KR100487330B1 - Apparatus for generating thumbnail image of digital video - Google Patents

Abstract

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a thumbnail image generating device among additional functions provided by a personal video recorder or a video browser installed in a digital TV, and more particularly, to an I picture of a series of input MPEG video streams. A thumbnail image is generated by using a DC value of. In this process, the complexity of the thumbnail image is analyzed by using a tilt value of the image to determine whether the image is blank. In this case, when the input image is determined to be blank, a thumbnail image is not generated from the determined image or the corresponding thumbnail image is controlled so that only a meaningfully meaningful image is provided to the user, not a blank.

Description

Apparatus for generating thumbnail image of digital video}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a thumbnail image generating apparatus and method among additional functions provided by a personal video recorder (PVR) mounted on a digital TV, or other system that provides a function such as a video browser. The present invention relates to an apparatus for automatically generating a thumbnail image only for an image that is meaningful to a user by calculating a complexity of a currently input image in a process of generating a thumbnail image.

In the environment where various and diverse digital multimedia contents can be accessed, many methods for allowing users to access contents more quickly and easily have been studied. In particular, since video is huge, a system with a lot of video data needs to spend a lot of time in order to access data desired by a user. For this reason, studies are being actively conducted to provide an environment in which desired contents can be easily and quickly accessed or searched within a limited time.

As a solution to this problem, numerous methods for implicitly summarizing video data have been studied. Among them, the thumbnail image displays a specific frame of the video smaller than the original size. The thumbnail image has a form of a still image, not a movie type, and provides one or more thumbnail images to the user. Can be.

These thumbnail images summarize the contents of the recorded video in a personal video recorder (PVR) into several thumbnails, or create a key frame for video indexing or program guides. There are various fields of application such as generation of a representative image for the purpose of preview or content summary.

The conventional method of generating the thumbnail image may be classified into a method of decoding a frame of video encoded by MPEG and then directly reducing a reconstructed frame and a method using a DC value of an I picture. In this case, the former method of directly reducing a frame of a video needs to decode all pixels of the frame, which requires a variable length decoder, which is a hardware burden. In addition, in the case of a B or P picture that is not an I picture, the B or P picture needs to be decoded and then converted into an intra-coded still picture using a compression method such as JPEG. On the other hand, the latter method of creating a thumbnail image using the DC of the I picture of the video does not require such a variable decoder, so it is less burdensome in hardware and the DC value is directly used as the pixel value of the thumbnail image. Therefore, this method generates the 1/64 size thumbnail image of the original image by a simple method without complicated decoding and re-encoding process, which has the advantage of ease of implementation and efficiency.

The thumbnail image generated by the above methods provides the user with a method of implicitly representing the entire contents of the video as a representative image of a smaller size, that is, a lower resolution than the original image, so that the video frame that is the source of the thumbnail image is noisy. There should be no blanks, overlaps due to scene transitions, etc., and at the same time, the delivery of information about the entire video to the user should be clear. Therefore, for the thumbnail image, an unclear image generated by camera movement such as a sudden zoom and blur due to the movement of the subject should be avoided.

However, the conventional thumbnail image generating apparatus does not provide an effective method of detecting and filtering an image that is not suitable for use as a thumbnail image. That is, in the conventional thumbnail image generating apparatus, a thumbnail is generated for an arbitrary I picture without analyzing the content of the image. In this case, a method of generating a thumbnail image for an I picture generated at a predetermined interval by a hardware preset or user input sample rate is used. However, in the digital video that is actually received, the original image may be noisy, scenes or other special effects may cause the image to be very dark or have a single color without the actual content, or may be caused by excessive zooming or close-up. Sometimes the image itself is meaningless or the amount of information contained in the image is weak. Thumbnail images generated from such images are very difficult for a user to identify the contents, and as a result, a problem arises in providing implicit information about the entire video.

