the above problems, an object of the present invention is to provide an image optimizing apparatus and an image optimizing method capable of improving the quality of generated images and maintaining the speed of image processing.
為達上述目的，依據本發明之一種影像優化裝置，包含一分割單元、一判斷單元及一優化單元。分割單元分割一影像資料為複數區塊。判斷單元與分割單元耦接，並依序判斷各區塊是否具有前景與背景。優化單元與判斷單元耦接，其中當判斷結果為是，優化單元對具有前景與背景之區塊之前景之輪廓進行優化。To achieve the above objective, an image optimization apparatus according to the present invention comprises a dividing unit, a determining unit and an optimizing unit. The dividing unit divides an image data into a plurality of blocks. The determining unit is coupled to the dividing unit, and sequentially determines whether each block has a foreground and a background. The optimization unit is coupled to the determination unit, wherein when the determination result is yes, the optimization unit optimizes the contour of the foreground of the block having the foreground and the background.
在本發明一實施例中，優化單元依據區塊之前景與背景的顏色資訊進行前景之輪廓的優化。In an embodiment of the invention, the optimization unit optimizes the contour of the foreground according to the color information of the foreground and background of the block.
在本發明一實施例中，優化單元具有一貝式分類器。In an embodiment of the invention, the optimization unit has a shell classifier.
在本發明一實施例中，影像資料為一深度圖。In an embodiment of the invention, the image data is a depth map.
承上所述，因依據本發明之一種影像優化方法，包含：分割一影像資料為複數區塊；判斷該等區塊其中之一區塊是否具有前景與背景；以及當判斷結果為是，對該區塊之前景之輪廓進行優化。According to the above, an image optimization method according to the present invention includes: dividing an image data into a plurality of blocks; determining whether one of the blocks has a foreground and a background; and when the determination result is yes, The contour of the front view of the block is optimized.
在本發明一實施例中，當判斷結果為否，判斷該等區塊中之另一區塊是否具有前景與背景，直至該等區塊判斷完成。In an embodiment of the invention, when the determination result is no, it is determined whether another block in the blocks has a foreground and a background until the determination of the blocks is completed.
在本發明一實施例中，當判斷結果為是，依據該區塊之前景與背景的顏色資訊進行前景之輪廓的優化。In an embodiment of the invention, when the determination result is yes, the contour of the foreground is optimized according to the color information of the foreground and background of the block.
在本發明一實施例中，前景之輪廓係藉由一貝式分類器進行優化。In an embodiment of the invention, the contour of the foreground is optimized by a Bayer classifier.
在本發明一實施例中，影像資料為一深度圖。In an embodiment of the invention, the image data is a depth map.
承上所述，因依據本發明之一種影像優化裝置及影像優化方法係藉由將影像資料分割為複數區塊，並判斷各區塊是否具有背景與前景，而對具有前景與背景之區塊之前景的輪廓進行優化處理。從而實現能夠提升所生成之影像的品質，並維持影像處理之速度。According to the above description, an image optimization apparatus and an image optimization method according to the present invention divide a video data into a plurality of blocks, and determine whether each block has a background and a foreground, and a block having a foreground and a background. The contour of the foreground is optimized. Thereby, the quality of the generated image can be improved, and the speed of image processing can be maintained.
以下將參照相關圖式，說明依據本發明較佳實施例之一種影像優化裝置及影像優化方法。Hereinafter, an image optimization apparatus and an image optimization method according to a preferred embodiment of the present invention will be described with reference to related drawings.
請參照圖1，圖1係為依據本發明較佳實施例之一種影像優化裝置1的示意圖。影像優化裝置1包含一分割單元11、一判斷單元12及一優化單元13。Please refer to FIG. 1. FIG. 1 is a schematic diagram of an image optimization apparatus 1 according to a preferred embodiment of the present invention. The image optimization device 1 includes a segmentation unit 11, a determination unit 12, and an optimization unit 13.
分割單元11係選取一儲存單元(圖未示出)所儲存之影像資料，並將影像資料分割為複數個區塊(block)。其中，各區塊為具有相同的大小。在本實施，例中，影像資料係為一2D影像之深度圖。此外，在實施上，分割單元11亦可以接收或選取一處理單元所生成之影像資料，並進行影像資料的分割。The dividing unit 11 selects image data stored by a storage unit (not shown) and divides the image data into a plurality of blocks. Among them, each block has the same size. In this embodiment, the image data is a depth map of a 2D image. In addition, in the implementation, the dividing unit 11 can also receive or select image data generated by a processing unit, and perform segmentation of the image data.
判斷單元12與分割單元11耦接，並依序判斷分割單元11分割後之影像資料的各區塊是否同時具有前景與背景。在實施上，判斷單元12係可藉由高通濾波器(High Pass Filter)與閥值分割法(Thresholding)而對影像資料進行處理，進而判斷各區塊是否具有前景與背景。The determining unit 12 is coupled to the dividing unit 11 and sequentially determines whether each block of the divided image data of the dividing unit 11 has both foreground and background. In implementation, the determining unit 12 can process the image data by using a High Pass Filter and a Thresholding method to determine whether each block has a foreground and a background.
