WO2012172706A1 - 動き画像領域判定装置またはその方法 - Google Patents
動き画像領域判定装置またはその方法 Download PDFInfo
- Publication number
- WO2012172706A1 WO2012172706A1 PCT/JP2011/077755 JP2011077755W WO2012172706A1 WO 2012172706 A1 WO2012172706 A1 WO 2012172706A1 JP 2011077755 W JP2011077755 W JP 2011077755W WO 2012172706 A1 WO2012172706 A1 WO 2012172706A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- block
- motion
- rectangular
- row
- motion image
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G5/00—Control arrangements or circuits for visual indicators common to cathode-ray tube indicators and other visual indicators
- G09G5/34—Control arrangements or circuits for visual indicators common to cathode-ray tube indicators and other visual indicators for rolling or scrolling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/215—Motion-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/254—Analysis of motion involving subtraction of images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G2320/00—Control of display operating conditions
- G09G2320/10—Special adaptations of display systems for operation with variable images
- G09G2320/106—Determination of movement vectors or equivalent parameters within the image
Definitions
- the present invention relates to a method for determining a moving image area, and more particularly to determination of a rectangular moving image area.
- Patent Document 1 discloses motion detection in units of blocks. Specifically, the gradation values of the same pixel in the previous and subsequent frames are compared, and if the gradation values are different, it is determined as a moving pixel, and the ratio of the number of moving pixels is calculated for each block, which is larger than the threshold value. In this case, it is determined that the block is a moving image block.
- An object of the present invention is to solve the above problems and provide a determination method or apparatus for determining a rectangular motion image area with a simple configuration.
- a rectangular motion image region determining method is a rectangular motion image region that determines a rectangular motion image region displayed in a part of a display region in which pixels are arranged in rows and columns.
- a method of determining a motion unit block wherein the display area is divided into unit blocks each having a predetermined number of pixels, and each unit block is a motion unit block in which motion exists.
- a set of unit blocks belonging to the same column as each unit block located at the top of the unit blocks is defined as a column block, and if there is at least one motion unit block in each column block, the column
- a motion sequence block determining step for determining a block as a motion sequence block among the unit blocks, the same as each unit block located at the leftmost stage
- a set of unit blocks belonging to a row is defined as a row block, and if there is at least one motion unit block in each row block, a motion row block determining step for determining the row block as a motion row block
- a first rectangular motion image region determination step for determining a rectangular region specified by a unit block belonging to both of the block and the motion sequence block as a rectangular motion image region; an outer periphery of unit blocks constituting the rectangular motion image region;
- each unit block is a motion unit block
- the motion row block and the motion sequence block are determined, and a rectangular region specified by the unit blocks belonging to both is determined as a rectangular motion image region.
- the motion image area can be determined.
- the region defined by the edge in the rectangular motion image region is defined as the rectangular motion based on each pixel in these unit blocks.
- the row direction in these unit blocks is determined for each row.
- a row having a different representative pixel value and a column having a different representative pixel value in each column in the column direction are determined as boundaries of the rectangular motion image region. Therefore, the boundary can be determined based on the representative pixel value for each row.
- step (b) in the second rectangular moving image area determining step, for each unit block constituting the four sides of the outer periphery, for each pixel in these unit blocks, between different frames
- step (b) it is determined whether the pixel is a moving pixel, and the boundary of the rectangular moving image area is determined. Therefore, the boundary can be determined based on the motion vector of each pixel.
- a row direction edge is determined for each pixel in these unit blocks.
- the pixels constituting the column direction edge are extracted, and the boundary of the rectangular motion image area is determined based on the total number of pixels constituting the row direction edge for each row and the total number of pixels constituting the column direction edge for each column. Therefore, the boundary can be determined based on the row direction edge and the column direction edge.
- the rectangular moving image area determining method is a scroll for determining whether or not the determined rectangular moving image area is different between frames by scrolling before the second rectangular moving image area determining step. Make a decision. Therefore, it can be distinguished whether it corresponds to scrolling or not. Further, when it is determined that the scrolling is performed, the second rectangular motion image region determination step is not performed. Therefore, it is possible to make a quick determination when the case corresponds to scrolling.
- the determination of whether or not scrolling is an edge for each pixel of the determined rectangular moving image area based on a difference in pixel values from surrounding pixels.
- the character gap existing area is recognized as a character gap existing area for a row or column in which a predetermined number or more of pixels that are not edges exist in the row direction or the column direction, and the ratio of the character gap existing area to the determined rectangular motion image area Based on the above, it is determined whether the scroll is appropriate. Therefore, it is possible to reliably determine the case where an image including characters is scrolled.
- the motion sequence block is determined before the second rectangular motion image region determination step for the rectangular motion image region determined in the first rectangular motion image region determination step.
- the determination step, the motion row block determination step, and the first rectangular motion image region determination step are repeatedly executed. Therefore, even when there are a plurality of motion image areas to be detected, accurate detection is possible.
- the rectangular motion image region determination device is a rectangular motion image region that determines a rectangular motion image region to be displayed in a part of a display region in which pixels are arranged in rows and columns.
- a motion unit block determination unit that divides the display area into unit blocks each including a predetermined number of pixels and determines that each unit block is a motion unit block in which motion exists;
- a set of unit blocks belonging to the same column as each unit block positioned at the top is defined as a column block, and if at least one motion unit block exists in each column block, the column Motion sequence block determining means for determining a block as a motion sequence block;
- a set of locks is defined as a row block, and when at least one motion unit block exists in each row block, a motion row block determining means for determining the row block as a motion row block, 4) the motion row block and A first rectangular motion image region determining means
- each unit block is a motion unit block
- the motion row block and the motion sequence block are determined, and a rectangular region specified by the unit blocks belonging to both is determined as a rectangular motion image region.
- the motion image area can be determined.
- the region defined by the edge in the rectangular motion image region is defined as the rectangular motion based on each pixel in these unit blocks.
- the second rectangular moving image area determining means includes a unit block that constitutes the four sides of the outer periphery, and the row direction within these unit blocks is determined for each line.
- a row having a different representative pixel value and a column having a different representative pixel value in each column in the column direction are determined as boundaries of the rectangular motion image region. Therefore, the boundary can be determined based on the representative pixel value for each row.
- the second rectangular moving image region determining means is configured to determine, for each pixel in these unit blocks, between different frames for the unit blocks constituting the outer four sides.
- step (b) it is determined whether the pixel is a moving pixel, and the boundary of the rectangular moving image area is determined. Therefore, the boundary can be determined based on the motion vector of each pixel.
- the second rectangular moving image region determining means includes, for the unit blocks constituting the outer four sides, for each pixel in these unit blocks in the row direction and The pixels constituting the column direction edge are extracted, and the boundary of the rectangular motion image region is determined based on the total number in the row direction for each row and the total number of pixels constituting the column direction edge for each column. Therefore, the boundary can be determined based on the row direction edge and the column direction edge.
