WO2013113191A1 - Method and system for rapid video image transmission - Google Patents

Description

 Method and system for fast transmission of video images

 The present invention relates to the field of video image processing, and in particular, to a method and system for fast transmission of video images. Background technique

 At present, with the continuous development of network technology, wireless networks have become popular in smart home appliances. Various terminal devices in smart home appliances can basically access the Internet and browse information through wireless technology. However, with the continuous improvement of smart home appliances, people have also put forward higher requirements, that is, to achieve timely screen sharing. The user browsing the information page on one terminal device can be timely transmitted to another terminal device that is closest to itself, which can greatly satisfy the user's needs and is convenient for the user to use. At present, there are several main sharing methods and technologies in multi-screen sharing technology: split screen screen sharing, server video streaming screen sharing, and screen fast capture technology screen sharing.

 However, the above sharing methods have the following deficiencies:

 Split screen screen sharing requires a VGA cable for wire connection, because the length of the cable is limited, thus limiting the distance between the two connected devices;

 Although the server video stream screen sharing solves the problem of the wiring length, one server device is only for one client device, and the problem that the playback content of one server device is simultaneously displayed on multiple client devices cannot be realized;

 Screen capture technology screen sharing handles full-screen data every time, the amount of data is huge, and a high machine configuration is required to quickly process the compression and transmission of data, which takes a long time. Summary of the invention

In view of this, the object of the present invention is to provide a method and system for fast transmission of video images, which saves bandwidth consumption and realizes images of multi-screen sharing of terminals in a smart home. Fast transfer.

 The method for fast image transmission provided by the invention comprises the steps of:

 A. The video sending end performs an image screen capture on its screen and caches it;

 B. determining an area of the screen image that is different from the buffered image of the previous frame of the screen image, and transmitting the image of the area;

 C. The video receiving end receives the area image and combines it with the buffered image as the received image, and buffers the combined image.

 From the above, only the areas where the image pixels are different are transmitted, which saves the bandwidth occupation, and realizes the fast transmission of the image when the multi-screen sharing of the terminal in the smart home.

 Optionally, the determining step of the pixel in step B includes:

 It is determined that the pixel is different from the pixel value of the previous frame image when the deviation of the position pixel attribute value exceeds a certain value.

 From the above, the judgment of the pixel difference is realized by judging the pixel attribute deviation.

 Optionally, the determining step of the area in step B includes:

 Rows and columns different from the image pixels of the previous frame are determined in units of rows and columns, respectively, and the regions formed by the intersection of the rows and columns are the different regions.

 From above, by determining different implementations of row and column pixels, an area that needs to be screened is determined, thereby avoiding full-screen screen capture and achieving bandwidth saving.

 Optionally, the step of determining a row and a column different from the pixel of the previous frame image includes:

 Performing a difference operation between the screen image and each pixel of the previous frame image to obtain a grayscale difference value, and storing the operation result as an array;

 Performing projections of the X-axis and the Y-axis on the array and respectively calculating the sum of the gradation differences of the columns and the rows;

 The rows and columns for which the sum of the gradation differences of the respective columns and the rows are greater than a threshold are respectively determined as the determined rows and columns.

 From the above, by comparing the gradations of the adjacent two frames of images, it is determined that the positions of the start and start columns of the image pixels are different.

Optionally, respectively, determining that the sum of the gray level differences of each column and each row is greater than a threshold value Includes:

 A step of judging that the sum of the successive set number of column and line gradation differences is greater than a threshold.

 From the above, the positions of the end line and the end column in which the image pixels are different are determined, thereby determining the regions in which the adjacent two image pixels are different.

 Optionally, it also includes:

 Counting regions with different pixels according to screenshot images of a certain number of frames in a certain period of time;

 The screen shot in step A is: Screen capture of the statistical area.

 From above, by setting the time interval of the screen shot image and the number of screen shots in the time interval, the screen cut rate is reduced, thereby further reducing the bandwidth occupation.

