TWI837469B - Method and device for removeing the background - Google Patents

Abstract

The present invention provides a method and system for removing background of an image, wherein a image acquiring device is utilized to acquire an initial image, an operation interface in the image acquiring device is utilized to display the initial image, a plurality of removal targets are selected from the initial image through the operation interface and a plurality of positioning information respectively corresponding to each removal target are generated, a processing module is utilized to determine a plurality of color features respectively corresponding to the plurality of positioning information in the initial image, and finally, the background corresponding to the color features in the initial image are removed by the processing module.

Description

Background removal method and background removal device

The present invention relates to an image processing technology, and more particularly to a background removal method and a background removal device for removing image background.

The development of the Internet and the advancement of audio-visual equipment have driven the development of many industries. In the audio-visual multimedia industry, audio-visual playback platforms provide amateurs with unique audio-visual content to gain the love and attention of viewers, thus creating a new generation of audio-visual industry model dominated by Internet celebrities or YouTubers. In this new type of online audio-visual industry, in order to attract subscribers, Internet celebrities or YouTubers need to invest in content planning and design. Audio-visual editing and processing, such as editing from static images to dynamic images, or adding special effects to images, are also important links to help increase subscription rates.

Background removal is an indispensable technique in video editing. Background removal means that part of the image subject is retained and the rest is removed. Background removal allows the image subject to change the background environment at will, increasing the diversity of the image or picture.

In practice, when users need to replace the background environment, they generally need to prepare a monochrome screen (such as green or blue cloth) as the background screen, and then shoot to obtain an image. Then, input the image into the image processing software, select the foreground part to be retained, and then delete the unselected background part and fill it with a transparent background before saving. Finally, the background-removed file is superimposed with the background to be replaced to obtain a new image, achieving the image effect of removing the background and replacing the background.

In the application of this technology, users must prepare their own blue or green curtains, and after shooting, they must upload the images to image processing equipment using another program before post-processing, which is quite inconvenient for users. In addition, because the curtains are affected by the light, the colors presented in the images or pictures are not uniform, so it is difficult for users to accurately select the color to be removed, and they need to adjust repeatedly, affecting the efficiency of background removal.

Therefore, in view of the shortcomings of the prior art, a background removal method and a background removal device are proposed to improve the shortcomings of the above-mentioned conventional means.

The present invention provides a background removal method and device. After a user captures an image with a handheld device, the user is allowed to select any multiple groups of areas through an operation interface. The color characteristics of the selected areas are then automatically detected, and the background removal is performed in real time according to the color characteristics, thereby achieving an effective background removal operation effect corresponding to the on-site light source.

The background removal method and device of the present invention can find the corresponding area for background removal according to the color characteristics of the area selected by the user immediately after capturing an image or picture. By calculating the color characteristics of the selected area, the problem of color difference caused by reflection caused by the angle of the captured image affecting the background removal effect can be solved. In addition, the method of directly selecting the area by the user can further improve the background removal image processing efficiency, which can be applied to live broadcasts, meetings or speeches.

The present invention provides a background removal method, which uses an image capture device to capture an initial image, and then provides an operation interface in the image capture device to display the initial image. Thereafter, multiple targets to be background removed are selected in the initial image through the operation interface, and multiple position information of the targets to be background removed is generated. Then, a processing unit determines multiple color features corresponding to the position information from the initial image. Finally, the processing unit removes the background of the initial image based on the obtained color features.

In one embodiment, the present invention provides a background removal device, including an operation interface, an image receiving port, and a processing unit, wherein the image receiving port receives an initial image, and a plurality of background removal targets are selected from the initial image through the operation interface, and a plurality of position information of the background removal targets is generated. In addition, the processing unit is electrically connected to the image receiving port, and the processing unit determines a plurality of color features corresponding to the position information from the initial image, and removes the background of the initial image according to the color features.