- the rectangular motion image region determination device is configured to perform the motion sequence block determination unit, the motion row block determination unit, and the rectangle for the rectangular motion image region determined by the first rectangular motion image region determination unit. Repetitive means for repeatedly executing the processing by the moving image region determining means. Therefore, even when there are a plurality of motion image areas to be detected, accurate detection is possible.
- the rectangular motion image region determination method is a rectangular motion image region that determines a rectangular motion image region displayed in a part of a display region in which pixels are arranged in rows and columns.
- a determination method wherein the display area is divided into unit blocks each including a predetermined number of pixels, and a motion unit block determination step for determining that each unit block is a motion unit block in which motion exists, Among them, a set of unit blocks belonging to the same column as each unit block located at the top row is defined as a column block, and if at least one motion unit block exists in each column block, that column block is defined as a motion column block A step of determining a motion sequence block as a unit block belonging to the same row as each unit block located in the leftmost stage among the unit blocks.
- a set of position blocks is defined as a row block, and when there is at least one motion unit block in each row block, a motion row block determining step for determining the row block as a motion row block, the motion row block and the motion block
- a first rectangular motion image area determining step for determining a rectangular area specified by the unit blocks belonging to both of the motion sequence blocks as a rectangular motion image area;
- each unit block is a motion unit block
- the motion row block and the motion sequence block are determined, and a rectangular region specified by the unit blocks belonging to both is determined as a rectangular motion image region.
- the motion image area can be determined.
- the motion sequence block determination step and the motion row block determination step are repeated for the determined rectangular region. Execute. Therefore, even when there are a plurality of motion image areas to be detected, accurate detection is possible.
- a scroll region determination device is a scroll region determination device that determines whether or not a determination target region in a screen is a scroll region, and for each pixel of the determination target region, a pixel with a surrounding pixel Edge pixel determination means for determining whether or not an edge is based on a difference in values, and determines a line or column in which a predetermined number or more of non-edge pixels exist in the same row or column as the character gap existence region Character gap existence area determination means, and scroll area determination means for determining whether or not the character gap existence area is a scroll area based on a ratio of the character gap existence area to the determination target area.
- a boundary determination device is a boundary determination device that determines a boundary of a rectangular region existing on one screen, and determines pixels that constitute an edge based on a pixel value of each pixel in a determination target pixel.
- Edge pixel determination means for determining, extracting pixels constituting the row direction edge as row direction edge pixels, calculating the total number of the row direction edges for each row, and based on the total number of the row direction edges of each row,
- a row direction boundary determining means for determining a boundary, extracting pixels constituting a column direction edge as a column direction edge pixel, calculating a total number of the column direction edges for each column, and based on the total number of the column direction edges of each column
- Column direction boundary determining means for determining a boundary in the column direction
- rectangular area determining means for determining the determined row direction boundary and the column direction boundary as a boundary of the rectangular area.
- the boundary determination device is a means for determining a pixel to be determined by the edge pixel determination means, and A rough area determining means having the following means is provided.
- Judgment means 2) Among the unit blocks, a set of unit blocks belonging to the same column as each unit block located at the top row is defined as a column block, and if at least one motion unit block exists in each column block, Motion sequence block determining means for determining a sequence block as a motion sequence block; 3) Among the unit blocks, a set of unit blocks belonging to the same row as each unit block located in the leftmost stage is defined as a row block, and when at least one motion unit block exists in each row block, A motion row block determining means for determining the row block as a motion row block; 4) First rectangular motion image region determining means for determining a rectangular region specified by a unit block belonging to both the motion row block and the motion sequence block as a rectangular motion image region; 5) Target pixel determining means for determining
- pixel value includes not only luminance values but also numerical values for specifying image information such as RGB values.
- rectangular motion image area refers to a rectangular area that has different pixel values between a plurality of frames and constitutes a motion image area.
- “Column block” refers to a set of unit blocks belonging to the same column as each unit block located at the top of the unit blocks. In the present embodiment, one unit block located in the uppermost stage is identified and a column block is determined. As a result, if one unit block located in the uppermost stage is included, a unit block other than the uppermost stage is included. You may make it identify from a line. “Row block” refers to a set of unit blocks belonging to the same row as the unit blocks located in the leftmost stage among the unit blocks. As a result, the specific method is not particularly limited as long as it includes one unit block located in the leftmost stage.
- the “first rectangular moving image region determining step” and the “second rectangular moving image region determining step” are respectively performed by a moving image region determining process (step S7 in FIG. 3) and a boundary determining process (step S9 in FIG. 3). Applicable.
- the “column direction” refers to the direction in which each row is arranged, and in the embodiment refers to the ⁇ direction.
- “Row direction edge” refers to an edge in a direction parallel to the row direction, and “Column direction edge” refers to an edge in a direction parallel to the column direction.
- FIG. 3 is a functional block diagram of the rectangular motion image region determination device 1.
- FIG. It is an example of a hardware structure at the time of comprising the rectangular moving image area determination apparatus 1 using CPU. It is the whole flowchart.
- a display area 100 in which motion image areas 110 to 112 exist is shown.
- FIG. 1 shows a functional block diagram of a rectangular motion image region determination device 1 according to an embodiment of the present invention.
- the rectangular motion image region determination device 1 is a device that determines a rectangular motion image region to be displayed in a part of a display region in which pixels are arranged in a matrix in the row direction and the column direction. 5, a motion sequence block determination unit 7, a motion row block determination unit 9, a first rectangular motion image region determination unit 11, a repetition unit 13, and a second rectangular motion image region determination unit 14.
- the motion unit block determination means 5 divides the display area into unit blocks each including a predetermined number of pixels, and determines that each unit block is a motion unit block in which motion exists.
- the motion column block determining means 7 defines a set of unit blocks belonging to the same column as each unit block located at the top of the unit blocks as a column block, and at least one motion unit block is included in each column block. If it exists, the column block is determined as a motion column block.
- the motion row block determining means 9 defines a set of unit blocks belonging to the same row as each unit block located in the leftmost stage among the unit blocks as a row block, and at least one motion unit block is included in each row block. If it exists, the row block is determined as a motion row block.
- the first rectangular motion image area determining means 11 determines a rectangular area specified by a unit block belonging to both the motion row block and the motion sequence block as a rectangular motion image area.
- the repeater 13 repeatedly executes the processing by the motion sequence block determiner 7, the motion row block determiner 9, and the rectangular motion image region determiner 11 for the rectangular motion image region determined by the first rectangular motion image region determiner 11.
- Let The second rectangular motion image area determining means 14 is configured to determine the rectangular motion image area based on each pixel in the unit blocks constituting the outer four sides of the unit blocks constituting the rectangular motion image area. An area defined by an edge inside is defined as the rectangular motion image area.