That is, in the conventional thumbnail image generation method, there is no method for automatically determining the validity of the thumbnail by analyzing the thumbnail image, so that it does not effectively cope with the meaningless image, which has many problems in effectively extracting the representative image.

The present invention is to solve the above problems, an object of the present invention is to calculate the complexity of the input image to generate a thumbnail image by not generating a thumbnail image for a low complexity image, the user means more An object of the present invention is to provide a thumbnail image generating device for digital video that allows only a clear thumbnail image to be viewed.

An apparatus for generating a thumbnail image of a digital video according to the present invention for generating the above object generates a thumbnail image from an I picture of an MPEG encoded video stream, but has a low complexity or almost no texture in the process. Display process of thumbnail images generated by these images after detecting images with no background only or images with low luminance and single luminance dominated by poor lighting conditions, and meaningless images caused by excessive zoom or close-up By blocking the feature, it provides a more meaningful and clear thumbnail image to the user.

In the present invention, an image that is too dark or simple colors, or because of the complexity and blur due to the cause of blur and zoom, the content or information that can be delivered to the user is very small, the image that is not suitable for generating a thumbnail image blank ( blank) Defined as an image.

In accordance with an embodiment of the present invention, a thumbnail image generating apparatus according to the present invention may include a memory for parsing only a DC value of an I picture from an input video stream and storing a thumbnail image generated from the DC value, and the parsed DC. Whenever the DC value of the DC image generated by the value is input, the slope of the horizontal and vertical components for each DC pixel is obtained, the two slope values are added, the slope for the corresponding DC pixel is calculated, and the slope of the previous DC pixel is accumulated. A blank image determination unit for determining whether the I picture is a blank image by comparing a predetermined gradient value after obtaining a total gradient value of the current DC image by repeating the process, and according to a determination result of the blank image determination unit And a controller for controlling whether to use the corresponding thumbnail image stored in the memory.

The total slope of the DC image is an average value obtained by dividing the total sum of the slopes of each DC pixel of the DC image by the size of the DC image.

The blank image determining unit may determine the corresponding I picture as a blank image when the total tilt value of the DC image is smaller than a preset threshold.

The inclination of the horizontal and vertical components with respect to each DC pixel in the blank image determiner is a change amount of each of two adjacent DC pixels in the horizontal and vertical directions.

The inclination of the horizontal and vertical components with respect to each DC pixel in the blank image determiner is an absolute value of a difference from the DC pixel immediately before in the horizontal and vertical directions.

If it is determined that the input I picture is a blank image, the controller controls the thumbnail image generated from the next I picture to be overwritten to a position in the memory where the thumbnail image generated from the I picture is stored.

The control unit includes as many blank vectors as the number of thumbnail images that can be stored in the memory, and then stores all the thumbnail images in the memory regardless of whether they are blank or not, and according to the determination result of the blank image determining unit, corresponding thumbnail images to the blank vectors. It is characterized by displaying the presence or absence of the blank.

Other objects, features and advantages of the present invention will become apparent from the following detailed description of embodiments taken in conjunction with the accompanying drawings.

Hereinafter, with reference to the accompanying drawings illustrating the configuration and operation of the embodiment of the present invention, the configuration and operation of the present invention shown in the drawings and described by it will be described as at least one embodiment, By the technical spirit of the present invention described above and its core configuration and operation is not limited.

The present invention analyzes the DC value of an I picture to generate a thumbnail image to determine whether the corresponding I picture is a blank image, and not to use a thumbnail image generated from the I picture determined as the blank image. In particular, the present invention is characterized in that it is determined whether a blank image is obtained by obtaining a slope by a DC image unit generated by the DC value of the I picture and comparing the slope with a preset threshold.