Various exemplary embodiments will be more fully described below with reference to the accompanying drawings, some of which are shown in the accompanying drawings. However, the concepts of the present invention may be embodied in many different forms and should not be construed as limited to the exemplary embodiments described herein. Rather, these exemplary embodiments are provided so that the present invention will be detailed and complete, and will fully convey the scope of the concepts of the present invention to those skilled in the art. Similar numbers always indicate similar elements. The background removal method and background removal device will be described below with a variety of embodiments in conjunction with drawings, however, the following embodiments are not intended to limit the present invention.

Please refer to FIG. 1A, which is a schematic diagram of an embodiment of the background removal device of the present invention. The background removal device 3 includes an operation interface 300 and a processing unit 32. In the present embodiment, the operation interface 300 is provided on the image capture device 30. In the present embodiment, the image capture device 30 is a portable device, such as a smart phone, a camera, or a tablet computer, but is not limited thereto. The image capture device 30 also has a computing and processing capability, and can generate a user operation interface 300 by executing an application or an APP. It should be noted that the operation interface 300 can provide the user with the ability to perform specific functions on an image or picture, for example, selecting a specific position or area on the image or picture so that the application or APP can obtain information about the image characteristics of the selected specific position or area through calculation, such as brightness, contrast or color. In this embodiment, the image capturing device 30 captures the image of the subject 4 and its background environment 40 via a built-in lens 30a to generate an initial image P0.

In the present embodiment, the processing unit 32 is disposed in the background removal device 3. In the present embodiment, the background removal device 3 is a live broadcast device with a network communication function, which has a first connection interface 330 electrically connected to the image capture device 30. The first connection interface 330 in the present embodiment is an image receiving port, which can be a wireless communication interface, such as: Wi-Fi wireless communication or Bluetooth wireless communication interface, or can be a wired interface, such as: various forms of USB connection interfaces, HDMI, DVI, etc., but not limited to this. The image capture device 30 can transmit the captured initial image P0 to the background removal device 3 through the first connection interface 330. In addition, the first connection interface 330 can also transmit the background removal information (eg, location information of the background removal target) obtained after operating the operation interface 300 to the background removal device 3, and then transmit it to the processing unit 32 to remove the background of the initial image P0.

In this embodiment, the processing unit 32 further includes a pixel conversion module 320, an image analysis module 321, and a background processing module 322. The pixel conversion module 320 determines a plurality of corresponding color features on the initial image P0 according to the position of the target to be removed from the initial image P0 selected by the user on the operation interface 300. The target to be removed can be a single point or a range of a specific size. The image analysis module 321 combines these color features into a color range. The background processing module 322 removes the colors with these color features from the initial image according to the color range.

The following uses the embodiments shown in FIG. 1A and FIG. 3A and FIG. 3B to illustrate the method of selecting a position from an initial image. An initial image P0 is displayed on the operation interface 300 of the image capture device 30, and the user selects the target 90a~90e to be background-removed on the initial image P0. It should be noted that the selection method is to select on the display screen of the image capture device 30 by touch. Each target 90a~90e to be background-removed can be a single pixel or a collection of multiple pixels. The image capture device 30 transmits the position information of the selected target 90a~90e and the initial image P0 to the background-removing device 3.

The background removal device 3 receives the position information and transmits it to the pixel conversion module 320. The pixel conversion module 320 obtains the color characteristics corresponding to the selected targets 90a~90e on the initial image P0 based on the received position information. In this embodiment, the color characteristics are YUV information, but are not limited to this. For example, RGB information or CMYK information is also acceptable. The following is explained using YUV information. The YUV at position 90a is (149,53,19), the YUV at 90b is (122,41,21), the YUV at 90c is (110,43,30), the YUV at 90d is (166,69,42), the YUV at 90e is (160,60,30), and so on. Then, the image analysis module 321 combines the color features corresponding to the selected targets 90a-90e into a color range R1 (Max: YUV (166, 69, 42), MinYUV (110, 41, 21), that is, Y is between 166-110, U is between 69-41, and V is between 42-21. Afterwards, the color range R1 is transmitted to the background processing module 322, and the background processing module 322 removes the colors in the initial image P0 that meet the color range R1 according to the color range R1.