- FIG. 2 shows a hardware configuration of the rectangular motion image area determination device 1.
- the rectangular motion image area determination device 1 includes a CPU 23, a RAM 25, and a flash memory 26.
- the flash memory 26 stores a program 26p.
- the program 26p performs a rectangular motion image region determination process as will be described later.
- the RAM 25 stores calculation results and the like.
- the frame memory 27 holds image data for one screen.
- the CPU 23 determines whether or not it is a moving image area composed of pixels with motion based on the pixel values constituting the display area stored in the memory 27, and stores the result in the RAM 25.
- pixels are arranged in a matrix in the row direction ⁇ and the column direction ⁇ .
- CPU 23 performs block division (step S1 in FIG. 3).
- 32 * 32 pixels are used as one block, and the display area 100 shown in FIG. 4 is divided into a plurality of blocks in a matrix.
- the block is divided into n + 1 blocks in the ⁇ direction and m + 1 blocks in the ⁇ direction.
- CPU 23 determines representative values for all blocks (step S3 in FIG. 3).
- the representative value instead of using the average value of the pixel values in one block, the top pixel value, and the value as it is, a hash value such as CRC may be used as the representative value.
- the average value can be expressed in 18-bit length in a 32 * 32 * 8-bit image, but the upper and lower 10 bits may be rounded down and only the middle 8 bits may be used.
- (n + 1) * (m + 1) block representative values are stored in the RAM 25.
- CPU 23 determines a motion block among (n + 1) * (m + 1) blocks (step S5).
- the motion block determination process will be described with reference to FIG.
- CPU 23 initializes processing block numbers i and j (steps S11 and S13 in FIG. 5).
- the block (0, 0) is compared with the representative value of the previous frame (step S15).
- the representative values at time t and time t ⁇ 1 are compared.
- the CPU 23 determines whether or not the difference between the representative values exceeds the threshold value thb (step S17). If the difference between the representative values exceeds the threshold value thb, it is determined as a motion block (step S19). On the other hand, if the difference between the representative values does not exceed the threshold thb, it is determined as a non-motion block (step S21).
- step S23 if the processing block number j is final, the process proceeds to step S27, and the CPU 23 determines whether or not the processing block number i is final. In this case, since it is not final, the processing block number i is incremented (step S29), and step S13 and subsequent steps are repeated. If the process block number i is the last in step S27, the process is terminated.
- a motion block is determined for (n + 1) * (m + 1) blocks.
- blocks (4, 2), (4, 3), (4, 4)... are determined as motion blocks.
- the CPU 23 performs a motion image area determination process (step S7 in FIG. 3). Details of the moving image region determination processing will be described with reference to FIGS.
- the CPU 23 determines that the column is a non-motion column (step S37).
- the CPU 23 determines whether there is a temporary motion sequence that has been added and stored (step S39). If there is, the CPU 23 determines whether the adjacent set has a width that exceeds the threshold thw (step S39).
- the threshold thw is set to 2 blocks or more. In this case, since the temporary motion sequence stored in addition exists from blocks (0, 2) to (0, 8), the threshold value thw is exceeded, so the temporary motion sequence is set as a motion sequence (step S45).
- the CPU 23 determines whether or not the processing block number j is final (step S46).
- the CPU 23 determines that the column is a non-motion column (step S37). The CPU 23 determines whether or not the detected temporary motion sequence exists (step S39). In this case, since it does not exist, the CPU 23 determines whether or not the processing block number j is final (step S46).
- the CPU 23 determines that the column is a non-motion column (step S37). The CPU 23 determines whether or not the detected temporary motion sequence exists (step S39). If there is, the CPU 23 determines whether or not the adjacent set has a width exceeding the threshold thw (step S41). ). In the present embodiment, since the threshold thw is 2 blocks or more, the width of the temporary motion sequence does not exceed the threshold thw, so the temporary motion sequence is set as a non-motion sequence (step S43). Thereby, it is possible to prevent an image area such as a mouse from being erroneously recognized as a moving image area.
- step S49 it is determined whether or not the provisional motion sequence that has been added and stored exists, and if it exists, the processing from step S41 is executed.
- Fig. 8 shows the motion sequence after detection.
- the region 121 is determined as a motion sequence.
- the regions 122 and 123 to which the motion blocks (6, n-3) and (m-3, n-6) belong do not have a width equal to or greater than the threshold thw in the direction of the arrow ⁇ . Is not determined.
- the CPU 23 initializes the processing block number i (step S51 in FIG. 9).
- the CPU 23 determines whether or not there is at least one motion block in the row of the block (0, 0) (step S53). In this case, as shown in FIG. 6, since there is no motion block in the block (0,0) and its horizontal blocks (0,1) to (0, n), such a row is a non-motion row. Is determined (step S57).
- step S57 The CPU 23 determines whether there is a motion row that has been added and stored (step S59). In this case, since there are motion lines that have been added and stored, it is determined whether or not the set has a width that exceeds the threshold thw (step S61).
- the threshold thw is set to 2 blocks or more. In this case, since there are detected motion rows from block (4,0) to (8,0), the threshold thw is exceeded, so the temporary motion row is set as a motion row (step S65).
- step S69 is the same as that of step S49, description thereof is omitted.
- Fig. 10 shows the motion line after detection.
- the area 131 is determined as a movement line.
- the motion block (6, n-3) belongs to the region 131.
- the region 132 to which the motion block (m ⁇ 3, n ⁇ 6) belongs does not have a width equal to or larger than the threshold thw in the arrow ⁇ direction, and is not determined as a motion row.
- the CPU 23 determines a block belonging to both the motion sequence and the motion row as a motion image area (step S70 in FIG. 9).
- a block belonging to an area 140 where the area 121 and the area 131 shown in FIG. 11 overlap is determined as a moving image area.
- the block (4,5) and the like are determined as motion image regions. . In this way, it is possible to prevent spillage in relation to the surrounding blocks.
- step S9 in FIG. 3 the boundary of the motion image area in the block composed of 32 * 32 pixels is obtained.
- step S9 the boundary 150 of the motion image area in units of one pixel can be obtained. Details of step S9 will be described with reference to FIG.
- step S80 extracts the upper block (step S80 in FIG. 12). In this case, a total of seven blocks (4,2) to (4,8) shown in FIG. 13 are extracted.
- the CPU 23 initializes the processing row number P (step S81), extracts all the pixels in the P row of the extracted block, and calculates the representative value (step S83). In this case, 32 pixels in the 0th row of the block (4,2), 32 pixels... And 7 * 32 pixels in the 0th row of the block (4,3) are extracted, and their representative values are calculated. In this embodiment, the representative value is the average value of the extracted pixels.
- the CPU 23 extracts all the pixels in the processing row number P + 1 row of the extracted block and calculates the representative value (step S85).