1 is a block diagram showing an embodiment of the overall configuration of a thumbnail image generating apparatus according to the present invention, a DC decoder 101 for parsing only the DC value of an I picture in an MPEG encoded video stream, the parsing A blank image determination unit 102 determining whether a thumbnail image currently generated from the DC value is valid, a temporary buffer 103 temporarily storing the parsed DC value in line or slice units for thumbnail image generation, and the The DC value is stored in the temporary buffer 103, the DC value stored in the temporary buffer 103 is transmitted to the thumbnail dedicated memory through the thumbnail memory interface 105 in units of slices, and the blank image determination unit 102 is also used. The buffer controller 104 controls the overall flow of the system, such as controlling the use of the thumbnail image stored in the thumbnail dedicated memory in accordance with the determination result of the It is sex.

That is, the buffer controller 104 may store data stored in the buffer 103 through the thumbnail memory interface 105 at a specific location of the thumbnail dedicated memory.

The DC decoder 101 has a structure that can parse only the DC value and image information parameter of the I picture in a very simple structure compared to the VLD of a general MPEG video decoder. In the present invention, since a thumbnail image is generated, the DC decoder 101 does not need to decode the P picture and the B picture, and removes the variable decoding table and hardware parts necessary to obtain other AC values.

That is, the DC decoder 101 decodes an information parameter of an image such as a picture size and an address of a macro block among the I picture streams of MPEG video and a DC value of each block to decode the blank image determination unit 102. The thumbnail is sent to the temporary buffer 103.

In this case, in order to simplify the function of the blank image determining unit 102, the analysis may be performed only on the luminance component of the image. In this case, a DC value corresponding to blocks 0 to 3 of each macro block, that is, a Y component Only the DC values of the transmission are transmitted to the blank image determiner 102.

Here, each DC value has a value from 0 to 255, which is represented by 1 byte of data. In general, in an MPEG video having a 4: 2: 0 format, a DC value means an average value of each block, and since there are six blocks per macroblock, six DC values are input from one macroblock. DC values of the first four (that is, blocks 0 to 3 of each macro block) are DC values of the Y component, and the remaining two (ie blocks 4 and 5) are the Cb and Cr components, respectively. It corresponds to DC value.

On the other hand, the temporary buffer 103 must transmit all DC values necessary for the configuration of the thumbnail image. That is, the thumbnail image is stored in a thumbnail-only memory in a complete form, and under the control of the buffer controller 104, the data corresponding to one slice is stored in the temporary buffer 103, and when the slice ends, the temporary image The data in the buffer 103 is transferred to the thumbnail dedicated memory.

Here, the slice refers to one horizontal line of the macro block according to the ATSC standard. The reason for generating the thumbnail image based on the slice is that the buffer 103 is controlled based on the slice based on the characteristics of the input MPEG video stream. Because it is efficient.

The buffer controller 104 controls the input and output of the temporary buffer 103, the temporary buffer 103 is implemented in the form of a double buffer for the efficiency of the input and output. That is, when the slice ends, the buffer controller 104 performs a process necessary to transfer the thumbnail data stored in the temporary buffer 103 to the thumbnail dedicated memory together with the thumbnail dedicated memory interface 105. The thumbnail dedicated memory interface 105 controls a process of directly writing data stored in the temporary buffer 103 to a thumbnail dedicated memory, and controls all other processes for accessing the thumbnail dedicated memory.

The thumbnail image generating apparatus of the present invention shown in FIG. 1 is a block diagram in consideration of scalability, and the thumbnail dedicated memory interface 105 may be used in connection with another external unit to access the thumbnail dedicated memory. That is, in addition to the thumbnail image generating apparatus according to the present invention, a block for using a thumbnail dedicated memory may be connected to the thumbnail dedicated memory interface 105. In such a case, the thumbnail dedicated memory may be expanded to enable input and output of other data in addition to the thumbnail data.

The blank image determiner 102 determines whether the current image is blank by performing a series of processes to be described later using the input DC value. As described above, in the case where the blank image determining unit 102 intends to analyze an image using only the luminance (Y) component of the image, that is, to determine whether the blank is zero, the number 0 of each macro block corresponding to the luminance component of the image is determined. Uses only the DC value from block 3. In addition, if a blank is to be determined using a specific component or a combination of several components (including chrominance components) of the other image, the blank determination function may be performed using all DC values of the corresponding components.