 Preferably, the method further includes:

 When it is judged that the application of the screen image of the video transmitting end outside the area screen shot is activated, the full screen image is taken as the screen shot of the step A.

 The application being activated includes: determining that a mouse, keyboard, or other application window is activated.

 From above, by setting the activation state, when not activated, the screen is not screened, and the time interval for increasing the screen capture is increased, thereby further saving the bandwidth occupation.

 Optionally, the attribute value includes a color attribute and a color depth attribute of the pixel. The invention also provides a system for fast transmission of video images, comprising a video transmitting end and a video receiving end,

 The video sending end includes:

 a screen capture module, configured to perform an image screen capture on a screen of the video sending end;

 a video sending end storage module, configured to cache the image of each frame cut by the screen capture module;

 a comparison module, configured to compare the current screen capture image with a previous frame image buffered by the storage module to determine a different area of the image pixel;

 The sending module is configured to output different areas of the image pixels in a wireless form.

The video receiver includes: a receiving module, configured to receive a different area of an image pixel sent by a sending module of the video sending end;

 An integration module, configured to combine different regions of the received image pixels with the cached image as the received image;

 The video receiving end storage module is configured to cache the integrated image of the integrated module.

 From the above, only the areas where the image pixels are different are transmitted, which saves the bandwidth occupation, and realizes the fast transmission of the image when the multi-screen sharing of the terminal in the smart home. DRAWINGS

 1 is a flowchart of a method for quickly transmitting video images according to the present invention;

 2 is a flowchart of a method for establishing a wireless shared connection between a video sending end and a video receiving end in step S10 of the present invention;

 3 is a flowchart of a method for intercepting a shared video file by a video sending end in step S20 of the present invention;

 4 is a flow chart of a method for obtaining a grayscale deviation region by comparing a current frame image with a previous frame image in step S40 of the present invention;

 FIG. 5 is a schematic structural diagram of a system for fast transmission of video images according to the present invention. detailed description

 The main principle of the present invention is: a wireless sharing connection is established between the video transmitting end and the video receiving end, and when the video transmitting end encodes the video image to the video receiving end through the screen capture, the video transmitting end simultaneously caches the image, and The buffered image is subjected to filtering processing. Comparing the current frame image with the image of the previous frame of the buffer, acquiring the grayscale deviation region (ie, the region different from the image pixel) in the two frame images, and encoding only the image portion corresponding to the grayscale deviation region. send. The video receiving end integrates the grayscale deviation area into the previous frame image and buffers it. Thereby, the bandwidth consumption and the encoding speed are saved, and the image fast transmission when the multi-screen sharing of the terminal in the smart home is realized.

The specific method and system for fast image transmission according to the present invention will be described below with reference to FIGS. The embodiment will be described in detail.

 As shown in FIG. 1, the method for fast image transmission includes the following steps:

 Step S10: Establish a wireless shared connection between the video sending end and the video receiving end. As shown in FIG. 2, the step S10 includes:

 Step S101: The video sending end sends the video sharing request information to the video receiving end through a wireless communication network, such as a WIFI network.

 Step S102: The video receiving end receives the video sharing request information, determines whether the video sharing is acceptable according to the value of the shared identification bit stored by the video receiving end, and feeds back the corresponding information according to the information.

 When it is detected that the shared identifier bit is set to 1, it indicates that the video receiving end is sharing video with other video transmitting ends, and can no longer share with the video receiving end, and proceeds to step S103; when the shared identification bit is set to 0. , indicating that the video receiving end does not share video with other video transmitting ends, and is in an idle state. At this time, video sharing with the video receiving end may be performed, and the process proceeds to step S104.

 Step S103: The video receiving end sends a feedback feedback rejection request message to the video sending end, and the process ends.

 Step S104: The video receiving end feeds back the information that allows the sharing request, and establishes a video sharing connection.

 After the wireless sharing connection is established between the video sending end and the video receiving end, the video receiving end stores the address information of the video sending end, and changes the shared identification bit to 1.