- 32 pixels in the first row of the block (4, 2), 32 pixels... And 7 * 32 pixels in the first row of the block (4, 3) are extracted, and their representative values are calculated.
- step S87 determines whether or not the representative value obtained in step S83 is different from the representative value obtained in step S85 (step S87). If both are different in step S87, it is determined that the row of the pixel (P, 0) is a boundary (step S93). If they are not different in step S87, the process row number P is incremented (step S89), and step S83 and subsequent steps are repeated until the final pixel of the extraction block (32 pixels in this case) is reached (step S91). If both are not different in step S87 even for the last pixel, it is determined that the end of the extracted block is a boundary (step S95).
- FIG. 12 illustrates the case where the boundary of one pixel unit in the upper block is extracted, but the same applies to the lower block.
- pixels for one column are extracted in the vertical direction, and the representative values may be compared with adjacent columns.
- This device can automatically detect the motion image area.
- the determination can also be detected in several frames. Therefore, even if the moving image area itself dynamically changes on the monitor, the moving image area can be detected almost in real time.
- step S7 in FIG. 3 is further repeated for the determined motion image area, so that the motion image area is detected even when there are a plurality of motion image areas 201 to 207 as shown in FIG. be able to.
- step S7 in FIG. 3 Once the process of step S7 in FIG. 3 is performed, motion image areas 301 to 304 as shown in FIG. 15 are determined.
- the region 207 is not detected as a motion image region because it does not have a width equal to or greater than the threshold thw.
- the motion region 301 originally includes a region that is not a motion image region.
- the CPU 23 regards the areas 301 to 304 as the entire image area, and detects the motion image area again. Thereby, for example, in the case of the area 301, the area 311 defined by the block coordinates (y1, x1) (y3, x11) and the area 312 defined by the block coordinates (y1, x12) (y3, x13) are detected. Is done. The same applies to the other areas 302 to 304.
- the region 311 is regarded as the entire image region, and the motion image region is detected again. Thereby, an area defined by the block coordinates (y1, x1) (y11, x11) is detected. The same applies to the region 312.
- Such repeated detection may be performed until it can no longer be divided, or the upper limit number of repeated detections may be determined in advance.
- CPU23 extracts the block except the edge part among upper blocks (Step S100 of Drawing 17). In this case, a total of five blocks (4,3) to (4,7) shown in FIG. 13 are extracted. The reason for removing the end portion in this way is that there is a possibility that some of the blocks at both end portions do not have an edge. However, this end block may also be adopted.
- the CPU 23 initializes the processing row number r and the processing pixel number g (steps S101 and S102 in FIG. 17), and determines whether the g-th pixel of the extracted block of the row number “0” is a horizontal edge pixel (row direction edge). It is determined whether or not (step S103). Whether or not the pixel is a horizontal edge pixel may be determined based on a luminance difference from an adjacent pixel. In the present embodiment, a filter and an arithmetic expression as shown in FIG. 18A are employed.
- the CPU 23 determines that the Pth pixel is a horizontal edge pixel, the CPU 23 increments the total number Et (step S105 in FIG. 17). The CPU 23 determines whether or not all lines have been completed (step S107). If not, the processing pixel number g is incremented (step S109), and steps S103 to S107 are repeated.
- the CPU 23 determines whether or not the total number Et of horizontal edges in the line is larger than a preset threshold value ths (step S111). This is to eliminate the horizontal edge pixels as contours in the moving image area.
- the CPU 23 sets it as a boundary (step S113). As a result, a boundary where the set of horizontal edges exceeds the threshold ths can be obtained.
- the threshold ths may be a fixed threshold, or a ratio may be set in advance and calculated according to the size of the detected area (number of blocks).
- step S111 if “NO”, the CPU 23 determines whether or not all lines have been completed (step S113). If all lines have not been completed, the CPU 23 increments the process line number r and repeats steps S102 to S111.
- the boundary can be determined on a pixel-by-pixel basis.
- vertical edges may be detected in the same manner using the filters and arithmetic expressions shown in FIG. 18B for the unit blocks located at the rightmost and leftmost stages.
- boundary determination is performed based on the total number of vertical edges and horizontal edges. Therefore, even a moving image region on a messy background can be detected. In addition, detection accuracy is improved and shaking of the moving image area is reduced.
- the pixels of all the blocks except for the end portion among the blocks obtained in FIG. 11 are adopted, but it may be determined by a part thereof.
- the comparison with the threshold value ths is a total of one line.
- the boundary is a boundary.
- edge images may exist even in natural images. Therefore, edges that are not continuous over a certain number may be removed from the calculation of the total number.
- the present invention is applied to the case where the rectangular area is determined and the boundary of one pixel unit for the block on the outer periphery thereof is determined.
- the present invention is applicable to general motion estimation and other fields.
- a boundary of a plurality of pixel units (for example, 2 * 2 pixel units) may be obtained instead of one pixel unit.
- the boundary is determined by extracting edge pixels. However, for each pixel in the target block, it can be determined whether there is a motion vector between different frames, and a motion vector can be detected. You may make it judge that the boundary of a pixel and the pixel which cannot be detected is a boundary.
- the boundaries of each frame may be stored in sequence, and if the boundary cannot be detected in a certain frame, the accumulated boundary may be adopted.
- the number of frames to be accumulated is set to ten or more frames.
- the CPU 23 initializes the target row number Q and the number of matching rows k (step S121 in FIG. 20), and measures the total number of edge pixels Enq in the Q row of the extracted block (step S123). Whether or not the pixel is an edge pixel may be determined based on a luminance difference from surrounding pixels.
- the CPU 23 determines whether the total edge pixel number Enq is larger than the threshold value thm and the immediately preceding total edge pixel number En (q-1) is smaller than the threshold value thn (step S125). When the determination in step S125 is “yes”, the CPU 23 increments the number of matching lines k (step S127).
- CPU 23 determines whether all rows have been determined (step S131). If all the rows have not been determined, the CPU 23 increments the processing row number Q (step S133) and repeats step S123 to step S127.
- step S131 determines whether the number of matching rows k is greater than the threshold thg (step S135). If the number of matching lines k is greater than the threshold thg, it is determined that the movement is due to scrolling (step S137).
- step S9 Such scroll determination is performed between step S7 and step S9 in FIG. 3, and when it is determined that the determination is scroll determination, the process of step S9 may be omitted.
- scroll determination may be performed on a part of lines instead of all lines in the area.
- the determination can be similarly made even if the text data is written vertically.
- the scroll determination is not limited to this method, and may be a well-known scroll determination.
- This specific area is not limited to a moving image area by moving image area detection, and may be a specific area in an image regardless of whether it is a still image or a moving image.
- the luminance value is adopted as the pixel value, but it may be an RGB value or the like.