FIG. 2 is a flowchart illustrating an operation of the blank image determiner 102 according to the present invention. The embodiment of the blank image determiner using only luminance signal components may be described, and other components may be similarly implemented.

In the present invention, whether the image is blank is determined based on the complexity of the image, and the complexity of the image is calculated based on a derivative value or a gradient of two adjacent DC pixel values.

That is, the blank image determiner 102 calculates the slope Gr_Pix for each DC pixel by using the DC value input from the DC decoder 101. The slope of each DC pixel is defined as the sum of the magnitudes of the slopes of the horizontal and vertical components.

To do this, first calculate Gr_H representing the slope of the horizontal component and Gr_V representing the slope of the vertical component. The method of calculating Gr_H and Gr_V is shown in Equation 1 below. When the current input DC pixel is f (i, j), Gr_Pix indicating the slope of the DC pixel is calculated as the sum of Gr_H and Gr_V.

That is, in Equation 1, f (i, j) denotes a pixel value (eg, DC value of each Y block) of the input DC image, and i and j denote vertical and horizontal coordinates, respectively.

In addition, in Equation 1, the slope Gr_H of the horizontal component and the slope Gr_V of the vertical component with respect to each pixel value may be calculated in various ways. That is, for each DC pixel in the image, another value representing the derivative value in the horizontal and vertical directions or the slope of the pixel value may be used. The inclination means a change amount of adjacent pixel values.

In an embodiment of the present invention, the slope, that is, the amount of change of two adjacent DC pixel values is simply calculated as in Equation 2 below.

That is, the absolute value of the difference from the immediately preceding pixel value in the horizontal and vertical directions is used.

As described above, each time the DC value is input, the slope Gr_Pix is calculated for each pixel, and these values are added and stored in the variable Gr_Img_tmp in units of DC images. For example, the gradient value Gr_Pix of the first pixel obtained from the DC image may be set to Gr_Img_tmp, and then, whenever the DC value is input, Gr_Pix of the corresponding DC pixel may be obtained and accumulated in the variable Gr_Img_tmp. Therefore, the sum of the slopes of the DC pixels input so far is stored in Gr_Img_tmp. Here, Gr_Img_tmp for one DC image is obtained, and when the determination of the blank image is finished, the Gr_Img_tmp value is initialized for the next DC image.

That is, when the DC pixel value f (i, j) at the (i, j) th position of the DC image is input (step 201), the slope GR_H of the horizontal component of the pixel value f (i, j) and the slope of the vertical component are input. After GR_V is calculated, the slope Gr_Pix of the pixel value f (i, j) is calculated by adding two values (step 202). Then, the slope Gr_Pix of the pixel value f (i, j) is accumulated in the variable Gr_Img_tmp (step 203).

Then, it is checked whether the line is the end of the input DC image (step 204). If it is not at the end of the line, the horizontal coordinate j value is increased by 1 to obtain the slope of the horizontal component of another pixel in the line, and the process returns to step 201 and the above steps are repeated (step 205). Check if it is the end of step (206). In step 206, the j value is initialized to obtain the slope of the vertical component of another pixel on another line that is not the end of the DC image, and the vertical coordinate i value is increased by 1, and then the process returns to the step 201 and the above process is repeated. (Step 207).

If it is determined in step 206 that the end of the DC image, that is, if all DC values of the current image is input, the average value Grad_Img of the slope of the entire DC image can be calculated by dividing Gr_Img_tmp by the size of the DC image as shown in Equation 3 below. (Step 208).

In Equation 3, f (i, j) denotes a pixel value of an input DC image, i and j denote vertical and horizontal coordinates, respectively, and h_size and v_size denote horizontal and vertical sizes of the DC image. . Also, , to be.

At this time, the magnitude of Grad_Img of each DC image means an average value of the slopes of all pixels of the corresponding image, which is a value representing the complexity of the corresponding image.