 From above, the video transmitting end and the video receiving end complete the establishment of the video sharing connection.

Step S20: The video sending end performs a frame-by-frame screen capture on its entire screen, that is, the desktop. Common screenshots include GDI functions, DirectX functions, or Windows Media API functions. In this embodiment, the GDI function is used to perform screen capture on the desktop. The GDI is based on "the desktop is also a window, and the desktop also has a window handle (HWND)", and the time for one screen capture is only 4 sec. GDI is an executable program that accepts access requests from Windows applications, but the application does not have direct access to the output device (screen), so the operation of the output device is done through the device context (DC, Device Contex). DC is a data structure in Windows that contains the GDI functions needed. All description fields about the type of output device and the status of the display interface. The image captured by the GDI function is not directly output by the output device, but the image is copied to the DC. Each window on the screen corresponds to a DC, and the operation of the DC is reflected on its corresponding screen window.

 As shown in FIG. 3, the screen capture step in step S20 includes:

 Step S201: Acquire a window of the current screen desktop.

 When taking a desktop screenshot, first point the pointer to the current screen desktop window. Therefore, the function needs to be called to obtain the handle of the desktop window. In this embodiment, the window handle of the desktop is obtained by calling the GetDesktopWindow function.

 Step S202: Obtain a DC of a current screen desktop window.

 This step calls the GetDC function to get the DC of the desktop window, which is used to get the contents of the desktop window. This includes the bitmap image of the desktop window corresponding to the entire desktop and the width and height of the current entire desktop window.

 Step S203: Create a DC and a bitmap compatible with the window DC, and select the bitmap into a compatible DC.

 In this step, the CreateCompatible function is first created by creating the compatible DC ( CreateCompatibleDC ) and the bitmap ( CreateCompatibleBitmap ).

The DC is used to get the pixel value of the entire desktop window. The bitmap is created based on the size of the current screen, the same size as the entire desktop window. In addition, bitmaps can be read and written to store full-screen pixels of memory. Then, call the SelectObject function to select the created bitmap into a compatible DC.

 Step S204: Copy the contents of the desktop window DC to the compatible DC.

 This step calls the BitBlt function to copy the desktop window DC in step S202 to the step.

On the bitmap created in S203, the bitmap on the compatible DC is the image at the time of the screen capture, thereby completing the screenshot of the current screen desktop.

 Step S205: Release the created DC.

 This step calls the Release function to release the created DC, which frees the memory to ensure that other programs run smoothly.

Step S30: Filtering and buffering the screen image. In the process of taking a screen, due to uneven illumination or environmental changes, the brightness of the image of the screen is not uniformly changed, so the image generates a lot of noise. By filtering the screen image, image noise is eliminated to overcome image interference.

 Step S40: Comparing the current frame image with the pixel attribute of the previous frame image to obtain a grayscale deviation region.

 As shown in FIG. 4, in the embodiment, the image of the adjacent two frames has a grayscale deviation area as an example. Step S40 includes:

 Step S401: Converting the color map of the acquired image RGB numerical matrix into a grayscale image and performing a difference operation between the current frame image and the previous frame image to obtain a grayscale difference value, and the operation result is stored in the same array I (xi, yj ).

 In this embodiment, in order to speed up the processing speed of the image contrast, the color attribute of the pixel is first changed, and the screen image is converted from the color image to the gray image. Converted Grayscale Each pixel of the image represents the grayscale value of that pixel with one byte. The gray value is between 0 and 255. The larger the value, the whiter the pixel (ie, the higher the brightness). The more the value is 'J, the darker it is. In this embodiment, the comparison of the pixel color depth attributes is performed.

The current frame image is compared with the gray value of each pixel of the previous frame image, and the operation result is stored in an array I ( _X i, yj ). Where X in the array represents the abscissa and y represents the ordinate; i and j represent the ordinal numbers of the abscissa or ordinate pixel respectively.

 Step S402: The array obtained in step 401 is separately projected on the X-axis Y-axis and integrated.