- step S15 in FIG. 5 it is determined whether or not the t-th frame and the t + 1-th frame are blocks constituting a moving image area. For example, it may be determined whether there is movement between the t-th frame and the t + 2 frame. Further, instead of comparing the two frames, for example, 16 frames may be adopted by adding the front and rear frames.
- the boundary line determination process in step S9 in FIG. 3 is arbitrary, and may not be performed when the boundary line in pixel units is not necessary.
- all the pixel values in one row or all the columns are used for all the blocks located on the outer periphery. Moreover, you may make it judge with some blocks instead of all the blocks.
- step S15 in FIG. 5 the past representative values at the same position may be held and compared.
- one block is composed of 32 * 32 pixels, but the present invention is not limited to this.
- step S9 When the value of any pixel itself (for example, the pixel value at the upper left corner of the block) is used instead of the average value or hash value of the pixels, the processing of step S9 in FIG. Then, the boundary may be obtained. For example, when 7 * 5 blocks are detected as a motion image area in the process of step S7, the process of step S9 is performed with 9 * 7 blocks added by one block around the block.
- the set top box may be configured as a switching hub for switching the output destination monitor of input data.
- a rectangular motion image area can be detected with one pixel accuracy. Furthermore, a plurality of rectangular motion image areas can be determined. In addition, it is possible to remove a small area movement with a small width. For example, it is possible to take measures against pseudo movements such as a mouse pointer. Moreover, since these can be distinguished with respect to the movement area
- the display area is divided into unit blocks each including a predetermined number of pixels, and the pixel values of the predetermined number of pixels belonging to each unit block are set. Based on this, the representative pixel value of each unit block is calculated and compared with the representative value of the unit block at the same position in the comparison target frame for each unit block.
- Judged as a motion unit block that exists the present invention is not limited to this, and a method that can determine whether each unit block is a motion unit block in which motion exists may be employed. For example, as already described, it is possible to employ a well-known motion unit block determination step, such as comparing the sum of representative values of the same blocks of a predetermined number of frames.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Image Analysis (AREA)
- Controls And Circuits For Display Device (AREA)
- Picture Signal Circuits (AREA)
- Studio Devices (AREA)
Abstract
Description
(21)本発明にかかる境界決定装置は、前記エッジ画素判断手段が判定対象とする画素を決定する手段であって、以下の手段を有する荒領域決定手段を備えている。
2)前記単位ブロックのうち、最上段に位置する各単位ブロックと同じ列に属する単位ブロックの集合を列ブロックとして定義し、各列ブロックに少なくとも1以上の動き単位ブロックが存在する場合には当該列ブロックを動き列ブロックとして決定する動き列ブロック決定手段、
3)前記単位ブロックのうち、最左端段に位置する各単位ブロックと同じ行に属する単位ブロックの集合を行ブロックとして定義し、各行ブロックに少なくとも1以上の動き単位ブロックが存在する場合には、当該行ブロックを動き行ブロックとして決定する動き行ブロック決定手段、
4)前記動き行ブロックおよび前記動き列ブロックの双方に属する単位ブロックで特定される矩形領域を矩形動き画像領域として決定する第1矩形動き画像領域決定手段、
5)前記矩形動き画像領域を構成する単位ブロックの内、外周の四辺を構成する単位ブロックに属する画素を判定対象として決定する対象画素決定手段。
図1に、本発明の1実施形態にかかる矩形動き画像領域決定装置1の機能ブロック図を示す。矩形動き画像領域決定装置1は、行方向および列方向に画素が行列配置された表示領域のうち、その一部に表示される矩形動き画像領域を決定する装置であって、動き単位ブロック判断手段5、動き列ブロック決定手段7、動き行ブロック決定手段9、第1矩形動き画像領域決定手段11、繰り返し手段13、第2矩形動き画像領域決定手段14を備えている。
図2に矩形動き画像領域決定装置1のハードウェア構成を示す。矩形動き画像領域決定装置1は、CPU23、RAM25、フラッシュメモリ26を備えている。フラッシュメモリ26には、プログラム26pが記憶されている。プログラム26pには、後述するように、矩形動き画像領域決定処理を行う。RAM25は演算結果等を記憶する。フレームメモリ27は1画面の画像データを保持する。
図2に示すプログラム26Pによる処理について図3を用いて説明する。以下では、図4に示すように、モニタの1フレームである画像領域100に、3つの矩形動き画像領域100~112が存在する場合を例として説明する。
上記実施形態では、検出対象が1つの場合を例として説明した。第2実施形態では、決定した動き画像領域についてさらに、図3ステップS7の処理を繰り返すことにより、図14に示すような動き画像領域201~207が複数存在する場合でも、動き画像領域を検出することができる。
第1,第2の実施形態においては、ブロック単位で検出した後、外周の四辺を構成する単位ブロックについて、行方向であれば各行について代表値を求めて、隣接する行と異なる場合に、境界であると判断している。列方向についても同様である。かかる検出については、以下のように境界判断をすることもできる。図17を用いて説明する。