In addition to the aforementioned method, in another embodiment, the image analysis module 321 defines a plurality of color ranges for each color feature. The background processing module 322 removes the colors having the color features in the initial image P0 according to the color ranges. In this embodiment, as shown in FIG. 1A , FIG. 3C and FIG. 3D , the operation interface 300 of the image capture device 30 displays the initial image P0, and the user selects the target 90f~90h to be removed from the initial image. In this embodiment, each target 90f~90h is a range drawn by the user on the initial image P0 using a touch screen. The image capture device 30 transmits the position information (e.g., range boundary) of the selected target 90f~90h to the background removal device 3, and then the pixel conversion module 320 obtains the color features corresponding to the range boundary of the selected target 90f~90h on the initial image P0 according to the received position information. Then the image analysis module 321 determines the corresponding color range R2~R4 according to the color features corresponding to the range boundary of the selected target 90f~90h, for example: YUVmax(149,53,19), YUVmin(138,51,17) of R2; YUVmax(122,41,21), YUVmin(115,25,18) of R3, and YUVmax(110,43,30), YUVmin(108,32,25) of R4. It should be noted that the method of generating the color range R2-R4 is as described in the previous Figures 3A and 3B, which will not be repeated here. Afterwards, the color range R2-R4 is transmitted to the background processing module 322, and the background processing module 322 removes the colors in the initial image P0 that match the color range R2-R4 according to the color range R2-R4.

In addition to the aforementioned method, in another embodiment, the image analysis module 321 defines a plurality of color values for each color feature. The background processing module 322 removes the pixels having the color features in the second initial image according to the color range. In one embodiment, as shown in FIG. 3E and FIG. 3F , the operation interface 300 of the image capture device 30 displays the initial image P0, and the user selects the target 90i~90l to be removed from the initial image P0. In this embodiment, each target 90i~90l is a position selected by the user in the initial image P0 by touch. The image capture device 30 transmits the position information of the selected target 90i~90l to the background removal device 3, and then the pixel conversion module 320 obtains the color features corresponding to the selected targets (pixels) 90i~90l on the initial image P0 according to the received position information. Then the image analysis module 321 determines the corresponding color values Ca~Cd according to the color features of the selected targets 90i~90l, for example: the YUV value Ca of position 90i is (149,53,19), the YUV value Cb of 90j is (122,41,21), the YUV value Cc of 90k is (110,43,30), and the YUV value Cd of 90l is (166,69,42). Afterwards, the color values Ca~Cd are transmitted to the background processing module 322, and the background processing module 322 removes the colors in the initial image P0 that match the color values Ca~Cd according to the color values Ca~Cd.

Returning to FIG. 1A , the processing unit 32 further has an image superimposition module 323, which superimposes the background-removed image after the background processing module 322 removes the background and at least one background image to form a superimposed image. The background image can be a picture or a dynamic image. The background image can be stored in the background removal device 3 in advance, or obtained by connecting to the cloud storage device 7 via the network. In addition, the background removal device 3 further has a third connection interface 332, an image output module 34, and a fourth connection interface 333, wherein the third connection interface 332 is electrically connected to the electronic device 5 as an interface for outputting audio and video signals, and transmits the multimedia information obtained after processing, such as audio and video information, image information, or image information, to the electronic device 5. The electronic device 5 can be a smart phone, a laptop, or a desktop computer, etc. The third connection interface 332 can be a USB interface of various forms, an audio and video interface such as HDMI or DVI, or a wireless or wired network interface, but is not limited thereto. In one embodiment, the electronic device 5 has an application that can upload the multimedia information output by the audio and video processing device 5 to a remote network audio and video streaming platform or a cloud storage device.