 Assume that the image taken by the screen is an arrangement matrix of M*N, and the above array is projected on the X-axis.

 M N

The integral is: F _{x =} ∑ ( ', ;yj'), and the integral of the projection on the γ axis is: F _{yj =} ∑li, yj).

 j=l '=1

Where /( , }/') represents the above-mentioned gradation difference value of the pixel, represents the gradation difference summation of all the pixel points of the i-th column (ie, the integral), and ^F yj represents all the pixels of the j-line. The gamma difference sums (ie the integral).

Steps S403 to S414 are based on the integral judgment process, and ^F yj and ^F _{xi are} taken from Bottom up, from left to right, compared with the preset integral threshold F0, when the current row (column) ^F yj (or ^F xi ) is greater than F0, it indicates the gray scale deviation of the row (or column), Continue to judge whether the next line (column) is greater than F0, until ^^ (or ^ d ) of a certain number of consecutive rows (columns) is greater than F0, indicating that the gray scale deviation area appears. Further, the above judging method is continued to judge that the ^3⁄4 (or ^d) of the continuous line (column) is smaller than F0, indicating that the gradation deviation area ends. After judging the integration results of all the rows and columns, the rectangular grayscale deviation region of the current frame image and the previous frame image, that is, the region of the image to be transmitted, is determined. Specific steps are as follows:

 Step S403: Determine the integration result of the current line, that is, whether ^^ is greater than F0. If yes, go to step S405, otherwise go to step S404.

 Step S404: Add 1 to the ordinate j of the current line (i.e., move to the next line), and clear K1 to 0, and return to step S403 until the integral operation of all the lines included in the image is completed. Where K1 represents the number of rows in which the grayscale deviation occurs continuously.

 In this step, if the integration result of the current line is less than F0, it means that the gray level deviation between the current frame image and the previous frame image at each pixel of the line is not enough to be recognized by the naked eye, thereby being regarded as the current frame image and the previous frame. The image has the same gray level of pixels in this row, increments the ordinate j of the current line by 1, compares the integration result of the next line with F0, and clears K1.

 Step S405: When the integration result is greater than the preset F0, K1 is incremented by 1. In this step, if the integration result of the current line is greater than F0, it means that the grayscale deviation of each pixel point of the current frame image and the previous frame image can be recognized by the naked eye, and K1 is incremented by one.

 Step S406: It is judged whether K1 is greater than a preset number threshold. ^ If yes, go to step 408, otherwise go to step 407.

 For example, if K=3 is set, when K1 is greater than Κ, it indicates that the current frame image and the previous frame image have three consecutive lines of grayscale deviation, which constitutes the grayscale deviation region.

 By setting the threshold Κ, the accuracy of the grayscale deviation area recognition can be improved, and the area of the grayscale deviation can be avoided as long as there is a little difference.

Step S407: Add 1 to the ordinate j of the current line, and return to step S403 until The result of all the line integrations included in the image ends with the comparison of F0.

 In this step, K1 is less than κ, indicating that the grayscale deviation has not yet formed a degree deviation region, so the vertical coordinate j of the current row is incremented by 1, and the integration operation of the next row is performed and compared with F0.

 Step S408: Record the ordinate yj l (ie, the number of rows) of the current line, and clear the K1 value. In this step, K1 is greater than K, indicating that the current frame image and the previous frame image have gray-scale deviation regions, so the ordinate yj l of the starting line where the gray-scale deviation occurs is recorded, and the K1 value is cleared.

 Step S409: It is judged whether the integration result is smaller than a preset F0. If yes, go to step S411, otherwise go to step S410.

 At this time, from the ordinate yj l of the starting line where the gradation deviation occurs, it is judged whether or not the continuous line integration result in the integration result is smaller than F0, that is, whether the gradation deviation area is ended or not.

 Step S410: Add 1 to the ordinate j of the current line and clear K2, and K2 indicates the number of lines in which the gradation deviation does not occur continuously, and return to step S409 until the integral operation of all the lines included in the image is completed.