上記で説明したフレーム間の画素情報に基づいて動画領域か否か判断すると、特定の矩形領域でテキストデータをスクロールする場合にも、誤って動画領域と判断するおそれがある。以下に示すように、スクロール判定を行い、スクロール処理である場合には、境界判断をしないようにしてもよい。
本実施形態においては、画素値として輝度値を採用したが、RGBの値などであってもよい。
25 RAM
26 フラッシュメモリ
Claims (21)
- 行方向および列方向に画素が行列配置された表示領域のうち、その一部に表示される矩形動き画像領域を決定する矩形動き画像領域決定方法であって、
前記表示領域を、所定数の画素で構成される単位ブロックに分割するとともに、各単位ブロックが、動きが存在する動き単位ブロックであるか否か判断する動き単位ブロック判断ステップ、
前記単位ブロックのうち、最上段に位置する各単位ブロックと同じ列に属する単位ブロックの集合を列ブロックとして定義し、各列ブロックに少なくとも1以上の動き単位ブロックが存在する場合には当該列ブロックを動き列ブロックとして決定する動き列ブロック決定ステップ、
前記単位ブロックのうち、最左端段に位置する各単位ブロックと同じ行に属する単位ブロックの集合を行ブロックとして定義し、各行ブロックに少なくとも1以上の動き単位ブロックが存在する場合には、当該行ブロックを動き行ブロックとして決定する動き行ブロック決定ステップ、
前記動き行ブロックおよび前記動き列ブロックの双方に属する単位ブロックで特定される矩形領域を矩形動き画像領域として決定する第1矩形動き画像領域決定ステップ、
前記矩形動き画像領域を構成する単位ブロックの内、外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の各画素に基づき、前記矩形動き画像領域内のエッジで定義される領域を前記矩形動き画像領域とする第2矩形動き画像領域決定ステップ、
を備えた矩形動き画像領域決定方法。 - 請求項1の矩形動き画像領域決定方法において、
前記第2矩形動き画像領域決定ステップでは、前記外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の行方向については、各行の代表画素値が異なる行を、列方向については各列の代表画素値が異なる列を、前記矩形動き画像領域の境界として決定すること、
を特徴とする矩形動き画像領域決定方法。 - 請求項1の矩形動き画像領域決定方法において、
前記第2矩形動き画像領域決定ステップでは、前記外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の各画素について、異なるフレーム間で動きが存在する動き画素か否かを判断して、前記矩形動き画像領域の境界を決定すること、
を特徴とする矩形動き画像領域決定方法。 - 請求項1の矩形動き画像領域決定方法において、
前記第2矩形動き画像領域決定ステップでは、
前記外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の各画素について、行方向エッジおよび列方向エッジを構成する画素を抽出し、各行についての行方向エッジを構成する画素の総数、各列について列方向エッジを構成する画素の総数に基づき前記矩形動き画像領域の境界を決定すること、
を特徴とする矩形動き画像領域決定方法。 - 請求項1~4のいずれかの矩形動き画像領域決定方法において、
前記第2矩形動き画像領域決定ステップの前に、決定した矩形動き画像領域がスクロールによってフレーム間で画像が異なるか否かを判定するスクロール該当判定をし、スクロールであると判断した場合には、前記第2矩形動き画像領域決定ステップを行わないこと、
を特徴とする矩形動き画像領域決定方法。 - 請求項5の矩形動き画像領域決定方法において、
前記スクロール該当判定は、
決定した矩形動き画像領域の各画素について、周辺画素との画素値の差に基づいて、エッジであるか否かを判断し、
前記エッジではない画素が行方向または列方向に所定数以上存在する行または列について、文字隙間存在領域として認識し、
前記決定した矩形動き画像領域に対する前記文字隙間存在領域の割合に基づいてスクロール該当判定をすること、
を特徴とする矩形動き画像領域決定方法。 - 請求項1~6のいずれかの矩形動き画像領域決定方法において、
前記第1矩形動き画像領域決定ステップにて決定した矩形動き画像領域について、第2矩形動き画像領域決定ステップの前に、
前記動き列ブロック決定ステップ、前記動き行ブロック決定ステップ、および前記第1矩形動き画像領域決定ステップを繰り返し実行すること、
を特徴とする矩形動き画像領域決定方法。 - 行方向および列方向に画素が行列配置された表示領域のうち、その一部に表示される矩形動き画像領域を決定する矩形動き画像領域決定装置であって、
前記表示領域を、所定数の画素で構成される単位ブロックに分割するとともに、各単位ブロックが、動きが存在する動き単位ブロックであるか否か判断する動き単位ブロック判断手段、
前記単位ブロックのうち、最上段に位置する各単位ブロックと同じ列に属する単位ブロックの集合を列ブロックとして定義し、各列ブロックに少なくとも1以上の動き単位ブロックが存在する場合には当該列ブロックを動き列ブロックとして決定する動き列ブロック決定手段、
前記単位ブロックのうち、最左端段に位置する各単位ブロックと同じ行に属する単位ブロックの集合を行ブロックとして定義し、各行ブロックに少なくとも1以上の動き単位ブロックが存在する場合には、当該行ブロックを動き行ブロックとして決定する動き行ブロック決定手段、
前記動き行ブロックおよび前記動き列ブロックの双方に属する単位ブロックで特定される矩形領域を矩形動き画像領域として決定する第1矩形動き画像領域決定手段、
前記矩形動き画像領域を構成する単位ブロックの内、外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の各画素に基づき、前記矩形動き画像領域内のエッジで定義される領域を前記矩形動き画像領域とする第2矩形動き画像領域決定手段、
を備えた矩形動き画像領域決定装置。 - 請求項8の矩形動き画像領域決定装置において、
前記第2矩形動き画像領域決定手段は、外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の行方向については、各行の代表画素値が異なる行を、列方向については各列の代表画素値が異なる列を、前記矩形動き画像領域の境界として決定すること、
を特徴とする矩形動き画像領域決定装置。 - 請求項8の矩形動き画像領域決定装置において、
前記第2矩形動き画像領域決定手段は、外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の各画素について、異なるフレーム間で動きが存在する動き画素か否かを判断して、前記矩形動き画像領域の境界を決定すること、
を特徴とする矩形動き画像領域決定装置。 - 請求項8の矩形動き画像領域決定装置において、
前記第2矩形動き画像領域決定手段は、外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の各画素について、行方向および列方向エッジを構成する画素を抽出し、各行についての行方向の総数、各列について列方向エッジを構成する画素の総数に基づき前記矩形動き画像領域の境界を決定すること、
を特徴とする矩形動き画像領域決定装置。 - 請求項8~11のいずれかの矩形動き画像領域決定装置において、
前記第1矩形動き画像領域決定手段が決定した矩形動き画像領域について、前記動き列ブロック決定手段、前記動き行ブロック決定手段、および前記矩形動き画像領域決定手段による処理を繰り返し実行させる繰り返し手段、
を備えたことを特徴とする矩形動き画像領域決定装置。 - 請求項8~12のいずれかの矩形動き画像領域決定装置を含むモニタ。
- 請求項8~12のいずれかの矩形動き画像領域決定装置を含むセットトップボックス。
- 請求項14のセットトップボックスは、入力データの出力先モニタを切り替えるための切り替えハブであること、
を特徴とするもの。 - 行方向および列方向に画素が行列配置された表示領域のうち、その一部に表示される矩形動き画像領域を決定する矩形動き画像領域決定処理をコンピュータに実行させるためのプログラムであって、
前記コンピュータは下記手段として構成させる矩形動き画像領域決定プログラム。
前記表示領域を、所定数の画素で構成される単位ブロックに分割するとともに、各単位ブロックが、動きが存在する動き単位ブロックであるか否か判断する動き単位ブロック判断ステップ、
前記単位ブロックのうち、最上段に位置する各単位ブロックと同じ列に属する単位ブロックの集合を列ブロックとして定義し、各列ブロックに少なくとも1以上の動き単位ブロックが存在する場合には当該列ブロックを動き列ブロックとして決定する動き列ブロック決定手段、
前記単位ブロックのうち、最左端段に位置する各単位ブロックと同じ行に属する単位ブロックの集合を行ブロックとして定義し、各行ブロックに少なくとも1以上の動き単位ブロックが存在する場合には、当該行ブロックを動き行ブロックとして決定する動き行ブロック決定手段、
前記動き行ブロックおよび前記動き列ブロックの双方に属する単位ブロックで特定される矩形領域を矩形動き画像領域として決定する第1矩形動き画像領域決定手段、
前記矩形動き画像領域を構成する単位ブロックの内、外周の四辺を構成する単位ブロックについて、これらの単位ブロック内の各画素に基づき、前記矩形動き画像領域内のエッジで定義される領域を前記矩形動き画像領域とする第2矩形動き画像領域決定手段。 - 行方向および列方向に画素が行列配置された表示領域のうち、その一部に表示される矩形動き画像領域を決定する矩形動き画像領域決定方法であって、