優化單元13與判斷單元12耦接。在本實施例中，優化單元13係具有一貝式分類器(Bayes classifier)。當判斷結果為是，優化單元13將針對具有前景與背景之區塊的前景之輪廓進行優化處理。在實施上，優化單元13是依據區塊之前景與背景的顏色資訊，而去除前景之邊緣上的雜點，進而達成前景之輪廓的優化。此外，優化單元13亦可以針對前景之內部進行優化，例如是將前景內部所形成之小洞填滿。The optimization unit 13 is coupled to the determination unit 12. In the present embodiment, the optimization unit 13 has a Bayes classifier. When the result of the determination is YES, the optimization unit 13 optimizes the contour of the foreground for the block having the foreground and the background. In implementation, the optimization unit 13 removes the noise on the edge of the foreground according to the color information of the foreground and background of the block, thereby achieving optimization of the contour of the foreground. In addition, the optimization unit 13 can also be optimized for the interior of the foreground, for example by filling a small hole formed inside the foreground.
接著，以下請參照圖2之流程圖並搭配圖1所示，以說明本發明之較佳實施例之影像優化裝置的影像優化方法，其係與例如上述之影像優化裝置1搭配使用。影像優化裝置之影像優化方法的步驟係包含S01～S03。Next, please refer to the flowchart of FIG. 2 and FIG. 1 to illustrate an image optimization method of the image optimization apparatus according to the preferred embodiment of the present invention, which is used in combination with the image optimization apparatus 1 described above. The steps of the image optimization method of the image optimization device include S01 to S03.
步驟S01係分割一影像資料為複數區塊。在本實施例中，分割單元11係接收或選取一影像資料，並將影像資料分割為複數個大小皆相同的區塊。其中，影像資料係為一2D影像之深度圖。Step S01 divides an image data into a plurality of blocks. In this embodiment, the dividing unit 11 receives or selects an image data, and divides the image data into a plurality of blocks of the same size. The image data is a depth map of a 2D image.
步驟S02係判斷複數區塊其中之一區塊是否具有前景與背景。在本實施例中，判斷單元12係針對分割單元11所分割後之影像資料的區塊進行是否同時具有前景與背景的判斷。Step S02 is to determine whether one of the plurality of blocks has a foreground and a background. In the present embodiment, the judging unit 12 judges whether or not the block of the image data divided by the dividing unit 11 has both the foreground and the background.
步驟S03係當判斷結果為是，對該區塊之前景之輪廓進行優化。在本實施例中，優化單元13具有一貝式分類器，並針對具有前景與背景之區塊的前景之輪廓進行優化處理。其中，優化單元13是依據區塊之前景與背景的顏色資訊，而去除前景之邊緣上的雜點，進而達成前景之輪廓的優化。Step S03 is to optimize the contour of the foreground of the block when the judgment result is YES. In the present embodiment, the optimization unit 13 has a shell classifier and performs optimization processing on the contour of the foreground having the foreground and background blocks. The optimization unit 13 removes the noise on the edge of the foreground according to the color information of the foreground and background of the block, thereby achieving optimization of the contour of the foreground.
另外，值得一提的是，優化單元13亦可以針對前景之內部與背景之內部進行優化處理，例如是將前景內部與背景之內部中所形成之小洞，依據所述之小洞周圍之深度值將小洞填滿。In addition, it is worth mentioning that the optimization unit 13 can also optimize the interior of the foreground and the interior of the background, for example, a small hole formed in the interior of the foreground and the interior of the background, according to the depth around the small hole. The value fills the hole.
此外，當判斷單元12之判斷結果為否，判斷單元12將針對另一個區塊進行判斷，直到影像資料之所有區塊皆判斷完成。In addition, when the determination result of the judging unit 12 is no, the judging unit 12 will judge for another block until all the blocks of the image data are judged to be completed.
當影像資料之各區塊均判斷且優化完成之後，影像優化裝置1可將優化後之影像資料顯示於一顯示單元，或透過一輸出單元將優化後之影像資料傳送至另一處理單元，以進行後續之影像處理。After the image data is judged and optimized, the image optimization device 1 may display the optimized image data on a display unit, or transmit the optimized image data to another processing unit through an output unit. Perform subsequent image processing.
綜上所述，因依據本發明之一種影像優化裝置及影像優化方法係藉由將影像資料分割為複數區塊，並判斷各區塊是否具有背景與前景，而對具有前景與背景之區塊之前景的輪廓進行優化處理。從而實現能夠提升所生成之影像的品質，並維持影像處理之速度。In summary, an image optimization apparatus and an image optimization method according to the present invention divide a video data into a plurality of blocks, and determine whether each block has a background and a foreground, and a block having a foreground and a background. The contour of the foreground is optimized. Thereby, the quality of the generated image can be improved, and the speed of image processing can be maintained.