 Step S411: If the integration result is less than the preset integration threshold, K2 is incremented by 1. Step S412: It is judged whether K2 is greater than a preset number threshold K:, and if so, it indicates that the current grayscale deviation area ends, and the process proceeds to step 414, otherwise, the process proceeds to step 413. Among them, K can be the same as the value.

 Step S413: The vertical coordinate j of the current line is incremented by 1, and the process returns to step S409 until the result of comparing all the lines included in the image with F0 is ended.

 Step S414: Record the ordinate yj2 of the current line, and clear the K2 value.

 At this time, K2 is greater than K:, indicating that the current frame image and the previous frame image end in the gray deviation region of all the lines, the ordinate yj2 of the end line of the gradation deviation region is recorded, and the K2 value is cleared.

 At this time, the area between yj l and yj2 is the interval of the gray scale deviation on the ordinate.

 Step S415 (not shown): Returning to step S403, the integration is based on the X-axis projections xil, xi2.

依据 X轴投影的积分式 ^^^Σ ^^)所计算出 _Xil、 xi2, 计 i=l

According to the integral formula ^^^Σ ^^) of the X-axis projection, _X il, xi2, i=l

The calculation method is the same as the steps S403 to S414, and will not be described again. The rectangular area composed of xil, xi2, yjl, and yj 2 is the gray scale deviation area.

 From the above, the present invention determines the gradation deviation area in the above manner, and sequentially judges the pixel points as compared with each other, and the processing speed is faster.

 Step S50: calculating, in step S40, the grayscale deviation region of the current frame image and the previous frame image, that is, the rectangular region surrounded by xil, xi2, yj1, and yj2 (ie, the grayscale deviation region on the screen) corresponds to The area image is encoded.

 Step S60: The area image encoded in step S50 is sent to the video receiving end in a wireless form, wherein the transmitted information includes location information of the area image, such as the above xil, xi2, yj l and yj2 information.

 Step S70: The video receiving end receives the encoded area image, and displays it after decoding. After receiving the area image data and decoding, the image is integrated into the buffered previous frame image according to its position information, and the integrated image is buffered for integration with the next frame image.

 Playing a video file in a non-full-screen window at the video sending end, only when the image of the other area except the playing video window changes (for example, the user moves the position of the video playing window or adjusts the size of the video playing window by using a mouse or keyboard input command, Or activate a new window for other apps or files) to take a full-screen window capture. On the other hand, when the user does not input the command within a certain period of time, it is only necessary to take a screenshot of the window in which the video is played. Further, the current frame video window image is compared with the previous frame to calculate the grayscale deviation region.

 As described above, when the full-screen window is screened, the grayscale deviation area calculated by comparing the adjacent two frames of images is the window for video playback. When the video window is screened, the calculated grayscale deviation area is the video playback window (that is, the video screen is completely changed). If the video playback content is a lecture or the like (usually the background is unchanged, only the presenter is the main speaker). The action of the change occurs, and the calculated gray-scale deviation area can be further reduced, thereby saving bandwidth consumption and speeding up the encoding speed.

Therefore, before step S20, a step (not shown) is further included: video transmission terminal monitoring The interval at which the user inputs the command, and controls the desktop window screen capture or the video playback window screen according to the time interval control. Wherein, the confirmation of the position of the video play window may be judged according to the difference result of the cut-off image of the desktop window according to the consecutive consecutive frames (the number of frames is greater than 2, preferably 4 frames), and changes after multiple judgments The part formed by the area is relatively determined, that is, the position of the video playing window, and after the method confirms the position and size of the video playing window, the screen portion of the identified video playing window can be screened in the subsequent screen capture. And the judgment of the above step S40. Preferably, during the video playback window screen capture, the desktop window screen capture is performed again at intervals (for example, 5 seconds) to increase the accuracy of image transmission (for example, the playback progress bar of the video playback software is outside the video screen, thereby The background image is updated in time according to the playback progress bar as the background image.