前記表示領域を、所定数の画素で構成される単位ブロックに分割するとともに、各単位ブロックが、動きが存在する動き単位ブロックであるか否か判断する動き単位ブロック判断ステップ、
前記単位ブロックのうち、最上段に位置する各単位ブロックと同じ列に属する単位ブロックの集合を列ブロックとして定義し、各列ブロックに少なくとも1以上の動き単位ブロックが存在する場合には当該列ブロックを動き列ブロックとして決定する動き列ブロック決定ステップ、
前記単位ブロックのうち、最左端段に位置する各単位ブロックと同じ行に属する単位ブロックの集合を行ブロックとして定義し、各行ブロックに少なくとも1以上の動き単位ブロックが存在する場合には、当該行ブロックを動き行ブロックとして決定する動き行ブロック決定ステップ、
前記動き行ブロックおよび前記動き列ブロックの双方に属する単位ブロックで特定される矩形領域を矩形動き画像領域として決定する第1矩形動き画像領域決定ステップ、
を備えた矩形動き画像領域決定方法。 - 請求項17の矩形動き画像領域決定方法において、
前記第1矩形動き画像領域決定ステップでは、決定した矩形領域について、さらに、
前記動き列ブロック決定ステップ、および前記動き行ブロック決定ステップ、繰り返し実行すること、
を特徴とする矩形動き画像領域決定方法。 - 画面内の判定対象領域がスクロール領域か否かを判定するスクロール領域判定装置であって、
前記判定対象領域の各画素について、周辺画素との画素値の差に基づいて、エッジであるか否かを判断するエッジ画素判断手段、
同一行または同一列に、前記エッジではない画素が所定数以上存在する行または列を文字隙間存在領域として判断する文字隙間存在領域判断手段、
前記判定対象領域に対する前記文字隙間存在領域の割合に基づいてスクロール領域か否かを判断するスクロール領域判断手段、
を備えたことを特徴とするスクロール領域判定装置。 - 1画面に存在する矩形領域の境界を決定する境界決定装置であって、
判定対象画素における各画素の画素値に基づいて、エッジを構成する画素を判断するエッジ画素判断手段、
行方向エッジを構成する画素を行方向エッジ画素として抽出し、前記行方向エッジの総数を各行について演算し、各行の前記行方向エッジの総数に基づいて、行方向の境界を決定する行方向境界決定手段、
列方向エッジを構成する画素を列方向エッジ画素として抽出し、前記列方向エッジの総数を各列について演算し、各列の前記列方向エッジの総数に基づいて、列方向の境界を決定する列方向境界決定手段、
前記決定した行方向境界および前記列方向境界を、前記矩形領域の境界として決定する矩形領域決定手段、
を備えたことを特徴とする境界決定装置。 - 請求項20の境界決定装置であって、
前記エッジ画素判断手段が判定対象とする画素を決定する手段であって、以下の手段を有する荒領域決定手段を備えていること、
1)行方向および列方向に画素が行列配置された表示領域を、所定数の画素で構成される単位ブロックに分割するとともに、各単位ブロックが動きが存在する動き単位ブロックと判断する動き単位ブロック判断手段、
2)前記単位ブロックのうち、最上段に位置する各単位ブロックと同じ列に属する単位ブロックの集合を列ブロックとして定義し、各列ブロックに少なくとも1以上の動き単位ブロックが存在する場合には当該列ブロックを動き列ブロックとして決定する動き列ブロック決定手段、
3)前記単位ブロックのうち、最左端段に位置する各単位ブロックと同じ行に属する単位ブロックの集合を行ブロックとして定義し、各行ブロックに少なくとも1以上の動き単位ブロックが存在する場合には、当該行ブロックを動き行ブロックとして決定する動き行ブロック決定手段、
4)前記動き行ブロックおよび前記動き列ブロックの双方に属する単位ブロックで特定される矩形領域を矩形動き画像領域として決定する第1矩形動き画像領域決定手段、
5)前記矩形動き画像領域を構成する単位ブロックの内、外周の四辺を構成する単位ブロックに属する画素を判定対象として決定する対象画素決定手段、 を特徴とする境界決定装置。
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201180071038.2A CN103548055B (zh) | 2011-06-14 | 2011-12-01 | 运动图像区域判定装置或其方法 |
EP11867827.5A EP2722817B1 (en) | 2011-06-14 | 2011-12-01 | Motion image region identification device and method thereof |
RU2013154986/08A RU2592457C2 (ru) | 2011-06-14 | 2011-12-01 | Устройство для идентификации области движущегося изображения и способ |
ES11867827T ES2710860T3 (es) | 2011-06-14 | 2011-12-01 | Dispositivo de identificación de región de imagen en movimiento y método del mismo |
AU2011371064A AU2011371064B2 (en) | 2011-06-14 | 2011-12-01 | Motion image region identification device and method thereof |
US14/126,353 US9311557B2 (en) | 2011-06-14 | 2011-12-01 | Motion image region identification device and method thereof |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011131926 | 2011-06-14 | ||
JP2011-131926 | 2011-06-14 | ||
JP2011-232571 | 2011-10-24 | ||
JP2011232571A JP4980486B1 (ja) | 2011-06-14 | 2011-10-24 | 動き画像領域判定装置またはその方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012172706A1 true WO2012172706A1 (ja) | 2012-12-20 |
Family
ID=46678910
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2011/077755 WO2012172706A1 (ja) | 2011-06-14 | 2011-12-01 | 動き画像領域判定装置またはその方法 |
Country Status (8)
Country | Link |
---|---|
US (1) | US9311557B2 (ja) |
EP (1) | EP2722817B1 (ja) |
JP (2) | JP4980486B1 (ja) |
CN (1) | CN103548055B (ja) |
AU (1) | AU2011371064B2 (ja) |
ES (1) | ES2710860T3 (ja) |
RU (1) | RU2592457C2 (ja) |
WO (1) | WO2012172706A1 (ja) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5192597B1 (ja) * | 2012-04-03 | 2013-05-08 | 株式会社ナナオ | 動き画像領域判定装置またはその方法 |
JP5542981B1 (ja) | 2013-02-06 | 2014-07-09 | Eizo株式会社 | 画像処理装置、フレームレートコントロール処理判定装置またはその方法 |
EP3564917B1 (en) * | 2018-05-04 | 2020-07-01 | Axis AB | A method for detecting motion in a video sequence |
CN110278484B (zh) * | 2019-05-15 | 2022-01-25 | 北京达佳互联信息技术有限公司 | 视频配乐方法、装置、电子设备及存储介质 |
KR20210065447A (ko) | 2019-11-27 | 2021-06-04 | 삼성전자주식회사 | 전자장치와 그의 제어방법, 및 기록매체 |
CN113691721B (zh) * | 2021-07-28 | 2023-07-18 | 浙江大华技术股份有限公司 | 一种缩时摄影视频的合成方法、装置、计算机设备和介质 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07334616A (ja) * | 1994-06-06 | 1995-12-22 | Sharp Corp | 画像分類装置 |
JPH0879767A (ja) * | 1994-07-05 | 1996-03-22 | Ricoh Co Ltd | 動きベクトル符号化および復号化方式 |
JP2001076161A (ja) * | 1999-09-02 | 2001-03-23 | Canon Inc | 画像処理方法及び装置並びに記憶媒体 |