After the step 208 is performed, the average value Grad_Img of the slope of the entire DC image is compared with the preset threshold Grad_Thr, and when the condition of Equation 4 is satisfied, the current image is determined to be a blank image, and a thumbnail generated from the corresponding image The image is not used (step 209).

Grad_Img <Grad_Thr

That is, if the Grad_Img representing the average slope value of the entire image is smaller than Grad_Thr, which is a preset threshold, the blank image determiner 102 determines the corresponding image as a blank and sets the blank signal as '1'. 104) (step 210). If the blank signal is '1', the buffer controller 104 controls not to use a thumbnail image corresponding thereto (step 211).

In an embodiment of the present invention, the threshold value Grad_Thr is set to 5.0 based on the results of experiments on the actual MPEG video stream, which may be different by the designer, and thus will not be limited to the embodiment described above. .

On the other hand, the blank image determining unit 102 determines that the corresponding image is not blank if Grad_Img, which represents the average slope value of the whole image, is not smaller than the preset threshold Grad_Thr, and sets the blank signal to '0' to control the buffer control unit. Output to 104 (step 212). If the blank signal is '0', the buffer controller 104 controls to use a thumbnail image corresponding thereto (step 213).

In this case, the blank image determination unit 102 needs a temporary buffer for calculating the slope of each pixel of the image. That is, since the previous DC value generated in each direction is required for the calculation of the horizontal and vertical slope of each pixel, the size of the temporary buffer used in the blank image determining unit 102 is (h_size + 1). * 8 bit.

As described above, when the I picture ends, the blank image determiner 102 calculates the current image (ie, the complexity is calculated based on the complexity calculated during input of DC values of the I picture, that is, the average slope value of the entire picture). By determining whether the I picture is blank and outputting a corresponding signal BLANK, the use of the thumbnail image generated from the I picture determined as the blank is controlled.

That is, if the current image is determined to be blank through the above-described process, the thumbnail image generated from the image determined as the blank should not be displayed through a software or hardware method in a system using a PVR or other thumbnail image. For example, when the current thumbnail image is blank, a method such as overwriting the next thumbnail image in a location where the current thumbnail image is already stored in the thumbnail dedicated memory may be used, or by using a blank vector or the like. Only thumbnail images with a blank value of '0' can be software-selected for use in thumbnail image display or other applications.

3 is a view illustrating an embodiment of a method for controlling a thumbnail image using an address value of a thumbnail dedicated memory when the current image is determined to be a blank image, FIG. 3A is an operation flowchart, and FIG. 3B is a thumbnail dedicated memory. The map state is showing.

That is, in step 301, it is determined whether the BLANK signal is '1', and if it is '1', the current image is determined to be a blank image, and if it is '0', it is determined that it is not a blank image.

In step 302 and 303, ThmAddrStart means a start address in a thumbnail dedicated memory in which a thumbnail image to be generated next is stored. CurrThmStart represents the start address of the thumbnail image currently stored in the thumbnail dedicated memory. ThmPictSize refers to the size of each thumbnail and represents the size of the space occupied by each thumbnail image in the thumbnail dedicated memory. That is, when the current video is finally determined as a blank, the thumbnail for the current video is already stored in the thumbnail dedicated memory. This is because thumbnails are stored in slices. In other words, it is not possible to determine whether to blank the image until the last slice of the current image is finished, and the DC values corresponding to the slices of the previous image are already stored in the thumbnail dedicated memory.

Therefore, when the current image is determined to be blank in step 301, step 302 overwrites the next thumbnail image to a thumbnail dedicated memory space in which the current thumbnail image is stored, thereby using the thumbnail image generated from the image determined as blank. (ThmAddrStart = CurrThmStart). In this case, only thumbnail images, not blanks, may be stored in the thumbnail dedicated memory, thereby increasing memory efficiency. For example, assuming that the second and fourth images are determined to be blank images, the thumbnail image corresponding to the third image is overwritten at the position where the thumbnail image corresponding to the second image is stored, and the original fourth image is overwritten. The thumbnail image corresponding to the fifth image is overwritten at the position where the corresponding thumbnail image is stored. As a result, as shown in FIG. 3B, the thumbnail images for the second and fourth images are deleted in the thumbnail dedicated memory.