 For example, when the user browses the video, if the command from the mouse or keyboard is not detected within the expected time interval (for example, 30 seconds), the video sender considers that the user has no instruction input, thereby only taking a screen shot of the video play window. Otherwise, if there is an instruction input within the expected time interval, a full window screen shot is taken. The screen capture method of the video play window is the same as that of step 20, and will not be described again.

 The system for quickly acquiring and transmitting images is described below. As shown in FIG. 5, the video transmitting end 51 and the video receiving end 52 are included.

 The video sending end 51 is configured to send a video sharing request to establish a wireless shared connection with the video receiving end 52, and view and cache the broadcasted video. The current frame image is compared with the previous frame image to calculate each gray scale deviation region, and each gray scale deviation region of the two frame images is encoded and output in a wireless form.

 The video transmitting end 51 includes a first sharing module 511, a screen capture module 512, a video transmitting end storage module 513, a filtering module 514, a comparison module 515, an encoding module 516, and a transmitting module 517.

 The first sharing module 511 is configured to send a video sharing request, and determine whether to perform a shared connection according to the feedback information.

 The screen capture module 512 is connected to the first sharing module 511 for performing frame-by-frame screen capture according to the shared trigger information.

The video sending end storage module 513 is connected to the screen capture module 512 for receiving the window cutoff Screen images are cached. The video sending end storage module 513 caches at least two screenshot images of the current frame and the previous frame. When the video sending end storage module 513 receives the next frame window image, the previous frame window screen image that is cached earlier according to the buffering timing. delete.

 The filtering module 514 is connected to the video sending end storage module 513 for retrieving the buffered window screen image and performing filtering processing to eliminate image noise.

 The comparison module 515 is connected to the filtering module 514, and is used for comparing the current frame image with the previous frame screenshot image to calculate each grayscale deviation region.

 The comparison module 515 includes:

 a pixel gradation conversion unit for converting a color map of an image RGB numerical arrangement matrix into a grayscale image;

 a differential operation unit, connected to the pixel gradation conversion unit, for performing a difference operation between the current frame image and the previous frame image, and combining the difference operation results into the same array;

 The integral operation unit is connected to the difference operation unit, and is configured to respectively perform the X-axis Y-axis projection on the array and obtain the integral thereof;

 The integral result comparison unit is connected to the integral operation unit for determining the magnitude of the integration result and the preset integration threshold. If the integration result is greater than the integration threshold, the grayscale deviation is indicated, and vice versa, the grayscale deviation is not present;

 The gray-scale deviation area starting point determining unit is connected with the integral result comparing unit, and is used for judging the number of gray-scale deviations of the continuous line (or column) and the preset number of thresholds, if the continuous line (or column) appears gray The number of degrees of deviation is greater than the threshold of the number, indicating that the gray scale deviation area appears, and vice versa, that the gray scale deviation area does not appear;

 The gray-scale deviation area end point determining unit is connected to the gray-scale deviation area starting point determining unit for determining the number of gray-scale deviations of the continuous line (or column) and the size of the preset number threshold, if continuous lines (or columns) The number of grayscale deviations is less than the threshold of the number, indicating that the grayscale deviation region ends, and conversely, the grayscale deviation region is not ended;

 The row and column coordinate accumulating unit is connected to the integral result comparison unit for adding 1 to the coordinates of the image row (or column).

The encoding module 516 is connected to the comparison module 515 for encoding the image corresponding to each grayscale deviation region calculated by the comparison module 515 to form pixel data. The transmitting module 517 is coupled to the encoding module 516 for outputting the encoded pixel data in a wireless form.

 A user command monitoring module (not shown) is used to monitor the frequency with which the user inputs a command signal via a mouse or keyboard. The monitoring result is sent to the screen capture module 512. If the user does not input a command signal within a predetermined time, the screen capture module 512 performs image capture only on the video play window.

 The video receiving end 52 is configured to respond to the video sharing request of the video transmitting end 51 according to the shared identifier bit, and after receiving the wireless shared connection, receive the encoded pixel data, decode and integrate the broadcast data.