The image output module 34 is electrically connected to the image superposition module 323 and the fourth connection interface 333. The fourth connection interface 333 is electrically connected to the display device 6. The fourth connection interface 333 can be various forms of USB interfaces, HDMI or DVI and other audio and video interfaces. Through the image output module 34 and the fourth connection interface 333, the user can directly see the image effect after post-processing through the display 6. In another embodiment, the background removal device 3 can further include an audio connection interface 334 for electrically connecting to an audio input device. The audio input device can be a microphone or other broadcasting equipment. It should be noted that although the background removal device 3 and the image capture device 30 are two separate devices in this embodiment, the background removal device 3 and the image capture device 30 can be integrated with each other in another embodiment. For example, the image capturing device 30 may be a smart phone, wherein the background removal device 3 is integrated into the image capturing device 30 .

Please refer to FIG. 1B , which is a schematic diagram of another embodiment of the background removal device of the present invention. The background removal device 3 includes a first image capture device 30 ′, a second image capture device 31 and a processing unit 32. In the present embodiment, the first image capture device 30 ′ is a portable image capture device, such as a smart phone, a camera, or a tablet computer, but is not limited thereto. The first image capture device 30 ′ also has a computing and processing capability, and can generate a user operation interface 300 by executing an application or APP. It should be noted that the operation interface 300 can provide the user with the ability to perform specific functions on an image or picture, such as selecting a specific position or area on the image or picture so that the application or APP can obtain information about the image characteristics of the selected specific position or area through calculation, such as brightness, contrast or color. In this embodiment, the first image capturing device 30' captures the image of the subject 4 and its background environment 40 via the built-in lens 30a to generate a first initial image P1.

In this embodiment, the second image capture device 31 is an external photographic device, such as a camera or a video recorder, but is not limited thereto, and may also be a smart phone, a camera, or a tablet computer. In another embodiment, as shown in FIG1C , the second image capture device 31 may be integrated into the background removal device 3. Returning to FIG1B , the second image capture device 31 captures the image of the subject 4 and its background environment 40 through the lens 31a to generate a second initial image.

In this embodiment, the processing unit 32 is disposed in the background removal device 3. In this embodiment, the background removal device 3 is a live broadcast device with a network communication function, which has a first connection interface 330 and a second connection interface 331, which are electrically connected to the first image capture device 30' and the second image capture device 31 respectively. The first connection interface 330 and the second connection interface 331 can be wireless communication interfaces, such as Wi-Fi wireless communication or Bluetooth wireless communication interfaces, or can be wired interfaces, such as various types of USB connection interfaces, HDMI, DVI, etc., but are not limited thereto.

The first image capture device 30' can transmit the background removal information (e.g., the location information of the background removal target) obtained after the operation of the operation interface 300 to the background removal device 3 through the first connection interface 330, and then transmit it to the processing unit 32 for image processing. Similarly, the second image capture device 31 transmits the captured second initial image to the background removal device 3 through the second connection interface 331 and then to the processing unit 32. The processing unit 32 removes the background of the second initial image according to the background removal information determined by the user.

In this embodiment, the processing unit 32 further includes a pixel conversion module 320, an image analysis module 321, and a background processing module 322. The pixel conversion module 320 determines a plurality of corresponding color features on the second initial image according to the position of the target to be removed from the first initial image selected by the user on the operation interface 300. The target to be removed from the background can be a single point or a range of a specific size. The image analysis module 321 combines these color features into a color range. The background processing module 322 removes the pixels with these color features in the second initial image according to the color range.

The following is an example shown in FIG. 1B , FIG. 3G and FIG. 3H . A first initial image P1 is displayed on the operation interface 300 of the first image capture device 30 ′, and the user selects the target 90a~90e to be background-removed on the first initial image P1. It should be noted that the selection is made by touch on the display screen of the first image capture device 30 ′. Each target 90a~90e to be background-removed can be a single pixel or a collection of multiple pixels. The first image capture device 30 ′ transmits the position information of the selected target 90a~90e to the background-removing device 3.