However, this method should provide a space for storing IDs of each thumbnail separately. That is, a list indicating which thumbnails are currently stored in the thumbnail dedicated memory must be separately prepared. The thumbnail ID indicates from which image in the original video the generated thumbnail is extracted.

If the current image is not determined as the blank image in step 301, the start address (ThmAddrStart) in the thumbnail dedicated memory in which the next thumbnail image is stored in step 303 is the start address of the thumbnail image currently stored in the thumbnail dedicated memory. (CurrThmStart) is a value obtained by adding the size (ThmPictSize) of the space occupied by each thumbnail image (ThmAddrStart = CurrThmStart + ThmPictSize).

On the other hand, as shown in FIG. 3, if a blank vector is simply placed without using a list indicating which thumbnails are stored in the current thumbnail dedicated memory, thumbnail ID management may be easy.

4 is a view illustrating an embodiment of a control method using a blank vector when a blank image is generated in the thumbnail image generating apparatus according to the present invention, FIG. 4A is an operation flowchart, and FIG. 4B is a map state and a blank of a thumbnail dedicated memory at this time. A vector is used to indicate whether each thumbnail image stored in the thumbnail dedicated memory is blank. That is, as shown in FIG. 4B, all of the generated thumbnail images are stored in a thumbnail dedicated memory, and a value of a blank signal corresponding to each thumbnail is stored together in a blank vector. If the N thumbnail images are stored in the thumbnail dedicated memory, a blank vector of size N bits is required. Each bit indicates whether a thumbnail image corresponding to the sequence is blank. For example, if the n th bit of the blank vector is '0' and the m th bit is '1', the n th thumbnail image is available and the m th thumbnail image is not used because it is a blank image.

This process is illustrated in Figure 4a. In FIG. 4A, the blank vector value is set to 1 when the current image is determined to be blank, and the blank vector value is set to 0 when it is determined that the image is not blank.

That is, in step 401, it is determined whether the i-th blank vector is 1 or 0, and if it is 1, a thumbnail image corresponding to the i-th blank vector is not displayed on the screen (step 402). In addition, if it is determined as 0 in step 401, a thumbnail image corresponding to the i-th blank vector is displayed on the screen (step 403). After the step 402 or 403 is performed, the value of i is increased by one (step 404). If the value of i is not N (step 405), it is not the end of the thumbnail dedicated memory. Extract the black vector value. If it is determined in step 404, the end of the thumbnail dedicated memory is moved to the beginning or the program is terminated.

As described above, when using the method of FIG. 4, the system that wants to use the thumbnail image may know whether to use each thumbnail image by referring to the blank vector when reading the thumbnail image from the thumbnail dedicated memory. In this case, all thumbnail images generated according to a rule determined in advance by the system are stored in a thumbnail dedicated memory regardless of whether each image is blank, so that separate thumbnail ID management is not required.

Accordingly, the present invention provides a summary of contents for a video content provided by a PVR or video storage system, a video library and a browser mounted on a digital TV, a program guide and a representative image generation function for quick searching of a video, a personal program guide, and a video. When applied to other application fields related to implicit information transmission for the user, it is possible to provide an effective and automatic thumbnail image to the user.

As described above, according to the apparatus for generating a thumbnail image of a digital video according to the present invention, a thumbnail image is generated by using a DC value of an I picture of a series of input MPEG video streams, and in this process, a tilt value of the image is used. The complexity of the thumbnail image is analyzed to determine whether the image is blank. In this case, when the input image is determined to be blank, a thumbnail image is not generated from the determined image or the thumbnail image is controlled so that only a meaningfully meaningful image, not a blank, is provided to the user.