 The video receiving end 52 includes: a second sharing module 521, a sharing detecting module 522, a receiving module 523, a decoding module 524, an integrating module 525, and a video receiving end storage module 526.

 The second sharing module 521 is coupled to the first sharing module 511 for receiving a video sharing request and outputting detection information of the identification bit.

 The sharing detection module 522 is connected to the second sharing module 521, and is configured to detect the sharing status of the video receiving end 52 according to the detection information of the identification bit (ie, whether video sharing is being performed with other video transmitting ends) and generate feedback information.

 The second sharing module 521 is further configured to receive feedback information and transmit the feedback information to the first sharing module 511.

 The receiving module 523 is coupled to the transmitting module 517 for receiving encoded pixel data. The decoding module 524 is coupled to the receiving module 523 for decoding the encoded pixel data.

 The integration module 525 is coupled to the decoding module 524 and the video receiving end storage module 526, and is configured to integrate the image corresponding to the decoded grayscale deviation area into the cached image of the previous frame according to the row and column pixel coordinates. in.

 The video receiving end storage module 526 is connected to the integration module 525 to buffer the image integrated by the integration module 525.

 A display module (not shown) is coupled to the integrated module 525 for displaying the integrated video image.

The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Claims

Claim A method for fast transmission of video images, characterized in that it comprises the steps of: A. The sending end performs an image screen capture on its screen and caches it;  B. determining an area of the screen image that is different from the buffered image of the previous frame of the screen image, and transmitting the image of the area;  C. The video receiving end receives the area image and combines it with the buffered image as a received image for buffering.  2. The method according to claim 1, wherein the determining step of the pixel in step B comprises: determining that the pixel is different from the pixel value of the previous frame image when the deviation of the pixel attribute value exceeds a certain value.  3. The method according to claim 2, wherein the determining step of the area in step B comprises:  The rows and columns in which the pixels of the previous frame are different are determined in units of rows and columns, respectively, and the regions formed by the intersection of the rows and columns are the different regions.  4. The method according to claim 3, wherein: the step of determining rows and columns different from the image pixels of the previous frame comprises:  Performing a difference operation between the screen image and each pixel of the previous frame image to obtain a grayscale difference value, and storing the operation result as an array;  Performing projections of the X-axis and the Y-axis on the array and respectively calculating the sum of the gradation differences of the columns and the rows;  The rows and columns for which the sum of the gradation differences of the respective columns and the rows are greater than a threshold are respectively determined as the determined rows and columns.  The method according to claim 4, wherein, after determining that the sum of the gradation differences of the columns and the rows is greater than a threshold, respectively, the method further comprises:  A step of judging that the sum of the successive set number of column and line gradation differences is greater than a threshold.  6. The method according to claim 1, further comprising: Counting different areas of pixels according to screenshots of a certain number of frames in a certain period of time area;  The screen shot in step A is: Screen capture of the statistical area.  7. The method according to claim 6, further comprising: determining that the full-screen image is used as the screen shot in step A when the application of the screen image outside the area screen shot is activated by the video transmitting end.  8. The method of claim 7, wherein the application is activated comprises: determining that a mouse, keyboard, or other application window is activated.  9. The method of claim 2, wherein the attribute value comprises a color attribute and a color depth attribute of a pixel.  10. A system for fast transmission of video images, comprising a video transmitting end and a video receiving end, wherein  The video sender includes:  a screen capture module, configured to perform an image screen capture on a screen of the video sending end;  a video sending end storage module, configured to cache the image of each frame cut by the screen capture module;  a comparison module, configured to compare the current screen capture image with a previous frame image buffered by the storage module to determine a different area of the image pixel;  The sending module is configured to output different areas of the image pixels in a wireless form.  The video receiver includes:  a receiving module, configured to receive an image of a different pixel of the image sent by the sending module of the video sending end;  An integration module, configured to combine different regions of the received image pixels with the cached image as the received image;  The video receiving end storage module is configured to cache the integrated image of the integrated module.