The background removal device 3 receives the position information and transmits it to the pixel conversion module 320. The pixel conversion module 320 obtains the color characteristics corresponding to the selected targets 90a~90e on the second initial image P2 based on the received position information. In this embodiment, the color characteristics are YUV information, but are not limited to this. For example, RGB information or CMYK information is also acceptable. The following is explained using YUV information. The YUV at position 90a is (149,53,19), the YUV at 90b is (122,41,21), the YUV at 90c is (110,43,30), the YUV at 90d is (166,69,42), the YUV at 90e is (160,60,30), etc. Then, the image analysis module 321 combines the color features corresponding to the selected targets 90a-90e into a color range R1 (Max: YUV (166, 69, 42), MinYUV (110, 41, 21), that is, Y is between 166-110, U is between 69-41, and V is between 42-21. Afterwards, the color range R1 is transmitted to the background processing module 322, and the background processing module 322 removes the colors in the second initial image P2 that meet the color range R1 according to the color range R1.

In addition to the aforementioned method, in another embodiment, the image analysis module 321 defines a plurality of color ranges for each color feature. The background processing module 322 removes the pixels having the color features in the second initial image P1 according to the color range. In this embodiment, as shown in FIG. 1B , FIG. 3I and FIG. 3J , the operation interface 300 of the first image capture device 30 ′ displays the first initial image P1, and the user selects the target 90f~90h to be removed from the background on the initial image. In this embodiment, each target 90f~90h is a range drawn by the user on the first initial image P1 using a touch screen. The first image capture device 30' transmits the position information (range boundary) of the selected target 90f~90h to the background removal device 3, and then the pixel conversion module 320 obtains the color features corresponding to the range boundary of the selected target 90f~90h on the second initial image P2 according to the received position information, and then the image analysis module 321 determines the corresponding color range R2~R4 according to the color features corresponding to the range boundary of the selected target 90f~90h, for example: R2's YUVmax(149,53,19), YUVmin(138,51,17); R3's YUVmax(122,41,21), YUVmin(115,25,18), R4's YUVmax(110,43,30), YUVmin(108,32,25). It should be noted that the method of generating the color range R2-R4 is as described in the previous Figures 3G and 3H, which will not be repeated here. Afterwards, the color range R2-R4 is transmitted to the background processing module 322, and the background processing module 322 removes the colors in the second initial image P2 that match the color range R2-R4 according to the color range R2-R4.

In addition to the aforementioned method, in another embodiment, the image analysis module 321 defines a plurality of color values for each color feature. The background processing module 322 removes the pixels having the color features in the second initial image according to the color range. In one embodiment, as shown in FIG. 3K and FIG. 3L , the operation interface 300 of the first image capture device 30 ′ displays the first initial image P1, and the user selects the target 90i~90l to be removed from the initial image. In this embodiment, each target 90i~90l is a position selected by the user using a touch screen in the first initial image P1. The first image capture device 30' transmits the position information of the selected target 90i~90l to the background removal device 3, and then the pixel conversion module 320 obtains the color features corresponding to the selected targets (pixels) 90i~90l on the second initial image P2 according to the received position information, and then the image analysis module 321 determines the corresponding color values Ca~Cd according to the color features of the selected targets 90i~90l, for example: the YUV value Ca of position 90i is (149,53,19), the YUV value Cb of 90j is (122,41,21), the YUV value Cc of 90k is (110,43,30), and the YUV value Cd of 90l is (166,69,42). Afterwards, the color values Ca~Cd are transmitted to the background processing module 322, and the background processing module 322 removes the colors in the second initial image P2 that match the color values Ca~Cd according to the color values Ca~Cd.

Returning to FIG. 1B , the processing unit 32 further has an image superimposition module 323, which superimposes the background-removed image after the background processing module 322 removes the background and at least one background image to form a superimposed image. The background image can be a picture or a dynamic image. The background image can be stored in the background removal device 3 in advance, or obtained by connecting to the cloud storage device 7 via the network. In addition, the background removal device 3 further has a third connection interface 332, an image output module 34, and a fourth connection interface 333, wherein the third connection interface 332 is electrically connected to the electronic device 5 as an interface for outputting audio and video signals, and transmits the multimedia information obtained after processing, such as audio and video information, image information, or image information, to the electronic device 5. The electronic device 5 can be a smart phone, a laptop, or a desktop computer, etc. The third connection interface 332 can be a USB interface of various forms, an audio and video interface such as HDMI or DVI, or a wireless or wired network interface, but is not limited thereto. In one embodiment, the electronic device 5 has an application that can upload the multimedia information output by the audio and video processing device 5 to a remote network audio and video streaming platform or a cloud storage device.