JP2004157818A (ja) * | 2002-11-07 | 2004-06-03 | Panasonic Communications Co Ltd | 画像処理装置及び画像処理方法 |
JP2006024147A (ja) * | 2004-07-09 | 2006-01-26 | Foundation For The Promotion Of Industrial Science | 背景画像生成方法及び装置 |
JP2009064199A (ja) * | 2007-09-05 | 2009-03-26 | Casio Comput Co Ltd | ジェスチャー認識装置及びジェスチャー認識方法 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5715325A (en) * | 1995-08-30 | 1998-02-03 | Siemens Corporate Research, Inc. | Apparatus and method for detecting a face in a video image |
JP3873554B2 (ja) * | 1999-12-27 | 2007-01-24 | 株式会社日立製作所 | 監視装置、監視プログラムが記録された記録媒体 |
RU2315357C2 (ru) * | 2003-11-27 | 2008-01-20 | Корпорация "САМСУНГ ЭЛЕКТРОНИКС Ко., Лтд." | Способ детектирования объекта |
JP4416607B2 (ja) * | 2004-09-08 | 2010-02-17 | 富士通株式会社 | 移動体画像抽出方法 |
CN101477687B (zh) * | 2009-01-22 | 2011-05-04 | 上海交通大学 | 复杂背景下的棋盘格角点检测方法 |
US8411738B2 (en) * | 2009-03-12 | 2013-04-02 | Samsung Electronics Co., Ltd. | System and method for identification of vertical scrolling regions in digital video |
CN101854466B (zh) * | 2010-05-13 | 2012-05-30 | 北京英泰智软件技术发展有限公司 | 一种运动区域探测方法及装置 |
CN101859440A (zh) * | 2010-05-31 | 2010-10-13 | 浙江捷尚视觉科技有限公司 | 基于块的运动区域检测方法 |
-
2011
- 2011-10-24 JP JP2011232571A patent/JP4980486B1/ja active Active
- 2011-12-01 EP EP11867827.5A patent/EP2722817B1/en active Active
- 2011-12-01 US US14/126,353 patent/US9311557B2/en active Active
- 2011-12-01 RU RU2013154986/08A patent/RU2592457C2/ru not_active IP Right Cessation
- 2011-12-01 WO PCT/JP2011/077755 patent/WO2012172706A1/ja active Application Filing
- 2011-12-01 CN CN201180071038.2A patent/CN103548055B/zh active Active
- 2011-12-01 ES ES11867827T patent/ES2710860T3/es active Active
- 2011-12-01 AU AU2011371064A patent/AU2011371064B2/en not_active Ceased
-
2012
- 2012-01-30 JP JP2012016343A patent/JP5006479B1/ja active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07334616A (ja) * | 1994-06-06 | 1995-12-22 | Sharp Corp | 画像分類装置 |
JPH0879767A (ja) * | 1994-07-05 | 1996-03-22 | Ricoh Co Ltd | 動きベクトル符号化および復号化方式 |
JP2001076161A (ja) * | 1999-09-02 | 2001-03-23 | Canon Inc | 画像処理方法及び装置並びに記憶媒体 |
JP2004157818A (ja) * | 2002-11-07 | 2004-06-03 | Panasonic Communications Co Ltd | 画像処理装置及び画像処理方法 |
JP2006024147A (ja) * | 2004-07-09 | 2006-01-26 | Foundation For The Promotion Of Industrial Science | 背景画像生成方法及び装置 |
JP2009064199A (ja) * | 2007-09-05 | 2009-03-26 | Casio Comput Co Ltd | ジェスチャー認識装置及びジェスチャー認識方法 |
Non-Patent Citations (1)
Title |
---|
See also references of EP2722817A4 * |
Also Published As
Publication number | Publication date |
---|---|
EP2722817A1 (en) | 2014-04-23 |
JP4980486B1 (ja) | 2012-07-18 |
EP2722817B1 (en) | 2018-11-28 |
CN103548055B (zh) | 2016-02-17 |
JP5006479B1 (ja) | 2012-08-22 |
US20140126827A1 (en) | 2014-05-08 |
AU2011371064A1 (en) | 2014-01-16 |
EP2722817A4 (en) | 2015-08-19 |
JP2013020605A (ja) | 2013-01-31 |
RU2592457C2 (ru) | 2016-07-20 |
US9311557B2 (en) | 2016-04-12 |
AU2011371064B2 (en) | 2015-04-16 |
RU2013154986A (ru) | 2015-07-20 |
CN103548055A (zh) | 2014-01-29 |
ES2710860T3 (es) | 2019-04-29 |
JP2013020602A (ja) | 2013-01-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5006479B1 (ja) | 動き画像領域判定装置またはその方法 | |
JP6088792B2 (ja) | 画像検出装置及び制御プログラム並びに画像検出方法 | |
JP6809613B2 (ja) | 画像前景の検出装置、検出方法及び電子機器 | |
US20150187051A1 (en) | Method and apparatus for estimating image noise | |
JP2020170252A5 (ja) | ||
JP5690800B2 (ja) | 動き画像領域判定装置またはその方法 | |
JP6591349B2 (ja) | 動体検知システム及び動体検知方法 | |
JP5542981B1 (ja) | 画像処理装置、フレームレートコントロール処理判定装置またはその方法 | |
US9430959B2 (en) | Character region pixel identification device and method thereof | |
JP5656964B2 (ja) | シーンチェンジ判定装置またはその方法 | |
CN106934814B (zh) | 一种基于图像的背景信息识别方法及装置 | |
WO2015166518A1 (ja) | アノテーション線判断装置、アノテーション線除去装置、医療用の表示装置およびそれらの方法 | |
CN115240067A (zh) | 一种利用隔帧差分法自动检测高空坠物的方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201180071038.2 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 11867827 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14126353 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2013154986 Country of ref document: RU Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 2011371064 Country of ref document: AU Date of ref document: 20111201 Kind code of ref document: A |