以上所述僅為舉例性，而非為限制性者。任何未脫離本發明之精神與範疇，而對其進行之等效修改或變更，均應包含於後附之申請專利範圍中。The above is intended to be illustrative only and not limiting. Any equivalent modifications or alterations to the spirit and scope of the invention are intended to be included in the scope of the appended claims.
1．．．影像優化裝置1. . . Image optimization device
11．．．分割單元11. . . Split unit
12．．．判斷單元12. . . Judging unit
13．．．優化單元13. . . Optimization unit
S01～S03．．．影像優化方法的步驟S01～S03. . . Image optimization method steps
圖1為依據本發明較佳實施例之一種影像優化裝置的示意圖；以及1 is a schematic diagram of an image optimization apparatus according to a preferred embodiment of the present invention;
圖2為依據本發明較佳實施例之一種影像優化方法的流程圖。2 is a flow chart of an image optimization method in accordance with a preferred embodiment of the present invention.
1．．．影像優化裝置1. . . Image optimization device
11．．．分割單元11. . . Split unit
12．．．判斷單元12. . . Judging unit
13．．．優化單元13. . . Optimization unit
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW099134103A TWI495354B (en) | 2010-10-06 | 2010-10-06 | Image refinement apparatus and image refinement method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW099134103A TWI495354B (en) | 2010-10-06 | 2010-10-06 | Image refinement apparatus and image refinement method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW201216721A TW201216721A (en) | 2012-04-16 |
| TWI495354B true TWI495354B (en) | 2015-08-01 |
Family
ID=46787310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW099134103A TWI495354B (en) | 2010-10-06 | 2010-10-06 | Image refinement apparatus and image refinement method |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI495354B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI585715B (en) * | 2015-12-23 | 2017-06-01 | 粉迷科技股份有限公司 | Method and apparatus for processing border of computer figure to be blended into background image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1246085A2 (en) * | 2001-03-28 | 2002-10-02 | Eastman Kodak Company | Event clustering of images using foreground/background segmentation |
| TWI281126B (en) * | 2002-03-19 | 2007-05-11 | Intervideo Digital Technology | Image detection method based on region |
-
2010
- 2010-10-06 TW TW099134103A patent/TWI495354B/en active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1246085A2 (en) * | 2001-03-28 | 2002-10-02 | Eastman Kodak Company | Event clustering of images using foreground/background segmentation |
| TWI281126B (en) * | 2002-03-19 | 2007-05-11 | Intervideo Digital Technology | Image detection method based on region |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI585715B (en) * | 2015-12-23 | 2017-06-01 | 粉迷科技股份有限公司 | Method and apparatus for processing border of computer figure to be blended into background image |
| US9959637B2 (en) | 2015-12-23 | 2018-05-01 | Framy Inc. | Method and apparatus for processing border of computer figure to be merged into background image |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201216721A (en) | 2012-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI595771B (en) | Optimization method of image depth information and image processing apparatus | |
| US8472746B2 (en) | Fast depth map generation for 2D to 3D conversion | |
| US9621869B2 (en) | System and method for rendering affected pixels | |
| KR102146560B1 (en) | Method and apparatus for adjusting image | |
| TWI512680B (en) | System and method for optimizing image quality in a digital camera | |
| US20100315488A1 (en) | Conversion device and method converting a two dimensional image to a three dimensional image | |
| US20120001902A1 (en) | Apparatus and method for bidirectionally inpainting occlusion area based on predicted volume | |
| US20130300735A1 (en) | Mesh boundary smoothing | |
| US9747694B2 (en) | Method and apparatus for optimizing depth information | |
| JP2004164624A (en) | Method and apparatus for low depth of field image segmentation | |
| CN113160161B (en) | Method and device for detecting defects at edge of target | |
| JP6703314B2 (en) | Focus detection | |
| JP2018515743A (en) | Method and apparatus for depth image processing | |
| US20140354645A1 (en) | Relightable texture for use in rendering an image | |
| TW201606705A (en) | Method and image processing apparatus for image visibility restoration using fisher's linear discriminant based dual dark channel prior | |
| CN111179189A (en) | Image processing method and device based on generation countermeasure network GAN, electronic equipment and storage medium | |
| EP3110149B1 (en) | A system and a method for depth-image-based rendering | |
| TW201710998A (en) | Method and apparatus for determining a depth map for an image | |
| CN110599453A (en) | Panel defect detection method and device based on image fusion and equipment terminal | |
| CN110728692A (en) | Image edge detection method based on Scharr operator improvement | |
| CN104378619A (en) | Rapid and efficient hole filling algorithm based on foreground and background gradient transition | |
| KR20220127718A (en) | Electronic device and controlling method of electronic device | |
| TW201541937A (en) | Method and apparatus for generating depth information | |
| CN106028020B (en) | A kind of virtual perspective image cavity complementing method based on multi-direction prediction | |
| CN103942756A (en) | Post-processing filtering method for depth map |