The image output module 34 is electrically connected to the image superimposition module 323 and the fourth connection interface 333. The fourth connection interface 333 is electrically connected to the display device 6. The fourth connection interface 333 can be an interface in various forms of USB, an audio and video interface such as HDMI or DVI. Through the image output module 34 and the fourth connection interface 333, the user can directly see the image effect after post-processing through the display 6. In another embodiment, the background removal device 3 can further include an audio connection interface 334 for electrically connecting to an audio input device. The audio input device can be a microphone or other broadcasting equipment.

In one embodiment, the background removal method 2 of FIG. 2A is performed by the background removal device 3 of FIG. 1A. First, the initial image P0 is captured by the image capture device 30 in step 20. Then, step 21 is performed to provide an operation interface in the image capture device 30 to display the initial image P0. Then, in step 22, one or more background removal targets are selected from the initial image P0 through the operation interface 300, and a plurality of position information corresponding to the selected background removal targets is generated. In step 22, any of the three methods shown in FIG. 3A, FIG. 3C or FIG. 3E can be used to achieve the effect of selecting the target position. Afterwards, step 23 is performed to determine a plurality of color features corresponding to the position information from the initial image P0 using the processing unit 32. In step 23, any of the three methods shown in FIG. 3B, FIG. 3D or FIG. 3F can be used to achieve the effect of obtaining color features. Finally, step 24 is performed, and the processing unit 32 removes the background of the initial image P0 according to the color features. In the background removal process, the processing unit 32 removes the color of each pixel in the initial image P0 that meets the color feature, and makes the area where the color is removed transparent to form a background-removed image. The background-removed image obtained in step 24 can be further superimposed with another target image or picture through the image superposition module 323 to form a superimposed image.

In another embodiment, the background removal method 2a of FIG. 2B is performed by the background removal device 3 of FIG. 1B or FIG. 1C. First, the first initial image is captured by the first image capture device 30' and the second initial image is captured by the second image capture device 31 in step 20a. Then, step 21a is performed to provide an operation interface in the first image capture device 30' to display the first initial image P1. Then, in step 22a, one or more background removal targets are selected in the first initial image P1 through the operation interface 300, and a plurality of position information corresponding to the selected background removal targets is generated. In step 22a, any of the three methods shown in FIG. 3G, FIG. 3I or FIG. 3K can be used to achieve the effect of selecting the target position. Afterwards, step 23a is performed to use the processing unit 32 to determine a plurality of color features corresponding to the position information from the second initial image. In step 23a, any of the three methods as shown in Figure 3H, Figure 3J or Figure 3L can be used to achieve the effect of obtaining color features. Finally, step 24a is performed, and the processing unit 32 removes the background of the second initial image based on the color features. In the background removal process, the processing unit 32 removes the color of each pixel in the second initial image that meets the color feature, and makes the area where the color is removed transparent to form a background-removed image. The background-removed image obtained in step 24a can be further superimposed with another target image or picture through the image superposition module 323 to form a superimposed image.

In summary, the background removal method and background removal device of the present invention can find multiple corresponding color features according to multiple regions selected by the user immediately after capturing an image or picture to remove the background. By selecting the calculation of the color features of multiple regions, the problem of color difference caused by reflection caused by the angle of capturing the image and affecting the background removal effect can be solved. In addition, the method of directly selecting the region by the user can further improve the background removal image processing efficiency.

The above only records the preferred implementation methods or examples of the technical means adopted by the present invention to solve the problem, and is not used to limit the scope of implementation of the present invention. That is, all equivalent changes and modifications that are consistent with the scope of the patent application of the present invention or made according to the scope of the patent of the present invention are covered by the scope of the patent of the present invention.