In particular, the present invention is generated from an image that is difficult to deliver the content as a representative image, such as a low complexity image due to the meaningless image due to excessive zoom or close-up that may occur in the video stream received in the actual broadcast, low brightness image due to lighting conditions, etc. There is an effect that can prevent the use of the thumbnail image, which enables the user to automatically obtain a more meaningful and clear thumbnail image.

Those skilled in the art will appreciate that various changes and modifications can be made without departing from the spirit of the present invention.

Therefore, the technical scope of the present invention should not be limited to the contents described in the embodiments, but should be defined by the claims.

1 is a block diagram showing an embodiment of a thumbnail image generating apparatus according to the present invention

2 is a flowchart illustrating an embodiment of a blank image determining unit of a thumbnail image generating apparatus according to the present invention;

3A is a flowchart illustrating an embodiment of a method for controlling a thumbnail image using an address value of a thumbnail dedicated memory when a blank image is generated in the thumbnail image generating apparatus according to the present invention.

FIG. 3B is a diagram illustrating a map state of a thumbnail dedicated memory in FIG. 3A; FIG.

4A is a flowchart illustrating an embodiment of a control method using a blank vector in generating a blank image in a thumbnail image generating apparatus according to the present invention;

FIG. 4B is a diagram illustrating an embodiment of whether each thumbnail image stored in the thumbnail dedicated memory is blank by using the map state and the blank vector of the thumbnail dedicated memory in FIG. 4A; FIG.

Explanation of symbols for main parts of the drawings

101: DC decoder 102: blank image determination unit

103: temporary buffer 104: buffer controller

105: thumbnail memory interface

Claims (9)

A digital video thumbnail image generating device for parsing only a DC value of an I picture from an input video stream and generating a thumbnail image from the DC value, A memory for storing the thumbnail image; Whenever the DC value of the DC image generated by the parsed DC value is input, the slope of the horizontal and vertical components of each DC pixel is calculated, the slope of the corresponding DC pixel is calculated by adding the two slope values, and then the previous DC pixel. A blank image determining unit to determine whether the I picture is a blank image by repeating the process of accumulating the slopes of the current DC image to obtain a total slope value of the current DC image and comparing the threshold value with a preset threshold; And And a controller configured to control the thumbnail image stored in the memory not to be used when the blank image determination unit determines that the image is a blank image. The method of claim 1, wherein the overall slope of the DC image is And an average value obtained by dividing the total sum of the slopes of the respective DC pixels of the corresponding DC image by the size of the DC image. The method of claim 1, wherein the blank image determination unit And determining an I picture as a blank image when the total slope value of the DC image is smaller than a preset threshold value. The method of claim 1, wherein the blank image determination unit And calculating the inclination of each DC pixel in the horizontal and vertical directions using only the DC value of the parsed Y component. The method of claim 1, wherein the blank image determination unit And calculating the inclination in the horizontal and vertical directions of each DC pixel by using a combination of the parsed DC value of the Y component and the DC value of the CbCr component. The method of claim 1, The inclination of the horizontal and vertical components with respect to each DC pixel in the blank image determining unit is a change amount of each of the two adjacent DC pixels in the horizontal and vertical direction. The method of claim 1, And wherein the inclination of the horizontal and vertical components with respect to each DC pixel in the blank image determining unit is an absolute value of a difference from the immediately preceding DC pixel in the horizontal and vertical directions. The method of claim 1, wherein the control unit If the input I picture is determined to be a blank image, the thumbnail image of the digital video is controlled so that the thumbnail image generated from the next I picture is overwritten to a position in the memory where the thumbnail image generated from the I picture is stored. Generating device. The method of claim 1, wherein the control unit After having as many blank vectors as the number of thumbnail images that can be stored in the memory, whether or not blanks exist for the corresponding thumbnail images according to the blank image determination unit while all the thumbnail images are stored in the memory regardless of blanks Thumbnail image generating device for digital video, characterized in that for displaying.