2: Background removal method 20~24: Steps 3: Background removal device 30: Image capture device 30’: First image capture device 30a: Lens 300: Operation interface 31: Second image capture device 31a: Lens 32: Processing unit 320: Pixel conversion module 321: Image analysis module 322: Background processing module 323: Image superposition module 330: First connection interface 331: Second connection interface 332: Third connection interface 333: Fourth connection interface 334: Audio connection interface 34: Image output module 4: Subject 40: Background environment 5: Electronic device 6: Display device 7: Cloud storage device P0: Initial image P1: First initial image P2: Second initial image 90a~90l: Target R1~R4: Range Ca~Cd: Color value

FIG. 1A is a schematic diagram of an embodiment of the background removal device of the present invention. FIG. 1B and FIG. 1C are schematic diagrams of other embodiments of the background removal device of the present invention. FIG. 2A is a schematic diagram of a flow chart of an embodiment of the background removal method of the present invention. FIG. 2B is a schematic diagram of a flow chart of an embodiment of the background removal method of the present invention. FIG. 3A to FIG. 3L are schematic diagrams of different embodiments of selecting color features of the present invention.

2: Background removal method

20~24: Steps

Claims (10)

A background removal method is applicable to the background removal situation where the color of the background curtain is uneven. The background removal method includes: capturing an initial image by an image capture device; providing an operation interface in the image capture device to display the initial image; selecting a plurality of background removal targets in the initial image by the operation interface, and generating a plurality of position information of the background removal targets; providing a processing unit to determine a plurality of color features corresponding to the position information from the initial image; and the processing unit removes the background of the initial image according to the color features. The background removal method as described in claim 1, wherein the image capture device includes a first image capture device and a second image capture device, the initial image includes a first initial image and a second initial image, wherein the first initial image and the second initial image are captured by the first image capture device and the second image capture device respectively, the operation interface is provided on the first image capture device and displays the first initial image, wherein the background removal targets are selected from the first initial image through the operation interface, wherein the color features are determined from the second initial image by the processing unit, and the background removal is performed on the second initial image according to the color features. The background removal method as described in claim 2, wherein the background removal of the second initial image according to the color features further includes combining the color features into a color range, and removing the pixels having the color features in the second initial image according to the color range. The background removal method as described in claim 2, wherein the background removal of the second initial image according to the color features further includes defining a plurality of color ranges from the color features, and removing pixels having the color features in the second initial image according to the color ranges. The background removal method as described in claim 2, wherein the background removal of the second initial image according to the color features further includes defining a plurality of color values from the color features, and removing the pixels having the color features in the second initial image according to the color values. A background removal device is suitable for removing background when the background curtain has uneven color. The background removal device includes: an operation interface; an image receiving port, which receives an initial image, selects a plurality of background removal targets in the initial image through the operation interface, and generates a plurality of position information of the background removal targets; and a processing unit, which is electrically connected to the image receiving port, determines a plurality of color features corresponding to the position information from the initial image, and removes the background of the initial image according to the color features. The background removal device as described in claim 6, wherein the image receiving port includes a first image receiving port and a second image receiving port, the initial image includes a first initial image and a second initial image, the background removal targets are selected from the first initial image through the operation interface, the processing unit determines the color features from the second initial image, and removes the background of the second initial image according to the color features. The background removal device as described in claim 7, wherein the processing unit further includes: a pixel conversion module, which determines a plurality of corresponding color features on the second initial image according to the selected background removal targets; an image analysis module, which combines the color features into a color range; and a background processing module, which removes pixels having the color features in the second initial image according to the color range. The background removal device as described in claim 7, wherein the processing unit further includes: a pixel conversion module, which determines a plurality of corresponding color features on the second initial image according to the selected background removal targets; an image analysis module, which defines a plurality of color ranges from the color features; and a background processing module, which removes pixels having the color features in the second initial image according to the color ranges. The background removal device as described in claim 7, wherein the processing unit further includes: a pixel conversion module, which determines the corresponding color features on the second initial image according to the selected background removal targets; an image analysis module, which defines a plurality of color values from the color features; and a background processing module, which removes the pixels having the color features in the second initial image according to the color values.

Images